The rounds now come as the rational surfaces their classes give, closed ones as two faces split
at their seams, the touching pipes' as four: one splines entry a face, each of the class's degrees.
Their KNOWN_FAILING lines go.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A plane cutting a cylinder or a cone at an angle, a sphere or a cone coaxial with the other face, a
cone and a cylinder touching one ball, two pipes touching at a point: their rounds' centre lines
are rational, so the rounds are rational surfaces of degrees fixed by the construction ((4, 2),
(2, 2) revolved, (6, 2), (12, 2)). The tee's centre line is not rational; its round is graded on
its volume. Volumes from the divergence theorem over the exact pieces, at rel 1e-6. The six that
SMLib 0cc48bd traces are listed in KNOWN_FAILING with what they come out as.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A loop that runs out along a chain of edges and straight back (a meridian from a sphere's pole
stopping short of the other, out and back) encloses no (u, v) area, yet on a closed surface it
bounds all of it less a cut of none. The reader only read such a face when the slit already ran
pole to pole (and moved the sphere's frame onto it only then), so the 2 mm ball whose meridian
stops at 2.5 rad was refused for its empty loop.
The slit is now made the face's seams, so its loop bounds the surface's whole domain: on a
sphere, a revolution whose profile ends on its axis, or a spline closed round u that collapses
at both ends, the slit's meridian is run on to both poles, out and back (a sphere's frame put
on the slit first, whichever two points it joins); on a torus, the circle the slit lies along is
closed and the other circle through its start is the second seam. The seams on a sphere are its
own meridians (angle 0 on the equator), so their pcurves are exact lines on its form.
Tests: non_antipodal_seam_sphere_imports_as_a_whole_ball passes (its known-failing line goes);
seam_only_torus_imports_as_a_whole_ring adds a torus bounded by a slit along its tube.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A loop that uses each of its edges once each way (a whole torus within its tube and ring seams)
is the same set of uses run backwards, so it says nothing of the face's side. One written
clockwise in (u, v) turned the face over, against its same_sense: the torus came in inside out
with a warning. Such a loop is now reversed and the face keeps its side.
Test: seam_only_torus_imports_as_a_whole_ring, both orders of the two seams, volume 2π²Rr².
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Its seams grade needs the continuity check to skip the body the round is not on. Fillet cases
wait for the new algorithm.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A square cut is the slanted plane's ordinary crossing; the cam's half turn is the quarter turn's
contact at a symmetric angle; the r15 slab stands inside the same coincidence tolerance as the r10;
the 90 degree unfold and the 40 degree flange repeat the 72 degree and 20 degree cases' formulas.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The sloped-basin chamfers and rounds at 0.5 mm rims repeat the 1 mm rim's regimes; two of the
three knife-edge chamfer tips, the r1 and box r3 curvature corners, the r1 tangent corner, the
1 mm pocket-rim and hole-rim chamfers short of the floor and the r1 together-round each repeat a kept case's regime
and grades. #854 moves to 851-chamfer-arc-top-edge-9-5, and the cases that cited a dropped one
cite the kept one or state its formula.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The ring cut frame's and the up-down join's shells build now, and so do the up-down join's and
modern-plastic's chamfers (SMLib smlib-blend-refusals). #616's lid in the corpus still fails
SMLib's loop check, the shelled box shelled again at an inner wall is unchanged, and the
oblique cut of a cylinder's fillet still fails SmSurfOfRevolution's Test #7: 3 actions of 7.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib keeps the skirt top's offset where its loops wind the other way under the opening, the
cavity's floor (the owner's ruling of 2026-10-31: build it; a lip thinner than the wall is no
break-through).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib's inset now keeps only the regions that are cavity, not the one the offsets close where
they overlap (between the two faces of #545's 0.5 floor, round the up-down join's two empty
octants). The frame's core under the recess is hollowed to a sealed void, so the case says two
shells and its closed form, 21044 mm³.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib's merge now intersects the offsets of an edge's faces where they could not be mated along
it (the hole's offset poking out through the half-round side's; tangency-labs/solidcore#1795):
one closed valid solid with the one face open, as the owner ruled.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib now offsets a vertex where four faces meet (two walls and two facets at the floor's
corners, four facets at the drain) to two vertices joined by a new edge (tangency-labs/smlib
branch smlib-shells). Both volumes match their closed forms. The pocket's lining, hanging from
the top's rim round the pocket, stands apart from the slab's outer walls with the cavity rising
between them to the open top: the case says two shells in the one body.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
926 (the chamfer ending on the boss's wall), the oblique cut's r0.1
round, the every-block r1 round (its seam with the cylinder is no
contact seam, after the blend_seams fix), the up-down squares' concave
chamfers and the shelled box's lip chamfer build with SMLib branch
smlib-blend-refusals (f8650b8 and later). 1207 r0.5 passes SMLib's check
on its rails now and stays listed for the next cause: its end rib on the
boss, which starts where the round and the boss are tangent.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A blend face's name holds the edge it replaced, <face>|<face>, and a side
of that edge can be a blend face whose own name holds a |. Whether a face
is a contact of the blend was read as the edge's name starting or ending
with the face's name next to a |, so the r2 round's own face F2:side read
as a contact of the r1 round F4:blend:F1:+y|F3:blend:F1:+z|F2:side. The
edge's name is now split into its two faces, a blend side read whole as
the two faces of its own edge.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The r1 round of a block's top edge, chained onto the end curve of the r2
round of its arc at a cylinder (rig fillets/every-block-cylinder-crease-
top-edge-after-arc-round-r1), ends where it runs into the cylinder. That
edge is no seam of the r1 round: the cylinder is a face of the r2 round,
not of the r1. blend_seams takes it for one today (the r2 round's name
F3:blend:F1:+z|F2:side holds a | of its own), and the rig reads it as a
contact seam at 60 degrees.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner's ruling: a round or chamfer that runs past an outer edge of its face into open air
builds, tangent where it has room and cut by the face beyond where it runs out. SMLib's branch
smlib-blends-past-rims builds them by the rig's analytic answers. Off the known-failing list:
the disc's rim past its 2 mm wall (190), the hole's top rim R12 past the plate's bottom, the
chamfer past the cylinder's bottom (2893), each closing on the wall continued as itself, and
926's chamfer ending on the boss's wall, which now builds to its answer too.
The sloped basins' R4 rim rounds (1080, mitred rounds built together where one overflows) build
to their analytic volumes (to 6e-9), and stay listed with that reason: their continuity check
(g1 at every blend seam) takes in the mitres' creases, G0 by the mitre ruling.
2893-chamfer-two-distances-elliptical-rim is graded as a solid (the owner: it builds) and stays
listed, waiting on the ruling on unequal chamfers along a rim whose angle varies.
5158-fillet-round-just-past-a-10mm-face stays listed with its cause found: the round passes
5e-6 under the box's top back edge, inside the body's tolerance.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A projected spline edge of several Bezier spans came into the sketch as one
curve per span (#1508), so a G2 round's bottom edge was 16 curves and a path
naming the corner named one span of it (#3604). The kernel now reads such an
edge (a spline, or an elliptical arc of several spans) as the one clamped
B-spline it is over the edge's interval, its knots and weights its own, and
the sketch takes it as one control spline whose ends sit on the edge's
vertices. Planar regions still read edges span by span. The 3604 case builds
with its original path and leaves KNOWN_FAILING.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A sketch spline over control points was always the clamped uniform B-spline
over them, so a free-form edge whose knots are not evenly spaced (or whose
weights are not 1) could only come into a sketch as one curve per Bezier
span. A control spline now may save `nurbs`: its clamped knots and its
weights. Its points, polyline, exact pieces (one a knot span; conics for a
rational quadratic), the solver's distance and curvature, links, parallel
re-projection and inspect.sketch all read them; a loaded design's are
checked. One pending format file (spline-knots).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The tangent and curvature run-offs of the quarter cylinder r 10 strip, 5 across the surface,
are its section swept 20 along z: exactly 10 pi/2 20 + 5 x 20 = 414.159265. measure is SMLib's
integration on the exact surface (1e-8), not a mesh, so the run-off check is 1e-6 relative and
the 'measured on meshes' notes go. Red until SMLib's extension is solved to its distance
(smlib-extend-converges 590d8a1): eaa998a's curvature run-off is 4.986 long, area 413.882.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face tool cut with that face's surface alone, so a face of a surface body made of spans joined
tangent (a fit-point spline extruded, #3830) cut nothing where its span ends inside the body.
Split Body by a face of a surface body now takes the faces joined to it tangentially (the exact
check Extend uses, within the one angle for smooth) and slices with them as they are, through
SMLib's slice by a sheet (smlib_split_body_by_sheet, tangency-labs/smlib 7a01f1b), its sides those
of the picked face. A sheet that only nicks the body cuts nothing, as before; a lone face, or a
face of a solid, still cuts with its surface extended.
3830 comes off the booleans list. Its answer counted the two halves of the cup and not the surface
that cut it, which stays as a body of its own (4879 removes its surface for that reason): three
bodies, the volume unchanged.
Needs SMLib 7a01f1b pinned (SMLIB_REV, prebuilt; its SMLIB_ABI_VERSION 6, which crates/smlib now expects) to link.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A surface body of more than one face (a fit-point spline's spans extruded, #3830) was refused.
Each face with a picked edge is now extended alone by SMLib's extension of its surface, and the
faces are sewn back together, each the face it was. A side that lengthens onto a face meeting it
tangent takes that face along, past its edge continuing the picked one; across a crease the
neighbour grows only when its edge is picked too, otherwise the extension is refused. Faces whose
lengthened sides no longer meet are refused. Up to a target keeps the pieces every grown face
lies in.
3830's extensions now build; its split still fails (the tool point picks one span, and a face
tool cuts with that face's surface alone), so it stays on the list with that reason.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A sheet's side is extended exactly the distance asked, or the call says it cannot (#71);
NVIDIA's -Werror build passes again (#72); a body splits by a sheet of several faces as it
is, smlib_split_body_by_sheet (#73). The C API is ABI 6 (the new entry), so the binding
expects 6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The path chain oriented its first piece, not its first curve: a first curve
of several exact pieces (a fit spline, an ellipse, a projected free-form
edge) was never turned round, so a path whose next curve meets it at its
start was refused as "not connected end to end". The chain now orients
curve by curve: the first runs toward the second where they meet at one
end only (else from the end nearest the profile), each next one starts
where the previous ended, with one joint tolerance for both questions.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Develop 38fd4a4f6's Reference tests fail 34, 35, 77, 78 and 81 (STEP export finds fillet edges
4.6e-5 to 6.1e-5 off their surfaces against 2e-5), 37 (the apex blend 0.193 off), 80 (the
taper-to-fit area) and 83 (thicken status 1001). Each expected.json names its cause in
known_failure, which the oracle already reads: a known failure that passes fails the job.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Develop 38fd4a4f6 (CI run 41157648345, after the SMLib eaa998a pin) fails 10 cargo tests; each
is listed with the cause its failure message gives. Seven are SMLib refusals or results off
their derived volumes (fillet, chamfer and shell), the cli action matrix and naming corpus
among them; the two corner_g2 tests are the fitted G2 corner patch bulging past the box, which
the explicit G2 corner fixes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Native Linux tests and the GPU path test shards no longer fail on any failed test: nextest's
"tests failed" exit (100) goes to `cargo xtask known-failing --junit`, which fails on a failure
the list does not name and on a listed test that passes. Any other nextest exit (a build error,
a bad filter) still fails the step at once. The tests skill says to list a failing test with its
cause, in place of ignoring it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
`cargo xtask known-failing --junit PATH` reads nextest's junit (a file, or every .xml under a
directory, so all of a run's junit artifacts at once) and fails on a test that failed and the
list does not name, and on a listed test that passed (FIXED: take its line off, as the kernel
rig's gate does). A listed test that did not run is neither, since a pull request tests only the
packages it touches. The list has the rig's KNOWN_FAILING format (`<test> <cause>`, sorted, the
rig's own parser), with `<test>` nextest's `<binary id>::<test name>`; `cargo xtask lists`
checks it is sorted. The list starts empty.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The round's centre (32 - r, 42 - r) leaves the hollow's corner (30, 40)
outside its circle, opening the wall, where sqrt2 (r - 2) = r: the case
expects the refusal naming 6.828. 2852's R30, refused naming 25.24, now
passes too: both come off the known-failing list.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner's rule of 2026-10-30: a round running past a rim of its faces
into open air builds; one whose section would remove material through to
a hollow behind its edge (a slit into the hollow, a cut through the part)
is refused as too large, so the fillet names the largest radius that fits.
Decided exactly in the section of a straight convex edge between planar
faces: the round's centre is at depth r below both planes, and it opens
where its arc uncovers a boundary point strictly inside the wedge with no
material beyond it toward the centre (the body's exact point
classification). Looked at: straight edges parallel to the round's (a
hollow's corner), cylinders about parallel axes (a rounded corner), and a
picked concave parallel edge's own round, which fills the corner.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
"If too large: Stop and say so / Largest that fits / Narrow where needed"
becomes "Face limits: Exact radius / Largest radius / Narrowed radius", in
every language of the catalog, with the error and the tip that name them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The gate on aarch64 Linux reports 16 fillet cases FIXED with the pinned SMLib: the sloped and
equal-slope basin rim chamfers and rounds past their rims (1080-*, 5835-*), the pocket rim
chamfer past the floor, and the three pocket rim rounds now mitred at the pocket's corners
(every-pocket-*). No other area's gate reports a case fixed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
chamfer_wider_than_sloped_basin_rim_builds and a_rim_loop_chamfer_running_off_its_rim_builds
waited for SMLib to build the cliff overflow (CloseRemovalRegion refused it,
SM_FILERR_NO_ROOM). The pinned SMLib builds it (eddf0ede4, bba9914b6): both pass on aarch64
Linux and are no longer ignored.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib's point classification at the body's own tolerance, on the exact faces and edges, in
place of a mesh at a chord of 2e-3 of the body's size (about 0.4 mm on a 200 mm part). A point
within the tolerance of a face, edge or vertex is OnBoundary, never decided by a ray. A mesh
body's triangles are its exact shape: on them within the record's tolerance, else by ray
crossings. The kernel's own probes move to it; Body::contains goes once its last callers have.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NvoxKQ2rEZvcXNQnhE3Sgs
Binds the fork's smlib_brep_classify_point (SmBrep::Point3DClassify at the body's own zone
tolerance): inside, outside, or on the boundary within that tolerance of a face, an edge or a
vertex. A box and a cylinder just off its curved wall in measure's tests.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NvoxKQ2rEZvcXNQnhE3Sgs
1080-fillet-sloped-basin-1-mm-rim-r0-5-4-edges asked G1 at every blend seam. Its front wall
leans 19 over 20 and the others 10 over 20, so at each rim corner the two mitred rounds meet the
sloped corner edge at different heights (0.155625 and 0.276393 below the top): the deeper round
runs on past the corner edge and is cut by the plane of the other's wall, a crease by
construction, the gap below the shallower round filled on that plane. The check now measures
the rounds' contact seams, on the top and on each wall, where the rounds are tangent. The
volume is unchanged.
With SMLib smlib-batch-4-mitre (1c15364): the case passes (V 144530.427946 against
144530.427909, 15 faces, sound); before it, only the four corner creases failed the check.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Owner ruling d6: the letter t of "tres long" was drafted inward by SMLib (its
loop's normal, summed from the cross products of its chords, pointed down;
fixed in SMLib). With the fix every letter grows, and the join builds at 3°
and 15° by the Round rule and by the default rule (tangent for spline
outlines), where before 15° was expected to be refused.
Exact where the outline derives it: a letter's section at height z is its
outline offset by e = z tan a, by the Round rule of area A + P e + C e² while
the offset trims nothing (Steiner; C half each loop's turning less
tan(t/2) - t/2 at each reflex corner), so the letter is
A h + P k h²/2 + C k² h³/3. Each extruded alone, the t at 3° and s, l and o
at 3° and 15° are checked to 1e-6 (SMLib measures them within 1.3e-9; its
mass properties integrate to 1e-8). The concave joins of r, n, g and e are
tighter than e already at 3° (their swallowtails trimmed: r 1.1e-3 under the
formula), the t's by 15° (9.9e-5 over), and on the block the letters grow
into each other, so the whole word is checked for what holds exactly: one
closed body with the block, and half-way up every point of each straight
letter's outline inside the solid (the t drafted inward failed that).
With SMLib's fix (smlib-batch-4-text-loops 82133b2).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib's trim took a fillet meeting its face at more than 78 degrees for one with no side
(smlib-batch-4-rulings 971e567); the chord of the knife edge's 0.5 chamfer meets the flat face at
81 degrees. The case comes off the known-failing list.
Needs SMLIB_REV at that SMLib once it is pushed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib (smlib-batch-4-rulings 0e83cca, SMLIB_ABI_VERSION 5) rounds the edges whose rounds overflow
their faces first and then what is left of the others; smlib_blend_dropping says which picks'
edges the overflow removed whole. The smlib crate binds it (Brep::blend_dropping), the kernel
returns those picks (fillet_removing), and a fillet notes them as its warning: "the edge at
(0, 0, 10) was removed by the larger round, so it is not rounded" (a 50 x 50 x 20 block, R21 on
its bottom front edge with the front left upright: the block is R21's alone, in closed form).
Ruling d8: [21, 5, 5] on a 20 cube builds as R21 alone (6107.529370); the test is no longer
ignored. R25 with R2 at a corner (blend_corner) builds the same way, R25's alone in closed form,
where it was refused. The rig's 1080 notched block outline at R8 builds now and comes off the
known-failing list.
Needs SMLIB_REV at that SMLib once it is pushed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With SMLib's smooth limit at half a degree (422e6f1) the 1 degree crease rounds R2 and R20
(the test is no longer ignored), and with a cusp's chamfer the ball of radius d (d231547) the
The 0.5 one stays, for its own cause (the chamfer face left apart from the body). Two corner
cases the same SMLib (replay/1.0.1 and those commits) builds come off too.
Needs SMLIB_REV at that SMLib once it is pushed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
R21 on a 20 mm cube's edge builds (6107.529370, closed form, to 2e-12). With R5 on the
corner's other edges the R21 takes the faces they lie on, so the body is R21's alone; SMLib
fails that blend (status 1001, fillet error 4, Unable to Trim Rail Curve) and the test is
ignored until it builds. It replaces the refusal test that expected TooLarge.
Ruling d8 (2026-10-30).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
smlib_chamfer turns the distance into a ball radius d sqrt((1 - c)/(1 + c)), c the faces'
binormals' dot (capi chamfer_radius); at #851's cusp the binormals agree, the radius is 0 and
the call is SM_ERR_INVALID_INPUT before the fillet executive runs (gdb: chamfer_edges sees
radius 0; the dump's why is empty). SolidCraft hands every chamfer to SMLib, so the
ball-contact chord (the ball of radius d: 3.822 wide at d = 2, 159.2365 removed) is SMLib's to
build.
Ruling d4 (2026-10-30).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The blend's tangent angle is the shaded view's half degree (smooth_edges::ANGLE), one value
for smooth. A constant round lies r (sec(d/2) - 1) inside the crease it replaces (about
r d^2/8); below the edge's tolerance it changes nothing and is refused before SMLib (which
answered SM_ERR_INVALID_INPUT), naming the smallest radius that rounds,
tolerance / (sec(d/2) - 1). d is the least turn along the edge
(tangent_edge::least_angle_along, at tangent_all_along's stations).
Ruling d3 (2026-10-30). The block's tolerance is 2e-5 mm, not 0.1 um: at 1 deg the smallest
radius is 0.526 mm, so R2 and R20 reach SMLib, which still calls an edge within 2 deg smooth
(SmEdge::GetBlendEdgeuse); that test is ignored until SMLib uses 0.5 deg.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib measures an extension's distance across the surface (smlib_extend_sheet corrects
CreateExtendedSurface until the parameter line through the side's middle is that long), so each
run-off of the cylinder strip, swept 20 along z, adds 5 x 20: 10 pi/2 20 + 100 = 414.159 for
both. The parabola-length formula (418.18) was truck's.
Ruling d2 (2026-10-30).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Batch 4: drafts and loops, mitre corners, rim chamfers, rulings, tight box, stitch, point
classification, eased corners. Batch 5: a face's tight box divides its cells at the surface's
knots (thread boxes exact); a cylinder, cone, sphere or torus is extended along its own circles.
Batch 6: a revolution's extension closes its turn; a dropped point's foot is the converged one.
The C API is ABI 5 (smlib_blend_dropping and smlib_brep_classify_point are new; nothing existing
changed), so the binding expects 5.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Model cache format 7 (f3cf0c4f3) records a `rounds` list in every saved model, and reading one
without it is refused ("saved model: rounds"), as is a model without any of its other lists:
the states, the preview and the thinned model all came back empty. The io tests' saved_model
and timeline_model wrote format-6 records, so six tests (the round trip, the repeated part
stored once, and the four too-big-for-the-file cases) read nothing. Their bodies are none made
by rounds: the list is empty, and every expectation stays as it was.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
three_rounds_at_a_corner bounded the patch's extra cut at 3 mm³ beyond the strips, a figure
with no derivation; SMLib's all-at-once body at radii 2, 3, 4 takes 3.099 (V 7878.579574
against strips 7881.678). That body is the strips' outside the block its patch lies in: its
rounds are the strips' cylinders ending at a's section y = rc, b's x = 20 − rc and c's rails
where they meet a's and b's (the patch's four corners, now asserted), so the patch takes at
most what the strips leave in that block (11.86 at 2/2/3, 32.69 at 2/3/4). The lower bound is
that block; the upper one, at most the strips' volume, stays.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A design gave another body when opened from its model cache: the rounds a body was made of
lived only in memory, so a later round at their mitre was built sequentially (or refused, as
too large) after opening and blended with them when evaluated anew. The saved model now
records, per kernel body made by rounds, the body they were made on (a place in the bodies,
stored once though no state holds it) and the edges with their radii; reading it attaches
them back down each chain, and a thinned model keeps those bodies with the bodies made from
them. Model cache format 7.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The rounds that made a body are over the body its history is over: a new history (an
operation carrying a step's result back to its own operands) leaves the record out, so a
later round never reads one history against another body. Body::rounds gives the record and
Body::with_rounds_of takes one back, checked (edges of the body before them, a fillet's
radii, the body's history over that body), for a saved model to keep it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two box edges rounded r2 in one fillet, then the upright edge at their mitre r3 in a second
fillet (tangent chain on, so the seam comes with it): 10 faces, 7928.743041 mm³ as
fillet_radii of the box gives, the r2 rounds' names kept, the r3 round and the corner patch
the second fillet's.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner's ruling of 2026-10-30: a later fillet on an edge at a vertex where earlier
rounds already meet in a mitre makes the corner rounding all of them at once makes, one
smooth patch, the mitre gone. A body made by constant-radius rounds keeps what they were
made on and which edges they rounded at which radius; a later round reaching such a corner
blends the earlier rounds' edges (from the earliest of those meeting there on) with the new
edges in one SMLib blend of the body before them. The result's history is put over the body
the later round was given: an earlier round's face is that face trimmed anew (its name
kept), the corner patch and the new rounds are the later fillet's. A picked mitre seam has
no round of its own: it is refused unless a new edge of its radius meets it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A circle drawn as two half arcs now extrudes to one full-turn wall with its
seam edge, and each rim is one closed edge (tangency-labs/smlib#67). The tests
keep their intent on that body: a rim clicked once is one item, measured and
highlighted whole and picked from either side; a wall picked either side of
its seam is the same surface, lit whole, and toggles off from the other side;
every face is its own patch.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The L's six sides, its two caps and the bore: since tangency-labs/smlib#67 the
full circle the conversion fits extrudes to one full-turn wall, not two halves.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since tangency-labs/smlib#67 an extruded profile's arcs on one circle sweep one
full-turn wall face, made from all of them (Profile 4, 5 and 6 here), where it
was a face per arc. The geometric check of the record finds a face over several
segments by its triangles' centres and vertices on them; the revolved arcs keep
a face each.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
"After every step: checks each sweep, boolean and blend." / "Once per
feature: checks each feature's result once." / "Off (experimental): no
check.", in the order the options are shown, in en, ar, he, ja, zh, de
and ko.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Design preference surface_extrude_symmetric_per_side, which gave new
symmetric surface extrudes their distance on each side, is removed with
its Surfaces section, its session field and its labels. A symmetric
surface.extrude given no convention splits the total around the start
plane, as solid.extrude does; timeline.edit and timeline.redefine give a
one-sided extrude made symmetric the same. default_extrude_convention,
which only existed to read the preference, goes too.
An explicit `convention: legacy_per_side` and the marker saved designs
carry are unchanged. A saved preferences file that carries the key loads
as before (the prefs test now covers both removed keys).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Design preference metric_tap_drill_major_less_pitch, which gave new
metric tapped holes the major diameter less the pitch (M8: 6.75 mm), is
removed with its Holes section, its session field and its labels. A new
tapped hole given no diameter drills the tap maker's drill (M8: 6.8 mm);
a custom metric size the table lacks still takes the major less the pitch.
A saved preferences file that carries the key loads as before: serde
ignores keys Prefs doesn't have, and the key is not written back.
The preference's UI test becomes an engine test of the default drill
(M8, 1/4-20 UNC, a given diameter) and a prefs test that a saved file
with the removed key still loads.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SavedModel::thinned wrote `"validation": null` into the records of a model
that records no check mode, so a thinned model was never the model it came
from, even with room for every state, and its manifest grew by that key
(design::tests::a_thinned_model_keeps_the_final_state_then_the_slowest_that_fit,
a_model_cache_too_big_for_the_file_keeps_fewer_states). It copies the mode
when there is one and adds nothing otherwise; the tests' timeline model records
one, as every saved model does now, and a thinned model is asserted to keep it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An imported solid's faces were mapped to the file's only when SMLib's export numbered them as the
record did (native.rs, numbering). The export is in canonical order, which can list edges
otherwise: the two periodic-knot tubes of #1374 came back with their seam and rim edges listed
[2, 0, 1, 5, 3, 4], so the sheet, a valid B-rep, was refused as "SMLib renumbered the solid's
faces". The export's order says where SMLib's own face and edge i went; put back in SMLib's
order it numbers them as the record does, and SMLib's i is the record's i (an import lists them in
the record's order). numbering now takes that too, checked the same way; an edge split into
pieces is no renumbering and stays unmapped.
The tube test asserts what it should: one surface body, not a mesh, its faces the file's two
faces in order, every face meshed, in under a minute, and the area exact: each face is the file's
quadratic circle (corner weights sqrt(2)/2 to 12 digits) moved straight along its axis, a
cylinder's side, so 2 * 2 pi r h, which SMLib's measure meets to 1e-6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A solid SMLib refused came in as a surface body of its faces; where SMLib could not stitch those
either, the importer meshed them with SolidCraft's planner and kept the triangles as a mesh body
(step_in/parallel.rs, mesh_part). A build without SMLib showed it: every STEP part of the import
rig came in as a mesh of its faces. That fallback is gone: the solid is not imported, and its
warning names it and gives the kernel's reason ("solid 'Cube': not imported: ..."); a file whose
every solid is refused fails with the same words.
The import report tells a part the kernel refuses (Refused) from one the reader cannot read
(NotImported), and its summary says "N parts couldn't be made into solids and were not
imported" in place of the mesh sentence (en, ar, de, he, ja, ko, zh). STL, 3MF, OBJ and glTF
still come in as mesh bodies: that is what those files are.
The two-tube sheet of #1374 had come in only through that fallback: SMLib's export lists its
edges in another order, which the next commit maps. Until then its test asserts the refusal and
meshes the faces as read at the body's coarse display tolerance, the regression it guards.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since the body check mode became part of the evaluation context (9acbcf924),
evaluate digests the context with the mode, but Model::adopt still took the
context without it and the default mode: the first evaluation after opening a
design from its saved model saw another context and recomputed every feature
(undoing_an_edit_after_opening_reuses_the_saved_results,
rolling_the_marker_back_keeps_results_after_cache_eviction). The saved model
now records the mode it was evaluated under (thinned ones too), adopt takes
it and digests the context as evaluate does, and a session opening a design
under another mode evaluates it again rather than showing that mode's results.
Model cache format 6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner's bar running tangent into a cylinder rounds its concave arc and
its top edge both together and one after the other in either order, at
r 1 to 7. Two more peg cases come off the fillets known-failing list.
The ladder's shelled dome ring now has 9 edges, not 7: each of its two
cylinder walls is one full-turn face with its seam edge (one wall face per
extruded circle), where it was two halves. Faces, vertices, area and volume
are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NvoxKQ2rEZvcXNQnhE3Sgs
Tests that asserted an older design rebuilds as the build that saved it made it
now assert what still holds: it opens, is upgraded, evaluates, round trips and
names alike every time, and a feature that lost a reference asks for it again.
- naming corpus: two builds still name every face and edge alike and find the
same faces with their stored names; a stored name that finds nothing must
leave its feature failed (it asks to reselect), never a feature that builds.
A sketch's face is found by its name as the sketch finds it.
- design corpus: the volumes, areas and bounds the saving builds recorded are
no longer compared; a corpus design may fail only by a request to reselect,
and undo leaves the features that failed on opening failed.
- fixed robustness seeds: no panic, valid bodies and volumes moving the right
way; a step the kernel refuses is not a bug.
- a format 28 design's pieces and a format 13 design's extra bodies: marked as
they open and named alike, without promising the numbers or names saved.
- action matrix baseline: the two corpus designs whose fillets, chamfer and
offset face held names saved before format 14 re-recorded.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Face and edge names stored before format 14 were worked out from geometry and
open marked `?`. They were found among the names geometry gives today, as the
builds that saved them found them, pieces by the one nearest the pick. They
find nothing now: the feature holding one fails, asking for its reference to be
picked again, which stores its exact name. The legacy face and edge names a
body worked out for them are gone.
The tests that found such names as saved now assert the request to reselect:
a saved strip projection, a version 9 emboss on an annulus, a format 10 face
sketch and the corpus design named with numbered pieces.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face or edge name a feature stored finds only the face or edge it names: the
very name (or a synonym of that face), or the named piece told apart by the
faces around it. A face split into several pieces, merged into another face,
or left only as a face of the same stem is not the face that was picked, and a
name found on several bodies is told apart only by the body the reference's
point lies on. The fallbacks are gone: no piece nearest the pick, no merged
face, no edge at the picked point, no lost link that keeps its last geometry.
A feature whose stored reference finds nothing fails with DocError::Reselect:
blends, face points, sketch planes, sketch projection links, construction
references, hole reference edges, sheet bend edges and cylinder emboss.
The tests that asserted a fallback now assert the request to reselect: split
and removed fillet edges, a merged or rejoined face, a projected edge that was
split or is gone, a projection whose body went, saved strip projections after
wall coalescence, a fillet saved on a piece of a tapered side that is whole
now, a cut that starts from a piece that is no longer that piece, and a fillet
named on the lower of two stacked boxes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A side drawn in collinear sketch pieces is one face the extrude made from each
piece (#1788): its first role is its name, and the others were kept as aliases,
like the names of faces merged into it by a later join. They are names of that
very face, not of faces merged into it: the face keeps them as synonyms, which
a stored name matches as exactly as the face's name (a piece of it, once it is
split), through copies and later features, and the face whose name a merge
keeps keeps its synonyms too. An extrude-side certificate made on one of the
pieces finds the joined face the same way, and the saved model records the
synonyms (model cache format 5).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each component's bodies lie in its own coordinates in the state a feature's
points are named in, so a part inserted twice has a face at the same point in
each copy: a thread on the second copy was named after the first copy's face.
The points are named on the feature's own component's bodies first, then on
every body for a point that lies on none of them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
FeatureKind::face_body did not name Untrim's body, so its face was looked up by
name on every body and could be taken from another one (a copy has the same
face names).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
DocError::Reselect names the feature and the stored reference (a face or edge
name) that no longer finds what was picked: "<feature>: the reference `<name>`
no longer exists as it was picked; reselect it". Evaluation fills in the
feature's name; reselect_parts reads both back from a feature's error text, and
the UI shows the message translated (ar, de, en, he, ja, ko, zh in
assets/locales/pending/reference-reselect.json).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Design page's Body check section sets the validation preference
(each_feature; each_step keeps the old checks after every step; off is
experimental), applied to the session like the hole and surface extrude
preferences, to the sessions a design is opened in off the window thread,
and kept in step with diagnostics.validation. ui.prefs refuses an unknown
mode; a saved unknown one reads as the default.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
solidcraft-cli run --validation MODE, kernel-cases --validation=MODE and
cargo xtask kernel-cases --validation MODE (with --ladder too), so the rig
runs under each mode.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With input and output schemas. The mode is the session's user setting:
kept when another design is opened, started, recovered or imported (each
evaluated under it), and by timeline.compute_all. Linked and extracted
designs are evaluated under it too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The model carries the mode (Model::set_validation) and runs each feature
under it. When a step left its check, every body the feature made or
changed is checked by SMLib before the state is kept: a body the check
refuses fails the feature with the check's reason and the body's name.
The mode is part of the cache context, except each_step, whose results
are develop's.
Validation::EachStep checks every solid sweep, boolean and blend result
in SMLib as before; EachFeature leaves those checks to the end of the
feature and says a check was left; Off skips them. The evaluation installs
the mode around a feature (validating), as watch installs a watch; outside
it every step is checked as always, and the kernel's worker threads run
under the caller's mode.
Under EachFeature a body decoded from a step keeps that step's SMLib
result (for its own record only), and check_kept checks it as the step's
own check would have, then lets it go; any other kept body is checked as
SMLib imports its record. The record can pass where the result fails: a
fillet's surface of revolution whose polar curve SMLib's check refuses
(fillets/1080-fillet-sloped-basin-1-mm-rim-r0-5-4-edges) is rebuilt on
import, and its round meets a wall at 72 degrees; checking the record
alone kept that round.
The largest-size search of a blend checks every candidate as it is built,
whatever the mode: a candidate fits only as a body the check accepts.
Surface operations and stitching keep their checks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillets/1080-chamfer-runs-into-openings-refused already expects the
9 mm chamfer built and cut by the openings (V = 3350); its id said it is
refused. It becomes 1080-chamfer-runs-into-openings-cut-by-them, and
blend_fit's chamfer_through_the_openings_is_cut_by_them names it so.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner's ruling of 2026-10-30 (d7): a chamfer that reaches or runs
past a rim builds, everywhere, as SMLib's cliff overflow builds a convex
blend: on its faces' continuations, then cut by what the cliff cuts
away. The earlier rule refused a chamfer whose leg runs off its face.
SolidCraft has no such refusal of its own; SMLib refuses these, so the
cases fail on SMLib 985faaee and are listed with its reasons (traced in
gdb at SmFilletErrorInfo::NoteError):
- the sloped basin's single front edge past its rim (1 mm rim 2 and 3,
0.5 mm rim 4): CloseRemovalRegion, "a cliff overflow's fillet boundary
lies on the surfaces of two faces": the sheet's boundary along the end
wall's top edge line from the edge's vertex to the top's contact
(x = 20, y from -1 to 1, z = 30) lies on the top's plane and the end
wall's both.
- the equal-slope basin's rim loop 2 and the 5 mm pocket's rim 5.5:
CloseRemovalRegion, "a cliff overflow's fillet boundary closes on one
face" (every edge of the loop blended).
- the sloped basin's front and right edges 2: status 1001,
SmFilletNx2Corner, CalcFilletIntSideFace.
- the cylinder's rim 2 across the top and 25 down its 20 mm wall: status
1001, SmFilletNx1ClosedCorner cannot compute its geometry.
Expectations, each from a closed form or an exact quadrature:
- chamfer_wider_than_sloped_basin_rim_is_refused becomes
chamfer_wider_than_sloped_basin_rim_builds, ignored until SMLib builds
it: the triangle clipped to y >= 0, times 30 + z (shoelace):
73.450475, 142.112912, 184.526613; the rig's 1080-chamfer-sloped-
basin-*-past-the-rim cases: V = 144459.882858, 144391.220421,
144082.140054.
- blend_corner's a_rim_loop_chamfer_running_off_its_rim_is_refused
becomes ..._builds, ignored likewise: sliced at depth t the mitred loop
takes the pocket's rectangle grown by w(t) = d (1 - t / delta) on each
side, clipped to y >= 0, less the rectangle, Simpson exact on each
quadratic piece: 375.673069, V = 139957.660264 (the rig's
1080-chamfer-equal-slope-basin-1-mm-rim-loop-2 too).
- 5835-chamfer-sloped-basin-1-mm-rim-2-two-edges: the same slicing with
the front and right sides grown, 161.279149, V = 144372.054184.
- every-chamfer-pocket-rim-5-5-past-the-floor: 50 d^2 + (4/3) d^3 at
5.5, the groove 0.5 deep round the floor included, V = 19265.666667,
15 faces.
- 2893-chamfer-two-distances-past-the-bottom: the frustum r 8 to 9.6
over 20, pi 20 (64 + 76.8 + 92.16) / 3 = 4879.102831, 3 faces.
Each drops "-refused" from its id (the 5.5 pocket's says past the floor).
The same slicing reproduces the existing 5835-...-4-mm-rim-2-2-edges
(176.006388), equal-slope loop 0.5 (24.455545) and half-width
(5.426545) values.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner's ruling of 2026-10-30 (d7): a chamfer that reaches or runs
past a rim builds, everywhere. The earlier rule refused a full-width
chamfer, naming the largest size that fits. SolidCraft has no such
refusal of its own (a blend's TooLarge is SMLib's SMLIB_ERR_NO_ROOM,
and the doc layer only names the largest fit), and SMLib already builds
these, so only the expectations change, each from a closed form or an
exact quadrature:
- chamfer_wider_than_sloped_basin_rim_is_refused's full widths become
chamfer_as_wide_as_sloped_basin_rim_builds: the whole triangle,
d drop / 2 (60 - drop / 3), drop = 20 d / sqrt((20 - margin)^2 + 400)
(shoelace over the section, its length 30 + z between the sloped end
walls): 21.662379 (1 on 1 mm), 5.359321 (0.5 on 0.5 mm). The basin and
the removal move into sloped_basin and sloped_basin_rim_chamfer_removed.
- fillets/1080-chamfer-sloped-basin-{1,0-5}-mm-rim-as-wide-as-the-rim:
V = 144511.670954, 144261.307346.
- fillets/every-chamfer-pocket-rim-5-full-depth: 50 d^2 + (4/3) d^3 at
d = 5, V = 19583.333333, 11 faces (the walls used up).
- fillets/every-hole-top-rim-chamfer-10-full-depth: the triangle 10^2/2
revolved at rho = 5 + 10/3 (Pappus), V = 16000 - 3250 pi / 3 =
12596.607959, 7 faces.
- fillets/1113-chamfer-arc-top-edge-full-wall-height: the 9.5 case's
Int rho asin(21/rho) h(rho) drho at s = 10, by Gauss-Legendre in
u = asin(21/rho), 1303.400331, V = 8287.993768.
Each drops "-refused" from its id and its KNOWN_FAILING line goes: all
five pass on SMLib 985faaee.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A joint drive with contact on tested each sample and each of 14 bisection
steps with a brute-force mesh-against-mesh overlap check, its cost the product
of the two display meshes' triangle counts under a fixed budget. Since the
display mesh puts nodes inside planar faces, a cam and its tappet or a block and
a sphere exceed that budget ("motion exceeds contact sweep collision budget"),
and whether a motion may run depended on display tessellation.
Now, at each sample, bodies whose exact boxes meet are intersected exactly
(SMLib boolean, exact volume); a shared volume is a contact, a touch (a cam and
its follower) is none. Between the last clear sample and the first overlapping
one the drive closes in by conservative advancement: from a clear placement it
moves no point of the pair further than the pair's exact gap (body_distance),
until the gap is within 1e-9 of the bodies' size. Only clear placements are
measured, so no intersection is asked to tell touching from overlapping within
its tolerance (SMLib's boolean fails there, status 1001, on two cubes 1e-4
into each other). The budget is 2048 intersections per drive.
Fixes cmd::joints::tests::cam::a_cam_drives_a_closed_loop_and_stops_at_a_contact
and contact_sweep ordinary_curved_bodies_stop_at_their_contact_within_the_sampling_budget,
and the other contact, cam, motion, interference and clearance tests pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The batch brings history Kept for a face only cut by a vertex, revolve of a
curve touching its axis, loops touching at a vertex, a torus rail on its axis,
and the fillet causes behind the owner's bar running into a cylinder: a
tangent edge past a tangent side edge is the corner's edge, the rails of a
round collapsing at its end end at one point, and a trace from a tangent point
starts along the curve.
The two peg cases with the sketch's solver residues come off the fillets
known-failing list: both build.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NvoxKQ2rEZvcXNQnhE3Sgs
The test it replaces probed for a polyline by placing a 0.2 um square in the
gap between the rim and a 1 um polyline: a micro feature in a millimetre part.
This one asserts the property itself, in millimetres: every rim segment of the
flat pattern is an arc of radius 10, the DXF draws one circle of radius 10, and
a 1 mm cut 0.5 mm inside the rim stays one hole.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A G2 corner patch's side is G2 to its round except near the face it ends
on, and SMLib holds each side as two edges. The smooth-join test judged
each edge alone, so the G2 half was hidden and the line stopped at the
vertex in the middle of the seam. Edges between the same faces that
continue each other at a shared vertex now share one verdict: smooth only
where every one of them is.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
patch_continuity_and_points_cap_a_tube listed "g3" among the values surface.patch must refuse,
from #1521 when a patch built G0 to G2. Since cb8712aa0 Construction::Patch builds G0 to G4
(continuity_order.rs: `Construction::Patch => (G0, G4)`), so g3 is accepted, as it must be;
the order above what a patch builds is g5.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The block solve kept its sparse entries in hashed maps and summed the coupling of each patch's
interior onto the shared unknowns in the map's order, which a hashed map draws afresh each
time: the same loop filled to surfaces differing in their last bits, so a design read back
rebuilt another cap (engine patch_continuity_and_points_cap_a_tube, "the same surface read
back"). The entries are now in ordered maps; a fill twice of the tube's ring and of a hexagon
about a centre are bit for bit the same.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A tube's end has no corner at all; the kernel gives its circle in four quarters, and one patch
on them has a corner on each quarter joint where the loop runs straight on: flat, its two
derivatives parallel, which SMLib's check refuses (Bad Surf FlatCorner within 0.5°). Any
layout with a patch corner on such a loop has one.
Such a loop is now the ring of structured meshing (an O-grid): four patches each on one side of
the loop, two meeting at every point of it, about one inner patch (Quilt::ring). The spokes are
now any curve two patches share (Seam), each glued G1 with linear gluing functions: at an inner
vertex of valence m the characteristic value ±2cos(2π/m) (the centre's spokes as before, the
ring's inner corners with m = 3), at the loop what its two sides' derivatives there ask
(foot_gluing; 0 and their speeds' ratio where the loop runs straight on). IsoFill says where each
spoke starts and ends (SpokeEnd) and holds the inner vertices; the kernel builds them so.
The tube end caps G1 (normal within 0.0157° of the wall, sealed 10458.232 against the tube and
dome's 10458.232) and G2 (curvature across 4.6e-5 /mm against G1's 0.30), five faces with no flat
corner. A crown on the ring lifts the inner patch whole (the ring rises as v(2 − v)).
Saved designs: a general patch on a smooth closed loop (a tube's end) rebuilds as five faces
where it was one (which SMLib refused).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The crown moved the free points along the patch's own normal, whose sign is only the way the
patch happens to be parameterised: the pentagonal window's crown 0.1 sank 1.56 mm into the
cylinder (the middle 9.959 from the axis plain, 8.395 crowned) where the kernel documents it
filling out to the side the G1/G2 neighbours' normals face. The crown now takes that side: the
sum of the patch's normals against the neighbours' over the stations sets one sign for the fill.
The pentagon's middle goes out to 11.526.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The crown's bump went onto the first net with any control point clear of the loop; a coarse
net could leave points supporting the box's middle without it, so the middle rose by only a
share of crown · size (the pentagonal cup: 7.6558 of 13.3877). The net is now refined (knot
insertion, the surface unchanged) until every point whose basis is not zero at the middle takes
the bump, where the middle is inside the loop: the bubble is then exactly 1 there, as fill_loop
documents.
The kernel test measured against the face's domain, which SMLib holds as the loop's extent; the
documented box is the height field's whole domain, the loop's box and 1% more each way (as the
engine's crown test says, 28 × 28 + 1%). Its diagonal is 1.02 times the face's, so the
documented rise is 0.25 × 53.5506 = 13.387661 = 1.02 × 13.125158, which the patch now meets to
1e-6 for crowns ±0.25.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face's domain and pcurves are on Face::parameters() (its NURBS form when it has one, the
record's documented contract); the general fill tests' quadrature and the pick-order and crown
comparisons evaluated Face::surface instead. A fill SMLib recognises as an analytic surface (the
240° window's G0 patch is the cylinder itself) then read its arc-length domain as angles: the
quadrature came out 4188.79 for the window's 418.879 (10 × 41.888 of angle against 4.189).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A vertex where the loop runs straight on (SMLib's seams: a cylinder's at 35.12° splits a half
circle into two arcs) was a corner of the fill's region, so a patch had a 180° corner there:
flat (its two derivatives parallel), degenerate, and refused by SMLib's check (Bad Surf
FlatCorner, within 0.5°). Half a post's side, six edges on four corners, was six patches about
a centre and folded.
The region's corners are now the vertices where the loop turns by more than the kernel's corner
test (1°, IsoSpec::corner_turn); a side is the chain of edges between two corners, held as one
exact B-spline (each edge carried exactly to the side's degree on its stretch, joined at a knot
of full multiplicity, which the knot-line rows keep tangent across). Each edge's pcurve is still
a straight line, now on its stretch of a patch's side; a spoke says where on the loop it starts
(IsoFill::feet). A loop with fewer than three corners keeps its edges as sides.
The half post is one face again (a_loop_that_does_not_bound_a_height_field_fills_with_one_face),
the 240° window is one patch, the exact cylinder (area 418.879), and the torus window fills G0
and G1.
Saved designs: a general patch on a loop with seam vertices rebuilds with fewer faces (the half
post's 6 become 1), its corners where the loop's are.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Across each spoke the two patches' derivatives away from it were held opposed (parametric C1).
At the centre that asks the spokes' tangents d_{k-1} + d_{k+1} = 0, which for N other than a
multiple of four only d = 0 meets: every patch got a flat, singular corner at the centre (a
pentagon on a wavy rim: derivatives of 3.3e-5 there; the hexagonal rim's control net doubled
back and SMLib's check refused it as NonMonotonic).
The join is now G1 with the linear gluing function of Loop, Peters and Hahn:
d/du S_k + d/dv S_{k+1} = alpha(t) d/dt S along the spoke, alpha(t) = 2 cos(2 pi / N) t. At the
midpoint alpha is 0 (the edge's two halves leave it at the same speed, opposed); at the centre
it is the value any affine image of the regular N-star meets, d_{k-1} + d_{k+1} = 2 cos(2 pi/N)
d_k. N = 4 is the old row. The pentagon's centre corners now open 72° with derivatives of 16.2
(N = 5 to 8: 360°/N each).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A tensor patch's cross derivative along a side is, at a corner, the tangent of the side it
meets there (the two share the corner's next control point): the compatibility condition of
Coons and Gregory interpolation. The weak rows asked t·c′ = 0 and w·c′ = s right up to every
corner, which no patch can meet where the next side is not square to the edge; at a smooth
vertex (a tube's end in four quarter arcs) the next side runs along the edge itself. The least
squares spread the contradiction into the first spans and folded them: the tube end's worst
normal went 13.8°, 17.9°, 89.8° (folded), 85.6°, ... 25.1° (folded) with refinement, and the
symmetric loop came back asymmetric.
The target is now the Coons start's own cross derivative, compatible at every corner by
construction, turned about the edge into the neighbour's tangent plane (its part along the
edge kept, the rest laid along the leave direction at its length). The tube end converges
21°, 3.8°, 0.76°, 0.18°, 0.045°, symmetric, and does not fold.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An edge SMLib reports as new was given the origin "where its two faces meet", each side what
that face came from. A split line lies between two pieces of one face, so both sides are that
face and the edge read EdgeOrigin::Meet { top, top }: a face meeting itself, which no two faces
did. Such an edge now has no origin, as the history's EdgeProv documents for a new edge between
faces new to the operation (the doc now names the split line too).
New test: a_split_line_comes_from_nothing (one face split in two, its new edge has no origin).
modify_history_tests::split_a_face checks this at its last line; it still fails earlier, on its
counts ([3, 2, 2, 0, 0] for [5, 0, 2, 0, 0]) on develop's SMLib pin, until SMLib batch 3's
history fix (03bbc88) is pinned.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The closed-box certificate took exactly twelve triangles, two a face. The one mesher puts
interior points on planar faces (a 20 x 20 x 2 box comes in more than twelve), so
render.preflight refused a box as "unsupported dielectric boundary tessellation"; a box is a
box however its faces are triangulated.
The certificate now reads the mesh's geometry exactly: every triangle in the plane of one of
the faces of the box its points span (corners on it, normal exactly that axis) facing out,
and the triangles a closed, consistently oriented surface (each edge, by its ends' exact
coordinates, run once each way). Such a surface is a 2-cycle on the box's boundary, a sphere,
so it covers every point of it the same k >= 1 times, and its area is k times the boundary's:
k = 1 is decided by the area being under 1.5 times the boundary's. Up to 65536 triangles. A
thin sheet is still one or two triangles; a general closed mesh is still refused.
New test: a box with its top fanned from a point in it is certified, with the same bounds;
the point 1e-9 off the plane, or the top covered twice, is refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A tapered extrusion of a region with holes checks, at its two ends, that each hole's section
stays inside the outer loop's and clear of the others. Those sections were made with the
start section's rule, which refuses a line the offset swallows ("draft extrusion that starts
before its profile where the taper closes a side"). At the far end such a line is ordinary: a
short line between two concave corners whose wall closes on the way up. The letter g of
"tres long" (8 mm, joined 2 mm with a 3 degree taper) has one, and the whole text was refused
with a message about the start that does not apply.
The end check now takes the section as it is there (offset_loop's swallowed lines left out, their
neighbours meeting where they were); whether the closing wall is built stays the draft's to
say. The g builds at 3 and 15 degrees.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
raked_wedge (#1617, test from 7f2ae11a0 on 2026-10-17) rounds four end edges R4. Since
84bd2daad (2026-10-22) solid.fillet carries each picked edge round its tangent chain by
default, so the R4 now ran on round the end outline onto the R1.2 round's end arc, where the
round's rail on the end face would be an arc of radius 1.2 - 4 < 0, and it is refused as too
large. The issue's R4 rounds those four edges alone: the step says tangent_chain: false.
The test still fails, one step later, on SMLib: the issue's R0.854 round of both ends' outline
is refused by SMLib's own check (SmBSplineSurface Test #5, "close to singular at boundary"),
reported for the SMLib batch.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tiny_cut_inside_exact_circle_survives_unfold_refold_and_dxf cut a 50 nm square 0.2 um inside
the disc's rim. SMLib models at a zone tolerance of 1e-4 mm: its check refuses the cut's 50 nm
walls as planes "degenerated to a point (tolerance=0.0001)", so the extrude was refused. A
feature below the kernel's resolution is not one it can keep, on any path.
The test's point is that the exact circle, not its display polyline, bounds the flat pattern:
a cut between the two must survive unfold, refold and the DXF. The polyline at 1e-3 has 224
chords (sagitta 10 (1 - cos(pi/224)) = 0.98 um), so the square is now 0.2 um across, centred
0.4 um inside the circle on the first chord's bisector: outside the polyline, inside the
circle, its walls and its 0.3 um gap to the rim above 0.1 um. Its hole's area is
(2e-4)^2 = 4e-8 (was 2.5e-9 for the 50 nm square).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A strip's far curve, where its rulings turn along the edge, is a B-spline in the edge's own
spline space. A straight edge's space is one linear span, which holds only the chord between the
strip's corners: the rulings' ends (the closeness term) had no say, and the strip had no handles.
Off the patch over the cup's rim in a_strip_off_a_curved_face_is_ruled_to_its_far_curve the
normal along the straight rim at x = 14 turns by 5 degrees (x component -0.350 at the ends,
-0.262 at the middle; the patch's middle is at z 10.296, not the cylinder's 10), so the strip is
free and its rulings' ends lie up to 0.26 mm off that chord.
Such an edge is now taken as the cubic Bezier it also is (exact degree elevation: control points
at thirds, the same points at the same parameters) before the far curve is solved, so the far
curve has two interior control points: its handles. Exact strips on straight edges keep the line.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since 6cc759750 an extruded circle is a cylinder over the face's trimmed range, and the circle was
fitted through three points of that range. On the shallow strip of the extruded-ellipse test
(radius 3000, 6 mm wide: 0.002 rad round) the three points are 2 mm apart and the fit is
ill-conditioned: the centre came 2.5e-10 mm off the circle's plane (the frame's x axis tilted
8.3e-14 out of it) and the radius 8.1e-9 mm short. SMLib took that cylinder, but on the way back
(the import's SMLib round trip, then Body::smlib) SmCylinder::IsNurbSurfaceCylinder did not find
a cylinder in SMLib's own NURBS of it, and the strip was refused as needing repair
(SM_ERR_INVALID_INPUT, AnalyticOfNurbs). Before 6cc759750 the fit ran over the record's domain,
1.4e-6 rad wider on each side, and happened to come out level.
The patch's trims were right all along: they straddle the profile's seam (u = 4, the knot
between the spline's two quarter spans) without wrapping, [3.9992930395761980, 4.0007069604238020],
which is the exact 0.001 rad either side of it.
The circle is now fixed by three points on the arc's widest knot span (a quarter turn here),
where it is determined to rounding, and checked as before at 65 points over the face's range.
A revolution's profile circle is fitted as before.
Regression test (smlib import tests): a 0.002 rad patch of an extruded circle of radius 3000
across that knot is the cylinder of radius 3000 to 1e-10 (it was 2.8e-10 short), its axis
along z to 1e-15, over exactly the patch's range.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_long_setback_keeps_the_g2_patch_inside_the_box measured the patch's reach past the box on
the vertices of a display mesh at chord 1e-3, so where the tessellation's nodes fell decided
the result (#9307 moved them). It now brackets the highest max(x - 60, y - 40, z - 20) on the
patch within its trim from the exact B-spline: the surface cut into its Bezier patches, split
by de Casteljau while a patch's control hull can still beat the best surface point found
inside the trim (the pcurves), so the hull bounds it from above and the patch corners from
below.
Body::exact_bounds (SMLib's SmFace::CalculateTightBoundingBox) is not used: it gives the
patch 40 + 1.4e-14 in y at set back 12, while the surface at (u, v) (-1.509, 8.041), inside
the trim, is at y = 40.00133 (a mesh node there agrees).
The bound is 1e-9 mm: the body is convex, a convex face lies inside each tangent plane, and
the patch meets each box plane tangentially along its seam, so it reaches past none of them
but by rounding. The exact extent shows the patch does reach past the box at every setback,
each time 0.4 to 0.65 mm from a hole corner (where a round's section ends on a plane, inside
the fill's fade zone): 7.0776e-4 (s 11, z), 1.3298e-3 (s 12, y), 2.6338e-6 (s 13, y),
5.1293e-5 (s 14, x), 4.7891e-4 (s 15, x). The test fails with these numbers until the patch
construction keeps the fill convex there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Zebra, Draft, Curvature Map, Environment Map, Accessibility, Continuity,
Isocurve and Curvature Comb are one group: turning one on turns the one on
before off (Session::analyses_off), and the session says which is on
(Session::active_analysis). The new inspect.analysis names it and off: true
turns it off (typed contract and reply). The toolbar shows the active one's
button pressed, its tool running or not. Curvature Comb's button toggles:
on, the tool combs the edges clicked; again, off and the combs go; Esc or a
right click in the tool turns it off; opening another dialog leaves it on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
inspect.curvature_comb keeps what it shows in the session (Session::comb):
add: true combs the new picks together with those shown, so every comb
shares one scale and tooth lengths compare across curves (a curvature jump
where two edges meet shows); with nothing picked it draws those shown again
with new options; clear: true ends it. Each sketch curve keeps its own
sketch. The tool runs it once per pick and per Comb length / Teeth change,
and the viewport draws the session's combs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
comb_envelope and comb_teeth in the light and dark palettes (legible on each
view background), themable from a theme file, measured by the colour vision
tests.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This replaces the spline fallback of 64efbe7f6. A face SMLib holds as a rational spline that is
an extruded circle was recognised as a cylinder over the spline's whole parameter range: for the
upper half of the peg's wall after the front top round that range runs 393 degrees round (u in
[-15, 40.18], SMLib's support surface extended past the face), and SMLib refuses a cylinder past
one turn. The face itself, its loops' box, spans u in [0, 8 pi]: half a turn. The cylinder now
takes the face's trimmed range, the least and greatest u and v of its uses' pcurves over their
edges' intervals, found to rounding on each Bezier span (the values at the pieces' ends bound the
extreme from below, the control points from above; pieces that cannot beat the best are dropped,
the others halved). A face whose trims run more than one turn round is no face of a cylinder: it
is refused as an invalid record, never kept as a spline.
The peg (40 x 16 x 10 bar, R8 end, a disc R8 extruded 10 up on its top, joined), front top edge
rounded r2: the body measures to 1e-7 of its closed form, the three r2 corner rounds after it
build and measure, and the upper wall comes back from SMLib a cylinder of radius 8 (engine test).
The rig case of the round of the edge where that round meets the wall (tangent at one end) still
fails, for another reason: on the wall as its exact cylinder SMLib refuses the round as larger
than the faces beside its edges allow (it fades to nothing at the tangent end) instead of
returning a body its own check refuses. Its KNOWN_FAILING line says so.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This replaces the five fixed stations of 0a166bf3d, which compared the faces' normals at the
edge's ends and quarters: a sampled guess that misses a crease or a tangency between stations.
Along each use's pcurve (on the edge's own parameter), the face's natural normal Su x Sv is a
polynomial in t on every piece where the pcurve stays in one span of its knots and one knot cell
of its surface: of degree (2p + 2q - 2) r for a B-spline surface of degree (p, q) and a pcurve
of degree r, (4p + 4q - 2) r for a rational one (its homogeneous form), 0 for a plane. The cross
product of the two outward normals is then a polynomial of degree D, the sum, and it vanishes
on the piece iff it vanishes at D + 1 points of it (the piece's Chebyshev points). Pieces end at
the pcurves' knots and where a pcurve crosses a knot line of its surface (periodic copies too),
each a root of U(t) - k W(t) on a Bézier span, isolated on its Bernstein coefficients. A point
where the normals differ is a witness of a crease; agreement at all of them is a proof of
tangency (crate::tangent_edge). Outward normals opposite (a cusp, a knife edge) is not smooth,
as SMLib's GetBlendEdgeuse has it since fork 81f411d; a seam, whose two uses are on one face,
is smooth.
The C API has no edge classification query; SMLib's own (GetBlendEdgeuse at t = 0.34,
IsTangentSector at five stations) is sampled too. A fork item to make it exact and expose it is
written up for the SMLib batch. The tangent-chain walk decides each edge once.
The bar whose front face runs tangent into a cylinder R8, its front top edge rounded r2: the
round's contact with the front face and the front face's tangent line with the wall are smooth;
the round's edge with the wall (tangent at (40, 0, 8) only) and a box edge are creases; the
knife edge of #851 is a cusp, not smooth (blend.rs test). The blend, fillet and chamfer kernel
tests and the fillets rig keep their outcomes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A curvature-continuous fillet (type curvature, equivalent radius 2) of the three edges at a top
corner of the owner's bar built, but its corner was SMLib's own patch, which is only tangent to
the rounds: Continuity Analysis measures a 35.8 / mm curvature jump across its seams (with the
rounds G2 to the faces everywhere else), and it shows as a dome sitting on the corner, the
vertical round's shading lost under it. All three edges were rounded; the corner was not
curvature-continuous, and nothing said so.
Such a corner, where three or more rounded edges meet, is now set back (1.5 r unless a setback
names it) and left open by SMLib with the curvature sections (smlib_blend_setback_open with
SMLIB_BLEND_CURVATURE), and filled with the G2 patch fillet_g2_corners already uses for circular
rounds (fillet_g2_with_g2_corners). The `corner` option no longer leaves a curvature fillet with
SMLib's patch: its corners are always G2, and setbacks on it set back the curvature rounds (they
were circular rounds before). Where SMLib or the patch cannot build it, it is refused.
On the reported geometry every seam of the rounds and the patch is G2 (rig
report-fillet-curvature-three-edges-at-bar-corner-solver-residues), and the rounds outside the
setback cube measure exactly: V0 - 27 - (the quintic section, A2 = 0.597884, along the front edge
from 3 to the wall and along 7 and 13) = 9541.116961 (...-outside-setback). With round coordinates
SMLib's open sheet fails its own check (a 2D curve cusp, SmBSplineCurve test 13): the fillet is
refused, listed as known failing (report-fillet-curvature-three-edges-at-bar-corner).
Test: corner_g2_tests::curvature_rounds_meet_at_a_g2_corner (box corner, r5: the patch's jump
under CURVATURE_SHARE / r clear of the hole's corners, SMLib's own patch over ten times that).
Changes a default: a saved curvature fillet with such a corner rebuilds with the G2 patch; there
is no preference or fallback to SMLib's patch.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The cases from the owner's reports in the app (no issue) were named and cited after issues they
only resemble (#9, #896, #1207). They are report-<name> now and list no sources; their `what`
says where they were reported. The README gives the convention (report- for a report with no
issue or pull request, cov- for a command with no source).
report-fillet-g2-corner-of-bar-running-into-cylinder was graded by a 3^3 volume band and a face
count, which say little about a G2 corner. It keeps the band (provable: the patch differs from the
rolling ball only within the setback cube) and adds the corner seams' curvature jump, as
fillets/9-fillet-corner-r1-curvature-default-setback scaled by r = 2 (at most 0.0025 / mm, clear
of the hole's corners by 0.48 mm, within 4.4 mm of the corner). A new
...-outside-setback case cuts the setback cube [-1, 3] x [-1, 3] x [7, 11] away and grades the
three rounds outside it exactly: V0 - 27 - (front spandrel from x = 3 to the wall + (1 - pi/4) r^2
over 7 and 13) = 9342.029541. The bare face counts are dropped.
The -solver-residues recipes now take the reported sketch's points exactly (the arc's end
x = 41.649711830726524); they fail as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A 220 x 11 two-distance chamfer of a tapered plate's bottom front edge (rig fillets/5181) runs up
the front face past its straight side edges, l2|start and l6|start, and takes them up: no result
edge comes from them. SMLib leaves the blend face untagged, and it was given every blended edge
whose two faces lie next to it: the picked edge l1|start and both taken-up side edges. Its name
came from the first in edge order, F3:blend:F2:side:l6|F2:start, an edge the user never picked.
A blend face now comes from the picked edges it lies between; only a face that lies between no
picked edge's faces (a corner's piece over a short edge the corner took up) comes from the
taken-up edges it lies between. The chamfer is F3:blend:F2:side:l1|F2:start everywhere on it.
Persistent names: a blend face that also lay between the faces of an edge the blend ran over is
now named by its picked edge only. There is no alias for the old name (5181's
F3:blend:F2:side:l6|F2:start): a saved reference to it is reported lost.
Also corrects 5181's "what": the chamfer builds (335399.749), it is not refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An r2 round of the arc adds the spandrel ring clipped by the bar's sides, integral over rho in
[8, 10] of [2 - sqrt(4 - (10 - rho)^2)] rho 2 asin(8 / rho) = 18.455363. With round coordinates
it builds to that; with the reported design's sketch residues (~1e-10 mm) SMLib's blend fails,
status 1001 (SmFilletNx1Corner: unable to make topology at the arc's end), listed as known
failing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The edge where a round runs into a cylinder's wall it is tangent to at one end (the bar's front
top round meeting the wall at (40, 0, 8), where the round's contact with the front face meets the
wall's tangent line with it) was refused as smooth, "nothing to round": the faces' normals were
compared at the edge's start alone, which was that tangent end. They are now compared at five
stations from end to end, and an edge is smooth only where its faces meet tangentially at every
one: this edge is a crease everywhere but its end. The wall's seam with the front face, tangent
all along, is still refused as smooth.
The round of that edge with its tangent chain (the top's concave arc) now reaches SMLib, whose
result fails its own check (internal poles, coincident control points, cusps where the round
fades out): its rig case in fillets is listed as known failing for that SMLib defect.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A round of a bar's top edge, where the bar's side runs tangent into a cylinder standing on its
round end, seemed to build: SMLib's result holds the cylinder's wall as a rational spline whose
parameter range runs 393° round (u in [-15, 40.18]). Going back into SMLib (to measure the body
or change it again) the record's extruded circles become SMLib cylinders with the face's range on
them, and that range was 6.866 rad: SMLib refuses a cylinder past one turn
(SM_ERR_INVALID_INPUT from BrepFromFlat's face range check). The body could not be measured, a
second round on it failed, and the app said one of its freeform faces can't be used. An extruded
circle whose range runs more than one turn round now stays the exact spline it is.
The bar of the reported design (40 x 16 x 10 with a R8 end, a disc R8 sketched on its top and
extruded 10 up, joined): the front top round removes the spandrel up to the cylinder's wall,
integral over y in [0, 2] of (2 - sqrt(4 - (2 - y)^2)) (40 - sqrt(64 - (8 - y)^2)) = 32.306475;
the three rounds at the other top corner after it remove 53.285832 in all (spandrels outside the
corner cube, the cube less the rolling ball's eighth). Both now build and measure to 1e-7; the
curvature corner of the three edges at once (setback 3) builds too.
Rig cases (fillets) for the three reports: the curvature corner of the three edges, the front
top round, and the three corner rounds after it pass; the same corner rounds with the design's
~1e-10 mm sketch residues fail in SMLib (its corner's surface of revolution is close to singular
at its boundary) and are listed as known failing. (Their ids take the report- convention in the
rig commit later on this branch.)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
chamfer_wider_than_sloped_basin_rim_is_refused and
a_rim_loop_chamfer_running_off_its_rim_is_refused assert what the kernel gives: a
TooLarge refusal, the body unchanged. The largest size that fits is named by the doc
layer (with_largest_fit) and graded by the rig's 1080 basin cases through commands.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillets/851-fillet-knife-edge-tip-r2 passes on develop's SMLib (985faae): its line
leaves KNOWN_FAILING. The chamfers of the same edge still fail.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner ruled on 2026-10-30 that rounds meeting at a concave corner beside an
unpicked sharp edge are mitred, which fixes the basin loops' re-entrant corners.
Each loop case now grades its exact volume (rel 1e-6): each blend takes
Int width(z) length(z) dz, its section's width from its wall to its arc or chamfer
line (the front one clipped at the slab's face where it runs past the rim), its
length between its neighbours' walls extended; each corner of two picked edges adds
the overlap of their section prisms in the quadrant beyond both walls,
Int width_1(z) width_2(z) dz (r^3 (5/3 - pi/2) for vertical walls, as
every-pocket-rim-edges-r2). One edge alone reproduces the single-edge closed forms to
3e-7 of the removal. The equal-slope chamfer loop 0.5 passes; the others stay on
KNOWN_FAILING with the causes they had (SMLib status 1001, the surface-of-revolution
check, rounds past the rim not built by subtraction).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The 851 table rows grade continuity again, on each round's seams to the faces it
rolls on, picked by points at the middle of its edge (the top front round's at
z = 8 on the front and y = -19 on the top, the arc's at rho 23 and rho 21, the
uprights' at their contacts). The default `blend` selection read G0 there because
it pairs a blend face with the contact faces of every edge in the feature, so where
a round ends square on another edge's contact face (a crease by construction) it was
graded as a tangent seam: a gap in the grader's seam selection, worked around here
by naming the seams. All seven rows pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillets/1668-fillet-outer-top-edge-of-thin-wall-r7-past-inside-corner (and the
both-sides R7 case) now grade the volume at rel 1e-6 (0.04 mm^3), not the rig's
1e-3. The closed form is exact: V0 = 104 64 42 - 100 60 40 less the round's
section (1-pi/4) 49 along 104 plus its part over the hollow, the square
[25,30] x [35,40] outside the circle about (25,35), Int_sqrt24^5 (5 - sqrt(49-u^2)) du
= 0.0050679 mm^2 (a 0.0005 grid gives the same to 1e-9), along 100: 38458.896.
SMLib gives 38458.389, 0.507 mm^3 less: exactly that part over the hollow, which is
not material, taken anyway where the round opens its slit. The unit test's bound
was right; the case goes on KNOWN_FAILING with that cause. The both-sides case
passes at the tighter tolerance.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Unit tests assert the code's logic, the kernel rig asserts geometry through
commands with hand-derived answers; one example of each. The rest of the skill is
tightened to keep it under the skills check's 3600 characters.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The round trip's volume moves from the STEP unit test to an import rig case
(616.850, Pappus); the unit test keeps the writer's logic, the degenerate torus
written as one B-spline face and no toroidal surface or surface of revolution. The
case fails today where the revolve does (KNOWN_FAILING).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The horn torus quarter (a circle r5 touching the Z axis, revolved 90 deg) is a
sweeps rig case with its Pappus volume, pi 25 x 2 pi 5 / 4 = 616.850. SMLib's
revolve refuses the profile today (status 1001), so it is on KNOWN_FAILING.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A cylinder r10 x 30 with its top face deleted, the rim patched G1 and G2 and
stitched, becomes two continuity rig cases: one valid solid whose rim seam reaches
the order asked (the dome's shape is the fill's own, so no volume). They fail
today as the unit test did (the general fill refuses the cap: it folds) and are on
KNOWN_FAILING; the unit test and its helpers are removed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A 240 deg window through a cylinder's wall and a saddle window in a torus's inner
wall, their cut faces deleted, patched G0 and G1 and stitched, become surfaces rig
cases: the G1 cylinder seals to the whole cylinder's volume within 0.5 % with G1
seams on the window's edges, the G1 torus reaches G1 on its rim, the G0 fills are
one valid solid. They fail today as the unit tests did (the general fill refuses:
it folds) and are on KNOWN_FAILING; the unit tests are removed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The shell unit tests that checked geometry become shell rig cases built through
commands: the trough and the pocket whose floors fall 1 deg to a central drain
(the pyramid made as two triangular prisms intersected, in place of
convex_polyhedron; the cavity integrated exactly over the cells where each moved
facet is highest, sharp inner corners as the ladder's tier 09), the box opened at
a bottom rounded R2 and R2.4 (refused: the rim would have no thickness) and R1.5
(built, exact volume), and the plate with a hole through its arc side. Four
action-matrix shells that regressed (issue-616's lid, the ring-cut frame, the
filleted shelled box shelled again, the up-down join) get small cases too. The
ones that fail today are on KNOWN_FAILING with their cause, and the unit tests are
removed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The fillet and chamfer unit tests that checked geometry become fillets rig cases,
built through commands with answers derived by hand (each method field gives the
derivation): a plate outline and a disc rim as tall as or past their wall, a
notched block's outline at R7 and R8, a cube corner as wide as its faces (refused,
and reduced), the 1668 wall at R7 alone and both sides, the 2852 U shell at R30
(refused, naming 25.24), the 851 knife edge (R2, chamfers 0.5/2/5 on three
constructions), the arc top edge at the full wall height and at 9.5, the 851
table of selections, the 556 top edge ending on the boss wall, the 2893
cylinder-rim chamfer refusals and the elliptical rim, seed 33's full round, the
sloped basin rim blends (single edges with exact volumes, loops with the slab's
box), and three action-matrix chamfers and fillets. fillets/851's three-edge
corner gets its volume (67.051 removed, re-derived on a grid; the engine test had
67.05). Chamfers whose leg runs off the face beside the edge are refused and rounds
past a face on its tangent side build, as the owner ruled on 2026-10-30. The cases
that fail today are on KNOWN_FAILING with their cause.
The unit tests are removed, except two cut to what no command can express: the
per-edge-radius refusal at a cube corner and the chamfer of a cylinder's seam.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A colliding pair past the 16 pairs whose volumes are worked out, or past
the check's 256 body intersections, or whose intersection could not be
made, fell back to the sampled mesh depth for its deepest step. That
changed what deepest_step means without saying so. Such a collision now
answers deepest_step and volume_mm3 null and not_measured naming the
budget it ran past (or the intersection that failed), as compare does
since #9132; first_step, last_step and steps are exact and stay.
B pushed through A at 1000 samples overlaps it at about 330 samples, past
the 256 intersections: not measured, "the check's 256 body intersections
ran out". At the study's 16 samples it is measured, deepest at step 15.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
motion.interference picked a collision's deepest step by the depth sampled
on the contact meshes (vertices, edge midpoints, triangle centres), and
worked out the exact shared volume only there. That depth depends on where
the mesh puts its points: since planes mesh as their boundary's triangles,
the arm turning into a block (rig case
assemblies/cov-motion-interference-arm-into-the-block) sampled 4.64 mm at 60
degrees and 2.0 mm at 90, and the check answered step 6 with 536 mm3. The
exact depth is 5 mm at every step from 60 to 90 degrees (both slabs are
10 mm thick), so depth cannot tell those steps apart.
The exact shared volume is now worked out at each sample a pair overlaps,
and the deepest step is the first with the most (a tie keeps the earlier
sample). The arm and block share 536, 807, 952 and 1000 mm3 at 60 to 90
degrees, so it is step 9 with 1000 mm3, at 10 samples or 19. depth_mm is
reported as before and documented as an estimate from the contact meshes.
volume_mm3 is null, and the step falls back to the sampled depth, when the
volume was not worked out at every sample (more than 16 pairs, or past the
boolean budget, raised from 64 to 256 now that it covers every sample).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Where two curves of a face touch or cross by less than the chord tolerance, halving their
chords never parts them, and the face was left to the lenient fill: thin_face_tests::
a_face_touching_its_own_edge_meshes (#1648), an arch dipping 2.8e-7 mm below its face's
bottom line, came out 0.659 mm2 against its boundary's 0.567 at tol 0.3.
After the halving rounds, a face whose boundary still crosses itself has each node that lies
within the chord tolerance of a crossing chord of another edge (strictly between its ends) put
onto that edge as well, the same point (edges::snap; the node is marked as snapped). In a face
that runs both edges, the snapped node takes the other's parameters, so the boundary passes one
point of its (u, v) twice (a pinch) and the chords meet there instead of crossing; the edge's
other faces get the node too, within the tolerance of their pcurve, so the mesh stays
watertight. Where the curves run within the tolerance for a stretch, the nodes either side are
shared in turn and the boundary runs out and back along one polyline (a slit). A node is not
snapped where the pinch would turn part of its loop inside out by more than the tolerance (two
curves crossing at an angle: the face is refused, as before). Nothing is loosened: the
crossing test and the triangulation's boundary check are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A planar boolean leaves a cut-out's corner that touches another loop's corner as two vertices
a hair apart: in the sheet refold tests (doc sheet a_cut_out_on_one_side_of_a_bend_line... and
a_cut_out_with_a_corner_on_a_bend_line...), a triangle's corner 1e-8 to 5e-8 mm from the base's
corner on the bend line, both vertices of tolerance 1.05e-4. Their loops then cross by 1e-8 in
(u, v), which no constrained triangulation can keep, and the face was left to the lenient fill.
The boundary build already makes a vertex the face passes twice one point of its (u, v). Now a
vertex within the tolerance of another of the face's vertices (the larger of the two, at most
the chord tolerance) takes that one's parameters too, so the loops touch at one point, a pinch,
as at a shared vertex. Not for two vertices an edge of the face joins: the edge says they are
two points, however near (a_ledge_beside_an_edge_as_tall_as_the_setback_goes_with_the_strip
has a 1e-4 mm step, within the vertices' 1.18e-4 tolerance), nor at a pole.
Test changed, with the reason: thread_display_mesh_stays_on_its_surfaces checked every triangle
whose three corners lie at the core radius rc as one of the core cylinder, and wanted its centre
within 2 tol of rc. The flat end disc of the M10x1.5 thread (z = 12) has its corners a hair
apart too; joined, it is meshed strictly from its boundary (a plane, e39fbdebb), and one of its
triangles has its three corners on the rim, centre at r = 3.956, on the disc. The end-disc
exemption the test already makes for its radius check now covers this check as well.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
uv::divide started every curved face from ceil(range / (pi/2)) lines, as if the parameter were
an angle. A spline's is not: SMLib holds a 90 degree bend as one rational quadratic span with
v in [0, pi/2], and the far pcurve of its boundary gives v = -6.7e-13, so the range is a
rounding over a quarter, the count rounds up to 2 and the bend gets a line at 45 degrees
whatever the tolerance (display_mesh_tests::a_small_bend_on_a_large_sheet_is_meshed_round:
3 angles across the bend at the chord tolerance instead of 2).
The division now starts from the whole range, and the limits refuse an interval that spans
more than a quarter turn of its iso-lines, read from the points as an edge segment's is
(edges::QUARTER_TURN: the half chords of an arc turning theta differ by theta/2, a closed
iso-line's by pi). On a revolved surface the parameter is the angle, so the lines are the
same as before; an iso-line within a millionth of the tolerance of one point (a pole) has no
turn to read.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The comb tool passed only the pick, so a comb's length and tooth count
could not be changed. While the tool runs, a card at the top left of the
view has Comb length (0.25x to 8x the automatic length, default 1x) and
Teeth (4 to 200, default 40), with plain-word tooltips; moving either
runs the comb again on the curves and edges already shown. New strings
in assets/locales/pending/curvature-comb-options.json.
Test: comb_options_draw_the_combs_shown_again drags both sliders: the
shown comb's teeth grow to 8 times their length, then number 4; the
drags pick nothing, and the card goes with the tool.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
`inspect.curvature_comb` read the display mesh for model geometry: an
edge's comb fitted a circle through every three points of the nearest
mesh polyline, a face's did the same on its iso lines' chord points. The
comb was jagged, changed with the mesh tolerance and ignored `density`.
- geom: `CurveEval::jet`, a curve's point, unit tangent and curvature
vector kN from its first two derivatives (lines and conics in closed
form, B-splines and NURBS from the basis functions' derivatives, a
traced intersection from both surfaces' normal curvatures along its
tangent, by Meusnier: kN lies in the plane of the two normals and its
component along each is that surface's normal curvature).
- kernel (curvature_comb.rs): the edge nearest a pick by the exact
distance to its curve (a periodic face's seam and poles left out),
`density` teeth at even steps of the edge's parameter; along an iso
line, the surface's normal curvature n.S_tt / |S_t|^2 along its normal.
`IsoLine` carries the running surface parameter's span.
- engine: edges and faces use them (and `inspect.minimum_radius` on
edges); new optional `length` multiplies the automatic scale. The reply
keeps its shape.
Tests: a circle edge reads 1/R to 1e-9 at Coarse and Ultra display
quality alike, with `density` teeth; an extruded ellipse's edge matches
ab/(b^2 cos^2 t + a^2 sin^2 t)^(3/2) at every tooth, a/b^2 and b/a^2 at
the vertices; a cylinder face's rings read 1/R and its lines 0; `scale`
and `length` set the teeth's length; curve jets on every record kind.
The existing cylinder-edge check tightens from 0.5 mm to 1e-9.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
corpus_designs_rebuild_with_the_same_names_and_references compared every
corpus design with a text record of each body's face and edge counts and its
full name list, last written from truck's output (a55196e7a); 110 designs
differed only because SMLib keeps whole faces truck split in pieces. Those
counts and names are the kernel's, not an outcome.
The test, renamed corpus_designs_keep_their_references_and_name_alike_every_build,
now asserts invariants only, with no record:
- every design loads; one that does not is a failure naming the design and
the step (oracle/83-boundary-loop-thicken: surface.thicken, SMLib status
1001), not a record line;
- every name a design stores (fillet and chamfer edges, faces a feature
picks, a sketch's face, sheet metal edges) finds a face or edge: none
`gone`, no sketch face `lost`;
- two builds give the same name -> face map: each named face's area and
centre and each named edge's length and middle (to 1e-6 of their size), and
the same face or edge found for every stored name.
The design already exempt for unstable piece order
(regressions/closed_sweep_torus_cut) is still only printed for the second.
On develop it fails on two designs: oracle/83 above, and
regressions/cut_up_to_face_leaves_skin, whose Extrude13 face
`F47:top#[...]` finds nothing (the truck-era record had it `gone` too). All
159 designs name alike in both builds. The records in tests/naming/ are no
longer read; CONTRIBUTING no longer asks to re-record them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ordinary_curved_bodies_keep_a_practical_sampling_step asserted
ContactGeometry.step > 1.0, an internal sampling step derived from the
bodies' display meshes. What a user gets is the drive: it either stops at the
contact or is refused for running out of the sampling budget. The test,
renamed ordinary_curved_bodies_stop_at_their_contact_within_the_sampling_budget,
keeps that: the slide drives to -30 without a budget error and stops in
(-5.02, -4.9) (the sphere r 5 about x = 20 and the box from x = 15 meet the
block's x = 10 side after 5).
It still fails on develop with "motion exceeds contact sweep collision
budget": the planar faces meshed as grids, which the planar-tessellation fix
(#9307) addresses.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
open_in_easyprint_hands_the_host_a_3mf_or_an_stl_that_reads_back asserted the
STL was 84 + 50 * 12 bytes: twelve triangles, the tessellator's count for the
fixture's box (the fixture is a mesh import of box_solid's display mesh, 124
triangles today).
The STL is now parsed (solidcraft_io::read_stl): it has the imported mesh's
own triangle count, every side is shared by two triangles running opposite
ways (closed and consistently wound), it encloses the 10 x 20 x 30 box's
6000 mm^3 (divergence theorem, to 1e-6; the corners are exact in f32) and its
bounds are (0, 0, 0)..(10, 20, 30).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
free_form_profiles_extrude_exactly asked for an ellipse extrusion of exactly
3 faces and a spline-and-line extrusion of at most 5: how many pieces the
kernel cuts a closed side into. It now checks that the sides are exact.
Ellipse a = 10, b = 4, extruded 5: area = 2 pi a b + 5 P, P = 4a E(e),
e^2 = 1 - b^2/a^2. By the Gauss-Kummer series, P = pi (a + b) sum_k
(binom(1/2, k))^2 h^k, h = ((a - b)/(a + b))^2 = 9/49: P = 46.026225191330
(a 200000-point midpoint integral of sqrt(a^2 sin^2 + b^2 cos^2) agrees to
1e-13). A = 80 pi + 230.131126 = 481.458538 mm^2, asserted to 1e-4; a faceted
side would be short of it. Only the top and bottom are planes
(face_types.plane = 2, merged pieces counted once).
The fit spline closed by a line: the planes are its top, bottom and the line's
side (face_types.plane = 3); its volume stays checked against the profile
area.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_mirrored_or_patterned_sheet_body_is_sheet_metal_with_its_flat_pattern
(1171c4c2a) expected the mirrored copy's model turned over: the same outline,
its bend down and its pocket on top. That describes a SheetBody::moved that
never landed. The code on develop is SheetBody::copied (d9d6af51c, #3354): a
mirror moves the frame and mirrors the flat x -> -x with the same thickness
side, the opposite hand's blank with each bend turning the same way, which
a_mirrored_sheet_copy_is_the_opposite_hand_with_the_mirror_image_flat_pattern
asserts on an asymmetric bracket. The YZ mirror keeps z, so a pocket cut from
below stays below and the flange on y = 0 still turns up: the flat pattern
reads blank x 0..100 -> -100..0, pocket x 20..30 -> -30..-20, side bottom,
bend up.
The test now asserts that: the same flat size, the outline and the pocket
outline equal to the source's with x negated (to 1e-6), pocket side bottom and
bend up on both. The volume doubling, the bend table and the flange rebuilt
on the copy (point containment) stay.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_trays_faces_have_readable_sheet_names_that_follow_its_size asked for
exactly two unnamed faces (the hole's wall in two halves) and for the back
rim's named point at y = 199.5, the strip's centre line. The first is how many
faces the kernel makes a cylinder of; the second is where the face's roomiest
display triangle lies (naming::face_spot), any point on the face.
Now: there is an unnamed face and every one is the hole's wall (r 3.3), and
the back flange's far rim point lies on that rim: z = 30 and 199 <= y <= 200
(the flange 1 thick, its outside at y = 200), 0 <= x <= 400, to the 1e-3 the
test had. The right flange's top (the plane x = w - 1) keeps x = 319 and 399.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillet_the_knife_edge_at_the_tip (features_more_tests.rs) and
blend_a_knife_edge_where_a_curved_wall_runs_out (tests/robustness.rs) build
the same #851 part with the same script, round the same knife edge at
(62 1/6, 0, 5) with R2 and measure the removed volume through the same display
plan. The first integrated the removed area on a grid (5e-4); the second has
it in closed form. Drop the first and keep its tolerance on the second.
The ball 2 from the back face y = 0 and from the r 21 wall about (62 1/6, -21)
is centred at (62 1/6 - sqrt(168), -2). The removed section is the trapezoid
wall centre, tip, (kx, 0), ball centre (sqrt(168) wide, 21 and 2 high) less the
wall's sector theta = acos(19/23) and the ball's sector pi - theta:
10 (11.5 sqrt(168) - 220.5 theta - 2 (pi - theta)) = 119.648812 mm^3
the constant the kept test has; the dropped test's grid gives 119.6487 at a
0.004 x 0.001 step. The kept test's fillet bound goes from 0.5 to 5e-4 of
that (0.0598), the dropped test's. Its chamfer row (159.236451, the chord
triangle less the wall's segment) already says the chamfer takes more.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
cut_whose_end_touches_an_arc_side_leaves_no_sliver compared the touching cut's
face count with a clear cut's, a count of how the kernel splits its faces. It
now compares what the two parts are.
The plate's arc runs from (0, 20) through (-10, 10) to (0, 0): centre (0, 10),
r 10, so it reaches x = -10 only at y = 10. The r 5 cut along x about
(y, z) = (12.5, 12.5) spans x -10..7.5 (touching) or -10.1..7.4 (clear). Both
open the same patch of the arc; the slice x 7.4..7.5 lies wholly inside the
plate (y 7.5..17.5, z 7.5..17.5 in 0..20 x 0..27.5). So
V(clear) - V(touching) = pi 5^2 0.1 = 7.853982 mm^3
A(touching) - A(clear) = 2 pi 5 0.1 = 3.141593 mm^2 (the hole wall; the
floor disc is the same)
both to 1e-4 (the measure integrates the exact faces to 1e-8). A sliver of the
arc left where the end touches adds about a square millimetre of faces.
Seed 14's half (the fuzz generator's seed, then a shell) is pinned as its own
test, shell_after_a_cut_whose_end_touches_an_arc_side, with the commands the
generator gives today: the touching cut, then a 2 mm shell opening the top
face y = 20. It fails at the shell (SMLib status 1001), a kernel gap.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner's final ruling of 2026-10-30 reverses the swept corner 6539c070a
and 8dc344d22 used: rim rounds meeting at a concave corner beside an
unpicked sharp edge are mitred by default (a user who wants it swept rounds
the corner edge too; a swept corner with a sharp inner edge is a future
option). SMLib's swept result is the defect, so both tests now expect the
mitre and fail today on it.
rounds_that_fit_are_closed. At depth u the section is the outline moved in
by d = r - sqrt(r^2 - (r - u)^2) with every corner mitred: the area less
L d, plus d^2 at each of the six convex corners (the two strips overlap in
a d x d square) and less d^2 at each of the six concave ones (the mitre
also takes the d x d square beyond both strips), the overlap of the two
spandrel prisms being r^3 (5/3 - pi/2) per corner over the round. Six and
six cancel, so both ends remove 2 L r^2 (1 - pi/4) with L = 107.6:
11.545566 (the swept corners gave 11.514718).
fillet_tabbed_plate_outer_loop_and_partial_selection_after_hole. Where both
sides of a concave corner (+-4, sqrt(609)) move, the moved R25 arc and the
moved side x = 4 - d now meet where they cross,
y = sqrt((25 - d)^2 - (4 - d)^2), as where only one side moves; the swept
arc branch goes. Check: A(0,0) - A(d,d) agrees with the Steiner form for a
mitred outline, L d - d^2 (pi + theta - 2 tan(theta/2)), to second order
(17.190976 against 17.190968 at d = 0.1; the arc's curvature moves the
mitre point at third order, which the Green's-theorem area has exactly).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_strip_off_a_curved_face_is_ruled_to_its_far_curve lifted the free
strip's handles 1 mm and wanted the strip's bounding box to rise, which it
need not (the far curve's corners are fixed and the box's top may be
elsewhere). It now checks what the handles do: the plan's far curve takes
the lifted handles as its interior control points and keeps its corners,
so it rises by 1 - R0 - Rn inside; the built strip has that curve for an
edge (33 points of it within 1e-3 of the strip's edge polylines at 1e-4),
and its middle, risen more than 0.1, is more than half that from every
edge of the default strip.
It fails today on its premise: the strip's edge (the cup's straight rim at
x = 14) is a line, so its far curve, on the edge's degree and knots, has
no interior control points, and ruled_far_handles gives an empty list
(the old witness passed nothing to move). A real gap: the far curve of a
strip whose rulings turn along a straight edge is the chord between its
corners.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
rounded_slot_exports_cylinders required no B_SPLINE_SURFACE, four
CYLINDRICAL and two TOROIDAL surfaces; ramp_cut_slot_exports_sheets
_bounded_by_their_faces five CYLINDRICAL and two B_SPLINE (SMLib keeps one
more round an exact cylinder). Which kind a face is written as is the
writer's choice; what #612 and #1198 were about is that no surface reaches
past the part. Both keep that (every CARTESIAN_POINT's x within [0, 60],
control points and placements included) and their round-trip volume: the
first against Pappus, the plate (800 + 100 pi) 4 less (1 - pi/4) along
the straight edges and the arcs at 10 - r (10 - 3 pi) / (3 (4 - pi)), at
1e-3; the second against the exported body's at 1e-4. Both pass today.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three history tests pinned truck's sewing in tallies:
- shell_of_a_painted_body wanted changes [0, 2, 8, 0, 5], the side faces
split across the rim's level; SMLib's shell leaves them whole. It now
checks what paint follows: the painted face 0 comes through as exactly
one face (its geometry checked against face 0 by verified()), and each of
the five inner walls is one Generated Offset of the face it lines, the
open top none, as shell_lines_the_walls_with_offsets has it.
- fillet_after_box_cuts wanted the painted cuts to leave faces in pieces
and the fillet's edges to meet merged faces; it keeps the outcome for
every cut on both blocks: the round stands in place of the picked edge,
and composed with the cut the history holds over the block and the tool.
merged_meetings has no use left and goes.
- surfaces_keep_their_faces wanted every stitched face Kept; whether
sewing changes a face's class is not the contract. It now checks each
stitched face has one origin, the face it was sewn from (geometry checked
by checked()), the five walls the open box's faces 0-4 and the lid the
patch, and the four seams naming two edges each.
split_a_face is left as it is (the owner's ruling: a face whose boundary
only gained a vertex is Kept), failing until SolidCraft's split history
says so ([3, 2, 2, 0, 0] against [5, 0, 2, 0, 0]). The three tests above
pass today.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
rounded_shell_planar_roles_are_exact_distinct_and_transformable required
the four outer r6 rounds to be coalesced extrude sides (face_was_coalesced:
SMLib's round is not one any more), no floor-height seam between corner
faces, 19 faces in the cup and in its STEP copy, and the volume of a 1e-4
mesh within 0.2. It keeps what the roles are for: the right wall's role
resolves to the face at x = w/2 and the floor's to the face at z = t, both
still after the transform; each straight outer wall is one full-height
face, in the cup and in its STEP copy; and the cup's volume is exact.
Proof. The box w x 60 x 30 with its four vertical edges rounded r6 is
(60 w - (4 - pi) 36) 30; shelled t inward, open at the top, its inside is
the box moved in by t with its corner rounds at 6 - t:
((w - 2t)(60 - 2t) - (4 - pi)(6 - t)^2)(30 - t). Graded on SMLib's
volume (integrated on the surfaces, to 1e-8) at 1e-6. It passes today.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_small_bend_on_a_large_sheet_is_meshed_round also pinned the chord-only
mesh (exactly two stations across each bend, its sag over 0.29 r) and
required every flat side face to keep the chord-only mesh's triangle count:
facts about another mesh, not what the display shows (the chord-only mesh
now puts three stations across the r1 bend). It keeps what the display
mesh must do: each bend a quarter turn in more than five facets, each
turning at most DISPLAY_ANGLE, sagging at most r (1 - cos(angle / 2)) and
less than the chord tolerance. It passes today.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_five_sided_window_fills_g1_and_through_a_point required five faces and at
least 15 edges (five patches about a centre); a_hexagonal_rim_fills_through
_a_point... required (6, 18). How many patches a fill is made of is the
fill's own business: both keep their outcome checks (check_face: the rim's
own edges, within the face tolerance, a clean mesh; G1 to the wall within
0.1 degree; the point on the surface; the sealed cup holding more than the
plain fill's).
Both still fail on outcomes: the pentagon's fill through a point has a
folded triangle in its mesh, and SMLib's stitch refuses the hexagon's fill
through a point (its control net doubles back on itself, a cusp in a
curve).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_crowned_patch_fills_its_middle_and_keeps_its_edges checked only that NaN,
infinity, 3.01, -0.91 and 1e300 are refused; it now checks why: "a patch's
crown must be between -0.9 and 3" (patch_edges_crowned's range test).
The bump check is left as it is. The audit proposed measuring it in model
space along the loop plane's normal, but it already is: both patches are
height fields over the same plane on the same parameters, so the distance
between their points at the domain's centre is the bump along the normal,
and the record's domain is the fit's box in mm (diagonal 52.5). The crowned
patch's middle rises 7.656 where the documented crown x size is 13.125:
the crown's bump reaches 7/12 of what patch_edges_crowned promises on this
pentagonal cup (geom's own test of fill_loop, on a fine hexagon net, gets
it to 1e-9), a real gap, still failing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
collapsed_taper_mesh_stays_accurate_as_tolerance_decreases compared each
display mesh's volume with the taper law's integral,
6000 - (1 - pi/4) 40 integral r^2, at 1e-4: the built fillet removes about
10 mm3 more than that (5212.9 against 5223.1) at every tolerance, so it
measured the fillet, not the mesher. It now holds each mesh to the B-rep it
meshes: a mesh within its chord tolerance of the faces has its volume
within area x tolerance of the exact one (SMLib's, integrated on the
surfaces to 1e-8), the bound fillet_mixed_arc_and_line_removed_volume
_converges uses. It passes today.
taper_follows_the_room_along_the_edge keeps its law and closed-form
volumes (the 10 mm3 against the law is the failure it still shows) and
drops the two internal checks at its end (no unmeshed face, a face with
three vertices) for the outcome: the wedge's tapered fillet is valid.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
the_arc_rounded_after_the_back_edge_matches_both_together ran
one_then_the_other(BACK, arc_top()) as a test of its own. It is the same
check as every pair of a_round_ending_in_an_earlier_round_matches_both
_together (the round in sequence against both together, within the
display meshes' 5e-3), which had only the other order (arc_top(), BACK):
the pair joins that list and the separate test goes. Nothing is loosened.
Today the list fails before it reaches the pair: SMLib fails the arc's
round after the long front edge's (status 1001); the back-then-arc pair
failed the same way on its own.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
rounds_that_fit_are_closed counted faces (the input's plus one per picked
edge: 38, where SMLib makes 50 with its corner faces). It now keeps
validity and no crossed loops and holds the R0.5 rounds on all 24 end-face
edges to the exact volume.
Proof. The part (plate 14 x 21.5 x 10, block 14 x 9.948 x 10 on it, the
5.2 x 7.1 window through both along x) is
14 (215 + 10 (b - a) - 36.92). At depth u into an end face's rounds the
section is that face's outline moved in by d = r - sqrt(r^2 - (r - u)^2),
each edge parallel to itself: the area less L d, L = 2 (21.5 + 20) +
2 (5.2 + 7.1) = 107.6 (the stepped outline's perimeter is its box's), plus
d^2 at each of the six convex corners (two rounds mitred), less pi d^2 / 4
at each of the six concave ones (the two shoulders and the window's four
corners: rounds meeting at a concave corner beside an unpicked sharp edge
are swept round it, the owner's ruling of 2026-10-30). The wall between
window and block side is 1.256 > 2 r, so the offsets stay apart. Over the
round the integrals of d and d^2 are r^2 (1 - pi/4) and r^3 (5/3 - pi/2),
so both ends remove
2 (L r^2 (1 - pi/4) - (6 - 6 pi/4) r^3 (5/3 - pi/2)) = 11.514718.
Graded at 1e-7 of the part's volume. It passes today (aarch64, SMLib).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
issue2422_vertical_after_separate_bottom_rounds compared the face count and
volume (within 0.05) of the vertical round, picked with and without its
bottom mitre, against the same round picked alone, and required every face
meshed. It now holds each of the three picks, on the 40 x 80 x 27 box and
on the same shelled 1 mm open at the top, to the exact volume; the corner
probes stay.
Proof. Horizontal sections, s up from the bottom. Below s = r the four
separate R2 bottom rounds (mitred at the corners) leave the rectangle
moved in by d = r - rho, rho = sqrt(r^2 - (r - s)^2), and where the
vertical round meets them (a sphere octant) its corner is a quarter disc of
radius rho: (L - 2d)(W - 2d) - (1 - pi/4) rho^2. Above, LW - (1 - pi/4) r^2.
With the integrals over the bottom r of rho (pi r^2 / 4), rho^2 (2 r^3 / 3),
d (r^2 (1 - pi/4)) and d^2 (r^3 (5/3 - pi/2)):
V = LW r - 2 (L + W) r^2 (1 - pi/4) + 4 r^3 (5/3 - pi/2)
- (1 - pi/4) 2 r^3 / 3 + (H - r)(LW - (1 - pi/4) r^2) = 86174.445361;
with four such corners it is issue1557_all_vertical_corners_match_analytic
_volume's formula exactly. The shell's cavity, 38 x 78 x 26, is clear of
both rounds (along the corner's diagonal the R2 arc stays 2 - sqrt(2) in,
the cavity's corner is 1 in), so the cup is V - 77064. Graded at 1e-6.
Today SMLib's check refuses the bottom rounds' result (edge and vertex
tolerance 0.97 against 0.01), as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillet_tabbed_plate_outer_loop_and_partial_selection_after_hole only
checked that every subset of the R25 tabbed plate's seven top-outline edges
builds and stays inside the plate. It now first holds the hole's rim round
and the three selections that end only at the plate's sharp concave
corners (the R25 arc's two halves, the tab's tangent run of sides, R2
corners and top, and both) to their exact removed volume; the sweep over
all 127 subsets stays as it was.
Proof. The outline's area A(da, dt), the arc moved in by da and the tab's
run by dt, each element parallel to itself, by Green's theorem: an arc
about c of radius rho from a to b gives
1/2 (rho^2 (b - a) + rho (cx (sin b - sin a) - cy (cos b - cos a))), a
segment p -> q gives 1/2 p x q. At a concave corner (+-4, sqrt(609)) where
only one side moves, the elements meet where they cross (the round runs
into the other wall: the section is the union of the two sides' half
regions); where both move the corner is swept, an arc of radius d about the
corner, traversed clockwise from the arc's normal to the side's (the
owner's ruling of 2026-10-30 for rounds meeting at a concave corner beside
an unpicked sharp edge: the profile swept round the corner edge). Checks:
A(0, 0) = 312.5 (pi + 2 theta) + 4 y + 8 (cy - y) + 8 + 2 pi (disc sector,
triangle, tab, R2 corners), and A(0, 0) - A(d, d) = L d - pi d^2 by the
Steiner formula for the swept offset (total smooth turning 2 pi + 2 theta,
the two swept corners -theta each), agreeing to 1e-15 at d = 0.1, 0.3,
0.5; A(0, 0) - A(0, d) agrees with a 0.004 mm grid to 2e-6. The round's
section at height r sin(phi) into it is the outline moved in by
d = r (1 - cos phi), so it removes the integral of
(A(0, 0) - A(d)) r cos(phi) dphi over [0, pi/2] (midpoint rule, 8192
steps, converged to 1e-7 against 4096). The hole's rim round is Pappus:
(1 - pi/4) r^2 about the axis at 1.5 + r (10 - 3 pi) / (3 (4 - pi)).
Graded at 1e-5 of the removal, the bound the D prism's rounds are held to
(the blend surfaces are SMLib's fits).
Today the three selections build with these volumes (aarch64, SMLib); the
sweep still fails where a subset ends a round at a tangent joint (the x = 4
side without its R2 corner): SMLib cannot end a blend inside a tangent
chain, "the blend cannot end there".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner's blend-overflow ruling of 2026-10-30: a chamfer whose leg runs
off the face beside its edge over another face is refused as too large,
naming the largest size that fits (a 2 mm chamfer on the basin's 1 mm rim
would lower the slab's front face from 30 to 29.382); a full-width chamfer
stays refused. Two tests wanted those chamfers built and cut from the slab:
- chamfer_wider_than_sloped_basin_rim_is_cut_and_a_full_width_one_refused
becomes chamfer_wider_than_sloped_basin_rim_is_refused_naming_the_largest:
(margin, d) = (1, 2), (1, 3), (0.5, 4) and the full widths (0.5, 0.5),
(1, 1) are TooLarge, the body unchanged, and the message names a largest
size just below the rim's width: the full width is refused, so the fit
search (three significant digits of the size asked, step
10^(floor(log10 d) - 3)) ends one step below it, in
[margin - step, margin). The removal formula for the cut chamfer
(basin_rim_chamfer_removed) has no use left and goes.
- equal_slope_rim_loops_stay_inside_the_material: the 2 mm chamfer of the
1 mm rim's loop is refused the same way (largest in [0.999, 1)); the R4
round (tangent side, builds) and the 0.5 blends stay.
The half-width chamfer on each rim, which only had to take something off,
now takes its exact volume. As in chamfer_sloped_basin_rim: the triangle
with legs d on the top and d down the wall towards (20, 10) (run
20 - margin over the 20 drop) has area d * drop / 2 with
drop = 20 d / sqrt((20 - margin)^2 + 400), and the chamfer's length at
depth h is 60 - h between the end walls x = 20 + (30 - z) / 2 (planar:
x depends on z only), so it removes d * drop / 2 * (60 - drop / 3);
graded at 1e-8 as that test is.
Today SMLib refuses the chamfers as too large, but the message does not
name the largest that fits yet ("the chamfer is larger than the faces
beside its edges allow"), the SolidCraft change issue_1080's R30 test
waits for too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner's blend-overflow ruling of 2026-10-30: a round that runs past
its face on its tangent side is built on the face continued and cut from
the part, keeping its radius. loop_fillet_as_tall_as_its_wall_consumes_it
still wanted R2.001 and R2.5 on the R13 x 2 disc refused as taller than the
wall; they now must build, valid, with the exact volume.
Proof. The round's section in (rho, z) is the r square at the top edge
outside the circle about (13 - r, 2 - r), clipped to z >= 0. With
t = z - (2 - r) = r sin(theta), the round's inner radius is
rho(theta) = 13 - r + r cos(theta), theta from asin((r - 2) / r) to pi/2,
and the disc keeps pi (169 * 2 - integral (169 - rho^2) dz), dz =
r cos(theta) dtheta (midpoint rule, 20000 steps). At r = 2 it is Pappus,
2 pi 169 - (1 - pi/4) 4 * 2 pi (13 - c), c = 2 (10 - 3 pi) / (3 (4 - pi))
(checked to 1e-9 in the test); at r = 2.5 it is 957.692239 mm3, the
ruling's number, the same integral as
rounds_past_the_bottom_of_a_wall_are_cut_from_the_part's. Graded at 1e-6
as that test is.
Today SMLib fails the R2.001 round (status 1001): a convex round past its
face is not yet built by subtraction, the cause fillets/KNOWN_FAILING gives
4782 and 5158.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_round_into_a_hole_is_cut_by_it (issue_324_tests.rs) built rig case
fillets/324-fillet-over-hole-overlap's body, but cut its hole with a cylinder
longer than the block (z -1 to 11) where the rig extrudes a sketch circle
flush. The new case fillets/324-fillet-over-hole-cut-by-overlong-cylinder
builds it that way (solid.cylinder, base z -1, height 12, cut) and holds it to
the same hand-derived answer: block 40x20x10 less the r3 hole at (20, 14.5)
(282.74) less the R4 round on the top edge at y = 20, 40 (16 - 4 pi) = 137.35,
plus what both remove, the integral over y in [16, 17.5] of
(4 - sqrt(16 - (y - 16)^2)) * 2 sqrt(9 - (y - 14.5)^2) = 0.375830 (re-derived
with the substitution y = 17.5 - s^2, 200000 steps):
V = 8000 - 90 pi - 40 (16 - 4 pi) + 0.375830 = 7580.287315, and G1 on the
blend. It passes today (aarch64, SMLib), as the flush case does, so it needs no
KNOWN_FAILING line; the unit test, a duplicate of the two cases, is dropped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillet_mixed_arc_and_line_joint_preserves_both_section_radii put R0.5 on the
D prism's top outline and checked 258 points of each mitre edge against the
cylinder and torus to 1e-7 in r squared, through the record's edges. The rig's
fillets/485-fillet-d-prism-top-outline-arc-and-flat builds the same prism with
the same radius and grades the removed volume to 5e-5 (a 0.12 mm3 loss shows)
and G1 on every blend seam, which is what the mitre's accuracy is for.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A box edge chamfered 2 and by two distances (3 and 1, and flipped, told apart by the
centre of mass); a pocket's rim chamfered 1, by two distances, 4.9 (fits), and 5 and
5.5 (refused by the owner's rule on full-width chamfers: built at 5 today).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A box edge rounded, then the edge meeting it at the corner, in both orders (both the
three-edge rolling-ball corner); r3 then r2 (the r2 round's torus about the r3 round's
axis, by Pappus); an L-bracket's front inner outline after its bend round; a block's
top edge after the arc round at the cylinder (fails today: G0 at the round's end on the
crease); a curvature round after a curvature round with Tangent chain off.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Inside bend r3, outside r5, both concentric (r3 and r8), the inside r35 using up both
inside faces, r36 refused naming 35, and the inside bend Type: Curvature.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Top and both rims rounded r1 and chamfered 1; the top rim rounded r10 (as deep as the
plate), r12 (past it on its tangent side: built by subtraction, fails today with
status 1001) and Type: Curvature; chamfered 9.9 (fits) and 10 (the full depth:
refused by the owner's rule, built today). Volumes by Pappus.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A rectangular pocket's floor edges, rim edges and both rounded r2, the rim r5 as deep
as the pocket; a boss's root r2 and r10 (as tall as the boss) and its top rim r2.
The rim cases fail today: SMLib puts a corner patch where two rim rounds meet at a
concave pocket corner, where the rounds running into each other make a mitre (four
faces more; the volumes agree to 5e-4).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A block whose sides meet the cylinder at creases: the concave arc at the wall above
the block's top r2, the cylinder's top rim r2, the block's long top edge r2 with
Tangent chain off and Type: Curvature. A block as wide as the cylinder, its sides
running tangent into the wall: the arc and the long top edge each rounded alone
(built or refused as continuing tangentially). Volumes by Pappus or one-dimensional
quadratures of the exact sections, the round ended by the planes it runs into.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three edges at one box corner rounded r1 and r3 with the rolling-ball corner, r3 with
Corners: Curvature at its default setback and at 3.5, and refused at 3 (the radius);
the corner chamfered 2; all twelve edges rounded r2; one edge rounded Type: Curvature.
Volumes in closed form (strips plus sphere octants, half-space intersection, the
rounded box as a Minkowski sum, the quintic section's area 113/756 s^2); G1 at every
round's seam, G2 for the curvature round and the corner patches.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Stale since 84bd2daad (the Fillet feature rounds each pick's tangent chain). After a and b are
rounded at 2 mm and meet in a mitre, the mitre's seam leaves c's end along c, so c's chain is c
and the seam; rounded at 2 mm, as the issue1557 tests pick, the corner is a sphere octant.
Proof: the 20 mm cube less its three rounds, each (1 − π/4) r² over its 18 mm straight part, and
the corner's 2 mm cube less the octant, r³ (1 − π/6): 8000 − 3·(1 − π/4)·4·18 − 8·(1 − π/6) =
7949.834793498635. Measured 7949.834793513291 (1.8e-12 relative), 10 faces, valid; the check is
tightened from 0.05 mm³ to 1e-11 relative.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Stale since 84bd2daad, which gave solid.fillet each pick's tangent chain by default. After the
long top front edge is rounded, the front-left upright meets the round's end (the quarter circle
on the left face) along its tangent, and that circle meets the top-left edge along its tangent:
the second fillet rounds the upright and the top-left edge. The "together" fillet picks the
three edges ([20, -21, 10], [0, -21, 5], [0, -10, 10]); one by one and together remove the same
volume within the display meshes' 2e-4 (it read 60.649 vs 43.185 with two picks).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tangency-labs/smlib#64 builds a tapered wall that closes partway up (its neighbours meet above
its apex), so the 0.2 mm slot of a 10 mm square tapered 10° out over 2 mm now builds where it
was refused. The test, renamed for what it checks, asserts the slot's solid is valid with its
exact volume and keeps the ring's refusal (its 0.3 mm hole closes entirely).
Proof of the volume, t = tan 10°: every wall moves out of the material by z·t at height z. The
square grows to 10 + 2zt, so without the slot ∫₀² (10 + 2zt)² dz = ((10 + 4t)³ − 1000)/(6t).
The slot narrows to 0.2 − 2zt and keeps its length 8 (its end wall and the square's side both
move up by zt); it closes at z = 0.1/t = 0.567 and takes ∫ 8 (0.2 − 2zt) dz over 0..0.1/t =
0.08/t. Volume ((10 + 4t)³ − 1000)/(6t) − 0.08/t = 213.984095421783; measured to 1e-6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tangency-labs/smlib#63 raised SMLIB_CURVE_MAX_CURVES from 4096 to 16384; CURVE_MAX_CURVES
follows it, so a profile of 5000 segments (a modelled thread, a fine polygon) extrudes:
kernel tests::measure_of_a_body_with_many_faces passes (5002 faces, 15000 edges, 10000
vertices).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- #63: a knife edge (two faces tangent along it, folding back) rounds to the ball touching both
faces back from it; a curve list takes up to 16384 curves (SMLIB_CURVE_MAX_CURVES); a stitched
solid is a nested solid, its cavities void; booleans pair each face only with the faces whose
box comes near its own, the closure test of a new face searches from both its sides, and a new
edge from a faceuse shell to a wire keeps the faceuse shell: a row of Menger sponges joins in
near linear work.
- #64: a tapered wall that closes partway up ends at its apex, its neighbours meeting above (it
was refused as SMLIB_ERR_WALL_CLOSES); a round of revolution offset in past its radius has no
offset; a round along an arc turns from its start section, not from the arc's frame.
- #65: capi tests, the Menger row join counts its work (smlib_work_counters) instead of timing it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tapered_extrude_swallows_a_short_tip expected the prismatoid whose two tip
corners run straight to the top's tip (h/6 (A0 + 4 A_mid + A1)), the solid
the old offset-and-rule construction built: its tip triangle leans about 40
degrees, not the 5 degree taper. With every wall drafted 5 degrees the tip's
wall is a triangle that closes where the long sides meet across it, at
z* = 0.126237833 (offset 0.011044379), and above that the long sides meet
along a ridge. The section at height z is the profile's lines moved in by
z tan(a): a quadrilateral below z*, a triangle above, each area quadratic in
z, so Simpson's rule on each piece is exact: V = 72.883284979193. The old
value, 72.882657253174, is 8.6e-6 off, past the test's 1e-6. SMLib's
DRAFT_CORNER_ROUND path builds exactly this solid (72.883284979193, 6 faces);
the default curvature rule still refuses it (SMLIB_ERR_WALL_CLOSES at
(20, -0.01, 0)), so the test fails until the fork builds a wall that closes
partway up.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
shell_thicker_than_a_rim_is_refused expected the "rim or ledge" refusal from 9508ee749,
added because truck's face offset turned the top face inside out and returned a body of
about twice the volume. SMLib builds the exact sharp-edged offset, and its volumes are the
closed form, so the test now checks those (renamed
shell_of_a_basin_past_its_rim_keeps_the_rim_solid).
Proof. Slab [0,100]×[0,60]×[0,30]; pocket lofted from x 20..80, y 4..50 at z 30 to
x 30..70, y 20..40 at z 10, its sides the planes x + z/2 = 35, x − z/2 = 65,
y + 0.8z = 28, y − z/2 = 35 (normal lengths √5/2, √5/2, √1.64, √5/2). Shelled open at
z = 0 by t, the cavity is the slab shrunk by t (x t..100−t, y t..60−t, z 0..30−t) less the
pocket grown by t (each plane moved out by t·|n|, floor z = 10 − t). For t ≤ 6 the grown
pocket stays inside the shrunk slab in x and at the back (50 + 0.618t ≤ 60 − t); its front
reaches y = t at z0 = (28 − (1 + √1.64)t)/0.8, below 30 − t from t ≈ 2.70 (the rim stays
solid above z0). With W = 30 + z + √5 t,
grown(t) = ∫[10−t, z0] W (7 + 1.3z + (√5/2 + √1.64)t) dz + ∫[z0, 30−t] W (35 + z/2 + (√5/2 − 1)t) dz,
V(t) = 180000 − grown(0) − (100 − 2t)(60 − 2t)(30 − t) + grown(t),
grown(0) = 33866.667, so V(1) = 17105.6240415, V(4) = 63560.2492284,
V(6) = 90113.9202101. SMLib measures 17105.6242, 63560.2492, 90113.9203 (≤ 1e-8
relative). The test computes the closed form and checks each shell to 1e-6, with the
centroid inside its box and on x = 50 as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The test offset the sphere's top patch to get a fitted B-spline sheet. That held on truck,
whose sphere came in several patches and whose sphere offset was fitted. On SMLib the
sphere is one face: offsetting it closes up, so it is a solid by design (surface_offset.rs,
"Faces that close up make a solid", and its test a_sphere_is_fitted_and_refused_past_its_centre
asserts exactly that), and SMLib's offset is exact, a sphere of radius 22 with no control
points, which surface.fair refuses. Body::is_surface was right: the record lists the offset's
closed shell as the outer shell of a material region (Parts::canonical), so it is a solid
whether read from edge uses or from regions.
The test now fairs the patch on the non-planar rim of a box cut by a sideways cylinder (the
cover in patch_trim_thicken_and_stitch), a one-face B-spline sheet: energy 0.7185 → 0.6938 in one
step, edges unmoved bit for bit, undo/redo/reopen and the refusals as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
hostile_params_never_panic built a box, a sketch and a rectangle afresh for each of the ~3000
calls it makes per command. SMLib serialises every call behind one process-wide lock, so those
boxes were built one after another whatever the number of threads: 20 threads made 4000 boxes
in 7.8 s, one thread 200 in 0.31 s (1.5 ms each). The fixture was nearly all of the test's time
(8814 of 8823 thread-seconds over the first 72 commands on the aarch64 box; the commands
themselves 9 s, none over 0.5 s), and the test took 2195 s in develop CI.
It now builds that model once per command and gives each call a copy of the session with its
design, model, undo, redo, open sketch and log, sharing the bodies as undo does. Every command
still gets every parameter name with every hostile value, from the same starting state. A new
check runs Cancel Sketch, two undos and a redo on a copy and on a session built afresh and
finds the same designs, with the copied session untouched.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
encode_native_body turned the whole saved record into a serde_json
Value (a map with an owned key for every field of every face, edge,
curve and pcurve), checked it, counted it, then wrote it out. Building
that tree was most of the cost: on the Menger level 4 sponge (3 tools,
--save) the 101 bodies the model cache encodes, 535 MB of encodings,
took 24.1 s of encode_native_body, 20.8 s of it in to_value.
binary::write_serialized is a serde Serializer for the tagged binary
format: it writes each value as it is serialized and applies the same
checks as it goes (nesting, string and array budgets, floats within
1e12, the value budget, the size limit). An object's entries are put in
key order when it closes, as the JSON map held them; floats that are not
finite are null, as to_value made them. The bytes are the same: a test
encodes plain, blended, cut, extruded, painted, split, open, round and
far-placed bodies both ways and compares them, and all 101 Menger bodies
(6 to 28 412 faces, 15 kB to 68 MB) encoded both ways in one run were
byte for byte the same. The JSON-value writer stays for the tests that
build hostile records.
Menger L4 3-tool --save, two runs each: encode_native_body 24.1 / 25.3 s
before, 3.0 / 3.0 s after (the longest call 1.34 s -> 0.29 s); the run
41.2 / 41.4 s wall before, 39.1 / 39.2 s after (the encodings run on
16 threads).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A To Object extrude to a curved face runs its tool through the target and
cuts the target out of it. The SMLib boolean carries no extrusion face
sources (they stood for the untrimmed profile), so sheet_tool_span, which
read the tool's depth and footprint only off its extrusion caps, gave up
and the intersect was refused as unrepresentable.
The tool's current geometry says what it is: the kernel's prism
certificate (prism_measure, the one component copies are compared by)
reads off the exact record whether a body is a uniform prism along a
direction, its two cap levels and its section of lines and arcs. It is
now public as certified_prism_section, and sheet_tool_span uses it when
the caps don't certify the tool: the levels are the depth, the section the
footprint the To Object cut is bounded by. A tool trimmed to something
that is not a prism (a curved roof over it) is still refused.
partial_depth_intersect_curved_to_object_preserves_current_trimmed_cap_footprint
passes: the tool is the square 20 ... 30 less the pin's circle (radius 2),
z 0 ... 1, 100 - 4π, as the test's own witness extrude measures it. A
kernel test reads that section off the trimmed lump.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The refusal "capture the source occurrence position before generating a
legacy face profile across components" lost its test when Offset Face got
an exact history name (b6c3bf3c5). A half of a box's top that Split Face
divided still has neither a construction certificate nor an exact history
name, so Extrude from that face, in a component whose occurrence has an
uncaptured move, takes the legacy path and is refused there; the document
and the pending moves stay as they were.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
end_at_wall takes the plate's end strip off a wall that stands on part of
an edge of the fixed face: the strip's two corners and its two edges go.
They were filtered at their places in the face's lists, counted from the
bend's edge, which pairs corner j with edge j only while the places do
not run across the lists' start. SMLib merges the plate's end strip and
both walls at x = 100 into one L-shaped face whose loop starts on the
strip, so the corners left were each followed by the next one's edge:
the bend's edge took the wall end's face and the wall end took the
plate's. The bend then ran the wall's full width and no relief cut it.
Both lists are now kept counted from the bend's edge.
a_wall_on_part_of_an_edge_meets_a_wall_at_a_relieved_corner_and_is_relieved_at_its_end
passes: bends [22, 92] (was [26, 96]). From the geometry: T 2, R 2, the
setback R + T = 4 on the plate from y = 0 and from x = 100; the round
corner relief 8 across is centred on the setbacks' corner (96, 4), so it
takes 4 off the end of each bend there: the long bend from x = 0 to
96 - 4 = 92, the short one from y = 4 + 4 = 8 to 30, 22 long.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The kernel switch broke these tests only through their counts, which record how a kernel splits
faces rather than whether the result is right. Each now checks its intent:
- a fitted circle is the exact circle (centre, radius, a full turn), not "one segment";
- the tight bend sweeps a valid solid of Pappus' volume, πr²(40 + 5π/2), not a count of faces,
edges and vertices;
- the Pi case keeps each outer side one face, stays valid after every feature and rebuilds the
same topology at 32 whether built there or edited there, without a hard-coded table of totals;
- a certified paint stays on the wall's radius and covers as many faces after an edit, a copy
and reopening as when it was painted;
- the offset sleeve and the extended strip are checked by their exact areas only.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
saved_curved_strip_projection_is_lost_after_exact_wall_coalescence built its "old" strip by
cutting the cavity raw (paint turning off the construction shortcut), where truck sewed the cut
slab by slab and split the outer round at the cavity's floor. SMLib cuts without slabs: the raw
cut's round is one face, so there was no piece to save a projection on, and the test stopped at
"must have a piece suffix". The link is now what such a build saved, written out: the face name
`F2:blend:F1:+x|F1:+y#1` (the lower piece, numbered up its spread) and the piece's projection on
YZ, the rectangle y 24..30, z 0..2 (the quarter round from (40, 24) to (34, 30), its arcs and
straight sides), 12 mm³ when extruded 1. The current wall is checked to be one face with that
stem and no planar authority.
The test still fails: the link is found on the whole wall ("was split; using the nearest
piece"), not lost, and the strip's material grows to 6 × 30 × 1 = 180 mm³. project.rs takes a
missing piece as removed only on a certified planar or coalesced face, which truck's shell
made and SMLib's does not. That is a persistent-name decision for the owner (see the PR).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillets_on_tapered_bosses_saved_before_still_find_their_edges compared the part with
69454.885209 mm³, what the truck build that saved it measured. In closed form: the 80 × 40 × 20
block (64000); the 20 × 20 boss 10 high tapered 5° in, a frustum to a 20 − 20 tan 5° square
(3660.251034); the Ø16 boss tapered alike, a cone frustum to radius 8 − 10 tan 5° (1798.751833);
less the R1 round on a sloping corner, the corner's section r²(cot φ/2 − (π − φ)/2) at its
dihedral angle φ = 90.435° times its length between z = 20 and 30 (2.124401), and the R0.5
round all round the cone's top, the section at the rim's 95° turned about the axis by Pappus
(1.928463): 69454.950004. SMLib's part measures 69454.950004.
The test still fails before that check: Fillet2's saved edge `F4:side:c1#1|F4:top#[F4:side:c1]`
names a half of the cone wall truck built, which SMLib builds whole, so it is found with a
"was split" warning (a persistent-name question, see the PR).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since 0a216d692 tangent and curvature extensions are SMLib's G1 and G2 extension of the surface
(smlib_extend_sheet, CreateExtendedSurface), and the face stays one face on it, as
surface_extend's docs say; extend_a_cylinder_strip_every_way still counted truck's separate
run-off face (2). It counts 1.
The test still fails after this, on the curvature run-off's area (see the PR): SMLib's G2
extension of the quarter cylinder by 5 gives 413.882 (a section 4.986 long), where the test
works out the parabola's 418.182.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
cut_keeps_posts_joined_while_apart_in_one_body expected the reply's `bodies` as
["Case", "Case (1)"]. That list is the state's order, which is SMLib's lump order (the -x half
first); the names follow the pieces' faces (lineage::order_lumps, 08cf526fe): the piece whose
least face name sorts first keeps "Case". That is the +x half, whose least is `F1:+x`, before the
-x half's least, its piece of `F1:+y`. The test now checks the names as a set, that the +x half
(from x = 1) is "Case" and the -x half (from x = -30) "Case (1)", and each half's volume as
before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since 09c979a4b a circle is one closed arc, so an extruded circle's wall, a hole's wall and a
cylinder's side are one face each, closed along a seam. Tests that counted truck's two halves:
- certified_annulus_side_rows_follow_radius_edits_and_clone and
certified_side_transfer_preserves_geometry_and_unsupported_taper_refuses_atomically: the
painted outer wall is 1 face (every vertex of it still at the painted radius), not 2;
- offset_surface_of_a_cylinder_and_a_box_top: the offset sleeve is 1 face (its area 2π·12·20
unchanged);
- press_pull_then_curved_face_offsets_it: a click on the hole's wall offsets its 1 face;
- pi_case_faces_per_step_at_both_heights: each standoff (a ring, Ø6 round a Ø2.5 hole) adds 4
faces, not 6: its outer and inner wall, its top, the floor inside its hole. So 23 after the
first and 35 after the pattern of four; the port openings (3 × 4), the end cut-out (4) and the
five vents (5 × 4) are planar and add what they did: 47, 51, 55, 71.
- the_round_sides_of_a_thin_ring_are_named_by_the_circle_they_run_beside: a thin ring 2 thick
about the R10 circle has two walls, `F2:wall:c1:left` (radius 9: the counter-clockwise
circle's left is inside) and `F2:wall:c1:right` (radius 11), with no numbers; each name is
found again by a point on its own wall.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
box_and_fillet_have_stable_wire_records compared the encoding of a box and a filleted box
with records truck built (2f4d6acac). Since f11c0c571 an operation's result is the record
SMLib gives, which numbers the box's faces in its own order and the fillet's edges by its
history, so the golden differed in face order, face sources and edge history, not in the
encoding. Rewritten with NATIVE_BREP_BLESS=1 and checked against the geometry: the box 6
planes, 12 edges, 8 vertices, each face source's plane its face's; the filleted box 7 faces,
15 edges, 10 vertices (one edge replaced by a face: +1, +3, +2), the round a cylinder of radius
2 along x on the axis (y, z) = (2, 8), 2 from y = 0 and z = 10.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since 09c979a4b a full circle in a profile is one arc of a whole turn, so what is built from
it is one closed edge and one wall closed along its seam. Five kernel tests still expected
truck's halves:
- a_small_hole_in_a_large_spline_face_cuts_cleanly: the Ø1 hole adds one face (its wall),
not two; the removed volume (π · 0.25 · 40) is unchanged.
- fit_loop (two mesh_convert tests): a fitted circle is Loop2::circle, one segment.
- tight_bend_counts: a Ø10 tube swept up 20, round a quarter turn of radius 5, along 20. The
circle starts at (5, 0), which runs up to (5, 0, 20) on the bend's axis, where the turn
leaves it: the bend face has no seam and both straights' seams end there. By the ladder's
rule (each whole curve once, each seam): 5 faces (2 caps, one per path segment), 6 edges
(4 circles, 2 seams), 3 vertices (the seams' ends: (5,0,0), (5,0,20), (25,0,20)), the
record's own counts.
- extruded_circles_go_as_cylinders: the record holds the wall as the cylinder it is (one
face of three), no longer a rational spline; it still goes into SMLib as its cylinder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A folded edge's Hermite pcurve on a bend's form breaks where the edge's angle crosses a
quarter. At the 120 degree corner with flanges at 60 and 90 degrees under a tear, an edge
starts 3.25e-12 of its parameter short of a quarter: the crossing was kept as a break
(breaks within 1e-12 of the parameter merged), the pieces halved to 1.6e-12, and SMLib,
given knots 1.6e-12 apart, handed the pcurve back with 15 equal knots at degree 7 ("SMLib
export: pcurve: loop 0 use 1: given pcurve: portable B-rep: knot multiplicity"). A crossing
whose point on the edge lies within the pcurve's lift accuracy (FOLD_TOL of the edge's size)
of the previous break is that break; the end is never merged into the start.
Fixes sheet::tests::every_pair_of_angles_is_torn_or_keeps_its_notch (torn pairs 37, 37, 27).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two expectations encoded truck-era output:
- a_flange_bent_inside_keeps_the_base_sketchs_arcs_in_the_flat_pattern_and_the_dxf: the
DXF has 7 LINEs, 4 ARCs, 11 shapes (was 9, 4, 13). The flat outline is SMLib's exact 2D
union of the plate, the bend zone and the flange panel; a zone's side and its panel's side
meet in line, and a region's boundary has no vertex in the middle of a straight run, so
each flange side is one line: 6 outline lines (plate top, left, right; flange sides and
end) and the bend's centre line. The same reason as 5d25684a1 gave for
a_base_trimmed_for_a_flange_keeps_its_arcs (12 to 10 edges). Areas, the bend length and
the volumes are checked as before and hold.
- hole_beside_two_adjacent_edge_flanges_cuts_and_bends_stay_guarded: the folded volume was
14057.2 +- 0.1, a faceted measure; since 32ae0cd9c the measure is SMLib's exact integral,
14057.3837123. Exactly, a folded sheet holds its flat area A times t, less each bend's flat
strip L·t·(R + K·t)·π/2, plus its sector L·t·(R + t/2)·π/2: t·A + t²·(½ − K)·(π/2)·L. With
t = R = 2.5, K = 0.44, A = 5597.0353454 (the flat pattern) and L = 75 + 35, that is
13992.5883635 + 64.7953485 = 14057.3837120, within 2e-11 relative of the measure. The
test now checks that identity to 1e-9.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face's sheet name is the name 80% of its mesh triangles' centres agree on. SMLib's
mesher fans a 1 mm rim (30 mm tall) into four triangles about its middle; two of their
centres lie 1/6 mm from the panel's top or bottom face, inside the 0.2·t tolerance, and
took that face's name, so the rim got no name (rim.side0/side1/ext0/ext1 of the tray's
flanges: 6 faces unnamed). A flat face's name now needs its plane to face along the
triangle's normal (within 1e-3); otherwise the point is a rim's, a bend's or an ext rim's.
No face changes name: these faces had none.
sheet::face_names_tests::a_trays_faces_have_readable_sheet_names_that_follow_its_size now
passes its naming checks and fails later, on the point named_face_point gives for a rim
(the roomiest triangle's incentre, 0.497 mm from the rim's edge, not 0.5).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A hole drawn on a flange's face and cut through ("direction": "negative") was refused as
"below the supported 0.000001 mm extrusion resolution": the flange panel's frame, turned
up by the cosine and sine of 90 degrees, puts the panel's origin at -1.9999999999999998
along the sketch normal, so the cut ends 2.2e-16 mm short of the sheet's far face and
reads as a layer 2.2e-16 thick. A depth within four units in the last place of the
numbers it is worked out from (the frame origins, the tool's span and the thickness) now
lands on the face. A real thin layer stays: a 1e-8 mm layer of a sheet 1e5 mm from the
origin (bound 9e-11 there) is still refused
(partial_depth_intersect_sub_extrusion_resolution_layers_refuse_atomically passes).
Fixes sheet::tests::a_sketch_on_a_sheet_face_follows_the_sheet_when_its_thickness_changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since 09c979a4b a circle in a profile is one closed arc (Loop2::circle). The sheet's
faces passed it to the sew as one arc edge from a point back to itself, which the sew
refuses ("sew: a degenerate edge"): a sewn edge runs between two vertices, and an arc
edge is the circle through its ends and middle. A disc's flange failed (first with the
4-face minimum, then with the degenerate edge). Each closed curve of a cell is now split
into its two exact halves (Seg2::split_half) where the cell's edges are prepared, as a
hole's wall already came out: a disc is two caps and two half walls.
sheet::tests::tiny_cut_inside_exact_circle_survives_unfold_refold_and_dxf now builds the
disc flange and fails later, on SMLib's extrude of the 50 nm tool (an SMLib tolerance
defect, reported separately).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The case expected "drill 6.75 mm for M8", the major diameter less the pitch.
Since 05f1c6b9a (Tapped holes: the tap maker's drill for every form) a tabulated
size names the tap maker's drill, 6.8 mm for M8 x 1.25 (Yamawa YEU23 p.704), as
case 2115 already grades and engine thread_tests.rs asserts; major less pitch
is for a size no table lists. The refusal itself and its reason are unchanged.
The case passes and leaves threads/KNOWN_FAILING.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The zero-area triangle is dropped (previous commit): the box converts with six
faces and V = 1000 mm^3, as the case expects.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A triangle whose corners lie on one line (or weld into one another) adds no
surface, so welding now drops it, with any corner only it used, instead of
refusing the whole mesh as having a degenerate triangle. The closed and
consistently oriented check that follows still refuses a mesh that needed it to
close, and a mesh of nothing but such triangles is refused by name. Dropping its
unused corners keeps a dropped triangle from widening the part's extent, which
the prism search reads from every position.
Regression test a_zero_area_triangle_is_dropped: the 20 x 10 x 5 box with the
case's extra triangle (0,0,0) (10,0,0) (20,0,0) converts to six faces and
V = 1000 mm^3.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fixed_seeds_run_clean expected the join of a box touching the first only along
an edge to be refused with "zero-thickness edge contact". Since 94ee28d5c (doc:
a join keeps solids that only touch as separate bodies) such a join succeeds and
keeps the new box a body of its own with an along_edge notice, which
join_touching_along_an_edge_keeps_a_separate_body asserts in engine tests.rs.
The seed now expects that: the notice, two bodies, each valid, and the volumes
unmoved, 25 x 15 x 10 + 17.5 x 17.5 x 20 = 3750 + 6125 mm^3.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib's thickening of the test sheet makes a top face of 72 x 78 control points,
which the 4096-point cap refused (TooLarge) before the solver saw it. The band
cap (2^21 entries, 16 MB) is what bounds the system's memory and its
factorisation (at most n min(n, bw)^2 work); the 72 x 78 bicubic net needs
5320 unknowns 213 apart, 1.14 M entries, well within it. The point cap now
bounds only the per-step energy sweep and the net's copies: 2^14 points, above
the band's own limit for square bicubic nets (about 90 x 90), so a face is
refused for the system it needs rather than its size. The hostile-input test
asks for 129 x 129 and checks the refusal is TooLarge.
a_body_face_is_faired_with_its_edges_untouched_and_its_history_kept passes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The README demo's Raspberry Pi case stands four standoffs on its floor: Extrude2's ring of
radius 3 round a 1.25 hole, patterned by RectangularPattern1. Each is its outer wall, its inner
wall, its top and the floor inside it: four faces. The recorded counts were truck's, which made
each cylindrical wall two halves (six faces a standoff): 25 after feature 6, 43 after the
pattern, and 8 more than now at every step after it. On SMLib each wall is one face closed on
its seam, and the counts are 23, 35, 35, 47, 47, 51, 51, 55, 71 from feature 6 on (the feature
names show it: F6:side:c1, F6:side:c2, F6:top and the floor piece bounded by F6:side:c2 per
standoff). The outer sides stay one face each, and (b) and (c) still agree step by step.
Fixes cli rebuild_topology::pi_case_faces_per_step_at_both_heights.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- A modelled thread reopened with the designation 1/4-20 UNC on the M8 it was made as: the
thread is metric, and since b4549d270 (threads in the standard forms) an unknown designation
is named with its form: "unknown metric thread `1/4-20 UNC` (like M8 or M8x1)". The test looked
for the old wording, "ISO metric".
- Press Pull on a hole's wall opens Offset Faces with the whole wall: one face, closed on its
seam, on SMLib (a cylinder wall was two halves on truck).
Fixes ui-egui dialogs_thread::contribution_tests::expanded_catalogue_rejects_external_taper_holes_and_nonmetric_thread_features
and preselect_tests::press_pull_then_curved_face_offsets_it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face on a B-spline surface closed around u (first and last control columns the same) may lie
across the surface's start without winding round it: the strip on a radius-3000 rational
cylinder whose seam is 0.001 rad into the strip. A kernel's trims run inside the surface's
knots, so its pcurve came back clamped at the start, 2.28 mm off its edge, and the strip was not
imported.
Before seams are added, such a surface is started anew in the middle of the stretch of u the
face leaves free (each loop followed round, unwrapped across the start; a face winding round is
the seams' to handle): the new start inserted to full multiplicity and the piece after it put
before the piece before it, a period on. The same points, a period apart in parameter before
the new start: exact. The face then lies inside the knots.
Fixes kernel step_in::tests::explicit_outer_bound_crossing_a_closed_nurbs_seam_stays_trimmed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two tests counted the old 2D boolean's edges, which kept a vertex wherever an operand's corner
or edge end lay on another's edge. SMLib's planar boolean leaves no vertex that is no corner:
edges on one line within the tolerance are one edge. The results are the same regions with
fewer, whole edges, and each test's point stands (no edge shorter than the tolerance, no sliver):
- a quadrilateral cut away from below a square whose bottom edge it lies on within 1e-8 (tilted
1e-9 rad) leaves the square's four 10 mm edges (its corner at x = 4 is within 1e-8 of that
edge's line), and its union runs along y = 0 as one 14 mm edge (the quadrilateral's bottom
edge is within 1e-8 of y = 0 all along): [4, 5, 5, 10, 10, 14];
- the bent plate and the strip under it: their ends on the line x = 150 are one edge from
y = 47 to 80, six edges for every turn and offset tried, the shortest 3.
Fixes prism::planar_tests::overlapping_lines_a_hair_from_parallel_do_not_cross and
a_bent_edge_crossing_its_neighbour_by_their_corner_leaves_no_sliver.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Reading a planar sheet's faces back as regions, each hole went to the smallest outline whose
Loop2::contains held its middle: even-odd against the outline's polyline within 1e-3 mm. A
5e-5 mm square cut 1.7e-4 mm inside a radius-10 circle lies outside the polyline's chords there,
so no outline held it and the cut failed ("face 0 has a hole outside its material"). An
outline now holds a point by its polyline within a quarter of the point's distance from it (its
nearest point on the exact segments), which no chord can pass the point by.
Fixes kernel prism::planar_tests::tiny_tool_between_circle_and_display_chord_is_not_lost.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
94bd8719a (#3730) turned a BREP_WITH_VOIDS void inside out when it enclosed volume of the
outer shell's sign: SOLIDWORKS writes .F. over a void shell already facing into its cavity,
SolidCraft .T. That lived in the truck reader; the schema-4 reader (crates/step) that replaced
it took the flag alone, and the block with a 20 x 6 x 5 cavity read as 15600 (the cavity added)
instead of 15000 - 600 = 14400.
The reader now takes each void's sign from the faces themselves before it orders the record:
the divergence theorem on the record's surfaces and pcurves (sew's face_flux of p / 3, moved
to geom's brep::eval so both use it). A void enclosing volume of the outer shell's sign has its
faces turned over. Flipping on the finished record is not an option: its edges are ordered by
the faces' sides.
Fixes kernel step_in::tests::a_void_reads_as_a_cavity_whichever_way_its_shell_flag_points.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The expected 6703.5222 was a mesh measure (truck). The lid exactly: the 76.8 x 89.3 x 2.4
plate less its four 3 mm corner rounds ((76.8 * 89.3 - 36 (1 - pi/4)) * 2.4 = 16441.2344), the
skirt ring ((71.5 * 84 - 68.5 * 81) * 3 = 1372.5), less the window (68 * 61 * 2.4 = 9955.2) and
the slot (44 * 10 * 2.4 = 1056) through the plate, and the notch through the skirt's 1.5 mm wall
(22 * 1.5 * 3 = 99): 6703.534401. The window and slot lie inside the ring's opening and the
notch above the plate, so nothing else overlaps. SMLib measures 6703.534401575, 2.6e-7 off.
The test now computes that closed form and holds the body to 1e-6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A whole circle is now one closed rational curve (Loop2::circle is one segment), whose first
control point is repeated as its last: the mean of its nine control points sat r/9 towards the
seam, and a rail scaled each station about that point. A rail bulging from radius 10 to 15
(r = 10 + 5 sin(pi z/40)) then grew the circles as if r = 10 + 5.625 sin, volume
4632.8 pi + 9000 = 23553.9 against the solid of revolution's pi * integral r^2 dz =
4500 pi + 8000 = 22137.2 (+6.4%; the engine's arc rail +6.5%).
The middle is now the curves' centroid by length, integral x ds / integral ds (8-point
Gauss-Legendre per knot span on the rational point's speed): the shape's alone, however its
curves are written; a circle's is its centre. geom's GAUSS_8 is public for it.
Fixes kernel loft_rails_tests::a_bulging_rail_makes_the_solid_of_revolution_of_the_rail and
engine loft_rails_tests::a_rail_turns_a_loft_into_its_solid_of_revolution_and_stretching_is_an_edit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillets/4782-fillet-two-end-verticals-past-a-full-round passes: SMLib (90aabff) refuses
R10 on the 15 mm wide box's two end verticals as too large, and the refusal now names
the largest radius that fits, 7.5 mm (the full round). Run on aarch64 Linux with
--known-failing: every other fillets case as listed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Under the blend-overflow decision (237f0ae74) a convex round running over a face's far
edge (a cliff) is built and cut from the part, so the truck-era fixtures of the reduce
and taper fits (a 2 mm fin, a plate narrowing to 2 mm, a wedge, a 10 mm cube's edge at
R20) no longer refuse their radius and the fits have nothing to react to. Each test now
uses a case SMLib still refuses, and checks the cliff case builds:
- engine fillet_fit_reduce_keeps_the_radius_where_it_fits: a 14 x 6 x 10 post on a
40 x 30 x 20 block, its four verticals at 3.5 (two rounds would cross on its 6 mm ends)
reduced to 3 on 4 of 5 edges, the block's corner kept at 3.5: 24840 - (4 - pi) 90 -
(1 - pi/4) 3.5^2 20. New fillet_fit_reduce_keeps_a_round_over_a_cliff: the fin's R4
builds, nothing reduced, at the kernel's fillet_max_each answer.
- engine fillet_fit_taper_follows_the_room_along_the_edge, spline_fillet_takes_over_a_taper,
kernel taper_follows_the_room_along_the_edge, ui-egui fillet_spline_handle_tests: the
narrowing plate stands against a wall behind its slanted side (z -15 to 30), so its top
and bottom end in concave edges: R3 is refused and tapers 3 -> 1.8 as before. Without
the wall, R3 is cut over the 2 mm top's far edge and kept constant: at x the section
loses (1 - pi/4) 9, or 3 w - (F(3) - F(3 - w)) where the top is w < 3 wide, 6522.7745566.
- The wedge closing to a point: on its own R16 is cut over its cliffs (the section
integral, 4195.578821, SMLib within 4e-8) and kept; its room's taper is checked as a
spline fillet. Against a wall (new fillet_fit_taper_closes_to_a_point) the taper must
close to 0, which SMLib fails today (status 1001).
- kernel taper_mesh_tests: the collapsed taper is built from taper() as a spline fillet.
- ui-egui the_feature_menu_copies_a_warning_message: the 10 mm cube's four verticals at
6 are reduced to 5 (a full round), which the warning says.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The owner's blend-overflow decision (2026-10-25, 237f0ae74): a convex round or chamfer
whose rail runs off its face across a cliff (an opening's rim, a hole, the part's
outline) is built on the faces continued and what it removes is subtracted, so it never
bridges an opening; a round may use up a face; a chamfer of a face's full width stays
refused. These tests still expected truck's refusals. Each now expects the decision's
body, with its answer from the geometry:
- issue_1080 chamfer_into_a_pocket_is_cut_by_it (was ..._is_refused_with_the_size_that_fits):
chamfer 6 into the pocket 4 mm in removes 360 - 20 (its triangle less the part over
the pocket): 14660; 6 along the top and 2 down, 120 - 20/3; 2 and 6, 120.
- issue_1080 fillet_into_a_pocket_is_cut_by_it_and_one_that_fits_builds: R5 removes
(1 - pi/4) 25 along 20 less 10 (5 - (sqrt(24)/2 + 12.5 asin(1/5))): 14893.0344442.
- issue_1080 chamfers_crossing_each_other_are_refused: still refused (they cross);
chamfer_max and fillet_max find 2 mm.
- issue_1080 rounds_on_both_sides_of_a_wall_build_as_one_fillet: R7 alone opens a slit
where its arc passes the inside corner: V0 - (104 (1 - pi/4) 49 - 100 a), a = the square
[25, 30] x [35, 40] outside the circle about (25, 35), 5 (5 - sqrt 24) - (F(5) - F(sqrt 24)):
38458.8958265. Both at R7: the arcs about (25, 35) and (23, 33) cross only outside both
quarter arcs and the inner round's material holds the slit, so V0 - 4 (1 - pi/4) 49 =
39509.9380404 (was expected refused).
- issue_324 a_round_into_a_hole_is_cut_by_it (was the_radius_offered_beside_a_hole_fits):
R4 over the r3 hole: 8000 - 90 pi - (640 (1 - pi/4) - the round's height over the
hole's width, integrated with y = 17.5 - s^2): 7580.2873154.
- issue_851 arc_top_blend_uses_up_the_wall_and_refuses_a_collapsed_one: a round of the
wall's full height uses the wall up (one face for the wall, the removal integrated as for
the other sizes); the full-height chamfer stays refused.
- kernel tests: fillet_max_finds_the_largest_radius_that_works keeps the post (3.5 -> 3)
and the no-edge case; its disc and notch cases move to
rounds_past_the_bottom_of_a_wall_are_cut_from_the_part: the 2 mm disc's R2.5 turned
section, pi (338 - integral of (169 - rho^2) dz), rho = 10.5 + 2.5 cos(theta) (the same
formula gives SMLib's R2 994.15188 exactly), 957.6922389; the notched block's outline
R7, 11925 - (1 - pi/4) r^2 (140 - 8 c), c = r (10 - 3 pi) / (3 (4 - pi)) (SMLib's R7 within
5e-8), and R8 less its part over the notch, integral of 2 (5 g - g^2/2), g = 8 - sqrt(64 - t^2).
- kernel tests: chamfer_wider_than_sloped_basin_rim_is_cut_and_a_full_width_one_refused:
past the 1 mm and 0.5 mm rims the chamfer runs over the slab's front edge and is cut:
the clipped triangle's area (30 + z centroid), its length 60 - h between the sloped end
walls (the same formula gives chamfer_sloped_basin_rim's fitting chamfer to 1e-12);
a chamfer as wide as the rim is refused.
- blend_corner sloped_rim_fillets_stay_inside_the_material, equal_slope_rim_loops_stay_
inside_the_material: R4 and chamfer 2 over the thin rim build, every vertex inside the slab.
- engine blend_fit chamfer_through_the_openings_is_cut_by_them: the report's 9 mm chamfer
is cut by the openings, 3350 (the rig's fillets/1080 case).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since bcbb99005 a blend larger than its faces is SMLib's refusal (KernelError::TooLarge,
"the fillet is larger than the faces beside its edges allow"); truck's room check, whose
words these tests still matched, went with blend_fit. The refusals themselves stand:
- fillet_consuming_a_side_face_makes_a_full_round: R3.001 and R3.5 on the 14 x 6 x 10
box's four vertical edges, past the full round, are refused as too large (no longer
"larger than the neighbouring faces"); the body unchanged is its own volume before
(SMLib's measure of the box is 839.9999999999998, not 840 bit for bit).
- a_chamfer_that_would_grow_the_part_on_a_convex_edge_is_refused: refused as too large;
the word "convex" was the old check's.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When the kernel's room pre-check went (bcbb99005), a fillet or chamfer refused as
too large stopped saying what would fit: only SMLib's NO_ROOM words were left. The
rig expects the refusal to name it (fillets/4782-fillet-two-end-verticals-past-a-full-round:
"the largest radius that fits is 7.5 mm", fillet_max for the exact fit).
A constant fillet with the exact fit, and an equal-distance chamfer, refused as too
large now add "; the largest radius (distance) that fits is X mm", found by the same
bisection fillet_max runs (up to 16 blends, three significant digits), now shared with
a new kernel::chamfer_max. Where the search itself fails (a candidate failing for
another reason), the refusal is left as it was. The reduce and taper fits and the
kernel's own searches call the kernel blends directly and do not pay for it.
Tests: the 14 x 6 x 10 post's R3.5 names 3 mm; a 4 mm rib's crossing chamfers and
rounds at 3 name 2 mm.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
sweeps_along_paths expected one wall for the circle swept through 41 points of
a smooth curve: a point path is the polyline its points make, swept line by
line and mitred at each point since 43e8b0349 (sweep_path's own test expects
40 + 2 faces since d9274aaf1), so the circle's one segment has a wall along
each of the 40 lines.
The holed square along the L-shaped path expected two faces per segment. The
corner turns the profile about Y, so each line at a constant y (the outer
loop's segments 0 and 2, the hole's 1 and 3) sweeps the plane y = c it lies in
along both legs, and the two walls meet edge to edge on the mitre in that one
plane: one L-shaped planar face. The lines at a constant x sweep two planes,
two faces. check_sweep takes the count per segment.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
sweep_origins read a sweep_path tool's faces by the truck face order (walls in
profile order, then the start and end caps), and fell back to swept_origins,
which names a face without an edge in the profile's plane an end cap. SMLib's
sweep along lines and arcs is made piece by piece and mitred at each corner
(tangency-labs/smlib#43), so most walls start further along the path: 39 of a
44-face sweep came out as end caps (105 of 110 before the previous commit). The
sweep records each face's origin as it builds it (Body::history), and
sweep_origins now returns those, finding by geometry only a face the tool names
no origin for. swept_origins says what it holds for.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A sweep along a path with corners sweeps each run past its corners and cuts it
back to the plane halving the corner. The lead past the corner was the
profile's reach times tan(turn / 2), enough only for a profile square to the
path. A point standing s ahead of the path and w from it needs
w tan(turn / 2) - s past a corner (w tan(turn / 2) + s before one): a profile
at a slant to its path reaches across the mitre by its slant as well, and a run
carried short of it leaves a notch with a piece of the run's section in it.
The section keeps both distances along a run and through each corner (it
turns with the path; a twist keeps them, an end scale grows them by at most
its largest scale), so the lead is taken from the profile's points at the
start, for the plain and the controlled sweep. A profile square to the path
(s = 0) leads exactly as before.
The walls' history seeds sat at the middle section of each run, which for a
slanted profile on a run shorter than its slant lies past the mitres, on the
next run's wall: half the walls came out generated. Each piece's seed now sits
half way along the stretch its own point keeps between the mitres, carried
there exactly along lines and arcs (as SMLib's sweep turning with the path
carries the profile), at the station there in the controlled sweep.
A flat 4 mm square swept along 21 points of a rising sine (provenance test):
110 faces, 66 of them notch faces, and walls left unnamed; now 44 (two caps, the two
walls in the path's plane, a wall of each slanted side along each line), volume
16 cos(slant) L to 1e-6 (new test a_slanted_profile_meets_itself_at_every_mitre).
The square round a closed 122-line polyline (issue_2005
closed_polyline_sweep_shares_its_starting_section): the first section leaned
by half a chord's turn left a 6.2 mm^2 piece of it at the closing mitre, named
the start cap; now 246 faces, every one a wall.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
legacy_internal_capture_refuses_an_uncaptured_source_in_another_component
picked the top Offset Face raised as a face with neither a construction
certificate nor an exact history name (#494), to reach the legacy path
that refuses a profile across components while the source occurrence's
move is uncaptured. Offset Face has SMLib's history since 0f3d6f177 (shell,
offset faces and thicken through SMLib), so that top is now named exactly
(F3:+z#[F3:-x], verified) and capture_face_anchor gives it a lineage
anchor; the legacy path is not taken, and the capture is right to
succeed: an anchored attachment captures its path's pending positions
when it is made (capture_positions).
The test, renamed to what it now checks
(internal_capture_in_another_component_anchors_an_offset_face_by_its_history_name),
asserts the face still has no construction certificate, the capture is a
Lineage anchor on the occurrence's path, and capturing changes neither the
document nor the pending moves.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The test cut three regions from a plate that keeps its meshes and
expected the memo to hold one mesh afterwards: the old kernel's booleans
meshed their operands. On SMLib (ops::boolean -> smlib_boolean, since
42c41464a) a boolean reads the exact record and a tool apart from the
body is found from the exact boxes, so no boolean meshes anything and the
memo is empty after them. The test now asserts that, then that meshing
the kept body twice at one tolerance makes one memo entry, bit-equal to a
fresh mesh, which is the reuse the memo is for.
Fixes kernel body::mesh_memo_tests::regions_cut_from_a_body_keeping_its_meshes_are_unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Split Face records its lines on the body (split_lines): heal and the
certified-planar regularizer keep the faces apart along them, and they
make the body regularization-protected. On SMLib the boolean, the n-way
union and Shell carried the protection over but dropped the lines, so a
body with an authored split lost its record after a Union with partial
contact (issue #674/#782's case), and a shelled one after Shell. Each
result now keeps its operands' lines, each line once
(Body::inherit_split_lines), as the old kernel's coplanar join did.
regularization_protection_tests: SMLib's shell makes no section strips,
so the painted cup no longer has more faces than the ordinary one. The
ordinary cup is the box's five kept faces, their five linings and the
rim: 11. A paint over a whole face authors no boundary inside it, so the
painted cup has the same 11 faces; the split side's line stays an edge on
the wall and its lining: 13. The test now asserts those counts exactly
(it asserted more than 11 for both, the old strip route's witness) and
that the split lines survive.
Fixes kernel issue_674_tests::partial_contact_retains_an_authored_split_away_from_the_contact
and body::regularization_protection_tests::stamped_shell_keeps_authored_painted_and_split_boundaries.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The test helper side_by_side copied the second body's regions by hand,
dropping its voids and keeping its infinite region (region 0) as if it
were a bounded one. Its closed shell was then the outer shell of two
regions, which the portable record refuses ("a closed shell bounds two
regions and encloses one"). The helper now joins the two records with
geom's brep::part::compound, which puts each body's outer shells in the
one infinite region and keeps its bounded regions; the face sources are
still found by plane afterwards, so nothing the test checks changes.
Fixes kernel body::planar_sources::tests::coincident_competing_source_planes_and_tool_planes_stay_unknown.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Body::parts (and so Body::lumps) mapped each piece's faces through the
body's face-source table with a silent `get`, so a table shorter or longer
than the body's faces gave its pieces sources of other faces, or none. The
table is by face position: one whose length is not the body's face count
belongs to another body. The pieces are now refused with "lumps: face
source table does not match the body", as before the SMLib port (the check
went with the old lumps_with_clearance in 66e31bede).
Fixes kernel body::lump_tests::grouping_lumps_does_not_infer_absent_or_misaligned_sources.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Which input plane each result face lies on was found by scanning every input plane whose
normal x was within the resolution's angle: quadratic on axis-aligned parts (two thirds of the
faces have normal x = 0), so it was capped at 16 384 faces a body and 2 M comparisons, and the
join of the 20 level-2 sponges of a level-3 Menger sponge (7 326 faces) was refused past the
comparison budget.
The planes are now indexed by their unit normal, in cells twice the bound on two matching
normals' difference (asin(angular) < 2 angular), and in each cell by their offset from a common
centre, searched in the window linear + |a.o - c| |n_b - n_a| that
d_b - d_a = (b.o - a.o) . n_b + (a.o - c) . (n_b - n_a) gives, plus rounding slack. The index
offers a superset; same_support still decides each match, and candidates are taken in the same
order as before, so every answer is unchanged. A plane with a normal not of unit length is
offered to every query. Both caps are removed: results are bounded by inputs capped where they
entered. The descriptor-size budgets on kept roles are unchanged.
Rig, --known-failing, results compared case by case with develop's: booleans (129) and shell
(28) identical except booleans/5126-join-menger-sponge-level-3, which now passes in 38.9 s on
aarch64 and leaves KNOWN_FAILING.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
outer_area, which decides whether SMLib's normal is the record's for each
face, copied every edge of the body into a record for each face before
computing the pcurves of that face's outer loop: O(faces x edges) curve
clones, on every operand and every boolean result. pcurves now takes an
edge lookup instead of a slice of edge records, so outer_area reads the
loop's edges where they are. The pcurve routine gets the same curve,
interval, tolerance and trims as before, so every area, face turn, record
and numbering is bit for bit what it was.
On the level-4 Menger sponge with 3 tools, body_from_smlib_ordered goes
from 92.6 s to 1.3 s of self time.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
explode.show put the view's placements in the uncaptured moves, so occurrence.properties reported
the exploded placement and Capture Position (and reparenting, moving bodies between components,
copying or pasting in place, URDF export) wrote the view's offsets into the design. The session
now records which uncaptured moves are the view's (Session::exploded); what reports the design's
placements or works a new placement out from them uses the design's own moves
(uncaptured_moves, design_document), while the viewport and picks still show the view.
New case cov-explode-show-keeps-the-placements: the arm's properties are the same before, while
and after the view is shown, and Capture Position keeps nothing. The two explode cases that read
occurrence.properties while exploded now expect the design's placement, centre (20, 0, 5), as the
command says ("shown only; the design keeps its placements").
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The curves were held on a 1-2-5 ladder from the world origin, so count 3 on a 40 x 30 face gave 3
one way and 2 the other. They now follow the command's contract: the held parameter at
lo + k (hi - lo) / (count + 1), k = 1…count, over the face's extent. Faces of the body on the same
surface record are spaced together over their combined extent, so a face a boss splits still
shows the same lines straight across the split.
New case cov-isocurve-40-by-30-face-count-3: x = 10, 20, 30 and y = 7.5, 15, 22.5, to 1e-9 mm.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Both walked the picked curve's display polyline: a point a third along an R10 sketch arc was
0.87 µm off (8.6603, 5). They now evaluate the path's exact curves by arc length: lines and arcs
in closed form, a free-form span at the parameter its arc-length integral reaches (body edges by
their record curves, 3D sketch curves by their arc, spline, line or linked exact curves; only a
curve that exists as points alone, an intersection edge or a projected curve, runs along them).
The arc's point is now within 1e-12 mm of (10 cos 30°, 5, 0).
A t outside 0…1 was clamped silently, which the contract (`t: expr, 0…1 along it`) does not
say: it is refused ("t is 1.5: a point along a path is 0 (its start) to 1 (its end) of the way
along it"). The construction cases grade the arc and line to 1e-9 mm, past-the-end is refused,
and a new case takes the middle of a symmetric cubic spline, (20, 7.5).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face of degree 1 in one direction (an extruded spline) was refused with "a fairness term does
not fit the net's degrees": the thin-plate energy's S_vv asks for second derivatives a linear
direction does not have. On every span that term is exactly zero, so it is dropped; the energy is
the bending along the curve and the twist. A face whose held layers are its whole net (a
single-span cubic with its edges kept, an extruded curve's two rows) has no interior control
point: it is left exactly as it is, its energy reported unchanged, 0 steps.
cov-fair-single-span-face-has-nothing-free-to-move passes and leaves KNOWN_FAILING.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The portable record's 4,000,000-item budget refused a real result: the join of a level-4
Menger sponge's level-3 sponges ended with "SMLib export: the mapped B-rep does not validate:
portable B-rep: item budget". The budget guards records from outside the kernel, and a body
the kernel built in memory from inputs already bounded where they came in isn't hostile.
Brep::validate keeps the budget for records read from outside: native design caches
(decode_native_body, behind its own 256 MiB, 1M-per-array and 20M-value caps) and STEP import.
The new Brep::validate_built runs every other check, without the cap, on what the kernel
builds: SMLib's export of an operation's result, a body's record (Body::from_record, after
the decoder has checked a file's), a part of a validated record, the record handed back to
SMLib, and the validity check. Its checks are linear in a record already in memory.
SMLib's export bounded its counts by the same budget; a count is now bounded by SMLib's own
longest array (flat::MAX_LEN, checked on entry), which keeps every offset worked out from
the counts from overflowing, and the redundant topology-length cap goes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Where a round meets a plane it is only tangent to it: its curvature jumps by 1/r there, so no
corner patch can be G2 to both at that point. The cases graded the whole seam and could not pass.
They now bound the curvature jump at stations clear of each seam's turn, within reach of the
corner, scaled by r (ladder 07/019's bound at r5): r1 within 2.2 mm, 0.24 mm clear, 0.005 per mm;
r5 within 11 mm, 1.2 mm clear, 0.001 per mm. Develop's corner gives 0.00135 and 0.00027 per mm
(the jump at the vertex is 1.017 and 0.2035, the round's own 1/r), so both leave KNOWN_FAILING.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When a − b or b − a cannot be made (a kernel error, or a body that encloses no volume), the
region's volume and centre are null, measured_by is "not measured" and not_measured gives the
cut's error; its deviation, area, faces and box are still given, the side's total is null and
the verdict says the volume is not measured. The integrated-along-lines estimate is no longer
reported as a volume.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
inspect.compare integrated a region's volume along sampling lines: the
40 x 30 x 5 slab between a box and one 5 mm taller came out at 5976.49 mm3
against 6000. The regions are still found from the two surfaces; each one's
volume, centre and box now come from the pieces of the exact cut a - b (or
b - a) that lie there, measured as inspect.measure measures a body
(measured_by: boolean). Where a cut cannot be made the region keeps the line
integral, marked measured_by: lines, with a warning. The rig case
cov-compare-taller-box passes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
inspect.center_of_mass measured each body's display mesh, so a cylinder r 5,
h 10 came out at 784.13 mm3 against 250 pi = 785.40, and the answer moved with
display quality. It now takes the exact mass properties inspect.measure uses
(ModelBody::measure), skips bodies that enclose no volume as Measure does, and
lists what it skipped. The rig case cov-center-of-mass-box-and-cylinder passes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- #58: a fillet's tangent-point step no longer depends on the blends before it (solidcore #1173).
- #59: the fillet clean-up keeps the target's edges and faces by a mark, not by address (#1176).
- #60: capi test, a line projected through a sphere's poles meets itself smoothly at each pole.
- #61: offset faces removes a step wall the move closes to nothing and merges the two tops, so
shell/cov-replace-face-step-onto-the-higher-face builds.
- #62: thicken puts a flat side wall on its plane (solidcore #1599), so a cylinder sheet's end
rings are planes: surfaces/cov-thicken-cylinder-sheet-symmetric-tube passes.
Both cases leave KNOWN_FAILING.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
They fail on develop today (4 native tests, 8 of 83 reference cases), so running them on pull
requests would fail every pull request. Back on pull requests once develop is green. Reverts
0cdc25094.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The single results form compares a whole number exactly (a count). The construction and inspect
cases gave positions and volumes as whole numbers, which the form they were written for compared
within tolerance; they are decimals now. The NPT groove fix makes ladder 06/023 pass: its golden
entry is the passing run (51 faces, 8226.4042 mm3).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The shared core (crates/smlib, the grader, doc eval.rs, engine features.rs and the kernel's core)
is in PATHS-ALL; the new areas' copies of it are dropped, so an area reruns on its own files and the
core reruns them all. sheet-metal also watches the sheet commands, the sheet document and the DXF
writer. 116 commands a case now uses leave EXEMPT; 25 geometry commands are still to cover.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two branches added a results field in different forms ({step, json} and {command, pointer,
value}). The grader keeps one: a 1-based step, an optional JSON pointer into its answer, and the
value matched as files are, with an optional relative or absolute tolerance. The construction and
inspect cases are rewritten to it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The groove's section was cut square a fixed margin past the crest. The crest cone changes 1 in 32
along the axis across the section's own width, so for NPT (margin 0.015 mm against a 0.028 mm
rise) the section's deeper corner stopped under the cone and each tooth kept a sliver of its
flank: 0.94 mm3 too much on a 1/2-14 NPT. The section's edge now follows the cone at the same
margin.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
167/168: every new case passes against its hand answer but 06/023 (1/2-14 NPT external), whose
groove stops 0.015 mm past the crest at its section's middle, under the crest cone on the section's
deeper side, so a sliver of each tooth's flank stays (0.94 mm3, 1.1e-4). The 139 cases before keep
their statuses and counts; their volumes and areas re-measure within 2e-6 on this SMLib.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Base flange; an edge flange at 90 deg; four made together, the corners closed on the mitre plane
with the rule's gap, at 90, 72 (leaning out, the ends run on along a slant) and 108 deg (leaning
in, cut back); round, rectangular and tear reliefs; a hem; a fold, unfolded and refolded; the
tray's flat pattern (outline size and area, bend lines); a bracket with a hole and a flange on a
flange. K = 0.5, so every volume is T x the flat area. The 108 deg tray is a sheet-metal rig case too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
From the script's last sheet.flat_pattern result: the outline's size, its area less the
cut-outs', and the bend lines, each matched to its own one of the result's either way round.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
G 1/2 external and Rp 1/2 internal (ISO 228-1 / ISO 7-1, 55 deg) by Pappus turn by turn; R 1/2
and 1/2-14 NPT external and Rc 1/2 internal tapers (1 in 16 from the standard's diameter at the
free end, crests cut by the major or minor cone) by the screw sweep's Jacobian, integrated exactly
(ladder_lib.sweep_removed); an S20x4 run down to a shoulder with its steep flank toward the end and
toward the start, which moves where the groove stops; tip: start, and a Tr20x4 chamfered at its own
30 deg and 2.5 mm; refusals for a Tr chamfer without a depth and for the square form.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
1/4-20 UNC (#7, 0.201 in) and Rc 1/2 (17.6) through the stock, 1/4-28 UNF (5.45) blind with a
flat floor, 1/4-18 NPT (10.85) through: a cosmetic thread in a straight hole, so each volume is
the stock less that cylinder. The drills are the catalogue's (Guhring, Yamawa YEU23); each method
gives the standard's minor diameter beside it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The tap maker's 6.8 mm M8 drill, the chamfered tip's ISO 724 d3 with tip: both, and 07/016's
refusal reason were edited into the cases' JSON but not into cases.py, so running it undid them.
Its output now matches the committed cases (to the last rounding digit of two volumes).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A Default row, a 60 mm row, a row without the hole, a height column and rules
that suppress the hole or drive the height, each checked after activation;
preview leaves the design; refusals for a row that cannot build, unknown rows,
columns and features, and a rule a row breaks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Suppress a hole and the block is whole again; a round moved before the notch
runs the whole edge and is cut by it; deleting a sketch takes its cut and
deleting the block leaves the round and cut failing; roll to the middle gives
the part at that step; edits, compute all and folders, each with a refusal.
The new history area has its KNOWN_FAILING (every case passes) and PATHS.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A case's construction field names construction points, axes and planes with the point, line or
plane each must be, and results reads values out of the commands' own replies by JSON pointer.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A new plastic area with its KNOWN_FAILING (empty: every case passes) and the
sources whose change reruns it; the README lists the area and the new files and
results fields.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A case's expected.json may name files written to $OUT (json, csv rows, or a DXF's
entity counts and curve lengths per layer) and a command's answer by step, each
matched as a part of what was made.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
cargo xtask kernel-cases --coverage lists every built registry command that no recipe.json under
tests/kernel-cases runs and fails unless tests/kernel-cases/EXEMPT lists it. EXEMPT holds 252
commands with no geometry to grade and, under [to cover], 133 geometry commands still waiting for
a case; a listed command a case now uses, or one the registry no longer has, fails the check, so
the to-cover section only shrinks. The gate's build job runs it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The booleans area also reruns on sketch projection from the model. The selection self-test
fails on an area glob that matches a file PATHS-ALL already covers, so a whole-crate glob
cannot creep in.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A plan job (scripts/ci/kernel_rig_plan.py, with its self-test) maps a pull request's changed files
to the areas: tests/kernel-cases/<area>/PATHS for each area, tests/kernel-cases/PATHS-ALL for the
shared core, the ladder whenever any area runs. Areas come from the folders, so a new one needs no
workflow edit. Pushes, merge groups, the release gate and runs by hand run every area. The
kernel-rig check counts a skipped build or shard as passed only when the plan selected none.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The gate passes no secrets to release.yml, so the macOS and Windows signing steps skip as
they do on a fork (no signing secret, no SignPath quota), and the new packages_only input
skips the release job (checksums, update signing, publishing). The call still grants
contents: write, which GitHub checks for the skipped job before the run.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tests/kernel-cases/README.md describes the kernel rig gate for every area, how a pull request
updates its area's KNOWN_FAILING, and the ladder's --gate and --record. The tests skill and
CONTRIBUTING say which pull requests run the native and reference tests, and what the
release gate runs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
release-gate.yml, run by hand on a branch, calls ci.yml as Run workflow with
cross-compilation (every native suite in full, reference, doc and GPU path tests, the whole
action matrix, property tests, web-offline, wasm, Windows FXC and startup), kernel-rig.yml
(every area and the ladder) and release.yml as a dry run (every package and its smoke
tests), and passes only when all three pass. release.yml can now be called (workflow_call).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
kernel-rig.yml builds solidcraft-cli once (release, SMLib prebuilt; sccache on a sticky disk
that only pushes to main and develop commit) and hands it, with the rig's runner, to a matrix
of eleven shards: the ten areas of tests/kernel-cases/, each against its KNOWN_FAILING, and
the ladder against its golden. One path filter covers the kernel surface; drafts skip; a new
push cancels a pull request's old run; merge groups always run it, and the release gate calls
it. The areas other than fillets get a KNOWN_FAILING with only its header, to be seeded from
develop after the SMLib pin bump.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A pull request whose author is an owner, member or collaborator of the repository runs
Clippy, Native Linux tests, GPU path tests and the reference tests without the
approved-for-ci label, scoped to the affected packages as before; others still wait for the
label. The reference tests leave the pull request fast lane: they run when solidcraft-cli is
selected. ci.yml can also be called (workflow_call) by the release gate, and runs as Run
workflow does. Pushes, merge groups and the nightly run are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
`cargo xtask kernel-cases --ladder --gate` grades every ladder case against
tests/kernel-cases/ladder/golden.json and fails on a changed status or count, better or
worse, or a case with no golden; times stay report-only, since they are the machine's. A
right change is re-recorded with --record "why" in the same pull request.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib's parallel projection (CreateParallelProjectionCurves) gives every curve where the
projecting lines cross the face: a line projected down onto a ball came back as its top and its
bottom, each split at the pole, and the link's first wire was half the bottom. A projection onto
a surface is where those lines first hit it (owner's decision, as mainstream CAD does): a curve
is now kept only when the first hit on the body's mesh along the line through its middle is the
curve itself (within 4 chord tolerances), and the pieces that meet end to end are joined in
whatever order and direction they come, a piece run the other way exactly (its knots and
control points reversed; a line's, circle's or ellipse's parameter mirrored). The ball's top
is then one wire from x = -10 to 10.
three_d_sketch_curves now expects that exact curve: one wire, its ends at (±10, 0, √300), every
point of its exact curves on the sphere of radius 20 in y = 0 over the top, to the projection's
own tolerance (a millionth of the body's size), and sweeps_and_pipes_follow_3d_sketch_curves
pipes along the whole joined top. Both pass on the pinned SMLib (90aabff): the pieces lie 9.1e-6
off the sphere against a 4e-5 tolerance and meet at the pole within 4.7e-8 rad, since the fork's
8d95600 keeps the pole's exact tangent in StepBackFromSingularity.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
path_points took any wire with three or more fit points as a fitted spline, and a 3D arc keeps
its start, through and end points as its fit: the pipe along a quarter arc of radius 20 ran
along the cubic through those three points and came out 388.06 mm³, not π·2²·10π = 394.78.
A wire with its exact arc (Wire::arc) now goes to the kernel as that circle, from its start
counter-clockwise about its normal through its sweep (kernel::curve_path of the record circle).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Design defaults gains a Surfaces section with one setting,
surface_extrude_symmetric_per_side (off by default). When it is on, new
symmetric surface extrudes go their distance each way, as they did
before they matched solid extrudes. The app hands it to the session
before every command and each frame, and it stays with the user across
tabs and new designs, like the metric tap drill rule.
The Surface Extrude dialog now sends a feature's stored convention only
when the feature is already symmetric, so a one-sided one switched to
Symmetric takes the new reading, as in the solid Extrude dialog.
Labels in en, ar, he, ja, zh, de and ko.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A symmetric surface.extrude given no convention now splits its distance
evenly around the start plane (10 mm: 5 mm each side), the way
solid.extrude reads it, instead of going the distance each way. Both
commands take their default from one function, which a session setting
(surface_extrude_symmetric_per_side) can turn back to the per-side
reading for surface extrudes. timeline.edit and timeline.redefine give a
one-sided surface extrude made symmetric the same default.
Saved designs are unchanged: every surface extrude saved its convention
(legacy_per_side), and one saved before the marker existed reads per side
as it always did. The help of both commands says how the distance is
read. The 3830 kernel case, recorded with the per-side reading, now says
so explicitly.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tangency-labs/smlib#57 on top of 60b1c6a. It fixes the two engine tests the earlier pin broke:
press_pull_on_curved_faces and a_shell_fillet_does_not_follow_a_gone_edge_onto_a_smooth_seam,
with their expectations unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The barrel's arc has its centre across the axis, so its side is the inner (lemon) zone of a
spindle torus, which SMLib's torus does not describe (its meridian is on the X side): SMLib
(revolve-spindle) sweeps it as the exact rational surface of revolution. shell_of_a_revolved_barrel
asserted an analytic torus of major radius -10. As the owner decided, it now asserts what
matters: two flat caps and the curved side, the closed-form volume, 14 mm round at mid height,
and after the shell the two offset faces (the inner wall and floor; truck's side was in halves).
Needs SMLib with revolve-spindle (pin bump).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Drafting a cube's side about its bottom edge (on the neutral plane) tilts the side and moves the
edges it shares with the top and the two faces beside it: those four are trimmed differently
(Modified); the bottom and the opposite side are untouched (Kept). Truck called all six
Modified; SMLib's history called the moved neighbours Kept. Needs the SMLib fix "history: a face
whose edges an operation moved is Modified, not Kept" (SMLib branch devgreen-kernel-records,
cdaf205) and a pin bump.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
From 7e917f0 to 60b1c6a on tangency-labs/smlib replay/1.0.1, 45 commits in 43
pull requests:
- Booleans (#28, #29, #34, #39, #45, #47, #55, #56): a coincident edge split with
its edge, a plane grown to cover a face however far, a seam point classified to
its face, intersection loops closed only along edges that lie on each other,
curves inside one ProtoVertex split at its vertices, an intersection along a
face's own edge taken as that edge, a plane a hair into a cylinder touching it
along one line (and crossing it in two), faces and loops apart not intersected.
- Transforms (#48, #49, #50): a mirrored cylinder read back from its NURBS turns
the right way, a rotation or reflection keeps every tolerance and a scale
scales them, a scaled body's model size is measured from its geometry.
- Fillets and chamfers (#24, #30, #51): a chamfer round an arc is a cone, a gentle
corner in a variable radius splits the round, a seam round's corner ball sits
where the cylinders' axes meet.
- Shell and offset (#17, #22, #23, #25, #26, #32, #38): closed faces and closed
sheets offset, exact lines and conics between analytic faces, cones and
cylinders extended as themselves, a cone's apex offset as a sphere, a wall
that would turn a face inside out refused, neighbours extended to reach a
face moved past the body's size, an open edge's trim point on a closed curve.
- Draft (#27, #53): a cylinder along the pull drafts to a cone, a cone about the
pull to the cone its turned generators make.
- Sweeps (#14, #31, #36, #37, #40, #43, #44): a whole turn of a holed profile
leaves the void, a curve near a circle swept as itself (text outlines' Béziers
pass the check), an arc centred across the axis revolves to an exact B-spline
surface of revolution, a minor circle on the axis and a horn torus, line and
arc paths swept exactly piece by piece.
- Surfaces (#18, #19, #20, #21): a B-spline starter in Bezier form, extend by the
distance across the surface and as the same cone or cylinder, untrim in place.
- Intersections (#35, #46, #52): a curve into a pole keeps its tangent, a periodic
value just past the top wraps a period down, a curve from a singular point
starts inside.
- Checks, history, measure (#15, #16, #33, #41, #42, #54): a face whose edges
moved is Modified, SMLib's warnings are no failed test, sheet faces count in
the area, internal poles sought in the knot's row, a collapsed seam not
classified, a small loop oriented on the surface's scale.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib now sweeps a curve within the tolerance of a circle as itself, not as that
circle's cylinder (tangency-labs/smlib#31, 3cfbbdd), so the offset curves' pieces
meet with one curvature at every wall join of the ring (the largest curvature
jump 4.2e-7): all 512 are smooth, where 400 were and the other 112 jumped by
0.9-3.6%. The comments that said the pieces meet only tangentially, or at small
real creases, go.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With SMLib integration-overnight + b78bd4a (the fixes of 2026-10-29), three
expectations that held a refusal now build, each checked against its exact answer:
- rig fillets/1207-fillet-box-end-edge-into-unblended-round (was ...-refused): the end
face's top edge of a cube rounded r3 along its top back edge, rounded r1 alone
(tangent chain off), builds. Its round runs on along y through the r3 round's end
until that surface drops below it at y = 17 + sqrt(5): V = 8000 - 20 (9 - 9 pi/4)
- 17 (1 - pi/4) - 0.218236 = 7957.505202 (the last term the section under the round,
integrated exactly); SMLib gives 7957.505245.
- blend_smlib smooth_edges_and_unfinished_tangent_chains_are_refused_saying_why: the
same cube builds, at that volume.
- modify_history_tests::smlib::draft_tilts_a_side_and_keeps_its_name: the rod's side,
a cylinder along the pull, drafts 5 deg to a frustum, radius 5 at the neutral plane
and 5 - 10 tan 5 deg at the top, at its exact volume.
tests/kernel-cases/fillets/KNOWN_FAILING drops 1557-fillet-vertical-corners-after-
bottom-rounds, which passes now; the other eight still fail, so the gate holds
(81 pass, 8 known failing, no regression).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since an operation's result is its record (f11c0c571), a mirrored body is
SMLib's transform read back from SMLib's export, and two SMLib defects there
made it invalid: a mirrored cylinder came back inside out the wrong way
(SmCylinder::IsNurbSurfaceCylinder; its edges 10 to 30 mm off the surface),
and every edge's own tolerance was reset to the body's (SmBrep::Transform).
Both are fixed in SMLib (fix/mirrored-cylinder-from-nurbs e02e63e, c603293,
with e008e82 for a scaled body without vertices);
this test needs the SMLib pin moved past them.
A plate with a through hole and a box with two rounds meeting at a corner,
mirrored in x = 30 and read back from their records: SMLib finds both valid
and their volumes are the originals'.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
notch_into_a_ring_chamfer_is_cut_exactly measured the notch as the ring's mesh volume before less
after (0.2906 against the integral's 0.28665) and oblique_cuts_through_a_chamfered_ring_are_exact
compared the two mirrored cuts the same way (3.3e-4 apart). The cut is right: the ring less the
cut ring is a piece of 0.2866521 mm³, the integral to 2e-6. A difference of two 1149 mm³ volumes
is not: the re-tessellated ring's mesh volume moves by 4e-3, and SMLib's mass properties (at
their finest, 1e-4 asked) by 1.5e-6 of the ring, 1.7e-3. Each cut is now measured as the piece it
took away (the ring before less the ring after, an SMLib boolean), to 1e-5 of the integral and
1e-6 between the two mirrored cuts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib keeps a whole cylinder one face and the arm's round end and the boss over it (one
cylinder) one face, named for the arm's end. Three naming tests relied on truck's pieces:
- a_cut_through_one_of_two_bosses_names_only_its_own_walls finds the boss's wall by a point
on it (the centroid of truck's half cylinder lies on nothing now) and checks it keeps its
name through the cut and the fillets;
- split_pieces_keep_their_numbers_across_evaluations keeps the torus (its sweep's walls are
whole now, their names stable) and adds a disc whose top a round groove splits into
F1:top#1 and #2;
- a_design_saved_before_pieces_were_numbered_by_place_keeps_their_numbers uses that disc,
whose kernel order numbers the pieces the other way round from their place (the post's
side of multi-body-join-cut is one face now).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- shell_tangent_chain_that_opens_every_face_explains_how_to_keep_walls: the shell is the
rounded box (its 37 x 27 x 57 core grown by the 1.5 ball: 71753.436) less the sharp 36 x 56
x 28 cavity, 15305.436, to the measure's 2e-7 (SMLib: 15305.445); truck's was 15305.28.
- press_pull_on_curved_faces: the shell's inner wall behind a fillet is offset as a face
(SMLib builds it, truck refused it as tangent to its neighbours): pulled 0.5 into the
cavity it adds exactly the quarter ring π/4 (3² - 2.5²) 18 = 38.877.
- revolve_cut_round_a_box_corner_does_not_panic: the side face is picked by its point; the
report's face index was truck's face order (SMLib's face 1 is the opposite side).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib keeps a whole cylinder (and an arc of one past half a turn) as one face closed on its
seam, where truck split it in two; its holes are exact circles; its mass properties come
within 2e-7 of a closed form. The tests that counted truck's pieces or its polygon hole now
check the exact geometry:
- extrude_of_adjacent_profiles_has_whole_edges: the cut-off cylinder is 4 faces (arc side
4674 = 40·(2π − 1.2019)·23, chord side 1039.8, two ends);
- step_export_of_an_enclosure_with_a_screw_hole: 27 faces, as the file holds them;
- model_faces_gives_a_point...: the hole's wall is one cylinder, its centroid on the axis;
- draft_takes_a_cylinders_side_by_name: one side face (the draft itself waits for SMLib's
draft of a cylinder);
- convert_to_sheet_metal_takes_the_plates_face_by_name: 15000 − 62.5π to the measure's 2e-7;
- a_checkpoint_survives_file_new: a unit box measures 1 + 2e-16;
- creating_a_component_recomputes_nothing: SMLib's boolean keeps the face certificate the old
setup relied on losing, so the legacy face plane is made as a design saved before face
attachments were certified (its anchor removed) and reopened.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since the section hatch per part (#8529), a cut is tinted toward the hatch colour its body's
name picks; Body1 picks the steel one, so the cap test's warm pixels were none
(cpu_picture_section_caps_hide_the_inside). The test turns the part hatch off: it checks that
the cap hides the inside and shows its colour, which the hatch tests cover per part.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
solid.shell takes faces: [] since 1df330c79 (a closed hollow body), so its row in
shared_types_keep_the_handlers_messages no longer matched the shared FaceRefs message. The row
uses solid.draft, which still refuses it; FaceRefs says shell's exception.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib makes a torus one face, so the face at (25, 0, 0) was the whole torus,
not a patch within half a turn each way. The patch is now the torus's piece
with x, y > 0 and z > 1, bounded by iso lines both ways (planes through the
axis, and z = 1 round the tube), and its thickened volume is the formula's
over the angles its mesh spans. (The sphere cap above it builds once SMLib's
offset takes an open edge's second trim point away from its first: SMLib
devgreen-kernel-geometry; waits for the pin.)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib thickens a sheet across a convex sharp edge as other CAD does: each
face's offset is extended until the offsets meet (and trimmed where they
cross inward), where truck bridged the gap with a wedge. Owner decision:
keep SMLib's result. The exact values:
- three faces of a 10 box's corner out by 1: three slabs, a 1 × 1 × 10 block
along each edge and a unit cube at the corner, 300 + 30 + 1 = 11³ − 10³;
- into the inside corner by 1: 10³ − 9³ (it was refused); half each way:
10.5³ − 9.5³ (refused too);
- two faces across one edge: 2 · 100 + 1 · 1 · 10 = 210.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A sweep along points follows the lines between them (no wall is fitted through
the points, sweep_sheet_points' contract, as sweeps_along_3d_paths already
says): along a quarter circle of 41 points the sheet is a wall per line, 40,
its area 10 × the polyline's length (within 10⁻⁴ of 10 × 15π); the bent chain
is 2 × 40 walls sharing every seam, 2·41 + 3·40 edges.
SMLib's sweep takes guide rails along a cornered path when each rail turns
with the path at its corners; a straight rail along a bent path is refused
for that reason, not as "guide rails along a path with sharp corners".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib revolves the cone frustum's sheet over angles 0…1 as one face, where
truck made two of half a radian each; the face's own span is angles 0…1 by
profile parameters 0…0.8: √125 · ∫₀^0.8 (5 + 5t) dt = √125 · 5.6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib makes a cylinder's side one face closed round its seam, and a sphere one
face; the tests picked "one half" of the side or "one sixth" of the sphere, which
were truck's pieces. The offset test moves the whole side (one face, area
2π·12·20); the thicken test takes half the side from a cylinder split at x = 0
(π·(11² − 10²)/2·20) and the whole side inward (π·(10² − 9²)·20); the sphere
test takes the cap above z = 5 (solid angle π: π/3·(11³ − 10³)).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
prism_measure's uncertain-material check built a 20 x 20 x 1e-5 box and asked that it get no
cap certificate. SMLib refuses that box (its loops fail SMLib's orientation check): at 1e-5 it
is half a body's tolerance (2e-5) thick, no solid. Refused or built without a certificate, the
property holds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
edge_at (matchsurf) and the mixed fillet mitre check evaluated `e.curve` at parameters from
`e.interval`, which is the interval on the edge's NURBS form when it has one (SMLib's edges
do): they picked and sampled the wrong points. They now evaluate `e.parameters()`, as the fill
helpers do since aa534cc6d's fix on the kernel-geometry branch. (Both tests still fail, for
their own causes: the G2/G3 match leaves a curvature jump of 2 at the seam, and SMLib's mixed
blend lies 4e-7 off its section circle.)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Truck split what it swept: a big arc into two faces, a full circle into halves, a turn into
two or three parts, a sphere into six patches, a torus into six; it numbered boxes' faces in its
own order and sewed stepped cuts slab by slab. SMLib sweeps each profile segment into one face,
closes a full cylinder, sphere or torus along one seam edge, and cuts without slabs. The record,
face source, helix, display mesh, measure and boolean history tests still expected truck's
pieces. Each expectation now follows the geometry, checked by the tests' own independent
measures: every face's origin against the profile segment its mesh lies over (unchanged),
faces picked by where they lie instead of by truck's face numbers, and the merged counts by the
kernel-cases ladder's rule (each whole curve once, and each cylinder's seam): a cylinder 3
faces, 3 edges, 2 vertices; a ball cut from a slab 7, 14, 10. The truck-only stepped-prism
check (walls coalesced across slabs) becomes the property it stood for: the stock's walls stay
whole, one face each, Kept.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Seg2::closest_param starts Newton from the nearest of 65 samples. On a closed segment (a full
circle) the start and end samples are one point and the first wins the tie; for a point just
before the start (in the sweep's sense) Newton then wants a parameter below 0, the clamp holds
it at 0, and the answer is the start point, off by up to a sample spacing (0.15 mm on a 5 mm
circle). The record tests' face-to-segment matching read an SMLib cylinder's side as lying
over no segment because of it. A closed segment whose answer lands on an end now also tries
Newton from the other end and keeps the nearer; every other answer is bit for bit as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
picked_edges_give_the_sketch_s_table picked [E + R, 0, 5]: on SMLib's extruded circle that is the
vertex where the side's seam meets the top rim, a pick either edge answers, and it took the
seam. The pick is now a quarter turn round the rim, [E, R, 5]; the product's pick rule is
unchanged (owner's call).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_face_picked_by_point_on_a_block_with_a_modelled_thread_is_found guarded #6720 (the face search
charged the whole mesh once per face) with a modelled thread's 5 000 facets. On SMLib a modelled
thread is a few exact helical faces, so the block had nowhere near the 1000 faces the test
asserts, and the case no longer reached the old charge. A block drilled with 16 x 16 holes does:
the test now asserts its faces times its mesh's triangles is past the search's 2 000 000 budget
(the old charge) while their sum is well inside it (the new), then picks its top by point.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
intersect_keeps_its_exact_cuts pinned the tilted cylinder's section to 21 segments and one
segment's ends to 1e-4: the crossings of truck's mesh. The cut is the mesh's segments, as many
as the tessellation crosses (20 on SMLib's), each end within the section's reach of the surface,
the display tolerance (a thousandth of the mesh's size, doc::project::fit_segments; SMLib's ends
lie 6e-4 off at most, the test's limit is 0.051). The test now checks that every curve is a
line, at least 16 of them, and every end lies that close to the cylinder of radius 5.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ids_names_suppression_marker_and_failures_survive needs a feature that a parameter change makes
fail, to see the failure survive the script's replay. It made a 1 mm round 50 mm, which truck
refused as too big; SMLib builds a convex round wider than its faces by subtraction (the owner's
overflow decision), so that round now succeeds. A radius of -1 mm fails on any kernel, and the
replay still has to keep the failing feature, its id and its marker.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_whole_sketch_section_stays_whole_through_an_edit compared the ring member's volume before and
after turning its section 30° to 1e-6. kernel::measure integrates at SMLib's finest relative
accuracy, 1e-4 (measure::ACCURACY, the clamp SmApiBrepComputeMassProperties documents): the
exact π (20² − 17²) 600 = 209230.07 mm³ came out 209229.86 before and 209230.43 after. The
test compares to 1e-4; the filled circle it guards against is 3.6 times the ring.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- An SMLib circle edge's `interval` is on its NURBS form (`Edge::parameters`), and an analytic
face's `domain` on the form's (u, v). edges_are_on_smlibs_curves evaluated `curve` at the
form's interval and took the domain for angles; it checks the side's whole turn and height
on the form and a rim's turn by `curve_interval`. The tangent hole test's edge middles were
read the same way (the copy seemed to join faces on opposite sides of the line).
- kernel::chamfer blends the edges it is given; a feature's pick is extended by
kernel::tangent_chain first (doc eval). The chamfer tests do the same:
one_chamfer_pick_follows_the_tangent_chain_at_any_part_size removes its 54.66 mm³ again.
- Since the section hatch per part (#8529) a cut is tinted with the hatch colour its body's
name picks (Body1: steel); the GPU cap test turns the part hatch off, as the CPU one does
(2885f101b on devgreen-engine-builds).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
truck made a cylinder's wall and rims in halves; SMLib makes one wall face and one closed edge
per rim (one face per profile segment, a circle being one closed curve since 09c979a4b). The
tests of what joins a rim's or a wall's pieces for picking, measuring and the agent chat
(circular_rim_hover_click..., circular_items_measure..., curved_face_hover..._cross_the_seam,
a_rim_clicked_once_is_one_item_not_its_pieces) now build their cylinder from a circle drawn as
two half arcs, extruded, so the pieces they test exist. A ring's outer circle is one wall face
(configuration_annulus_picker...: 1, was 2). The certified small crease is no longer taken for
a seam (truck's coarse fit did; the seam flags compare exact surfaces now) and, a kink of a
millionth, is smooth for the shading while still a measured crease: sharp, drawn in wireframe,
picked (smooth_edges::Join).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- The shaft's rim was picked at its polyline's first point, the vertex where the wall's seam
meets it, as near the seam as the rim; SMLib's cylinder lists the seam first, so the pick
took it ("the edge is not circular", "the revolve axis does not lie in the sketch plane").
The rim is picked at its middle.
- The gear discs touch at x = 10, on both walls; G1 is picked on its far side.
- 366b90597 refuses a thread of some length picked in the middle of a face free at both ends
and asks for `from`; the cosmetic thread test gives it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Builds before format 14 named the pieces a cut splits a body into in truck's shell order, and
the later features of such a design name the piece they work on by that name. SMLib orders the
pieces otherwise, so split-body-pieces' fillet and chamfer landed on the wrong piece, its
reference "no longer exists" (design_corpus body_names_saved_before_exact_history_keep_their_pieces,
old_designs_open_and_evaluate_as_saved). For a feature saved so (`shell_piece_order`), each
name its pieces are known by goes to the piece the points the later features naming it picked
lie on; a name no later feature uses keeps SMLib's order, and so do all of them where the picks
do not tell the pieces apart. docs/compatibility.md says so.
That makes the feature's result depend on features after it, which neither its fingerprint nor
its cache key cover: a script replays the design one feature at a time, so the split was
evaluated before the fillet and chamfer existed, kept SMLib's order, and was reused once they
came (script_roundtrip corpus_designs_replay_from_their_scripts: 'Body1 (1)' 4765.663706234679
opened against 4765.663706134907 replayed). The state now keeps a digest of the picks it named
the pieces by (`ModelState::legacy_picks`), and every reuse (the last results, the cache, the
kept, rolled-back and primed results) takes a state for such a feature only while the design's
later features pick the same, so the replay names the pieces again as the file does when it
opens. A saved model's states carry no lineage and are taken as saved.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib's draft extrusion takes one loop, so a tapered extrude of a region with holes was refused
("SMLib draft extrusion of a profile with holes"), and designs that use one no longer opened
(modern-plastic's Rest: design_corpus, action_matrix a_small_hole_through_the_wall_of_a_plastic_rest).
As other CAD does, every wall now leans by the taper relative to the pull on the material's
side: the outer loop is drafted, and each hole, taken counter-clockwise, is drafted the other
way and cut from it, so a taper that narrows the outer boundary widens the holes. Exact: each
loop is SMLib's draft extrusion of it, the walls' roles are found as the foot's are. Refused,
saying why, where the taper would make a hole meet the outer wall or another hole at either
end of the extrusion.
Test: a 40 mm square with a 10 mm square hole, 30 up at 5° inward, is the one frustum less the
other to 1e-6 with ten faces; at 30° the hole would pass the outer wall and is refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
17-loft's expected values were FreeCAD's (OpenCASCADE's Part.makeLoft), which runs each side of
the square to its quarter of the circle at uniform angle (arc length). That correspondence has
no exact rational B-spline form, and SolidCraft keeps its surfaces exact: its ruled loft runs
each side to one rational quarter of the circle's B-spline by their own parameters, 35080.7251
mm^3, 0.13% less (the oracle failed on volume and the second moments). The owner chose this
correspondence. The case is now worked out in tools/oracle-analytic/cases_basic.py: the section
area is quadratic in t with the loops' mixed area, so volume and centroid are closed forms of it
(the mixed area and the wall area by Gauss-Legendre on the exact rational arcs); its note says
why it differs from OpenCASCADE and gives that volume, and its `compare` leaves out the surface
check against OpenCASCADE's reference.stl. All its checks pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
chamfer_the_top_lip_at_a_tangent_wall_junction measured the removed volume as the difference of
two meshes at chord 0.0005. The display planner meshes the two bodies differently, so their
chord errors no longer cancel as truck's meshes did: 37.404, 37.672, 37.761 and 37.759 mm³ at
chords 0.002, 0.0005, 0.0001 and 0.00003, converging on the expected 37.762 (the chamfer as the
chord between the contact lines of a 2 mm ball, integrated exactly). SMLib's chamfer is that
one; the test measures at chord 0.0001.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The display mesh is the tessellation planner's (8b6bfc48c: the mesher plans a schema 4
record), which grids a planar face with no side longer than a fifth of the body's extent
(docs/capabilities.md, "Display meshes"); truck's mesher gave a box face two triangles. The
tests that counted truck's mesh now count the planner's, worked out by hand from that rule:
- a 10 x 20 x 30 box (37.4 mm diagonal): edges in 2, 3 and 5 pieces, faces of 2 x 3, 2 x 5
and 3 x 5 cells, 124 triangles; welded, the box lattice's 3 x 4 x 6 nodes less the
1 x 2 x 4 inside it, 64 positions (obj_export_is_welded_and_keeps_names);
- a 20 x 15 x 10 box: 4 x 3, 4 x 2 and 3 x 2 cells, 104 triangles (an_opacity_class_change...);
- a 20 x 20 x 10 box saved as a mesh: its edges in 4 and 2 pieces, 40 segments per box
(imported_mesh_edges_draw_in_each_style...);
- the cylinder's top cap is under a third of its triangles (select_other_over_a_dense_cluster...).
An imported mesh's outline was one line per crease side, so a box exported at that spacing
showed each edge in pieces: a click on a 10 mm edge measured 5 mm, or hit the node between
two pieces (measure_imported_stl_edges_and_vertices..., mesh_outline_hover_cache...). Outline
sides that run straight on through a vertex no other outline side meets are now one line, as
the CAD model's edge was.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face's edges are projected exactly: lines, and arcs read off SMLib's NURBS within the
planar-edge tolerance (Body::planar_edges). An SMLib shell's offset corner is a NURBS whose
arc came out with its centre 2e-6 mm off and its ends up to 6e-7 mm from the lines they meet,
beyond the link's merge tolerance (1e-7): the inner loop of the sensor enclosure's case top
projected as sixteen loose curves, the offset of one found no chain to follow, and the Lid lip
extrude found no profile l13..a16 (the five acceptance scenarios, action_matrix
an_axis_through_a_corner_round_that_a_parameter_moves). Each edge's pieces now end on the
edge's vertices where they lie within the edge's and the reading's tolerance of them (the
vertex is the point both edges share), and an arc whose ends moved takes the centre nearest
its own that is equally far from both, so the sketch keeps both ends on its circle.
Body::planar_edges_tolerance names the reading tolerance planar_edges uses.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A round face's box (form::extent) unwrapped each periodic parameter from one boundary point to
the next. Through a pole the loop goes on from anywhere round the direction the pole collapses:
a hemisphere bounded by its rim and its seam put its second seam use on the first and its box a
quarter turn off the seam, while its pcurves ran from the seam. Past the first pole the loop is
now moved the whole periods that close it. Reference case 14's re-imported dimple had a
quarter of its area missing (9601 against 9706 mm2).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A field scrolled out of a dialog's field area still published its handle at the centre of its
whole rect. The tapped hole's dialog has grown (thread hand, tap drill, thread end rows) past
1000 px, so its section's Thread depth field scrolls under the footer, and a click on that
handle pressed OK: the edit was applied (tapped_thread_depth_field_keeps_focus_and_typing_...
saw the model change at the click; focusing a field commits nothing). A field's handle is now
the middle of its visible part and is left out when none of it shows, so a scenario can only
click where a user could; the rect keeps its whole extent. The test opens a window tall enough
for the section.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Sketch::bounds took the points of each segment's polyline. Since 09c979a4b a whole circle is
one closed arc whose chords start at its start angle, so none lands on the circle's left or
bottom: a radius 5 circle at (-300, 0) spanned x from -304.50, and Home and Fit framed short
of it (home_and_fit_frame_the_moved_body_not_its_old_place). An arc's bounds now add the
quarter points of its circle it sweeps through, which are exact.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
checked_field, checked_row and warning_caption (the fillet Setback check), text_field_tip (tap
drills), continuity_row, continuity_row_explained and continuity_tip (the continuity choice) and
steps (opening progress) are pub(crate) ui_kit functions with no gallery entry, which
gallery_covers_every_inline_element requires of every one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
help.version (help-menu-version) and prefs.section_hatch* (section-hatch-by-part) came with
ja and zh only; ar and he are complete catalogs too, so
every_complete_catalog_carries_every_english_message failed on them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
be859aa0d added `colors::ISOCURVE` to the engine palette; the GPU viewport draws iso curves in
the theme's accent (`scene_palette`, the isocurve pass), so the engine value is only the
headless snapshot's, like the body and grid colours. every_engine_colour_tag_is_placed lists
it with those neutral tags.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
9a7958857 (placing the model returns its error) passed the legend the whole viewport rect
instead of the bottom overlay rect that clears the floating History card, and slid the
placement note's doc comment in between the legend's comment and its function. The legend is
laid out in the overlay rect again, as the frame rate readout is, and each function has its
own comment back (bottom_readouts_clear_floating_history_on_the_first_frame).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_face_engraved_with_text_projects_its_letters_exactly_and_extrudes counted the letters' pieces
as control-point splines. A TrueType outline's curves are quadratic Béziers, which are parabolic
arcs, and the cut made on SMLib keeps them exact: the projection gives 222 conics and 154 lines,
no splines. The count now takes conics with the splines. (The cut itself needs SMLib 1a1bebc;
with it the test passes.)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
loft_cut_through_a_shelled_basin_in_either_order asserted one body. The loft's top section is
the basin's full 100 x 60 outline at z = 10, so the cut takes the walls' whole cross-section
there and leaves the walls above (z 10 to 30) touching the rest only along their outer rim, an
edge: two pieces, which a cut makes two bodies (eval's lumps rule). SMLib gives exactly that;
truck kept them as one. The test now expects two bodies, the upper one the walls' exact
924 mm² ring times 20 = 18480 mm³, and the total as before (41582.028).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_face_sketch_saved_with_fitted_letters_reopens_and_builds_the_same compared the part with the
5916.018831 mm³ its saving build measured, to 1e-3. The part is the 30 x 20 x 10 block less the
"S" 1 deep and an r 2 circle 3 deep clear of it: 6000 - 46.290121 (the letter's exact profile
area, Sketch2's smaller region) - 12 pi = 5916.010767. SMLib measures 5916.010768 (1.2e-6 off);
the saved figure was 8e-3 off. The test now checks that closed form to SMLib's 1e-8, and counts
the letter's exact pieces as conics or splines (its quadratic Béziers come in as conics).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
map_planar_sources took every plane of a cut's tool as competing with an input face on it,
facing either way, and dropped that face's exact role. A text pocketed 1 mm down from a block's
top has its tool's top on the top's plane facing up, as the top does; the top then lost its
CertifiedExtrudeCap, and the sketch on it reported "the top cap of Extrude1 no longer exists"
(a_face_sketch_saved_with_fitted_letters_reopens_and_builds_the_same). A cut's tool faces enter
the result turned round, as the walls of the hollow, so a tool face can only take the place of a
face on its plane facing against it (a pocket's floor, a face the tool sat on from outside). Only
that orientation competes now; the existing blocking cases (a tool's bottom on a pocket floor, a
tool face against a wall) are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
unsupported_face_has_explicit_fixed_alternative_and_no_live_projection needs a planar face with
neither a construction certificate nor an exact history name, so a sketch on it must be fixed.
Its face, a split top face offset up, now keeps its history through SMLib's offset and gets a
name. A rib reports no roles for the faces it builds (Traced::unnamed, eval_more), so a rib's
wall has neither: the test now picks the wall at y = 12 of a rib in a bracket's corner.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
interference_lists_a_pair_it_could_not_check needed a pair interference cannot check. Its
pair, an arc revolved up to a post's wall, was one truck could not intersect ("not verified");
SMLib intersects it exactly, and that reason is gone from the code. The pairs left unchecked
are a kernel failure, the time budget, and a mesh body: an imported mesh has no exact surfaces
to intersect (owner's call). The ring is now brought in as an STL beside the post, and the pair
is listed with "mesh body: not checked" instead of passing as clear.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
freezing_rebuilt_arc_borders_keeps_authored_dimension_entity_indices needs the projected
border of the sketch's face to come back with a different number of curves. Truck split the
sector's arc into pieces that changed with its sweep (270 to 300 degrees); SMLib keeps one arc
whatever the sweep, so the count stayed 3. A notch, clear of the sector until a parameter moves
it across the arc at 90 degrees, now splits the arc border into two arcs and the notch's sides.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
same_support took two planes as one support only within 4096 ulps (normals 9e-13 apart,
origins about 1e-12). Faces SMLib rebuilds through a boolean lie within its own resolution of
the plane they were made on, not within ulps: a cap at z = 10 notched by a cut came back with
its origin 2.1e-11 off and its frame turned 1.36e-12 rad, so its CertifiedExtrudeCap was
dropped and the sketch on it reported "the top cap of Extrude1 no longer exists".
Supports are now compared at SMLib's own coincidence rule, SmPlane::CoincidenceCheck: normals
within the angle the tolerance makes across the plane (`tolerance / size`) and each origin within
the tolerance of the other plane, the tolerance being the record's `tolerance` (SMLib's per-brep
XSectTol3d, the one its booleans test surfaces for coincidence with). The body's size stands
for the plane's extent (at least as large, so no looser); rounding stays the floor. The normal
window the candidates are searched in follows the same angle.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The truck booleans answered a tool clear of the body without intersecting (the operand kept,
its record shared). The SMLib boolean always rebuilt the body, so a cut that could not reach a
body still handed back a new record: a hole far from a second body reported that body changed
(#2904's test), and a body clear of a pocket's cutter was no longer the same solid. A cut or
intersection whose operands are apart (their exact boxes, grown by twice the tolerance as
union_apart takes them, do not meet) now gives the body itself with its faces and edges kept,
or nothing, with no SMLib call.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- A section's chains are walked from either end, whichever segment the mesh's triangle order
puts first at a node (doc::project::fit_segments), so a line's direction says nothing about
the cut. SMLib's box tessellates in another order and the outline came out clockwise: the
same four lines. The two box-outline tests now compare each line with its lower end first.
- The base rim of SMLib's cylinder is one closed circle edge (truck's was two seam halves), so
projecting it gives one exact ellipse. Checked against the closed form: a radius-5 circle in
a plane tilted 45° projects to semi-axes 5 and 5·cos 45° = 3.5355339 (the reported minor
radius), and every sampled point lies on x²/(5/√2)² + y²/5² = 1 within 1e-9, as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Linking a reference (sketch.include_3d, project, intersect) stored the link's wires as points
only: add_link_unless_present called add_link_with_wires and dropped the resolution's exact
curves, which only the next refresh of the sketch filled in (doc::project, since 7952516eb).
So an included box edge had no exact line until any edit refreshed it, and the wire changed
under an unrelated sketch edit. The exact curves are now set with the wires when the link is
made, as a refresh sets them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two readings of the SMLib record went wrong in emboss_cylinder:
- The footprint arcs were labelled on the sketch's radial frame. The assembly gives each new
circle edge the NURBS form on its label's quarter grid, and the wall face keeps its own form on
its frame's grid (a cylinder's seam lies at an odd angle, build::SEAM_ANGLE). The two grids
differ by a non-quarter turn, so the arc's form is not a straight line in the face's (u, v)
and no exact pcurve exists: "no exact pcurve of a NURBS on a spline surface". The arcs at the
wall's radius are now labelled on the wall face's frame (form::on_frame), so their forms lie on
the face's grid and the pcurve is exact.
- wall_angles took the edge's interval as the circle's angles. On an SMLib record the interval
is on the edge's NURBS form (2πr long for a whole turn), so a wall spanned some 63 rad: a
profile across the seam was accepted, and a wall of radius 0.1 was refused as off its seam.
It now reads Edge::curve_interval.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face that meets itself along a line (the Pinch case's boss side, turned 30 degrees: the hole's
wall touches it along a vertical line from the hole's bottom to the top rim) has its loop run up
that line and back down. Its motorcycle trace runs from the slit's tip to the bottom and the
iso-line grid meshes it, but the band between the grid cells and the boundary was filled by a
parity flood that flipped at every constraint, the slit's too: the band on one side of the slit
was taken for outside, the face did not follow its boundary, and the lenient fill took over.
The band now crosses a side the loops run along both ways without counting it, as the
triangulation itself already did. Delaunay's sag refinement had the same flood and left the far
side of a slit unrefined (the lenient fill's 0.52 mm^3 shortfall); it takes the slits too.
The Pinch case now meshes strictly: the boss side is 628.3177 mm^2 against 200 pi = 628.3185, and
hole_tangent_inside_a_stepped_boss_cuts_exactly passes. New tess test: a square with a slit
meshed by the cut grid follows the slit both ways and covers the square's area exactly.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Where a face meets itself along a line (a hole's wall tangent inside the wall round it: the
boss side touched by the hole along a vertical line, run up and back down), its loop runs along
that line both ways. The triangulation's region test flipped inside and outside at every
constraint side, so crossing the slit put the face's far side outside it; the follow check
wanted every boundary side under exactly one triangle, where a slit has one each side. Sides the
loops run along an even number of times no longer flip the region, and the follow check wants a
triangle per run. (The Pinch case's boss side still falls back to the lenient fill, now within
0.52 mm^3 of the exact 8717.92 where it was 340 short; the strict fill's remaining mismatch on
the seam-cut face is open.)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A ring whose inner circle touches the outer one (a boss standing on a disc, tangent inside its
rim) is one loop through the touching vertex, both circles leaving it along one tangent. There
the inner circle's first node falls on the outer circle's last chord whenever the outer chord is
twice the inner (exactly but for the next order: the inner circle at s rises s^2 / 2r, the outer
chord at s/2 of its length rises L^2 / 8R, equal for L = 2l, R = 2r), as the samplings fell
here. No crossing by the untangle test, but no triangulation can keep both as constraints: the
strict fill refused the face, and the lenient one, dropping constraints, filled the inner disc
(314 of the ring's 942 mm^2, the Pinch test's 1047 mm^3 shortfall). A node of one chord lying on
another (inside it, not at its ends, and not two chords along one line, as a seam's two uses a
period apart are) now has that chord halved. Regression test: the ring meshes strictly, to
300 pi.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillet_an_open_curved_edge, fillet_sharp_arc_and_arc_loop and
fillet_mixed_arc_and_line_removed_volume_converges took the volume a round removes as the
difference of two mesh volumes, which magnifies their sag (each mesh lies within its chord of its
faces, so the difference can be off by both areas times the chord: for the r 0.25 round at a
1e-4 chord 1.8e-3 of what it removes). Measured exactly (SMLib, 1e-8) the rounds are right:
the open curved edge's removal 25.7629198 against Pappus' 25.7629199, the sharp arc loop's against
its lens quadrature, the mixed joint's 2.7155830 and 0.6842666 against the reference 2.7155785 and
0.6842658 (1.7e-6 and 1.2e-6: the blend's own fit, whose section error the joint test reports).
The removals are now exact, the mixed one held to 1e-5 of the reference; the meshes' removal is
held to its chord bound of the exact one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Support::station reads the curve leaving a boundary from five samples: one-sided where the
support ends at the station, central where the surface goes on past it. The central five-point
formulas had the first and second derivatives' signs reversed ((-f0 + 8 f1 - 8 f3 + f4) and
(f0 - 16 f1 + 30 f2 - 16 f3 + f4) over 12h, 12h^2, the negatives of the standard ones), so the
curvature read there pointed away from where the surface bends. Truck's half cylinders ended at
every picked line and read one-sided; SMLib's cylinder is one periodic face, the seam is inside it,
and the central branch read the wall bending outward (+radial / R): the G2 and G3 matches to it
built sheets bending the wrong way, and the continuity check, reading the same frames, called
them G2. The central formulas are now the standard ones.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_split_in_the_plane_of_a_face_keeps_all_of_the_body held the two parts' volumes to the body's
within a millionth, each read from a mesh at a 1e-3 mm chord. Nothing was lost: the parts came
out at exactly the closed forms (780.3707 and 1035.2797 for the 5 mm groove, 1815.6504 together);
the whole body's own mesh sagged 0.0034 mm³ short of that round the hole, twice the millionth.
They are now SMLib's exact volumes (1e-8).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
measure now integrates to SMLib's finest relative accuracy, 1e-8 (da57f37e5, cherry-picked from
the engine-commands branch), not 1e-4: the mirrored bracket's union and twice its half, each an
integral good to 1e-8, agree to 2e-8. (The test stops earlier on this base, at the mirrored
operand SMLib's check refuses: the mirror-regressions agent's cause.)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
measure asked SMLib for a relative accuracy of 1e-4, after SmApiBrepComputeMassProperties'
documented clamp to [1e-4, 1e-1]. smlib_brep_mass_properties does not go through that call: it
calls the kernel's integration directly because the clamp "leaves curved faces off by 1e-5 to
1e-3", and takes [1e-8, 1e-1] (solidcraft_smlib::MEASURE_ACCURACY). Asking 1e-4 threw that
away: a swept cylinder's centre came out 5e-7 off its exact place, and a 0.0177 mm³ crescent
(a pin 0.59 µm off its hole) 0.27% off the lens of two circles. Measured at 1e-8, the sweep
centres are exact to 1e-7 and the crescents match the closed form to 1e-6; the engine-commands
batch runs in the same 20 s.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
cap_support_origin_is_not_a_trimmed_face_height read the drilled block's last face as a cap. The
record's face order is SMLib's: its last face can be the hole's wall, which has no cap loop, and
the test unwrapped None. The cap is now the first face on a plane square to z.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_lip_swept_on_a_round_rim_joins_it compared the union's mesh volume with the two operands' at a
1e-5 chord to 1e-3 mm³. Each mesh lies within its chord of its own faces, and the union is
tessellated by its own faces (the lip's foot merged into the top): two meshes of one solid differ
by up to their area times twice the chord, 0.02 here; they differed by 1.6e-3. The bound is now
that. overlapping_holes_and_mirror_joins held the mirrored union to 2v within 1e-9, below the
relative accuracy SMLib integrates volumes at (1e-4, measure::ACCURACY): the two integrals
differed by 2.3e-8. (That test now stops earlier, at the mirrored operand SMLib's check refuses,
the mirror-regressions agent's cause.)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
faceted_cut_of_a_helical_groove took the volume removed as the shaft's exact volume less the cut
body's mesh volume at a 0.01 mm chord. That mesh sags inside the shaft's round wall: the uncut
shaft's own mesh at 0.01 is 0.18 mm³ short of pi r^2 h, five times the 3 % the groove is held
to, so the groove read 1.326 where the screw integral gives 1.2214 (the section's 2 pi int r dA
over 4 turns plus the ramp at z = 0). The cut is a B-rep: its exact volume (measure) gives the
removed volume within the 3 %.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
faceted_boolean took the tool mesh's triangles by vertex index and refused any edge not shared
by exactly two triangles. A body's mesh (Body::mesh) gives each face its own vertices, the faces
along an edge the very same points, so a box's mesh came in as six open patches: "the mesh is
not closed". Corners at the very same point are now one vertex first (triangles keep their order,
so the history's triangle faces are unchanged).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
check_kept panicked on any edge origin that was not an operand's edge, but a new edge where two
faces meet has a Meet origin by the history's contract (EdgeOrigin::Meet), and SMLib's history
gives them: the raised band's and the thickened sheet's new edges. It now checks the edges an
operand still owns and passes over the meetings. The thickened sheet is the whole cylinder side
(SMLib's one face, truck's half): a wall stands along each of its free edges, the two rims, and
none along its seam, which the face meets itself along (2 + 2 faces, not 2 + its 3 edges).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A layout can leave triangles standing on their long side across the surface:
their corners on it, their plane across it. They count several times the area
under them (the front triangle's loft tubes 0.7% and 1.1% over the finer mesh)
and no split repairs them, since each half stands the same way.
After the sag refinement a post-pass flips a side inside the face when the
two triangles on it turn less from the surface normal the other way (by more
than a degree), keep their orientation in (u, v) and against the surface
normal (nothing folds), have area in 3D, and the new side and triangle middles
stay within the tolerance of the surface or no farther than before. Boundary
sides are never flipped.
Test: tess triangles_standing_across_a_curved_sheet_are_flipped_along_it; the
boolean_regression joined_loft_frames_reference_mesh_covers_each_face_once
case now passes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Refinement split the side of a triangle that sagged most. Where a long
inner side lies on the surface at its middle (a section with an inflection,
a loft tube of the bike frame's front triangle) and the triangle's short
sides sag, only the short sides were split: each split left a child with the
long side and its vertex nearer to it, and the triangle beside a split side
lost its shape the same way. The triangles flattened onto the long side
without end, refinement ran to its point budget, and the mesh of the face
counted 3.4 times its area (1.45 times on a second tube).
A triangle whose widest angle is past 150 degrees now splits its longest
side, and a split never goes across an obtuse triangle's shorter side: that
triangle's longest side is split first, along the path (Rivara's
longest-edge propagation, its lengths growing, so it ends). Only sides inside
the face are split, as before.
Test: tess sag_refinement_across_an_inflection_ends_without_slivers, a sheet
with an S section meshed with a triangle flat on the chord across its
middle; it ran to its budget of 20000 points before and adds 28 now.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A whole sphere as one face has its seam as its only edge, used both ways. In a body many times
its size the seam gets two segments at the body's chord (pole, equator, pole), so no node off
the poles follows a node off them, and the loop kept its first node, a pole: its closing step
crossed the other pole, which the drift unwrapped to a whole turn ("boundary winds around the
surface"). The sphere was left out of the mesh (cut_beside_a_sealed_void_keeps_the_body: a void
r 2 in a 75 mm part, "1 face(s) don't mesh"). The loop now starts, failing a node with an
off-pole predecessor, at a node whose predecessor is a pole on the same edge, so the step into
it is that edge's own and the step into the pole is unwrapped in the loop like any other.
Regression test a_small_sphere_void_in_a_large_block_meshes (tess tests): watertight, no face
left out.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
export_keeps_faces_split_on_purpose took a box whose top Split Face cut in two, dropped the
split record by hand (with_split_lines(None)) and expected the STEP file to have 6 faces: truck's
export joined pieces on one surface, "as it does for pieces a boolean left". No operation makes
such a body: the split record is only set (doc eval, Split Face), never dropped, and removing a
Split Face rebuilds the part without the split. SMLib's booleans join pieces on one surface at
their end (two touching boxes unite into 6 faces), and the local modifiers heal through
boolean_record::healed, so no operation leaves incidental pieces either. The file now has what the
body has (7 faces, the export merges nothing), and the heal every later operation applies
joins the unrecorded pieces: 6 faces in the body and in its file.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib's cylinder side is one face round the whole turn (truck's two halves): a line at z = 5
splits it into a quarter and three quarters of 2 pi 10 x 20, two side faces, not three. (The
test still fails further on, at the face-split tool's "invalid argument", engine-builds'
cause fixed on their branch in 50356dbf2.)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_modelled_thread_in_a_bore_meshes_inside_its_block held each face's triangle area to within 5%
of its surface's area. At the 0.1 mm tolerance, face 18 failed that: 1.777 mm^2 of triangles
over 1.522 mm^2 of surface. Face 18 is the groove's crest band where it runs out through the
block's end, a wedge at most 0.44 mm wide. Its boundary sides are the helix edge's chords, 1.37
mm long and sagging 0.062 mm (inside the tolerance). A flat triangle on such a chord bridges the
sail between the chord and the edge instead of following it, and the band is no wider than the
sag. The excess, 0.255 mm^2, is less than the sails along the face's boundary (sum of 2/3 L sag
= 0.62 mm^2). Every triangle is within the tolerance: the chord check passes. No triangulation
of the inside changes this; the boundary nodes are the shared edge nodes.
The check now allows 5% plus the boundary's sails at the tolerance, 2/3 L tol per boundary side.
A zigzag that multiplies a face's area, which this test exists to catch, is still far outside
that. On the run-out root band the bound allows about 40% at 0.1 mm and 5% plus 4 mm^2 at
0.01 mm, against the former 6x.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tapered_extrude_of_lines_and_arcs expected the quarter disc (its arc meeting both lines at a
corner) to be refused, "arc meets another edge at a corner". That was truck's limit. SMLib's
curvature-keeping taper builds it as mainstream CAD does: each wall continues as itself until it
meets its neighbour. The test now checks what was built. The section at d = z tan 8 deg is the
disc of radius r + d over x >= -d, y >= -d. Its area is in closed form, and the volume is its
integral by Gauss-Legendre: 458.5870559 against SMLib's 458.5870567 (2e-9). The body has 5 faces.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
thread_display_tests sewed its M10 and M12 threads from cubic Bezier pieces approximating
helices, which lie about 1e-5 mm off the core and crest cylinders. No exact (same-parameter)
pcurve exists on the cylinder for such an edge, and SolidCraft's pcurves are exact B-splines
only. Sewing refused the core face ("no exact pcurve of a B-spline on a spline surface"). Truck
had fitted one. The thread feature makes its helices exactly, with SMLib's screw sweep, so the
test now does the same: the ISO ridge's section is screw-swept and joined to the core cylinder.
The section reaches a tenth of a pitch into the core along its flanks. ridge_volume is still the
thread's exact volume, and both display tests run unchanged, from the body as made and after a
STEP round trip. Imported edges that lie off their surfaces stay refused; they are Repair Part's
job.
The quality check's sag is now measured at the surface point nearest each triangle's centre.
Before, it was read at the mean of the corners' parameters. A closed surface's parameters are
kept within one period, so a triangle across the core's seam had corners a period apart, and
their mean read 6.27 mm off a triangle that lies on the surface.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
isocurves_of_a_split_face_run_on_across_the_split split the box's front face at x = -2
and 22, leaving an 8 mm piece. Since a body's size counts its edges (the boss's rims
reach y = -32 and z = 50), the lines' spacing grew to 10 (x = -20, -10 on the left
piece), and the 8 mm piece had no line up it to compare. The boss now leaves pieces
x < -6 and x > 18, room for two lines and one: the test still checks that both pieces'
lines lie on one ladder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Body size (the box diagonal every size-relative tolerance comes from: mesh chords, reach,
fit tolerances) came from the record's vertices alone. SMLib closes a sphere, a cylinder or a
torus along seam edges, so those vertices can be two poles, two seam ends or a single point: a
torus was 1 mm across (its preview at a hundredth of that took 214 720 triangles and broke the
budget), a 5 x 8 cylinder 8 instead of 16.2. The size now also takes three points along each
edge: points on the edges, not their control hulls, which for a rational edge can lie far off
it (a sheared L-prism's outline box was 49 km across).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
sample_aliasing_does_not_certify_spline_caps built its trap as one polynomial of degree 17 with
a root at each of the 17 old sample locations k/16 (and a circle times such a polynomial, degree
19). A B-rep record holds splines up to degree 16 (MAX_SPLINE_DEGREE, #8618), so no record can
carry either and the evaluator refuses them: the test failed before testing anything. The trap
is now a C0 spline with a knot at every k/16: a quadratic on the line at each knot, bowing 1 mm
off it half way along the first span; and the quarter circle of radius 10 with its homogeneous
coordinates scaled by 1 + bump / 10, radius 10 at every knot and 11 at 1/32. Sampling at the old
locations still sees a line and a circle; the certifier, reading the spans' control points,
must certify neither, and doesn't.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
genuine_curves_and_oriented_traversal_are_certified built its degree-elevated line at degree
17. A B-rep record holds splines up to degree 16 (MAX_SPLINE_DEGREE, Brep::validate, #8618), so
no such edge reaches the certifier, and the evaluator refuses to build it. The line is now
written at MAX_SPLINE_DEGREE: still a line of the highest degree a record can carry.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The smooth-path sweep named walls only by the profile's edges in the start plane and took any
planar face nothing claimed for the end cap. A wall the start plane meets along no edge got
no name or became a cap: round a closed path, a square tube's planar top and bottom (whole
annuli, no seam) were both called end caps; along a line then an arc, the walls SMLib makes
along the arc were unnamed or caps. Like the cornered and controlled sweeps, it now also names
walls by the profile carried half way along, takes the end cap from where SMLib carries the
profile last, and round a closed path has no caps at all.
The issue #2005 tube check counted V - E + F without the ring term: SMLib's annular planar
faces have two loops each, so it checks V - E + F - (L - F) = 2 - 2g (Euler-Poincare). Along a
line then an arc SMLib makes a wall per profile segment per path segment (two, where truck
made three).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
transform_matrix moves a body by a matrix that is not a rigid motion through SMLib, which
numbers the moved body's faces afresh, then attached the input's face sources and paint in the
input's face order. Planar certificates landed on the wrong faces (a reflected, sheared plate's
hole walls and caps no longer lay on their certified planes). Each moved face now takes the
source and paint of the face its history names; a face with no known origin takes no source,
and with no history at all the paint stays as it was.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
efd0a577b kept the cap flatness tolerance under a quarter of the gap between parallel start
and end planes. A half turn's end plane is its start plane: the gap is 0, the tolerance became
0 and neither cap was found, so m0-revolve-primitives' two flat ends could keep no sketch
attachment (design_corpus every_flat_face_of_every_corpus_design_takes_an_attached_sketch).
Only planes apart by more than rounding bound the tolerance; coinciding ones are told apart by
the profile's edges, as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The roles of a swept body's faces take a face as a cap when all its points lie within a
flatness tolerance of the start or end plane, sized to the body (1e-5 of it). A box 0.1 µm
thick in a 14 mm body is thinner than that, so each cap lay in both planes and box_solid
refused "box role correspondence" (engine component_clone
boundary_material_proof_requires_resolvable_stock_and_void). Between parallel start and end
planes the tolerance stays under a quarter of their gap; a cap's points lie on its plane to
the evaluation's precision. Regression case: a 10 x 10 x 1e-4 box in
box_certificates_are_six_construction_roles...
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each wall of a swept body is known by a seed on its piece's edge in the start plane; every
start-plane edge within `near` of a seed took its piece, `near` sized to the body
(3 x 2e-4 of its size). On a side shorter than that the neighbours' edges are near too: a
0.3 mm wall of a 1 m label is 0.15 mm from them, near 0.65 mm, so the long walls got two
pieces, had no source, and box_solid refused "box role correspondence" (gpu depth
at_ten_metres..., engine component_clone boundary_material_proof); a ring of radii 10 and
9.999 lost its side certificates (ui configuration_annulus_picker...). A seed now marks only
the edges nearest it (within the flatness tolerance of the nearest), still within `near`.
Regression case: a 1000 x 0.3 x 400 box in box_certificates_are_six_construction_roles...
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A chamfer picked at (10, 0, 20) on a cylinder's top rim lies on the vertex where SMLib's
cylinder has its seam. The blend takes the rim there (blend_smlib::resolve: a seam or a smooth
join loses a tie), but the name stored with the feature came from the nearest mesh edge, the
seam (`F2:side:c1`). Once the stem grew, the name found the seam and SMLib refused the chamfer
(join_extrude_from_a_chamfered_top_face). edge_names_for names the edge each pick takes as a
blend takes it (kernel::blend_edges), the nearest edge only where that finds none.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
face_patch put the tool face's surface (a cylinder) over the domain of its NURBS form, which
SMLib parameterises otherwise (0..1, not the angle and height), so the patch handed to SMLib
was a sliver of the cylinder or empty and SMLib refused it as an invalid argument
(split_body_by_a_face). The domain is the analytic surface's: a whole turn, unbounded along
the axis, closed past the body as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
edge_circle took a circle edge's interval as its angle. A circle SMLib holds on its NURBS form
is parameterised on that form (a fillet's quarter circle over 0..1), so the fillet's edge
measured 57.3° and an arc length of R·1 (a_fillet_edge_is_an_arc). The span is the edge's
curve_interval, the angles of its ends on the circle, as the lip already reads it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The thread forms (b4549d270) parse designations in thread_forms, whose refusal for a plain
M28 said "give it: M28x1" and, for M0 or M1e9, still offered a pitch. The hint the hole
command gave (the two coarsest ISO fine pitches that suit the diameter: M28x2 or M28x1.5) is
now thread_forms::no_coarse_pitch, used by both; a diameter no metric thread has is an
unknown metric thread. The test's "unknown thread" follows the forms' wording ("unknown
metric thread").
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
#5191 read the header's preprocessor version and FILE_DESCRIPTION, so a SolidWorks file named
only there was stood up. The reader had moved to crates/step (p21.rs) before #5191 merged,
and the merge kept only the originating system: a file naming SolidWorks in its description
got no exporter_up (step_replies_name_a_y_up_exporter). The parser reads both fields again,
StepImport carries them, and exporter_up looks at all three.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The normal equations of patches about a centre were solved dense, so at most
500 free control points: the torus window, 1433 once its patches ran in their
own spaces, was refused. Each patch's interior (coordinates of points only it
holds, coupled by no row to another patch's) is a banded block now; the blocks
are eliminated exactly onto the unknowns the patches share (the spokes, the
centre, and the points rows across a spoke couple), whose Schur complement is
solved dense, and the interiors follow by back substitution. A solve takes up
to 6000 free points, as one patch does, and refuses past 4000 shared unknowns
(a 128 MB dense block) with that reason. The torus window and the pentagon now
solve (they fold, open).
The fill tests' own-edge check compares an edge's NURBS form to rounding, as
fill.rs's does.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
About a centre, every patch ran in one space both ways that held the breaks of
every half edge and spoke, so each patch carried every edge's knots: a window
cut in a torus, its edges SMLib's traced intersections with many knots each,
needed 13537 free control points. Patch k now runs along u in the space of
spoke k − 1 and along v in that of spoke k; the space along a spoke holds the
spoke and the two edge halves it meets beside its two patches, so the spoke
is one row of both, as before. The torus window needs 1433 (still over the
dense solve's 500, open). Refinement about a centre adds the middles of one
uniform grid to every spoke's space; a patch's conditions read its own
spaces' spans, and the stations along spokes the finest of them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The fold check refused a patch whose normal, anywhere on a grid, was a right
angle or more from the normal at its middle. A fill may turn its normal that
far without folding: a dome that meets a tube's wall tangentially is vertical
at its rim (90° from its top), a window round more than half a cylinder turns
past 120°. A fold is the surface turning back over itself, which reverses the
normal between neighbouring points of the grid (24 × 24, 3.75° steps on a
half cylinder); that is what is refused now, and a normal that vanishes
inside the patch. It may vanish at a corner where two sides run on straight
(a closed loop's pieces) and on a three-sided patch's pole.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A cup's rim of seven or eight ellipse arcs meeting at corners was too wavy for
the height-field fill: its finest grid, 32 spans across, stayed 6.4e-5 off the
loop where the tolerance is 4.4e-5 (a smooth bicubic follows the height's kink
at a corner only as closely as its cells are small), and the loop went to the
general fill, seven faces where one fits. The grids now go to 64 spans (the
fit's system is banded); the first within tolerance is still the fill.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
face_tolerance (and planarity_tolerance) was 10⁻⁶ of the body's bounding-box
diagonal, which changes as a body turns: the cup turned 0.7 rad about
(1, 2, 3) grew from 54.7 to 61.6, its fill sampled to another tolerance, a
different surface (area 974.543 against 974.527). The size is now the
diameter of the smallest ball holding the vertices (solidcraft_geom::Ball,
Welzl's algorithm, expected O(n)), the same wherever the body is and however
it is turned. Twice the farthest vertex's distance from the vertices'
centroid is that too, but up to twice the body's diameter where the
vertices crowd one end: a rounded box's set-back-13 G2 corner patch then
ran 2.0e-3 past the box (corner_g2
a_long_setback_keeps_the_g2_patch_inside_the_box, limit 1e-3).
The own-edge check of the fill tests compares an edge's NURBS form to
rounding (10⁻¹² of each coordinate): SMLib makes a conic's form again on a
round trip, the last bit of a coordinate apart.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The iso fill needs three sides or more, so a loop of one closed edge (a tube's
end) or two was refused as "a loop of 1 edges". Such a loop's edges are now
taken in equal pieces of their own parameters (one edge in four, two in two;
a break that falls on one of the curve's knots, a conic's quarter points, is
that knot exactly), each piece the edge's own curve segmented exactly; the
patch's edges are those pieces, their ends new vertices shared by the pieces
either side. A piece of an edge the loop runs against has its parameter
reflected about the edge's interval, as a whole edge's is.
The tube end is still refused: one four-sided patch on four quarters of a
circle has corners where its sides run on straight (180°), and the G1 dome
folds there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A record's tolerance is SMLib's intersection tolerance: SMLib holds an edge's
gap to its faces to half of it (the zone tolerance) and a vertex's gaps to a
quarter (SM_XSECT_TO_ZONETOL3D, SmBrep::ValidateAndUpdateTolerances). The fill
gave its edges and vertices the face tolerance itself, the gap its surface is
certified to, so SMLib's check refused every fitted patch (an edge/surface gap
of 2.6e-5 against 1.85e-5 where the fit holds 3.7e-5), alone or stitched to
its body. Its edges now take twice the gap and its vertices four times.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A ruled strip off an arc whose rulings keep their angle to the arc's axis
(normal or tapered mode on a 30 degree lean) is a piece of a cone, but was
built as the rational B-spline between the arc and its image, which SMLib
labels with no analytic type (its CopyAndAddAnalytics only sees the plane
strips). Such a strip is now the start ruling turned about the arc's axis
through the arc's angle by SMLib's revolve, as the round corners already are:
a cone face, the same surface. a_circle_in_every_mode passes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Kept parallel (the profile as drawn), a profile flattens to nothing where the
path turns to run along its plane. The check that refuses it (the path must
cross the plane by 5° or more all the way) lived in the stations of a
controlled sweep only; a plain parallel sweep went to SMLib unchecked and
built a degenerate solid. Both take the same check now, the plain sweep on
each segment's direction of travel at 33 points along it.
A circle profile is one closed arc, one edge in SMLib: a wall per path line,
not two.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Going into SMLib a sheet's shell holds both sides of each face, so the side the
record keeps as the sheet's front is lost, and SMLib thickens along the
surfaces' natural normals: Reverse Normal did not change which way a thicken
grew. A sheet used against its natural normals everywhere is thickened by the
negated thickness.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A record edge on a circle or ellipse carries the NURBS form SMLib holds it as,
its interval on that form. The patch built its loop edges from the form alone
(sew::PartEdge had no place for it), so a fill's rim came back from SMLib as a
plain B-spline rather than the body's ellipse. PartEdge now carries the form;
assemble keeps a given form and labels the edge with what it is.
Test helpers that sampled an edge on `curve` with the form's interval picked
points off the edge ("no edge at …"); they sample `parameters()`. The fill's
own-edge check compares the curve's kind and the exact form it is parameterised
on: SMLib renormalises a conic's frame axes in the last bit on a round trip.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib can give a NURBS edge an interval that starts or ends a rounding past its
curve's knots: the end of the variable round r1 -> r5 run on into a box corner's
R5 round (ladder 07-fillets/012), trimmed by the R5 cylinder, started 7e-15 before
the first knot, and the record failed validation ("parameter interval outside
the natural domain"). Such an interval is moved onto the knots, as a face's
domain side is onto its surface's natural range (`onto`); farther past them is
still an error. SMLib 7e917f0 (c0d74ee..6e77631: the trace's reach set before
the corners) moved these ends by 1e-11, which is how the case regressed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Ten cases the pin fixed leave the list (the 1207 corner patch and its together cases, 1390,
4782 reduce-fit, 896). 1557 joins it with its cause: since bodies are read back from their
records (#8625) its bottom rounds come back as SmCylinder where SMLib's blend made SmCone(0),
and the corner-ball trace starts 2.78e-4 off; its fix is in progress.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_shown_dimension_that_breaks_a_later_feature_reports_it_and_undoes took the
plate's 10 width to 3 under an r4 round of its back top edge, expecting the round
to fail. At the 7e917f0 SMLib pin a convex round wider than its face runs over the
face's far edge and is cut as if the face went on (the owner's overflow decision),
so it builds and the test failed. The test is about the error a dimension change
reports and its undo, so the broken feature is now a 2 thick shell open at the
top: 3 wide leaves no room inside its two walls, a refusal made before SMLib.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
#8676 added FeatureKind::Repair without a fixture, so
feature_walk_tests::the_record_covers_every_variant failed. The fixture sets
every field (body, gap, patch_holes); the records, written with
FEATURE_WALK_BLESS=1, show its one input (Gap, a length) and that copying with
bodies renamed rewrites its body.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
At the 7e917f0 pin SMLib builds the r50 round of a 20 cube's edge that
hostile_inputs_are_errors expected to fail: a convex blend that overflows its
faces' far edges is built by subtraction (blend on the faces extended, then cut),
as the chamfer past the faces two lines below already is. The body is exact: the
two ends, the two far faces and the round (5 faces, 9 edges, 6 vertices), volume
20 (F(40) - F(30) - 300) = 1094.85273 with F(u) = (u sqrt(2500 - u^2) +
2500 asin(u / 50)) / 2, the cube's section inside the r50 circle about
y = z = -30. The test checks that answer instead of an error.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The arc part's front edge rounded r2, then its left top edge with tangent
propagation (the Fillet feature's default): that edge runs on round the first
round's end arc and down the left front upright, so three edges end up rounded
(a sphere octant at the corner where they meet), not two. Both results were
exact: 9530.77 (meshed) is the part less three r2 strips and the octant,
9591.44 − (0.858407·(39.1667 + 19 + 8) + 8(1 − π/6)) = 9530.83; 9538.79 is
two strips less their overlap, 9591.44 − (0.858407·62.1667 − 0.7670) =
9538.84. The test compares the sequential rounds with the same edges of the
part rounded together: the picks' chains and every edge of the part the second
chain carried on into.
The other pairs still fail in SMLib (1001, or SM_ERR_INVALID_INPUT at the
knife-edge tip) where a later round ends on the first round's arc.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Fillet and Chamfer features already blend each picked edge's tangent chain
by default (tangent_chain, true unless saved false), as mainstream CAD's
tangent propagation does; the kernel blends exactly the edges it is given.
#8678 refused, before SMLib, every edge that runs on tangentially into one left
out, which also refused blends SMLib builds (a round ending on the round its
edge runs into: blend_corner's a_round_next_to_a_round_of_another_radius, the
same feature with propagation off). Now:
- propagation off: only the picked edges are blended; where SMLib cannot end
the blend at such a continuation, its refusal names the edge it runs into
and that propagation (tangent_chain: true) blends the chain as one;
- propagation on, where the chain cannot be built (an r3 round carried on
round an r2 round's end arc), the refusal says the chain was blended and
that with propagation off only the picked edges are.
A smooth edge is still refused before SMLib. No default changes: the feature's
default was propagation on already.
The kernel tests that stand for a Fillet feature (curvecorner's sequential
rounds, issue 2422) take each pick's tangent chain, as the feature does by
default.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A smooth edge (its faces meet tangentially) has nothing to round, and an edge
that runs on tangentially into one left out has no end SMLib can blend (it
rounds a tangent chain as one sequence and failed building the face there,
status 1001): both are refused before SMLib with that reason. SMLib's new
SMLIB_ERR_CORNER_UNSUPPORTED (a vertex no blend corner handles) is
Error::CornerUnsupported and says so instead of a bare status.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tangency-labs/smlib#11 (the Coons corner patch is tangent to its sides' surfaces), #12
(SMLIB_ERR_CORNER_UNSUPPORTED for a vertex no blend corner handles) and #13 (a torus found in a
NURBS has its major radius from its origin and a whole turn of its tube). Prebuilt hashes written
by scripts/fetch-smlib.sh --pin.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Patch dialog's continuity row keeps each order's hover line and adds one: the patch meets
the faces round it at the chosen continuity along each side, and at a corner where those faces
meet each other with different curvatures it stays tangent (G1) there. New text in
assets/locales/pending/patch-continuity-corners.json, in English and the complete catalogs
(Arabic, Hebrew, Japanese, Chinese), German and Korean.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Four continuity cases for the patch fill. A hole in one smooth sheet (a cubic Bezier profile
extruded, a square trimmed out) patched g2 and g3: every seam reaches the order, corners
included, since round the hole there is one surface. The hole a three-round corner's blend
face leaves, patched g2 and g3: exactly g1 over each whole seam, since at the hole's corners
the rounds meet each other G1 (their curvatures differ by 1/4 or 1/8 per mm) and a patch smooth
there has one curvature; the middle of each seam reaches the order (kernel test).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The patch fill meets its neighbours to G4, so Construction::Patch now offers g0 to g4: the
surface.patch command takes g3 and g4 and the Patch dialog's continuity row offers
"Curvature change" and "Smooth curvature change" by itself, from the one table.
The three-round corner test patches G2, G3 and G4 and reads the seams with the Continuity
Analysis too: the middle of each seam (30-70 %) at the order asked; the whole seam G1, since at
the corners the rounds' curvatures differ and no patch can meet both.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
Two command scripts in examples/clean-up: a round along a plate's edge running into a boss
wall, and a corner where three rounds meet. Each takes the blend's messy face off (face.delete
heal false) and patches the hole with Curvature (g2) and, beside it, Curvature change (g3).
Gauss-Legendre constants at f64 precision.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The corner blend of a box's R8 vertical round and R4 top rounds taken off and patched: both
patches meet the tangent planes and the curvature in the middle of each seam; the G3 patch
also meets the curvature's rate of change (4e-6 to 4e-5 /mm^2), which the G2 patch leaves at
0.27 to 0.63. G3 to an edge between two faces is refused; G3 is no longer refused as such.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A fill asked G^n is faired with the thin plate plus the energy of order n + 1 (each to unit
trace), so the change of curvature its conditions ask at the edges spreads over the fill
instead of being squeezed into the first span; Gram matrices on six-point Gauss-Legendre,
exact for quintics.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Rows across a patch's C0 knot lines (a side's break where it is only C0 in its parameter, such
as a circle's quarter joint in homogeneous form): the one-sided derivatives across in the
ratio of the crossing sides' speeds there, blended between them. Without them the fill rippled
along those lines (zebra stripes in steps) and SMLib's surface check refused it.
- Space::basis_left: the basis of the piece ending at a knot.
- Orders above a corner's agreement come back smoothly over the next share of the side
instead of switching on at once.
- Orders above G1 are best effort once refinement is spent: the fill is returned with its
jumps reported; the tangent plane, the points and an unfolded fill stay hard.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- One patch has a space along u and one along v, each holding its two sides that run that way;
a side's knot keeps its continuity when raised (multiplicity m + q - p), and each side is
carried in exactly by interpolation at the space's Greville abscissae.
- Refinement halves the spans only in the direction whose sides miss, and one patch's normal
equations are banded (the shorter direction numbered fastest), so a loop with a long
intersection curve refines where it needs to.
- At a corner whose neighbours agree to a lower order than asked (their tangent planes, their
twist between the two edges, or their forms), the stations near it ask no more than that.
- A station is held to the angle tolerance beyond the edge's own lean out of its neighbour's
tangent plane: a fill that holds the edge cannot come nearer.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
One order-n implementation in the linkage-curve formulation (Pegna & Wolter 1992;
Hermann, Lukacs & Wolter 1999): geom::linkage reads the neighbour as a height over its
tangent plane to order n (its jet, from the surface's exact partials of any order) and
gives each order's condition on the linkage curve, with exact gradients (dual numbers),
so fill_iso solves orders 2..n by Newton's method. Each order's row is weighed by h^(k-1)
and its jump is held, carried into the fill by Taylor's theorem, within the angle tolerance.
The iso fill takes rational edges (a circle's arc on its NURBS form): homogeneous rows,
each edge scaled so the corner weights agree, the inside weights the Coons blend and
fixed, so the fill stays a linear least-squares problem. Its degree is at least n + 1,
refinement adds simple knots, breaks of different sides within rounding are one, and the
weak Hermite rows act in the tangent plane at the Coons start's own speed, so they no
longer pull against the G1 rows.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
The command worked the repair out itself for its reply and the feature worked it out again
when it evaluated. The evaluation now keeps the kernel's report on the feature's result (as
an import keeps its import report, cached with it), and the command's reply is built from
that: one repair per call, and a part with nothing to repair fails the evaluation, which
leaves the design unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A new undoable feature (`repair`, one pending format file) and command with input and
output schemas. In order:
- A cylinder, cone, sphere or torus face SMLib refuses on its own is rebuilt on a NURBS
form of its surface laid on arcs from where the face starts
(`form::surface_on_own_arcs`), with SMLib's own trims: the import's quarter-grid form of
a face round most of a turn closes on itself, which SmTorus::IsNurbSurfaceTorus does not
recognise. Geometry unchanged.
- Free edges that pair up within `gap` are sewn by SMLib's stitch; GlueEdgesGeneral gives
each glued edge its own tolerance (at least twice its gap), the body's and the other
edges' stay. The result is checked against the input's own SMLib check, the
tolerance-bound tests (#5, #6, #59, #82) exempt.
- Loops of free edges left are holes: offered as surface.patch, and with patch_holes made
(flat only when the loop is planar and every face around it is a plane, else G1) and
sewn in.
The reply reports what changed face by face; nothing to do is refused with why.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
allow_errors_writes_anyway_and_a_saved_design_keeps_failing compared the box's volume
with 4000.0 to the bit, which the old kernel's closed-form planar volume gave. SMLib
integrates mass properties numerically (relative accuracy 1e-4 asked for) and gives
the 20 x 20 x 10 box within a unit or two in the last place (3999.999999999999), the
correct answer to the precision the measure has. The test now checks 4000 mm³ to a
relative 1e-9.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
a_combine_tool_stays_where_it_is_placed makes a part and then moves its occurrence
to check that a combine tool stays where it is placed. Since #2804 the design's first
part is grounded when it is made, so the occurrence.move refused and the test never
reached its question. The part is now made with grounded: false, as a part that is
meant to be moved is; the checks are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A body made by an SMLib operation kept that SMLib body beside its record (SMLib's
export). The export is not exact to the bit, so the kept body measured and operated a
little differently from the record's import, which is what a body read back from a
file has: a box measured 24000.000000000004 mm³ live and 24000.0 reopened, a drilled
box 21989.38056 against 21989.38013. A cut ring's volume therefore changed when it
was moved, the moved ring being measured on the SMLib body the move made
(measure_centre_tests::combine_cut_volume_does_not_depend_on_placement). Every body
now imports its record when an operation first needs it, so evaluated and reopened
bodies are the same.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The GPU path tests (and the Windows FXC job, cargo test -p solidcraft-gpu) build real
designs through solidcraft-engine, a dev-dependency without its smlib feature. Run on
their own (-p solidcraft-gpu), nothing in the build turns SMLib on, so every geometric
step answered "this build has no SMLib" (10 path::* tests on develop). The
dev-dependency now asks for engine/smlib, as the apps do, and both jobs fetch SMLib
before building.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillet_into_an_earlier_round picked the box's upright after rounding its top edge
and expected only that edge rounded. Since solid.fillet rounds each picked edge's
tangent chain by default (84bd2daad), the upright runs on through the first round's
end arc into the top right edge, and that is rounded too. The two-radius checks now
ask for exactly the picked edges (tangent_chain: false), and a new check rounds the
chain and expects what a round of the 38 mm top right edge takes off as well.
ribs_flush_with_a_window_change_nothing_away_from_them compared the lid with
6703.5222 mm³, the old kernel's chorded volume. The lid's closed form (the plate
with its rounded corners, the skirt, less the window, the slot and the notch) is
6703.5344 mm³, which SMLib gives; the test computes it, and the script's
description says it. Each rib's volume and area are checked to SMLib's mass
property accuracy (2e-7 relative) instead of 1e-6 absolute.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three expectations encoded the old truck planar booleans' incidental output;
each is replaced by the exact answer:
- a_base_trimmed_for_a_flange_keeps_its_arcs: the flat outline has 10 edges,
not 12. The flange's bend zone and panel meet in line, and the union of
the two is one line along each side (a region's boundary has no vertex in
the middle of a straight run). Lengths and areas are unchanged and still
checked to 1e-9.
- a_base_trimmed_through_a_free_form_curve_follows_it_with_arcs, renamed
..._keeps_it_exact: the trim cuts the parabola exactly, so the base keeps
one conic and three lines instead of tangent arcs within 0.001 mm. The
strip (183.564 = ∫ y from 197 to 200, or 600) and the bend's length
(62.364) are now checked to 1e-6 and 1e-9 instead of 0.22 and 2e-3; the
chorded comparison stays.
- pockets_with_free_form_outlines_are_machined_beside_free_form_cut_outs:
the machining record holds the pocket's four exact conics in their order,
started at another of them; the check is for the same closed loop, not
the same first segment.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Reading a planar sheet back from SMLib split each loop where two of its
segments started within ten times the record's tolerance (2e-5 mm), so
the two ends of an edge shorter than 2e-4 mm counted as one point and the
edge came out as a loop of its own, which the next planar boolean refused
("needs closed loops"). A loop passes a point twice where it passes one of
the record's vertices twice; the split now reads the vertices, not their
places. The sheet built from our own regions still splits by place, at ten
times the caller's tolerance.
Fixes sheet::tests::a_ledge_beside_an_edge_as_tall_as_the_setback_goes_with_the_strip
(the ledge 1e-4 above the bend line); new kernel test
a_ledge_shorter_than_the_record_tolerance_reads_back_as_one_region.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The SMLib planar sheet united every region's material and then cut every
region's holes from it, so a region lying in another's hole (the island
inside a ring cut) was cut away with that hole. Each region is now its own
material less its own holes, and the sheet is the regions united.
Fixes sheet::tests::ring_cut_retains_exact_central_material_island (the
flat solid lost the island's π·25·2.5 mm³) and the kernel's
compound_planar_extrusion_normalizes_overlap_and_retains_islands.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two tests rebuilt the old kernel's raw strips (a shell's wall split along its
opening, a stepped tab's -x wall split by the Join and the Cut) with today's
kernel and took the saved piece's name and projection from them. SMLib leaves
those walls one face, so there is no strip to rebuild: the tests now say so (the
wall is one face) and give the saved link as what an older file holds, its piece
name and its rectangle (y 10..20 x z 0..20 on the -x wall, the same 4-line
projection the strip had), and keep every check of what the model does with it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The SMLib port of shell dropped with_shell_planar_sources (then deleted it as dead
code), so a shelled body's kept walls and their inner linings had no ShellPlanar
certificate: a sketch or press-pull on a shell's floor lost its attachment (the Pi
case's Sketch2 and everything after it, press_pull_on_the_pi_case_floor) and the
shell certificate tests found no inner/outer face. The roles come from the input's
exact planar supports, so they apply to SMLib's result unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Mesh::ray_crossings and InsideIndex counted a triangle crossed by a
Möller-Trumbore test with half-open barycentric bounds (u in [0, 1), u + v < 1).
Those bounds depend on each triangle's own vertex order, so on an edge two
triangles share the ray was in both or in neither: the bracket example's hole
centre lies on a diagonal of the base's top triangulation and was counted 0 or 2
times instead of once (bracket_example_cuts_two_through_holes_and_reopens).
Both now use ray_crosses, a watertight test: each edge's side of the ray is the
signed volume dir · ((p - o) × (q - o)), which is exactly antisymmetric in
floating point, so the two triangles of an edge see it with opposite signs, and a
ray exactly on an edge's line is put on one side by the edge's endpoints in
lexicographic order. New geom test ray_crossings_count_a_shared_edge_once crosses
a grid at every inner edge, straight down and obliquely.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The seeded robustness goldens (tests/robustness/goldens.json, tests::robustness_goldens) recorded
what truck built for seeds 0-95, so they now fail on every seed SMLib builds differently. The
kernel-cases ladder (tests/kernel-cases/ladder, cargo xtask kernel-cases --ladder), whose cases
each carry their own right answer, supersedes them.
Gone with them: fuzz::golden, the CI goldens lane (the goldens output, SLOW_SEEDS, the re-record
hint), their nextest filter and priority entries, the mutants sample and the tests/robustness/
test-data mapping in affected.py. The robustness seeds and tests::robustness stay. The docs point
at the ladder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tess.wgsl sized its basis arrays for degree 7 by hand. It now holds 17 slots, degree 16 as the
CPU evaluator, and a test fails when the two drift apart again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A G4 blend curve is one Bezier of degree 9, and the face extruded from it a spline surface of
degree (9, 1). The record accepts splines up to degree 16 (Brep::validate), but the evaluators
behind pcurves, tessellation and Continuity Analysis capped the degree at 7
(MAX_SPLINE_DEGREE), so the SMLib export refused the face with "pcurve: not supported yet: a
spline surface of degree (9, 1)". The cap was not an algorithmic limit: every routine under it
is the general NURBS Book algorithm, sized by fixed arrays of MAX_SPLINE_DEGREE + 1 slots.
MAX_SPLINE_DEGREE is now the record's own bound, 16, and Brep::validate reads the same
constant. The display evaluator's first-derivative basis (tess::basis_ders) works on two
one-dimensional tables (A2.2 at degree p and p - 1, eq. 2.7) instead of a square one, as its
GPU twin does, so the larger bound costs a low-degree surface nothing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
cargo xtask kernel-cases --known-failing FILE grades a set against its
baseline: it fails on a case that fails and is not listed (a regression)
and on a listed case that passes (fixed: its line goes, so the list only
shrinks). The Fillet rig workflow runs the fillet set that way on pull
requests touching the blend, fillet, corner and continuity kernel files,
crates/smlib, eval*.rs, engine features.rs or the cases: one release
build of solidcraft-cli on a Blacksmith runner, SMLib prebuilt, sccache
on a sticky disk that main and develop commit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
New: an r5 curvature corner at a box corner (its patch must seam G2), an
edge running tangentially into a round left out (refused: continues
tangentially), a convex r10.01 round 0.01 past a 10 mm face (built by
subtraction: V = 7139.875, A = 2585.140). The bar and boss arc rounded
together say their crease (15.6 deg r2, 13.1 deg r1); 1390 expects SMLib's
unsupported-corner refusal, the honest reason today; past a full round
the exact fit's refusal names the largest radius that fits, 7.5 mm.
KNOWN_FAILING lists the 18 of 89 fillet cases failing today, each with
its cause.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A new `continuity` area: seams whose order is certain by construction, each graded
`exact` (reads its order and not the next) at picked edges with Continuity Analysis's
default stations and tolerances: a box's crease (g0), a chamfer in place of a round (g0),
an r5 round on planes (g1), and a profile closed by a g1, g2, g3 or g4 blend curve between
lines, extruded (g1 … g4). `continuity.edges` may now be a list of points, picked as
`inspect.continuity` picks them, and `continuity.exact` asks for the order and no more.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
expected.json may give `continuity: {order: g0…g4, edges?: "blend" | "corner"}`: the
seams of each round, chamfer face or corner patch with the faces either side of the edge it
replaced, and of each corner patch with its rounds, are measured as Continuity Analysis
measures them (seam_diagnostics at its default 33 stations and default tolerances), and the
case fails naming the worst seam, its faces, the order it reaches and the next order's worst
value. The seams come from the faces' persistent names (doc::naming::blend_seams). The older
`{near, within_mm, …}` form is kept.
Every fillet-area case with a body gets its order: constant rounds and rolling corners g1,
curvature rounds g2, chamfers g0, the corner: g2 patch g2 at its seams.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A torus face covering most of its tube off the tube's quarter lines (an
exhaust collector's bend: 326° of the tube from −113.8°) needed a fifth
quarter along the tube, which STEP counts from the outer equator: its form
ran past a full turn of the tube and SMLib refused the solid. #8496 turns
frames about the axis only, and the torus label had no tube reference.
The torus label gets a tube start (`tube_start`, radians; 0 and left out
of the record by default, so older records mean what they did): v counts
from there, in every evaluator, form, pcurve, iso curve, tessellation,
placement, SSI and SMLib transfer. A spilling torus face's tube start moves
to where its tube begins, or a turn before where it ends, whichever puts
its tube circles on the grid of a round face they are about the axis of
(a pipe's cylinder meeting the bend); circles round the tube are labelled
from the tube start (brep::form::on_tube), so their forms and pcurves lie
on the face's quarter grid exactly. The saved-format change is one file,
crates/doc/src/format/pending/torus-tube-start.rs.
Engine assembly: COLLECTOR_EXHAUST_VER_1 now comes in as a closed solid
(refused 3 -> 2); Poppy Humanoid unchanged (4).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SolveApp::header (toolbar::HeaderLayout) lets the app that hosts the UI
choose where the application bar puts three things; the default is the
current desktop layout, pinned by a test:
- command search: in the ribbon's tab row (default), or in the bar
immediately left of Settings, folding to its magnifier when the bar is
short of room (the ribbon's tabs then run to its end);
- Report a bug: a labelled button (default), or an icon drawn like Settings
and Help with its name as tooltip and accessible label;
- the version: in the bar (default), or as a Help menu row after About that
opens the release notes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An app that embeds the engine can install a check on the Session
(Session::set_command_check). Session::execute asks it before each command
with the command id, its parameters and the session (read only); a refusal
comes back as EngineError::Refused with the check's reason and leaves the
design unchanged. Without a check nothing changes.
The check belongs to the host: it survives opening and starting designs,
scratch copies (previews, off-thread evaluation) carry it, and the desktop
UI hands it on to every document tab and to designs opened in the
background.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An edge along a face seen edge on lies within a pixel of the nearer face beside it, which hides
part of its width; its hidden copy, drawn after the visible edges, laid a light line over the
visible edge of that nearer face (some 100 px on a box at yaw 1.565). Drawn first, it is covered.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
uv::closed tells a spline closed on itself (periodic, one knot interval per period) by
its first and last control rows and knots clamped at both ends, checked as k[0..=p] and
k[n..=n+p] all equal. Over its interval [k[p], k[n]] a spline of degree p never uses
k[0] or k[n+p]: it starts at its first control point and ends at its last when
k[1..=p] and k[n..n+p-1] repeat. A periodic quadratic circle written with knot
multiplicities 1,2,...,2,1 (as OpenCascade writes one) was taken as open, so its seam
edge's two uses got one pcurve and the face was refused as a loop across a closed
spline surface's seam without a seam edge (periodic_knot_tube_faces_import_and_mesh:
"pcurves failed on 2 of 2 faces"). The check now skips the outermost knots; the knots
the existing test calls unclamped (uniform) stay open.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A SPHERICAL_SURFACE face bounded by one meridian used both ways is a whole ball, but the
file's placement need not have its poles at the meridian's ends: a sphere's frame is any
right-handed frame about its centre (the same surface, its natural normal outward). With
the poles elsewhere, the seam is a slit across the face's form, not the form's own seam;
SMLib's export refused the record ("a closed shell bounds two regions and encloses one"),
and the ball was not imported
(seam_only_sphere_whose_placement_axis_misses_its_poles_imports_as_a_whole_ball).
Before anything is measured on the frame, such a sphere's frame is put on its seam: the
axis through the seam's ends (which must be antipodes on the sphere, within the
tolerances), the x axis through the seam's middle. A seam already between the frame's
poles keeps the file's frame.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A cylinder, cone, sphere or torus face of one whole turn has a NURBS form of the four
quarters from the one its seam starts in (brep::form::face_quarters). When the seam is
not on a quarter line of the file's frame, the seam's second use lies a period on, past
the form's end: the face's domain stopped at the form's end and SMLib measured a ball
whose seam meridian is 0.7 rad off its frame's x axis 2.2% short
(tilted_seam_only_sphere_imports_as_a_whole_ball: 32.03 vs 32.74 mm3), and a 5 mm ball
with its seam 1 rad off, holding a 2 mm void, 481.0 vs 490.1
(seam_only_sphere_void_is_turned_inside_out).
turn_quarters, which says whether turn_frames (ee4d4b5a5) turns a face's frame, counted
such a whole turn as four quarters, never a spill. It now counts the quarters the face's
angle interval needs on the grid (brep::form::quarters, as for a face of less than a
turn): five for a whole turn off the quarter lines, so the frame is turned about its
axis to put the seam on one, and the form's seam is the face's.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The schema 4 reader adds a seam to a face that winds around its surface without one
(3d7e465b1) and turns over a face whose loop disagrees with its same_sense, and notes
both (the spindle torus tests assert the seam note). The tests written against truck
expected no warnings:
- step_conformance_parts_read_back_their_exact_shape: each part's seam count (the cone,
dome, cylinder parts bounded by their rims alone); volumes, boxes, faces unchanged.
- extruded_ellipse_patch_crossing_profile_seam_stays_trimmed: same_sense .F. adds the
turned-over note.
- implicit_bound_can_still_describe_the_complement_of_a_patch becomes
implicit_bound_against_its_face_sense_turns_the_face_over: the strip's one bound on a
cylinder cannot be a hole (the cylinder is unbounded along its axis, so the face would
have no outer boundary); the loop bounds the face, turned over, the 42 mm2 strip.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The face-bounds limit lives in crates/step's topology (MAX_BOUNDS, 54b10b9a3) since the
truck reader that said 'boundary loops' went (ab8bc6625); it names the face and its
bound count. The test asserts that message; the time bound stays.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The reader (crates/step, f06581ca5) keeps distinct knots and their multiplicities and
never expands them; a multiplicity above degree + 1 is invalid (ISO 10303-42) and is
refused on its own as 'bad knot multiplicity' before the sum is compared. The test's
1,000,000 knots of multiplicity 64 on a degree-1 curve hit that first. Both checks are
asserted: multiplicity 64 per knot, and 1,000,000 knots of multiplicity 2 by the sum.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
unsupported_face_is_left_out_and_reported replaced a plane with an OFFSET_SURFACE on a
placement: the reader reads offset surfaces since d41ae8512, so the face was refused
for its basis (AXIS2_PLACEMENT_3D), not for being unsupported. It now uses
CURVE_BOUNDED_SURFACE, which the reader does not read. Since cfee4a720 and 4cb0d6ee9 a
solid with a face left out comes in as a surface body of the faces that read, warned
'imported as surfaces' and 'left out', with no 'open body' warning on top (as
a_refused_solid_comes_in_as_a_surface_body_of_the_faces_that_read asserts); the test
asserts that.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
8eb93ff06 gave every cube edge its own line through two vertices, with the lines'
directions and vectors at #600 to #623. two_instance_assembly writes its occurrences
at #600 to #614, so 10 ids were defined twice, the reader kept the cube's, and the file
had no assembly: one body where hand_written_assembly,
later_instances_show_the_first_instances_mesh_placed and
repeated_items_are_shared_and_one_use_items_are_not expect two. The directions move to
#340 to #363, which no fixture uses.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An edge that lies off its face's surface has its tolerance raised to cover the
gap before SMLib makes its trims (cover_gaps). Where such an edge runs out of
its patch, SMLib ends it where its drop onto the patch ends when that is within
the edge's cross-section tolerance of its vertex (SmEdgeuse::CreateUVTrimCurve),
and moves the edge's interval there. The moved end was checked against the
edge's tolerance before it was raised (1e-5 mm), so two faces of Poppy
Humanoid's left thigh, whose edges SMLib ended 0.08-0.1 um short with a held
tolerance of 0.77 um, were refused ("SMLib moved an edge's interval"). The end
is now checked against the tolerance SMLib held the edge at.
Poppy Humanoid: thigh_left comes in as a closed solid (refused 4 -> 3).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An edge on a circle or ellipse read as a full turn whenever its two vertices lay
within the file's uncertainty of each other. The engine assembly's water pump
housing has a 20.4 um arc between two vertices in a part whose uncertainty is
20.5 um: read as a turn, its cylinder face's loop wound once round the axis with
no second bound ("a face around an open surface of revolution with one bound")
and the housing came in as surfaces. A full turn is now an edge with one vertex,
or two at one point; two vertices apart bound the arc between them.
Engine assembly: HOUSING_MAIN_WATER_PUMP_01 comes in as a closed solid
(refused 5 -> 4). No other edge of the engine or Poppy Humanoid read this way.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each body a section cuts is hatched at 45 or 135 degrees in one of three
colours of one family (the cap's clay, a steel and a sage, theme tokens
`section_cap` and `section_hatch`). Bodies whose cut outlines touch or
overlap get different hatches, the angle changed first, then the colour
(solidcraft_render::hatch, a greedy map colouring of the cut); otherwise a
body keeps the hatch its name seeds, so it stays the same from frame to
frame and as the plane moves. Spacing and line width are unchanged.
The hatch rides on the palette (an Rg32Uint texel: colour, 1 + hatch), so
the viewport, headless GPU pictures and the CPU raster draw the same.
Preference Graphics > Section hatch: By part (default) or One style, the
look before this change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
Ten `refused` cases outside the fillets had no `refusal` since #8445 made one required. Each
was run; its refusal is the right answer for the reason its method gives, and that reason is
now its `refusal`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A cylinder, cone, sphere or torus face of less than one turn that does not
start on a quarter line of its frame, with more than three quarters left to
it, spilled into a fifth quarter of the quarter grid: its NURBS form ran past
one turn and SMLib refused the solid ("an edge lies 12-16 mm off the face's
surface"), which came in as surfaces.
Before its edges are labelled on it, such a face's frame is turned about its
own axis (brep::form::turned: the same surface, its angle parameter measured
from further on) so it starts on a quarter line and its form spans four
quarters. Round faces a circle edge about both their axes joins share one
quarter grid, so they turn together by one angle, chosen so every one of them
fits in four quarters; a group with a face that cannot be turned keeps the
file's frames. Each face's box on its label is measured once and reused.
Engine assembly: PULLEY_WATER_PUMP and MOUNT_SCAVENGE_PUMP_01 now come in as
solids (refused 5 -> 3); Poppy Humanoid unchanged (4).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
A body showing another's display (a later instance of a part an assembly
places several times) took that part's mesh, placed, but was planned for
the GPU viewport on its own: Poppy Humanoid's display planned all 247
bodies where its meshes came from 94 parts. Body::mesh_plan now plans the
original once (kept with the shared display, as its meshes are) and moves
the plan, so the copy's display is the part's triangulation, as its mesh
is. Body::unmeshed_faces needs no case of its own for a copy any more.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The STEP import's mesh check made a coarse mesh of every part (planned,
evaluated, its seams and smooth edges flagged) only to see which faces had
triangles, and threw it away. A face has triangles in the evaluated mesh
exactly where its plan has a triangle on its own nodes, so the count is now
read off the plan (TessPlan::faces_with_triangles, the condition evaluate
keeps a triangle by).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A window of planned solids ended at the second instance of a part not yet
read, so an assembly that places a part several times in a row converted
in windows of one or two solids, each waiting for its slowest read before
the next began (Poppy Humanoid: 14 windows of 1 to 30 reads on 8 workers).
The window now always takes MAX_WINDOW solids; an instance of a part an
earlier job of the window reads is placed after that job, from the
orientation it found, as the next window would have placed it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face whose boundary is only two small circles (a tube's bend) did not show where it is from
its ring alone: its triangles are now drawn over in the error colour at 35%, depth-tested like
the selected face's tint, under the ring.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
While a feature dialog says that a part's face can't be used for modelling, and while the
pointer is over that part's "needs repair" row in the browser, the face is outlined round its
edges in the theme's error colour, over the shading and dimmed where the model hides it, like
the selection's outline; it goes when the message or the hover does. Show face now frames the
face without selecting it (a selection in a dialog becomes its pick and cleared the message).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
The current step's spinner was egui's, sized to the text and drawn as a small partial arc next
to the tick and the ring. All three marks now share one box the label's height, centred on its
capitals (the first glyph's ink, not the line box), at the same left edge: the tick, a turning
three-quarter circle and the ring are the same diameter, the spinner with the tick's stroke,
whole at 100% and 200% UI scale.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
SMLib's import says only that it refuses a body, so the kernel now keeps a structured
refusal with the body's SMLib cache (Body::smlib_refusal / smlib_refused: the first face SMLib
refuses on its own, found by sending each face alone, and its surface kind). A STEP import sends
each part's first instance as an operation would, notes the parts that need repair in the
import report, and the browser marks them ("VALVE_COVER_01: needs repair").
A feature dialog whose operation fails on such a part says, instead of the kernel's words and
an issue report: "Couldn't join to VALVE_COVER_01. One of its rounded faces can't be used for
modelling.", with Repair part (body.repair, declared as planned, also in the part's menu) and
Show face (selects and frames the face).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The kernel's import warnings get a structure (ImportReport: the file's parts, how many came
in as solids, and each warning's kind, part, faces and count; the reads' notes are typed as
repairs or faces left out). The Import feature's result carries it (and the model cache keeps
it), and the timeline tooltip and the status line after an import word it from the catalogs:
"Imported 40 of 43 parts as solids. 1 part couldn't be made into a solid and is shown as a
mesh. 3 faces couldn't be repaired and are missing." with what may help. Entity numbers and
the kernel's wording stay in Copy Warning Message, which now lists every warning.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An open or an import lists its three steps (a STEP or IGES file: reading the file, checking
and repairing the parts, building the model; a design: reading, rebuilding the features,
preparing the view), the ones done ticked, the current one with a spinner the height of its
text and how far it is, and the bar for that step. The other file operations keep their one
line, led by the same small spinner. New kit control: ui_kit::steps.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Setback field's "More than 2 mm" is the message; the kernel's refusal of
that setback shows no text under the dialog and no Create Issue (it is the
user's value to change), and its raw text stays in Copy Details.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
The Setback field reached the command and the kernel used it: on a box corner
with r 2 rounds a 3 mm setback built, while 1.5 and 2 mm failed inside SMLib
(status 1001), shown as "could not be built" with nothing saying why. A setback
must reach past where the rounds meet without one (the radius on square faces).
The kernel now refuses such a setback before SMLib runs, naming the least
(the pre-check of aa2b09bc2 on fillet-kernel-cases), and exposes that least
(fillet_least_setback). The field shows the default, 1.5 x the radius, as an
editable length that follows the radius until the user types one, and outlines
a value at or below the least in the warning colour with "More than 2 mm"
under it (ui_kit::checked_row, warning_caption). Its hover text says what a
setback does, the default and the least.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
An open body (a surface body, a mesh that does not close) has no inside: its faces counted
with the rest took the cap from the closed body it lay in or behind. The body's closedness,
which the kernel knows (`ModelBody::is_closed`: no shell left open, a mesh body every side of
which two triangles share), goes with it to the renderer (`SceneMesh`, `PlannedBody`,
`InstancedMesh`, a block's closed runs, the loose open triangles), and the count draws closed
bodies only.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
The cap went to the nearer surface by depth: a back face (the cut body's far wall) or a front
face. Where bodies touch, the cut body's far wall and the other body's face lie at one depth,
and rounding settled each pixel either way: the other body showed through the cap in specks,
stripes and triangles (an engine's sleeves in their bores, pins in their bosses). The stencil
now counts the faces behind the plane instead, from a base of 128: a back face adds one, a
front face takes one away; above the base, the plane is inside a body and the cap is drawn in
the nearest back face's cut colour, at the plane's depth. Elsewhere the faces draw as without
a section, an open body's inside lit rather than hatched. All four draw paths (built, planned,
instanced, instanced flat).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
A SHELL_BASED_SURFACE_MODEL with both closed and open shells (Poppy Humanoid's
thighs: the part and a two-face sheet on it) came in as one body, open for the
sheet's edges. Each closed shell is now read as an item of its own, a solid in
the model's colour, and the model's open shells as a surface body.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
A face's form covers the quarters of its box, and the box is the extent of its
boundary's points, which run past a quarter's end by as much as the edges lie
off the surface (nanometres in a file of 1e-11 mm digits). A half tube's form
then took the quarter past each end and went round the whole tube, which SMLib
does not take for a torus: 16 parts of an alternator came in as meshes. An end
within the tolerance of a quarter's end (the angle on the face's widest circle
there) is now on it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5Lzv2n8fyKqD61NCHSF
A corner setback must reach past where the corner's rounds' contact lines
meet without one (smlib_blend_setback); short of it SMLib failed with its
own status (1001). The kernel now refuses it before SMLib runs, for the
rolling-ball corner (fillet_setback) and the curvature corner
(fillet_g2_corners) alike, naming the least: on each face at a corner of
planar faces and straight edges the two rails, r tan(psi/2) from their
edges, cross (d2 + d1 cos alpha) / sin alpha along an edge (r on faces
square to each other). Other corners stay SMLib's to refuse.
The rig's `refused` cases now name the reason (`refusal`, a part of the
message, required), so a failure for another reason no longer passes; a
`solid_or_refused` case may name one. Fillets: the setback cases and
783, 3435 pass; 1390, 896, 4782 past-a-full-round and the no-tangent-chain
case are refused for an unclear reason and fail.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
22 hand-written cases in tests/kernel-cases/fillets:
- 1207-fillet-bar-top-edge-and-boss-arc-together (r2, r1, fit reduce,
fit taper, curvature corners): a bar's top edge and the partial base
arc of a boss standing out over the bar's side rounded together. The
rounds' volumes away from the corner, the corner patch within (2r)^3.
SMLib's 3x2 mixed corner fails here (1001) until the fork's fix.
- 1207-fillet-bar-top-edge-after-boss-arc-round (r2, r0.5, no tangent
chain): the edge rounded after the arc; r0.5 fails SMLib's check.
- 9-fillet-corner-*: setbacks not past the radius refused (rolling and
curvature corners), the curvature corner at r1, and the curvature
corner with reduce and taper fits (not supported yet).
- 4782-fillet-*: full rounds (two end verticals at half the width, exact,
reduce, taper), a round at half the wall, radii past the full round
(refused; reduce gives the full round) and a round wider than both
faces (built by subtraction).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The old body was drawn as ghost whole, behind the new body by a fixed depth bias. A curved
face the change re-meshes (a boss face whose boundary a fillet moves) has a new triangulation
that crosses the old one by up to the chord error, far past that bias, so the ghost showed
through it in slivers. Old triangles lying on the new body's surface are now left out; the
removed material is still shown.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The held parameter was spaced over each face's own trimmed (u, v) box, so the pieces of a
face split by a boss showed different lines that missed each other at the split. The lines
now sit on a ladder of the surface: a bounded parameter's range in count + 1 steps, a linear
one (a plane's, a cylinder's or cone's axis) at a round world spacing from the world origin.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The dialog follows the options solid.fillet takes together: If too large
(fit) and Corners show only for Constant radius, a curvature corner and a
fitted radius rule each other out and the other resets visibly, and an
option that does not apply is never sent (it used to send `corner: g2`
with Reduce or Taper and get refused). Variable Radius shows only for the
types that take it; the Advanced section held only the fit and is gone.
Each option has a one-line hover text. Kernel refusals read in plain
words, with the raw text in the hover, Copy Details and the issue report.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib holds both sides of a sheet's faces and gives each back on its surface's
natural side, so a sheet whose file uses faces the other way came back
renumbered. The reader no longer turns them (that lost the file's side);
instead SMLib is given the sheet with those faces turned over exactly on the
record, and they are turned back on SMLib's body, each face mapped through.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A solid the reader or SMLib refuses was shown as a mesh of its faces (#8223):
exactness lost, cracks between faces, and no modelling. It now comes in as an
open surface body of the faces that read, sharing their edges as the file has
them; each face left out is named.
- The reader reads a solid with a set of faces left out (read_solid_without):
a sheet whose faces face their surfaces' natural normals, as SMLib gives a
sheet's faces back, so SMLib's body numbers the faces as the record does.
- Tier 1: the faces that cannot be read alone and the face each refusal names
(named_face) are left out, up to 8 reads. Tier 2, when a refusal names no
face of the solid: the one-face bodies stitched by SMLib. Tier 3, when SMLib
cannot stitch them: the mesh of them, as before.
- Report: "imported as surfaces: <why>" and "left out: face #..."; the
ImportedBody field shown_as_mesh is now refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face's boundary node on a patch side collapsed to within the tolerance of
the vertex there (capped at the chord tolerance), not to a point, is a pole:
its parameter along the collapsed direction is free, as the record check and
the STEP loop check take it (`uv::collapsed`, shared with
`SurfaceEval::pole_within`). A loop through such a corner no longer winds
round the surface as projected, and plans strictly (Poppy Humanoid's
connector, face 43: a quarter of a sphere cap on a closed rational spline).
A face the strict plan and the lenient fill both refuse no longer fails the
whole body's mesh: it is left out with no triangles, named in
PlanStats::unmeshed (logged by the display plan, counted by
Body::unmeshed_faces), and the rest of the body is meshed; a body none of
whose faces plan is still the first face's error. Body::volume_at gives no
volume for a mesh with a face left out.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When the reader or SMLib refuses a solid, each of its faces is read on its
own as a one-face surface body (solidcraft_step::topology::read_faces_with)
and meshed by the planner; the triangles together are a mesh body (the
existing kind STL/3MF imports make), marked ImportedBody::shown_as_mesh with
the refusal. The report says 'shown as a mesh: <reason>' and names each face
left out. Modelling operations refuse it as any mesh body.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The check that a loop does not jump across a spline surface's parameters took a
pole only within the face's tolerance. A corner patch written with one side
shrunk to a few µm (Poppy Humanoid's connectors: 5 µm in a 1e-5 mm file) was
refused, though the record check takes the same junction as a pole at the
tolerance of the vertex the uses meet at. The import check now uses that
vertex's tolerance; the loop runs along the collapsed side, its boundary iso
line. A jump that is still refused says whether the surface closes there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An edge whose curve runs a few um past its vertex and its spline face's
patch was trimmed back by SMLib when it made the trim, and the face was
refused ("SMLib moved an edge's interval"). Our own measure never saw the
gap: STEP measures an edge against its surface only where the nearest point
is in closed form, and a spline face's gaps only on the pcurves the kernel
returns, after SMLib has trimmed the edge. Each bare use's edge is now
measured by SMLib (smlib_brep_edge_face_distance, over the patch) after the
sheet's import, its tolerance raised to cover the gap with
SURFACE_GAP_MARGIN and the sheet imported again: SMLib keeps the edge whole.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An edge raised to cover its pcurves' gap could get past SMLib's limit
(Congé19: 0.98 mm off, tolerance 1.47 mm) and its whole solid then failed
as "SMLib rejected the input". The tolerance an edge or vertex would get,
not only its gap, is now held to that 1 mm: farther off is refused with
how far. The record mapping refuses a tolerance past it with a message
rather than handing SMLib what it rejects.
The margin over a surface gap moves to solidcraft_geom::brep::pcurve
(SURFACE_GAP_MARGIN) for SMLib's pcurves to use too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Binds smlib_brep_edge_face_distance: how far an edge lies from a face's
surface over the surface's own domain, the gap SMLib sizes an edge's
tolerance by.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tangency-labs/smlib e51c3f2: a drop that comes back in pieces is joined
where they meet, never stretched onto its edge (solidcore#1383), and
smlib_brep_edge_face_distance.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Pins tangency-labs/smlib 56f6e81 (release prebuilt-56f6e811bc84…): smlib_blend_setback_eased
and smlib_blend_setback_open_eased, an ease length before each set-back corner over which a
round's section blends from its circle to the curvature-continuous section (solidcore#1376).
New calls only; existing ones are unchanged (ease 0 is today's body).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Corners row keeps the control's words (Tangent, Curvature) and its hover text says what
each makes: a rolling-ball corner (G1), or a smooth patch set back along each edge by the
Setback, 1.5 × the radius when empty (G2). continuity_row_explained is the shared control with
a caller's hover text; the new lines are in en, ar, de, he, ja, ko and zh.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ui_kit::continuity_row is the one widget that chooses an order: a dropdown of the orders a
construction builds, labelled with the table's words and with a hover line per order naming
its code. Patch, Blend Surface, Boundary and the fillet's Corners row (tangent rolling ball or
curvature-continuous patch) all use it, and their dialogs keep a Continuity rather than an
index into a copied list.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each seam now also reports the jump of the curvature's arc-length derivative across it (G3)
and of its second derivative (G4), both from the faces' exact partials, and its order: the
highest one every station meets within the tolerances (two new tolerances, defaults in the
kernel). The verdict keeps its meaning (smooth is G2 or better). inspect.continuity replies
with the new metrics, each seam's order and a count per order; the viewport widens G3 and G4
seams, the legend and the marker labels name every order in words. The Patch, Blend and
Boundary rows offer the orders the kernel's Construction table says they build. G4's words
in en, ar, de, he, ja, ko and zh.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
EvalSurface and SurfaceEval give every partial derivative up to total order four: in closed
form on cylinders, cones, spheres and tori, by basis-function derivatives and Leibniz's rule on
rational splines, and by the rotation's own derivatives on a revolution (which had no second
derivatives before). normal_section turns them into the curvature of a normal section and its
first two arc-length derivatives by solving the section's Taylor series order by order.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
The kernel's Continuity moves out of bridge.rs into continuity_order.rs and gains G4. One
table holds each order's command word (g0-g4), its code (G0-G4, for tips and messages) and
the message ids of its label and hover text. Construction says which orders the patch,
bridge, network, match and sketch blend curve build today; Continuity::built_by is the one
refusal every command and kernel entry point uses. A design may now save g4 (format note).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
The fit-point spline's extruded side is two faces split at x = 20; the hole now
lies inside one of them, so the test clips to a real inner loop.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The analysis sliced the display mesh with planes: faceted, approximate, and
broken at triangle seams and trims; the window drew it as a 2D overlay without
depth while the tool was active. Now one kernel routine (kernel::iso) clips a
surface's u = const or v = const line to a face's pcurve loops; the curve through
a point (sketch projection) and the analysis grid (count each way, evenly across
the face's extent) both use it. The loop polygon follows the loop's order (a
pcurve the loop runs against is walked backwards), which the inside test got
wrong before, and closes over vertex-loop poles.
ModelBody caches the lines per face, count and display quality. The session
keeps the picked faces as view state (inspect.isocurve add/clear, hostile counts
refused); the viewport, ui.render and snapshots draw them depth tested in the
accent colour. Curvature combs on faces follow the same exact lines.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The nearest point on a spline (5cb015fae measured spline faces up front too) may settle on
another part of the surface, or on its boundary for a point a hair past it, and reported gaps
the edges do not have: 2.9 mm on Poppy Humanoid's leg shells, whose edges lie within 0.003 mm
of their faces, and widened tolerances past what SMLib takes (it then refused
models/vertices-off-curves.step). The up-front measure is again for faces whose nearest point
is in closed form (plane, cylinder, cone, sphere, torus); a spline face's gaps are measured
on the pcurves its kernel makes, as the record check measures them.
The corpus survey is back to 53 of 56 files; Poppy imports 238 bodies (10 solids refused,
most for SMLib's pcurves on spline faces: solidcore#1383).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
As the app's Display Quality setting, set before the design opens so every display plan is
made at it; a level that isn't one is refused with the levels. For close-up reviews of
imported faces.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two edges leaving a vertex almost tangentially (a cusp, as a round runs out on Poppy
Humanoid's creative_commons_license) separate by about Δκ·s²/2 at distance s while a chord of
length L sags by κL²/8: halving the chords never stops them crossing. untangle returned that
face's error for the whole plan, and the body's 1834 faces were not shown. It now returns the
faces still crossing, each with why, and plan_or_fill fills each leniently straight away and
names it in the stats (the rule of #7755): the other faces are planned as usual.
Test: a face named as still tangled is filled leniently and named with that reason.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A corner ball's edge may run from its equator over the sphere's pole to a vertex just past
it (Poppy Humanoid's creative_commons_license: 1.8° past, 0.016 mm off the pole). On the
sphere's parameters the pole is a line, and the edge's pcurve would have to jump half a turn
along it: "pcurve starts 2 from the end of the use before it, on another period". Each
edge that passes through a pole of a face it bounds (a sphere's, a revolved face's on its
axis) inside its interval is now split at the nearest point to the pole (golden-section
search from 16 samples), the new vertex placed on the pole; the loop's half turn is then at
a vertex on the pole, where a loop may run along it. Each solid's notes say how many edges
were split. The geometry check takes a vertex at a pole to within the vertex's tolerance, as
the reader already does (SurfaceEval::pole_within).
Without the split the edge's jump read as a 0.024 mm gap and was covered by a tolerance:
the measured gap now only covers what the file's geometry really has.
Test: a spherical triangle (radius 0.5) with an edge over the pole to 1.8° past it reads
with its spherical-excess area and its edge split.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
A range sampled in even steps ends at a + (b − a)·1, which may round one ulp past b, where
find_span gave no span and every B-spline evaluation there failed. The import's loop area
(the reader's sampling of each pcurve, used to turn over a face whose outer loop runs
clockwise) then had no value, the face was left as written, and the record check refused
it: "outer loop runs clockwise in (u, v)" on 7 of Poppy Humanoid's solids. The same step
had refused closed-form pcurves (426924937). find_span now takes a parameter within 8 ulps of
the domain's magnitude past an end as that end; farther is outside, as before. It fixes the
pattern for every caller rather than each sampling loop.
Poppy: 232 → 242 bodies, 15 → 7 solids not imported; the kernel tests as before.
Test: Poppy's interval's last step rounds past its end and finds the end's span; 1e-9
past is still outside.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
SMLib may end an edge at a parameter of its own: a knot its vertex lies at (-1.0 where the
vertex's angle gives -0.99999998, a 2e-8 step on the edge's quarter arc, far under 1e-6 mm).
We refused any change beyond 1e-9 of the parameter, a bound with no geometric meaning:
"SMLib moved an edge's interval" on 8 of Poppy Humanoid's faces. The ends SMLib chose are now
measured on the edge's curve: within the edge's tolerance, its pcurve is mapped onto our
interval by the affine change of parameter (exact for a B-spline: its knots move), and the
record's check measures it as any pcurve; farther is SMLib trimming the edge, refused with
how far.
Poppy: 227 → 232 bodies, 17 → 15 solids not imported, no interval refusals left.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
When an edge's tolerance is raised to the gap it has, the vertices at its ends now cover it
too, as in any tolerant record (SMLib takes the vertex's zone as at least its edges'). The
edge measures sample the interval's end as t1 itself: t0 + (t1 − t0)·1 may round past it.
(This commit also measured spline faces up front by nearest point; 04de97cd9 takes that out.)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The closed-form pcurve's lift check sampled t0 + (t1 − t0)·k/n, which at k = n can round
one step past t1 (1.2985843833752289 for an end of 1.2985843833752286), where the pcurve
on [t0, t1] has no point: the closed form was refused, and on a plane, which has no
parameter box for a kernel's pcurves, the face with it ("a face on an unbounded surface
without closed-form pcurves", Poppy Humanoid's Combine1 and others). The end is now t1.
Test: Poppy's interval on a spline on a plane; without the fix, "no exact pcurve of a
B-spline on a plane".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
Pins tangency-labs/smlib 1072d1b (release prebuilt-1072d1b857de…): an N-sided corner
patch now meets each round on the round's own exact section edge instead of its polynomial
approximation, which the stitch kept on a near tie at some corners (solidcore#1377).
SMLIB_PREBUILT rewritten by scripts/fetch-smlib.sh --pin.
The kernel's tess, display_mesh, step, corner and fillet tests fail the same 63 as with the
previous pin, none new.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Brep::validate lets a pcurve's interval differ from its edge's by PCURVE_INTERVAL (1e-9
of the magnitude: rounding room), but check_geometry then evaluated the pcurve at the
edge's own ends and samples, which may lie just past it: "the pcurve has no point at
2.0000000000082903" on four of Poppy Humanoid's solids, circle edges whose end vertex
sits on a quarter of their form, where SMLib's pcurve ends on the knot. Both of the
check's pcurve readings (the use's points and the loop's area) now take the edge
parameter on the pcurve's interval, through one reader.
Poppy: 195 → 199 bodies, 41 → 37 solids not imported; the corpus survey and the kernel
tests are as before.
Test: a pcurve whose interval ends 1e-11 short of its edge's validates and checks;
without the clamp, "the pcurve has no point at 1".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
A closed-form pcurve (a B-spline edge's control points projected onto its plane) is
accepted within half its edge's tolerance, and the edge carried the file's: Poppy
Humanoid has planes whose B-spline edges stand 9 µm off them in a file that states
10 µm. The closed form was refused, and a plane has no parameter box for SMLib to make
pcurves in: "a face on an unbounded surface without closed-form pcurves" on 11 of its
solids. On a plane, cylinder, cone, sphere or torus (nearest points in closed form)
each edge's distance from the surface is now measured along it first, and its
tolerance covers three times that; a spline face's gaps stay measured on its kernel's
pcurves, with the 1.5 × margin of a vertex's. The pcurve code reads each use's own
tolerance.
Poppy: 194 → 195 bodies, the unbounded-surface refusals 11 → 3 (the solids they let
through now stop at later steps); the corpus survey and the kernel tests are as before.
Test: a plane face whose B-spline edges stand 0.8e-6 mm above it in a file stating
1e-6 mm imports with its closed forms; without the measure it fails as Poppy's did.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
A TOROIDAL_SURFACE whose tube reaches its axis (major radius at most the minor, or
written with a negative major radius) and a DEGENERATE_TOROIDAL_SURFACE were refused.
They are read as what they are: the tube's meridian circle (centre C + R·X, its angle
from X towards Z) revolved about the axis, the same points and normal as ISO 10303-42's
formula, a negative R putting the tube's centre across the axis.
A revolved circle or ellipse that meets its axis is now its whole arc on the face's
side, pole to pole, in closed form (where s0 + M cos(t - phi) = 0), and its NURBS form
is cut there exactly (Boehm knot insertion, bspline::segment): the parameter stays the
quarter grid's, so the edges' forms and straight pcurves are unchanged, and the surface
ends at the poles instead of running on to whole quarters past the axis (which folded
the surface over itself, and left a face with one bound around the axis without its
pole). The face's boundary picks the apple or the lemon.
On Poppy Humanoid 22 solids were refused for these tori; with the SMLib fixes before
this none of their torus faces fails: Poppy 42 -> 27 solids not imported, 194 -> 215
bodies.
Tests (hand-written files, analytic answers): a horn torus round of radius 1 on a pin
(V = 8 pi + pi(5/3 + pi/2)); a spindle torus pin end (major 1, minor 2, its apex in a
dimple at z = sqrt 3, V = 3 sqrt3 pi + 8 pi^2/3 on the cylinder) as TOROIDAL and
DEGENERATE_TOROIDAL, with a seam to the pole and with one bound; a nearly round end
(0.005 / 0.345) in two halves with SolidWorks's apex vertex; the lemon of a spindle torus
as one seam-bounded face with major radius -1 and as a degenerate torus with its lemon
selected (V = pi(6 sqrt3 - 8 pi/3)). Volumes and areas within 1e-3.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
The pcurves close a face's loop through a pole to within the face's tolerance (a
point where crossing half a period moves less than that): there the periodic parameter
is free and the loop jumps across. The reader's check that no loop crosses a seam
without a seam edge took a pole only exactly (1e-9 relative), so it refused the loops
the pcurves had closed near one. Poppy Humanoid's nearly round pin ends (a spindle
torus of major radius 0.005, minor 0.345) have their apex vertex at the tube's top,
3.6e-5 mm above the true pole: their meridians end 1.7e-6 rad short of it, and 13
faces were refused. Both now use one test, SurfaceEval::pole_within.
Test: a point 2e-7 mm from a sphere's axis is a pole within 1e-6 mm, not within 1e-7.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
A revolved circle whose centre is as far from the axis as its radius (a horn torus, its
tube touching the axis) measured its major radius a few ulps above its minor from the
profile's control points, became SMLib's ring torus, and SMLib refused it
(SM_ERR_INVALID_INPUT: no ring torus of that NURBS). A ring torus now needs its tube
clear of the axis by more than rounding (1e-9 of the radius, the slack the same test
already takes for the circle's plane); a horn torus goes as its rational B-spline.
Test: a revolved unit circle 1.5 from the axis is a ring torus; at 1, 1 + 4 ulp and 0.5
it stays a revolution (develop calls the first two ring tori).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
A face on a placed revolution that carries its NURBS form (every STEP surface of
revolution, read since b6a3b31ff) was written as two NURBS blocks: the revolution's own
net, then the form, flagged FACE_NURBS_FORM on a NURBS face. SMLib reads one block per
face, so the flat arrays did not add up and it refused the body (SM_ERR_INVALID_INPUT),
in the main import and in the one-face sheets SMLib makes pcurves on. SMLib has no class
for a revolution: the form is the face, its range and trims already in its parameters,
so it now goes as that NURBS face with no form flag. A revolution the analytic pass
recognises (a plane, cylinder, cone, sphere, ring torus) is unchanged.
Test: surface_of_revolution_sense_follows_iso, which develop fails with the refusal,
passes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
Each edge is sampled on its own, and a short one within the chord tolerance is one
segment. A face bounded by two such edges (a lens of short arcs, a sliver strip of a
cylinder between two rulings) had two chords on one line: "degenerate boundary", and
the whole body's display mesh failed with it (one of Poppy Humanoid's bodies lost all
336 faces to one sliver). Such a loop's segments are now halved until it has three; the
new nodes are on the edges' curves and shared by every face that runs them.
Test: a lens of two radius-1000 arcs beside a 2 m face meshes; without the split it
fails with the same error.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
An edge's curve as a file writes it may lie off a face's surface by more than the
file's stated uncertainty: Poppy Humanoid's intersection curves lie up to 9 µm off
surfaces in a file that states 0.01 µm, and 24 of its solids were refused by the
record check ("pcurve … lifts off its edge's curve"). Each use's gap, between its
pcurve on the surface and its edge's curve, is now measured along the edge at 17
points as the record check measures it, and the edge's tolerance covers it (1.5 ×
the gap, as a vertex's covers its gap; more than 1 mm is refused). The pcurves are as
made, and each solid's notes say how many edges were raised and how far.
The pcurve fit's tenfold retries went in 6cc0401fa; this measures instead of
retrying. On the corpus (the repository's STEP/IGES files, the models repository,
Poppy): 50 → 53 of 56 files import; Poppy 132 → 194 bodies.
Test: a hand-written domed patch whose edges are written 0.2 µm above it (SMLib makes
its pcurves) imports with them covered; 1.5 mm off is refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
New metric tapped holes drill the tap maker's drill (M8 6.8 mm). The Design preference "Metric
tapped holes drill the major diameter less the pitch" (`metric_tap_drill_major_less_pitch`, off
by default) gives them the drill every metric size took before (M8 6.75 mm): the Hole dialog
shows and sends it, and `solid.hole` given a thread and no `diameter` drills it, through the
session as `auto_project` is (kept when a design is opened or started, and across tabs). Unified,
pipe and taper sizes keep their tabulated drills; a given diameter is unchanged. Saved designs
keep the drill they stored either way.
Labels in en, ar, he, ja, zh, de and ko.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
Every tapped hole now drills the tap maker's tabulated drill. ISO metric takes Yamawa YEU23's
drill size column for cutting taps (printed p.704, M coarse; pp.704 to 706, MF fine): M8 drills
6.8 mm and M12 10.3 mm, where the major less the pitch gave 6.75 and 10.25. A custom metric size
the table lacks keeps the major less the pitch. UNC/UNF (Guhring), G, Rp, Rc and NPT (Yamawa)
were tabulated already. The Thread command's suggested hole for a metric size follows.
The Hole dialog's Tap drill diameter is now a field, the tap maker's drill by default, and the
command's `diameter` with a thread is checked the same way: a drill is accepted from the
internal thread's standard minor diameter (or the tap maker's drill where that lies below it:
Rp, Rc, NPT) to 5% over the tap maker's drill. Outside that the command names the range and
why. The field's tooltip gives the drill, the minor diameter and the range. Presets keep a
chosen drill.
A saved design keeps the diameter it stored, so it builds as before, and editing such a hole
keeps its drill: no preference and no format change. Scripts that give a thread without a
diameter drill the new default; ladder 05/007, 11/003 and rig case threads/2115 follow
analytically, and their golden entries are re-measured from runs (tier 05 re-sorted by time).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
The Rc and NPT drills were rounded to a drill series from a third-party reproduction of DIN 345
and a fractional/letter inch series cited from memory. They now come from the same tap maker's
catalogue the G and Rp drills do: Yamawa YEU23, printed p.709 (Rc, the maximum straight bored
hole with non-full threads, Rc 1/16 to 4) and p.710 (NPT, the maximum straight bored hole, 1/16
to 3). A size its tables lack (Rc 5 and 6, NPT 3 1/2 to 12) is refused as a tapped hole, naming
the range. The dialog's tooltip keeps the standard's minor diameter where the thread ends.
Rc 1/2 drills 17.6 mm (minor 17.807), 1/2 NPT 17.60 mm (minor 17.473).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
An Rc or NPT hole was drilled to the exact minor diameter where the thread ends, a size no shop
stocks. It now takes the largest stocked drill not larger than that: metric drills of the DIN 345
series for Rc (to 100 mm), letter and fractional inch drills by 64ths (to 1 1/2 in, ASME B94.11M)
for NPT; past the series it is bored to the exact value. The Hole dialog's drill shows the inch
size for NPT, and its tooltip the exact standard value. Rc 1/2: 17.807 mm, drilled 17.75 mm;
1/2 NPT: 17.473 mm, drilled 11/16 in (17.4625 mm).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
A preset's thread options save `modeled` and `ease_out` (a named ISO 3508 runout or the fade's
length, kept portable as the other lengths are) and recall them into the dialog. A runout
without a modelled thread is no preset's. Recalling checked the thread with the metric parser
only, so a UNC, G, Rp, Rc or NPT preset was refused; it checks the tapped-thread catalogue now.
An older preset without these options recalls as before: listed, square.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
Ladder 06/017's golden entry was edited by hand when the chamfered tip moved to ISO 724's d3.
Re-measured by running the case (solidcraft-cli kernel-cases on SMLib): its counts, area and
volume are the ones written by hand; its time is the run's. The tier's order by time holds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The tapped-hole catalogue takes ISO 7-1's Rc and ASME B1.20.1's NPT internal taper threads,
every size of the tables a modelled thread cuts. Their drill is the internal minor diameter where
the thread ends (the reference plane's minor at the face less the 1 in 16 taper over the useful
or effective thread length), derived from the standards rather than a tap maker's table. An
external R is still no hole's.
A tapped hole's options mirror the thread's: `modeled` cuts its thread into the wall, from the
entry, by the same cut as `solid.thread` (now one function both call), and `ease_out` fades its
inner end out before the bottom (ISO 3508); without it the groove stops square at the thread
depth, clear of a blind hole's bottom. The Hole dialog's Tapped rows add Rc and NPT, Modeled
and the End choice (Square, Fade, Fade Length). The format notes in pending/thread-ends.rs and
pending/pipe-threads.rs record the new keys and sizes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The thread dialog offers the pipe threads: G (ISO 228-1), Rp (ISO 7-1), R and Rc (ISO 7-1) and
NPT (ASME B1.20.1). Their sizes are listed by designation without a pitch, one start and a hand
of right or left; a size picked for a hole the dialog sized Rc stays Rc.
Rp, ISO 7-1's parallel internal thread an R screws into, is a form of its own (`rp`): the
parallel profile (H = 0.960491 P) at ISO 7-1's gauge diameters, refused on a shaft. Its cut
meets the analytic material of its profile within 1e-5, as G does.
A taper thread takes a chamfered tip: each end's cone is measured from the major diameter the
taper has at that end. Pipe threads are not ISO metric, so they give chamfer_depth.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The thread's start is its free end: it runs off plain or chamfered (ISO 4753). Its inner end
stops square in its own profile or fades over a length; an empty Fade Length is ISO 3508's
runout (2.5 P, or 1.25 P for a thread saved short). A thread saved chamfered at both ends keeps
that. "Square" is its own message (thread.end_square), apart from the geometric square. The
new labels are in en, ar, he, ja, zh, de and ko; Extent, Ends, the chamfer labels and the ISO
4753 hint come from the catalog entries already on develop.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A thread's start runs off its free end, plain or chamfered; only its inner end fades out
(ease_out) or stops square. The `ease_in` field, the command parameter (refused now, naming
ease_out), the dialog's Ease In row and their docs are removed; `pending/thread-ends.rs`
drops a saved `ease_in` as the design opens. The runout test covers the short, standard and a
sub-turn fade at the inner end.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The free-end test looked for material 1 to 1.5 pitches past the face's end, beyond the first
pitch so that a chamfer the groove runs out through counts as free. A plate thinner than that
(a boss on a 1 mm plate under M10) had none there, so its top was taken for a free end and
the groove's one-pitch overrun cut into the plate (15 mm³ of a 1 mm plate).
An end is now closed when, over the overrun's whole reach, material lies outside the face's
cylinder (inside it, for a hole) at the radii the groove cuts beyond the face: a shoulder,
plate, head or blind bottom, however thin, steps out past the face, while a chamfer, neck,
countersink or counterbore steps the other way. The band test beyond the first pitch stays
for a cylinder that goes on past a split.
The regression test builds the boss as the app does (a rounded plate, a circle sketched on its
top face, extruded with Join) on 6 and 1 mm plates.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three forms: `bspp` (G, ISO 228-1, 55° parallel), `bspt` (R external, Rc internal, ISO 7-1,
55° taper 1 in 16) and `npt` (ASME B1.20.1, 60° taper 1 in 16). Their sizes are tabulated
from the public sources: ISO 7-1:1978 Table 1 (its gauge-plane diameters are the G sizes)
and NBS Handbook H28 (1957) Table VII.2; tests check every row against the standards' own
relations.
A taper thread starts at its free end with its standard diameter there (external: the gauge
diameter its gauge length in, internal: at the face). Its groove is the screw sweep with the
section moved 1/32 of the axial travel off the axis (the radial law the runouts use, added to
it), its section reaching just past the crest, and the material beyond the major cone (or
inside the minor cone) is cut with it, as a die or tap cuts the crests. An external one runs
its standard length by default.
The volumes are checked against the profile's material fraction integrated over the tapered
radii: G 1/2 to 2e-6, R 1/2, Rc 1/2 and 1/2-14 NPT within 0.2 % of the cut.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
`tip` was none|chamfer, chamfer cutting every free end. It is now none|start|both: `start`
chamfers the end the thread starts from, refused when that end is not free (a shoulder, or a
stop inside the face), and `both` every free end as `chamfer` did. The dialog's Tip pick offers
None, Start and Both; `tip: chamfer` in a command names its replacements.
The thread's end options (runouts, tip, chamfer angle and depth) came without a format
change file; `pending/thread-ends.rs` records them and rewrites a saved `tip: chamfer` to
`both`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ISO 4753's chamfered end bounds the end's diameter by the minor diameter. The default took the
modelled root (the basic d1) for it, so the cone's small circle lay on the root cylinder and
the chamfer ring pinched to nothing wherever the groove reached the end. The default is now
the external minor diameter d3 = d - 1.226869 P of ISO 724, below d1: the cone crosses the
root, each end is a whole disk of d3 inside a whole conical ring. ISO 4753 is for ISO metric
threads; the other forms, whose root is already their design minor, give chamfer_depth.
Ladder 06/017's analytic volume follows (1945.354522563); its golden entry re-measured.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A regression test for the groove stopping at the plate's top: the boss loses the
closed-form volume and the plate keeps all of its 9600 mm³, with and without a runout
and a chamfered tip.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Curvature continuous is the Type entry's name; Corners (Ball, Smooth patch) and Setback are
described with the command words they write. The codes (G1, G2) stay in the text, where they
explain the words.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
Patch, Boundary and Blend Surface list Position, Tangent, Curvature and Curvature change; each
order's code (G0 to G3) and what it means is in the row's hover text. The Continuity Analysis
legend, its seam markers and its status line use the same words (Crease for a G0 seam). The
segments now draw their labels translated. The new labels carry over each catalog's existing
translation with the code taken off, so no language loses its wording; a word another label
already translates keeps that translation.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
The Continuity G1|G2 control repeated Type (both set `type`) and G1 silently cleared the
corner; it is gone, and Type names the section in words: "Curvature continuous", its code (G2)
in the row's hover text. The vertex choice moves out of Advanced as Corners: Ball | Smooth
patch, shown only while the edges have a corner where three or more rounded edges meet (or the
fillet already has a smooth patch), enabled for Constant radius with Exact fit and saying why
when not. Smooth patch shows Setback: one distance for every corner, sent as `setbacks` at each
corner's vertex; left empty, the command's default (1.5 x the radius) applies. A fillet saved
with setbacks reopens with them: one distance at exactly the dialog's corners shows in Setback,
any other list is kept as saved. "Smooth" in Variable Radius is "Smooth radius change". The
stale "G2 corner construction is not supported yet" tooltip and the "Fit and G2 corners" note
are replaced by new messages (en, ar, de, he, ja, ko, zh).
No command parameter changes; saved designs open unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
A face whose boundary crosses itself in (u, v) (a cusp, a sliver whose sides
sag across each other) was an error, and the body had no display plan. It is
now planned once more with a lenient constrained Delaunay in (u, v): sides
crossing one already kept are left out, the rest is seeded and refined as any
face, and its points lie on the surface, where the GPU evaluates them. The
display has no hole, but the face is still a defect upstream, in its pcurves
or seams: PlanStats::lenient names it with the strict plan's refusal, and the
display plan logs a warning for each such face, so it is never silent.
The face boundaries keep coming from the exact pcurves (d61efb306); cdt::Fill
gets its `lenient` field back (dropped in 3f2585569). The old flat fill (#4021,
tess/fallback.rs) stays gone: develop removed it in 9138e8704. A face the
lenient fill refuses too (loops that do not close in (u, v), or wind round a
seam or pole) stays an error naming the face. Closes#4453's intent on develop.
(cherry picked from commit 84de5aa071)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
SMLib's out_failed_corner reaches Rust as Error::CornerRefused { corner }, and the
kernel names the corner by its point: a G2 fillet at a corner of mixed convexity
says where it is and that SMLib sets back only corners whose rounds are all
convex or all concave. Engine test checks the message through solid.fillet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
The inner corner of a 20 x 20 x 10 notch in the 60 x 40 x 20 box, three
concave r5 rounds with corner: g2: one valid body of 13 faces within the box,
the patch's twelve seams jumping at most 1e-3 per mm clear of the hole's six
corners (2.7e-4 measured; the ball corner jumps 0.2). The continuity check
looks within 11 mm of the corner: the vertical round, 2.5 mm long past its
setback, meets its faces 11.2 mm out with every round's G1 jump.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The inner corner of a notch, three concave r5 rounds, takes the set-back G2
patch as a convex corner does: one valid solid with the ball corner's 13
faces, seam jump 6.3e-4 per mm clear of the hole's corners against the
ball's 0.2. A corner of two convex rounds and a concave one is refused as not
supported: SMLib sets back only corners of one convexity.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
expected.json `continuity`: over the tangent seams wholly within `within_mm`
of `near`, the curvature jump across them (exact surfaces, 65 stations a
seam) at most `curvature_jump_per_mm` clear (`clear_mm` along the seam) of the
vertices where two of them turn, where a neighbour's own G1 seam ends.
07/015, the box corner's three r5 rounds with corner: g2: one valid body of 10
faces within the box, its patch's twelve seams jumping at most 1e-3 per mm
clear of the hole's six corners by 1.2 mm (2.7e-4 measured; the ball corner
of 009 jumps 0.2). Its golden entry is the passing run's.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
FeatureKind::Fillet with corner G2 is kernel::fillet_g2_corners, its setbacks
the corners' distances (the rest 1.5 r); setbacks now go with either corner.
The engine's corner test builds the G2 corner, default and set back 7, one
closed solid of 10 faces within 0.1% of the ball corner's volume.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A corner where three or more rounded edges meet, every edge there rounded, is
set back (1.5 r unless a setback names it) and left open by SMLib
(blend_smlib::fillet_setback_open: the sheet and each corner's hole, in the
sheet's orders); each hole is filled by fill::matched_patch, the height field
over its plane fitted to the position, tangent plane and normal curvature
across the boundary of every face beside it (exact second derivatives), rounds
and planes alike, each edge's pcurve its exact projection; SMLib sews the
sheet and the patches into one solid, which must close. The history is the
sewing's carried through the sheet's, each patch face made from the rounded
edges its neighbours come from.
corner_g2 is that construction now. The polar patch over three sections it
held (never sewn in) could not be: its boundary is a closed cubic through
stations, which rounds off a corner of the loop, and the setback hole has six.
The 60 x 40 x 20 box corner of r5 rounds (07/015): 10 faces, a valid closed
solid (record and SMLib's check), curvature jump across the patch's seams
6.3e-4 per mm clear of the hole's corners against the ball's 0.2; set back 8
between the hull and prism bounds; set back 11 to 15 inside the box but by
under 7e-4; all eight corners of a box filled; refusals named.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SeamReport.stations: each station's edge fraction, gap, normal angle and
curvature jump, alongside its point, so a seam's profile (where a jump holds,
where it fades) can be read and plotted, not only its worst value.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
smlib_blend_setback_open's binding: the sheet, and per corner a Hole, its
edges in order (each starting where the one before ends), each with the face
beside it and whether that face runs it forwards. The holes' buffer starts at
64 edges a corner and grows once to the length a first call names. Tested on
the r5 box corner set back 8: 9/30/22, a sheet SMLib's check passes, one hole
of twelve edges chained end to start; an unrounded edge at the corner is
Unsupported.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fill_loop's bicubic height field over the loop's plane, fitted to the
neighbours' derivatives as well as the loop: every condition of a graph is
linear in its heights (G1: the gradient from the face's normal; G2: the second
derivative along the face's leave direction from its normal curvature), so
one banded least-squares solve with the thin-plate energy decides the surface.
Its parameters stay plane coordinates, so an edge's pcurve is still the edge
projected onto the plane, exactly.
Where a round's section meets a plane at a corner of the loop, the section has
the round's curvature and the plane is flat across its own curve: no surface
is G2 to both there. The G2 rows fade in over 15% of the run from each corner;
asked over a few cells instead, the plane's flatness to the second order along
its curve pulls the surface off the whole side. Grids go to 128 spans, which
G2 to a plane along a curved line needs. The thin-plate weight is a thousand
times fill_loop's: with the slope and curvature rows acting a little inside
the loop, the weaker one let the surface rise past a plane it is G2 to (0.06
mm at a box corner of r5 rounds set back 8).
loop_frame is fit_once's plane, shared.
Tests: the hole a setback leaves at a box corner of r5 rounds set back 8,
within 2.2e-5 of the loop, 0.027 degrees and 4.7e-4 per mm of its neighbours
clear of the corners; set back 7.5 to 15, under its planes but by under 7.2e-4;
a hole in a cylinder filled with the cylinder's own second order; refusals.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
What each category records, self and total time, the table and the Chrome
trace; the CLI's --profile and --profile-out on scripts, designs and STEP
files; diagnostics.profile over the control channel.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
The profiler is process-wide, so the engine's other tests (the schema
check's failing diagnostics.profile calls among them) record into it while
this test does. It now shows the reset by the wall time starting again,
and checks only rows that are certain to be there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
- Graphics page: "Profile operations" (off by default) starts and stops the
profiler when it changes; an agent's diagnostics.profile in between stands,
and SOLIDCRAFT_PROFILE=1 is not undone by the default at start-up.
- The desktop app reads SOLIDCRAFT_PROFILE at start-up.
- Labels in en, ar, de, he, ja, ko, zh. The pending-messages test compares
right-to-left pending text in display order, as the catalogs give it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Session::execute is a command span; Model::evaluate and each feature computed
(by its type) are spans; io's import, export and design read/write too; the
STEP reader's topology, pcurves and solid record.
- diagnostics.profile (contract + output schema): report, start/stop, reset,
Chrome trace file.
- A STEP/IGES import's reply carries its own profile while recording.
- Every app compiles the profiler in by default (feature `profile`);
SOLIDCRAFT_PROFILE=1 starts it.
- solidcraft-cli run/eval/bench-open --profile print the table on stderr,
--profile-out writes the trace; run --profile works on designs and STEP files.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Booleans, blends, sweeps, lofts, shell/draft/offset, healing and sewing,
SMLib conversions, meshing and measuring; the import as parse, index,
per-solid read, placement and mesh check, convert batches and finish.
The existing per-step boolean/tessellate capture names its spans too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The C API logs each call with its steady_clock time; the log sink records
that as an smlib span (child of whatever runs on the calling thread).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
solidcraft-profile (L0, standalone): spans with self and total time, SMLib-timed
records, a bounded trace (200k events) and an aggregate that stays; compiled in
with the `enabled` feature, off at run time until switched on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An edge is left out of the shaded view only where its two faces meet
with one tangent plane (normals within 0.5 degrees) and one curvature
across the edge: at each station the faces' normal curvatures square to
the edge agree within 0.5 degrees' worth, |k1 - k2| <= 0.0087 x
max(|k|, 1/L), L the edge's length, so the continued faces turn apart
by no more than that over a radius of curvature or the edge's length.
Fillet boundaries (0 to 1/r), corner-patch joins and a plane tangent to
a cylinder are drawn again, straight ones included. A modelled thread's
flank pieces (one sweep) differ by at most 0.0046 and stay hidden.
Second derivatives of a spline surface now exist at a boundary point a
pcurve carries a rounding step past the knot range, as its point and
first derivatives already did.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Smoothness (normals within 0.5 degrees) and a crease are separate
verdicts. A crease is certified where the normals differ by more than
the evaluation can account for: each pcurve point's measured distance
off the edge's own curve times the faces' larger principal curvatures.
So a real crease too small to see is not drawn in the shading but stays
a pickable edge in the wireframe, overriding a coarse same-surface seam
fit, as before; exact tangent joins (fillets, seams, a thread's flank
pieces) are no crease.
The viewport's smooth-wall test expects the wireframe to draw
`edge_visible` edges: the offset ellipse's pieces meet at real creases
of 1e-4 rad that the old checker could not certify.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The M10 thread's 32 flank-piece joins meet with normals about 1.5e-6
degrees apart (measured independently with seam_diagnostics); each is
left out of the shaded view, stays pickable and stays in the wireframe.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An edge is left out of the shaded view when its two faces' outward
normals, evaluated on their exact surfaces at each use's own pcurve,
agree within half a degree at eight stations along it (a seam one face
meets itself across is the same test with its two pcurves); it is a
protected crease when they differ by more. Replaces the special cases
for ruled clamped Bezier patches, planes, cylinders and cones. Fillet
tangent joins and corner-patch joins are hidden too; they stay
pickable and in the wireframe.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The truck vendor/ crates went with the kernel's move to SMLib, so their
mapping and the test that selected the kernel for vendor/truck-shapeops
are gone; egui-wgpu is the one vendored crate left.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LUXeKMoxpM7Vuv8Rb6E3K2
A collapsible if (assemble.rs), two manual assign operations (sew.rs), a range loop indexing
the placement inverse's translation row (step_in), and two needless borrows (engine features),
which failed CI's clippy -D warnings. No behaviour change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Seven labels of the thread dialog had no English catalog entry, so they stayed English in every
language (every_ui_label_literal_reaches_an_english_catalog_entry): Extent, Ends, Chamfer Angle,
Chamfer Depth, the two ISO 4753 hints and the designation hint. They are in
assets/locales/pending/thread-dialog-labels.json with Arabic, Hebrew, Japanese and Chinese, the
catalogs that must be complete. The ISO 4753 hints are looked up by id, and the designation hint
takes its example as {example}, so a right-to-left language places it where it reads.
pending_messages_are_in_the_catalogs compared a pending message with the catalog's text, which
for Arabic and Hebrew is in display order since #3768; it compares with that order now, as the
catalog stores it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Main's Arabic and Hebrew catalogs (#3768) arrived with thread.direction, thread.left_hand and
thread.right_hand, the labels of #4414's screw direction, which this branch replaced with the
thread's hand and took out of every catalog.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each job fetched the prebuilt SMLib into target/ before Swatinem/rust-cache restored target/,
which replaced it, so every build that links SMLib stopped with "SMLib ... is not fetched"
(the Docker image of #7403). The fetch now follows the cache restore in every job.
develop (the SMLib kernel's trunk until it merges into main) gets main's push runs: CI runs on
pushes to it, a push compares against the last passing push to its own branch, and its runs save
the Rust caches the PRs into it restore.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Move/Copy's pick from the browser (#4774) read Session::world_state as the placed model; it now
returns why a model cannot be placed, so the browser reads it as the other panels do, through
open_meshing::session_world (no placed bodies when it cannot be placed).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
v0.24.0 (format 39) saved a left-hand Thread or tapped Hole as `handedness: "left"`; this
branch replaced main's handedness (#4414) with the thread's `hand`, and features don't refuse
unknown fields, so such a design opened right-handed without a word. The pending format change
thread-hand.rs renames `handedness` to `hand` on Hole and Thread features (a `right` is
dropped, anything else refused).
Tapped holes take `hand` again, right or left (never both, and only with `thread`): the Hole
feature saves it, its thread is cut and annotated that way, solid.hole takes `hand?`, the hole
dialog's tapped rows have the thread dialog's Hand choice (without Both), editing a hole keeps
it, and hole presets save it under v0.24.0's `thread_options.handedness` key, so presets saved
by v0.24.0 recall it.
Test: a format-39 design with a left-hand thread and tapped hole opens left-handed; a hole's
`hand` refuses both and needs a thread.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every job that builds the kernel with SMLib runs scripts/fetch-smlib.sh after the toolchain:
the host's prebuilt archive, the matrix target's for cross builds, wasm32's for the web jobs.
RISC-V has no archive, so its check compiles the sources with g++-riscv64-linux-gnu.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HoZgkCdAVrscuw2kjQTWDs
tangency-labs/smlib's prebuilt workflow builds the C API once per commit for every Rust target
SolidCraft builds (Linux x86_64 and aarch64, macOS arm64 and x86_64, Windows x64, x86 and ARM64,
wasm32 with wasi-sdk's C++ runtime) and releases it as prebuilt-<commit>. SMLIB_PREBUILT pins each
archive's SHA-256. scripts/fetch-smlib.sh now downloads the host's archive (or the named
targets'), checks its hash and the commit it was built from, and falls back to the sources for a
target with no archive; --source fetches the sources, --pin writes SMLIB_PREBUILT from the
release. build.rs links a fetched archive whose hash matches the pin, else compiles as before,
now with -ffp-contract=off like the archives: every platform does the same arithmetic. No
network access in build.rs.
SMLIB_REV moves to 8b57e7e: the workflow, and fixes for MSVC (no directive inside a macro
argument, SM_PI for M_PI), the no-fused-multiply-add build and a test that needed a POSIX shell.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HoZgkCdAVrscuw2kjQTWDs
On the web target the smlib build compiles SMLib for wasm32-wasip1 with wasi-sdk's clang
(WASI_SDK_PATH), with wasm exceptions in the exnref encoding that wasi-sdk's libc++ and Rust
use, and links wasi-sdk's libc++, libc++abi, libunwind and libc. The WASI calls libc makes are
defined in ffi.rs: stdout and stderr go to the log, the clocks are the page's (the C API times
every call), and there is no file system or environment. solidcraft-web gets an smlib
feature, off until CI has wasi-sdk.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HoZgkCdAVrscuw2kjQTWDs
Code main added since the last rebase reads the placed model, a camera or a scene
as values; on this branch they return Result. Picks and the edit dialogs read the
session's placed model through open_meshing::session_world, as the others do; the
revolve and circular-pattern axes, the model axis of a command and the warming of a
computed edit pass the error on; tests unwrap. A test meshing bodies uses
Body::mesh, the one mesher.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HoZgkCdAVrscuw2kjQTWDs
The reworked thread has its own direction, `hand: right|left|both`, on SMLib's
exact screw sweep. Main's #4414 added a second one, `handedness: right|left`, on
Hole and Thread, for the faceted groove this branch replaces. It is taken out: the
field, its pending format change (never released), the hole dialog's direction row,
its hole-preset option, its strings and its tests, which measured the faceted
groove. A tapped hole is right-handed, as before #4414.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HoZgkCdAVrscuw2kjQTWDs
A modelled thread never changes the size of the shaft it is on (#1018, #2853): a
shaft thicker than the thread is refused with the size to make it, as
tests/kernel-cases/threads/2705 expects ("this 14.6 mm shaft is too thick for a
modelled M14 thread; make the shaft 14 mm for M14"). This takes out the turn-down
of #4256: `thread_turn`, the fit that let a thicker shaft through, and its tests;
the 9 mm shaft refusing an M8 and the preview's M6 refusal on an 8 mm shaft are
back as they were before it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HoZgkCdAVrscuw2kjQTWDs
Main's reader (#1609, #3845, #3928) and the branch's met in the rebase, and the
tree kept main's gap closing: it moved a trim piece's end vertex onto the mean
of the gap's ends without changing its curve, and widened the STEP file's
stated accuracy to cover the move. SMLib checks each edge against its vertex at
the body's tolerance and refused the box (SmTrimmingTools Edge/Edge Gap 0.005 >
0.00425, rig 3845); the first piece of a trim curve with no start point was
also turned against the second (rig 3913, "edge loop is not connected").
The reader keeps main's structure (groups and subfigure instances as products,
names, placements, separate lumps as separate solids, the originating system)
and sews each group's surfaces the branch's way: each meeting point of two
pieces is one vertex, where most piece ends there agree, and a line or clamped
B-spline that misses it is rebuilt to end there exactly. Main's limits on which
gaps are one corner written loosely (0.1 mm, 5% of the shorter piece) decide
which meeting points are joined; the stated accuracy stays the file's. A trim
curve's first piece starts at its end away from the second when no start point
is given.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The IGES reader (#1609) writes each group and subfigure instance's placement as a
CARTESIAN_TRANSFORMATION_OPERATOR_3D_NON_UNIFORM. The branch's STEP reader took
only the uniform operator, and only its origin: every IGES assembly placement
was refused ("placement location: expected a reference") and put at the origin.
An operator, uniform or not, is now read as main reads it: its axes (the global
ones where omitted), its scales (1, and the first for the second and third) and
its origin, refused when singular. A placement may then scale, so the inverse
that maps one placement onto another inverts the linear part by its adjugate
instead of transposing it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The corner pick (#5173) counts the edges meeting at a picked point; seams already
did not count, and now a tangent join between two spans of one smooth curve does
not either (4332: a chamfer on a spline rim was refused as two edges).
3383: the first occurrence is grounded by default on main, so the recipe
ungrounds it before moving it. 5024: a hole's position must lie on a face, so the
centre is on the box's top edge; the volume is half of the drilled cut, exact.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The branch's placement change (placing the model returns its error) made
world_state, home_camera, bounds, open_sketch_plane, the scene builders,
to_local and auto_operation return a Result. Main added callers of them
since the branch left it; they now take the error as the branch's own
callers do:
- engine: commands pass it on (`?`); helpers answering yes/no or "found"
(snap points, a face's normal, the DXF export check) answer no for a
model that cannot be placed. Frames keep the session for main's lazy body
frames (#6356) beside the placed model the picks are on; a library apply
maps its targets on that placed model. Before an extract or a link is
replaced, the placed model is taken or the command fails.
- ui: panels, picks, dialogs and previews read the placed model through
open_meshing::session_world (none when it cannot be placed: the viewport
says why); Home and Fit frame nothing then.
- cli: the usage table is main's, with kernel-cases in it (a resolution of
the rig's commit had kept the branch's whole file).
- tests read the Results with unwrap, and main's tests of Body::tessellate
mesh through Body::mesh. Main's test of saved boolean retries goes (the
branch saves none), as does the thread half of its dialog rendering test
(the branch's thread dialog has its own); the tapped hole half stays.
- What no caller uses after the merge goes: a mesh body's span, the metric
thread refusal text, and the revolve's targets-as-shown model (main used
it for the coplanar run-on the branch removed). Main's SolidWorks Y-up
test reads the exporting system's name (the branch's STEP header has no
preprocessor or description fields).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Main gave the Thread feature two things the branch's thread rework (forms,
hands, starts, extents, runouts, tips on ThreadAsk) was written without:
- Tapped holes take catalogue sizes (ISO metric and the catalogued UNC/UNF,
G and Rp sizes): `ThreadAsk::tapped` reads the designation from the tapped
catalogue, names the thread as the catalogue does, and fits a size that is
not metric to the hole drilled for it (its tap drill). The hole feature
asks for its thread this way.
- A modelled external thread on a shaft thicker than it turns the shaft down
to its major diameter first (#4256): main's `thread_turn` makes the ring,
the shaft is cut by it, and the grooves and a chamfered tip are cut at the
turned radius. Such a shaft fits the thread.
ThreadInfo keeps the radius of the face it is on, which main's cosmetic
thread display draws at (0 in a state saved without it).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The branch rebased onto main keeps each of its commits; where both sides
changed the same code, these are the pieces neither side's commit had.
- Cargo: one [features] table in the CLI's manifest (two met in the merge).
- geom: a NURBS surface's `closed` set once in a placed copy (both sides
added it).
- kernel: the whole-face extrude tools (#2040) work around truck's boolean
on truck types, which left the workspace with the SMLib kernel; the
module, its tests and the doc crate's tool and tests go. The display
planner's flat fill (tess/fallback.rs) has no caller with the schema 4
mesher, and the tessellator's seam and uv modules are the branch's. A
body's boundary tolerance from a STEP file belongs to the truck body the
schema 4 record replaced: `with_boundary_tol` goes. `unmeshed_faces`
counts the coarse mesh as the branch does. A loft's section re-split
passes the lead's perimeter shares to main's `resample` (a circle breaks
there when the matched corners do not go round in order). Main's tests of
the mesh-based measure (the SMLib kernel measures exactly) and of the
IGES preprocessor name (no such field on the branch's import record) go.
- doc: a feature result carries no boolean diagnostics on the branch, so
the model cache neither writes nor reads them, and a pattern copy's miss
test runs the intersect as it is. A placed body's worker result is a
Result: a body taken from the previous placement is Ok. A thread's
clipboard remap takes main's arguments, the tapped catalogue names a
metric size by ThreadForm::Metric, and main's tests unwrap the placing
calls that return their error.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The r7 on the vertical edge runs on (tangent chain) up the side's arc on the
r5 round, a circle of radius 5: its contact line on that face would be the arc
offset 7 towards its centre, so no such round exists and the right answer is
a clean refusal. 018 picks a radius that fits (r2: 11 faces, inside the box).
cases.py now writes 013 (the corner setback, graded by its convexity bounds)
instead of marking it missing; its tolerance is the interval's exact half
width.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
003 and 004 passed on one valid body alone. 003 is now the flange less
its M30 groove (stopped clear of the root round, run out through the
chamfer) and three M8 tap drills; 004 is 10/008 less the M16 groove on
the shelled boss, which cuts the wall only. With bboxes. Both hold on
SMLib (worst 1.4e-7 relative).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
It passed on one valid body alone. Each d3 vent goes through the shelled
top skin into the cavity and meets nothing else, so the case is 09/011
less 4 pi 1.5^2 1.5; with the bbox. Holds on SMLib (1e-6 relative).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
010, 011 and 014 passed on one valid body alone. 010 (top rounds in
mitres, shelled open below) and 014 (drafted box with tilted rounds:
every level section loses the same elliptic corner) now check closed-form
volumes, exact counts and bbox (014 also its centre of mass). 011, the
running part shelled 1.5, checks the part less its cavity built feature
by feature, the offset faces meeting sharp at concave edges as in the
rest of the tier (a distance-based offset would differ by 0.3%). All hold
on SMLib (worst 1e-6 relative).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
It passed on one valid body alone. It now checks the chamfer's prism, the
two r2 rounds below it and the columns they carry on up into the chamfer
face (the section's first moment), with exact counts and bbox. Holds on
SMLib (8e-9 relative).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
010, 011 and 012 passed on one valid body alone. 010 (r7 then an r3 chain
round its arc) and 011 (two r5 rounds in a mitre) now check closed-form
volumes, counts and bbox; 012's variable round takes only its edge and
runs on into the vertical round's column, so its volume is graded as the
interval between that end carried on as a constant r5 tube and as its
growing radius. All hold on SMLib (worst 4e-8 relative; 012 at the
growing-radius end).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
010 passed on one valid body alone: it now checks the tier-05 part less
the M16 groove (stopped w above the boss's shoulder, run out at its free
top) and the bbox. Two new cases with closed forms: 018, an M10 cut both
hands, whose crossing grooves leave the material fraction squared
(2 pi L int r m^2 dr, exact for L a whole number of pitches); 019, a
two-start Tr40x14(P7), six pitches of the Tr40x7 section. All hold on
SMLib (worst 2e-6 relative).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
07, 08, 09, 11 and 12 passed on one valid body alone. Each now checks
its volume, bbox and (where derived) counts, area and centre of mass:
the three-section loft is the degree-2 interpolant r = 10 + (z-20)^2/40;
the square-to-circle loft a ruled prismatoid, its mixed area from the
sides ruled to rational quarter arcs; the periodic fit spline the cyclic
C2 chord-length cubic; the pleated skins star polygons on fixed rays,
their crossing integrated exactly per wedge; the cap's union integrated
as y-intervals over x and z. All hold on SMLib (worst 9e-8 relative).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The generator had fallen behind the tree: it no longer wrote 02/021-025,
03/010, 04/011-012 and 09/012-015, and its 07/016 lacked the second round's
tangent_chain. Each is now built in cases.py; the two reporters' scripts
(04/011, 04/012) are kept verbatim in reported.py. Regenerating reproduces
the tree exactly; the only file changes are the generator's own format
(volumes to 9 decimals, one-space indent, a final newline), all under 1e-8
relative.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three r5 rounds at (60,40,20) set back 8: the volume between the convex
hull of the body on the corner cube's faces and the rounds' prisms (both
bounds closed form or a sampled hull below the true one), the box's bbox,
and SMLib's setback-patch topology (15 faces, 36 edges, 23 vertices).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
FeatureKind::Fillet gains setbacks (serde default: none, omitted when empty),
each a corner vertex and a length expression; a fillet with setbacks is
evaluated by kernel::fillet_setback. Constant-radius exact fillets with
rolling corners only (refused otherwise, in the command and in evaluation).
One distance per corner: SMLib keeps one setback per corner, not per edge.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each (point, distance) sets back the corner at the body's nearest vertex: its
patch starts that far along every edge there. Tested on the 60 x 40 x 20 box
corner with r5 rounds set back 8 (volume between the convex hull and the
rounds' prisms bounds, 15/36/23 topology, the box's bbox) and on refusals.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
At a pole or apex `EvalSurface::eval` takes the normal from a step beside the
point, trying (+h, +h) first. At the far end of an interval (a sphere's north
pole at u = 1) that step lands past the end, where the extrapolated surface
folds back over the pole, and the normal came out turned over: the pole vertex
shaded dark with a light ring around it (ladder 01/005, every sphere and dome).
Steps now stay inside the active interval of a spline or a revolution's
profile. The pole test checked only that the normal was radial, either way;
it now checks the side too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Recorded on SMLib replay 2c5baae (screw sweep walls exact and within their
tolerance, the radial sweep, the point solve on long rational surfaces) with
the thread rework: 06/008 (two starts) and 06/009 (runout) now graded and
passing, 06/011-017 new (Tr, Acme, S, ASME buttress, an internal Tr nut,
left hand, the chamfered tip), and 06/003, 004 and 007 grading the extent and
the collar. 132/135 pass; left: 07/013 corner setback, 07/015 G2 corners
(unsupported) and 07/016 three sequential corner rounds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Sizes, designations and the face's size come from ThreadForm (sizes, parse,
format) and suggested_thread_in, per form; the dialog's own trapezoidal and
buttress lists are gone. Acme sizes are listed by designation. The square
thread, which the command refuses (no standard defines it), is not offered.
Editing a thread reopens its form, hand, starts, extent (length or up to, with
its pick), start end, runouts, tip and steep flank, and OK gives them back.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Ladder 06/009 had no way to ask for one: "solid.thread has no runout (vanishing thread)
length". With SMLib's smlib_screw_sweep_radial (replay e7acea1):
- smlib: the FFI binding and Brep::screw_sweep_radial, the profile moved off the axis by a
piecewise linear law of the turn; crate test: a tooth moved out 1 mm over three turns has
A 2π (c turns + ∫ offset dt).
- kernel: sweep_helix_radial, sharing sweep_helix's checks (helix_profile).
- doc: `ease_in`, `ease_out` (a length, or ISO 3508's standard 2.5 P or short 1.25 P) fade the
groove over that length, its section moving out of the material as it turns (outward for a
shaft, inward for a nut). ease_in is at the end the thread starts from. The runout lies past
the full-form length: beyond a square stop, before a shoulder up to its clearance; at a free
end inside the face. Runouts longer than the thread are refused. No end gets one by default,
so saved threads keep their geometry.
The cut is the integral over the turn of the clipped section's 2π∫ r dA: engine test (2.5 P
and 1.25 P between two collars) within the sweep's wall tolerance times the runouts' wall area
(0.029 against a 0.036 mm³ bound), ladder 06/009 (004's shaft, a 2 P runout before the
shoulder) graded in closed form (measured 1.7e-6 off).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
`tip: chamfer` cuts ISO 4753's chamfered end at each free end a modelled external thread
reaches, after the groove, so the last teeth taper into the cone instead of ending in a
ragged first turn: the material outside a cone `chamfer_angle` to the axis (default 45°)
reaching d_f = d - 2 `chamfer_depth` at the face's end (default the external minor diameter,
the largest the standard allows), a revolved ring cut by one boolean per end. Past the end the
ring keeps to the radii the groove's overrun already cuts. An internal thread is refused (its
entry takes the hole's countersink), as is a tip with no free end or on a cosmetic thread.
The cone is axisymmetric and cut after the screw-symmetric groove, so the volume is the stock's
chamfered section with the thread's material fraction: engine test (both ends 45°, and 30°
x 0.5 mm at a shaft's one free end above a collar) within 1e-4, ladder 06/017 (measured
2.4e-7 off).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Ladder 06/008 had no way to ask for one: "solid.thread has no start count". `starts: n` (1 to
8) cuts n grooves of lead n x P, the j-th a j/n turn round from the first (the same helix j
pitches along), in one boolean with their union. A designation that gives the lead
(M16xPh3P1.5, Tr40x14(P7)) gives the starts, and a `starts` that disagrees is refused. The
helix's self-clearance check and the turn limit use the lead.
Over the same length n grooves of lead nP take the volume of one start of pitch P: engine test
(2 and 3 starts, M10xPh3P1.5) within 1e-4 of the closed form, ladder 06/008 graded as 001
(measured 3.5e-7 off).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Only the ISO metric 60° thread existed. Forms now (crates/doc/src/thread_forms.rs, values and
sources in docs/threads.md): ISO metric (ISO 68-1), trapezoidal (ISO 2904 design profile, ISO
2902 sizes), general purpose Acme (ASME B1.5), metric buttress 3°/30° (DIN 513), buttress inch
7°/45° (ASME B1.9 basic profile). Each has its groove as a straight-flanked polygon, its standard
sizes (none for ASME buttress), its designations (M10, M16xPh3P1.5, Tr40x14(P7), S20x4,
1/2-10 ACME, 2.5-8 BUTT) and the size that fits a face. The square thread is refused: no current
standard defines it.
`hand: right | left | both`: a left-handed groove is the screw motion about the reversed axis
rising along it; both cuts the two grooves, crossing, in one boolean. `steep_flank` turns a
buttress. The design diameters are checked against the published tables (ISO 2904/DIN 103,
DIN 513 via IS 4696-3); every form's cut against 2π∫ r w(r) dr per turn to 1e-4 in engine tests
and to 1e-4 in ladder 06/011 to 016 (they measure within 5e-7).
FeatureKind::Thread holds every option the command has (form, hand, starts, ease_in, ease_out,
tip, chamfer_angle, chamfer_depth, steep_flank, all defaulted so saved threads load unchanged);
starts, runouts and the chamfered tip are refused as not built yet. suggested_thread_in takes the
form.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A thread of some length ran from "the end of the face nearest the pick", a coin toss for a
pick in the middle (rig 602's 4 mm thread meant for the free bottom end ran at the head end,
cutting 1.17 mm into the hex), and every thread end at a face end got a one-pitch overrun,
which at a shoulder gouged the material past it (ladder 06/004 cut 0.9% of its volume out of
the collar).
- kernel: material_in_ring, the exact intersection of a body with a ring about a cylinder
face's axis, by an SMLib boolean.
- doc: a face end is free when no material lies past it, over the span the groove's one-pitch
overrun reaches beyond its first pitch, within the radii the groove cuts (a cap, a chamfer
the groove runs out through, a counterbore). At a free end the groove runs out as before; at
a closed one (a shoulder, a head, a blind hole's bottom) it stops with its widest part clear
of it; inside the face it stops square.
- doc: a thread of some length starts at the end nearest `from`, else at the face's only free
end, else at the end the pick lies within a third of the face of; a pick in the middle of a
face whose ends are alike is refused, naming `from`. A length as long as the face runs the
whole face. `up_to` a plane or a point's place along the axis. New fields `from`, `up_to`
default to none, so saved threads keep their meaning where it was not a coin toss.
- engine: solid.thread takes extent (full|length|up_to), from and up_to.
- ladder 06/003 (now with `from`), 004 and 007 are graded by volume and centre of mass in
closed form (a square stop inside material removes exactly length/P turns, the section being
symmetric); rig 602's method says why the thread is at the bottom, tolerance 0.02%.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Form (ISO metric, trapezoidal, Acme, ISO and ASME buttress, square), size and
pitch from the form's list, hand, starts, extent (full, length or up to, with
the end to start from), ease in and out (standard, short or a length), the
chamfered tip and the buttress steep flank. Only the fields that apply are
shown; a form change keeps the options and a manual size's diameter where
the new form lists it. Defaults stay out of the command, so a plain thread
means what it did.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The ISO metric table, the thread on a cylindrical face, the groove shape and the thread
feature's evaluation move from eval.rs to eval_thread.rs, unchanged, before the extents,
forms and starts work changes them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face's boundary nodes are its edges', shared with the neighbouring face.
On a nearly straight stretch (the base of a thread flank's corner, clipped by
a hex head's flats, on an iso-line of the helix) refinement met slivers
between consecutive boundary nodes: their centroid, or the middle of their
inner edge, lies on the boundary chord to rounding. Inserting it split the
boundary constraint, and the face was refused ("boundary crosses itself";
rig threads/602, refused while the body was meshed to pick the fillet's
edges). A sliver whose centroid lies on the boundary is now split at its
inner edges' middles, and a refinement point that lands on the boundary is
not inserted.
The 602 engine test assumed the old kernel's faceted thread (over 1000
faces); with SMLib the thread is exact, and the test now checks that the
fillet finishes and removes a little material.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three new cases in closed form: two through holes touching at a point
(02/021), the stock shelled with no face open (09/012, a sealed void, two
shells) and a dome shelled from its flat base (09/013, its rim the circle
r18). Nine rig cases copied into their tiers: a cylinder grazing a hole's
edge (1279), lips 1e-4 off a base's walls (2928), wall halves merged into a
flange (494), flush chord layers (5197), the horn torus revolve (4007),
crossing pleated loft skins (4623), loft ribs joined to a rounded pad (4092),
a drafted box with rounds shelled (3576) and a shelled dome with its ring
joined back (4847). Golden recorded on SMLib replay 8017df9: 123/128 pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A notch cut into a sheet's base is a cut-out, so the base outline that
edge_at and the edge names read has no edge along it: a flange picked on
the notch's bottom edge took the name of the nearest outline edge (the
plate's far edge) and was built on the stretch of it the notch left. A
pick nearer a notch's edge than any outline edge now names no edge, and
the flange runs the base's outline round that notch first, keeping as a
cut-out only what it takes of the flanges, then trims and bends as on any
edge of the outline. The edge of a hole inside the base is refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Convert to Sheet Metal keeps a shell's (and a plate's) own thickness, but
the rule's values were evaluated with `Thickness` the rule's: under the
default rule (2.5 mm, bend radius `Thickness`) an L 2 mm thick with its
bend modelled R 2 was rebuilt bent R 2.5, losing 8.58 mm3 (2659.911
against 2668.496), and its gap and reliefs followed 2.5 mm too. The rule
is now evaluated for the sheet's thickness (Document::rule_values_at):
in the conversion once the shell's thickness is known, for a converted
plate, and for the flanges, hems and folds added to an existing sheet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
solid.mirror and solid.pattern of bodies pushed each copy as a plain body:
no sheet model went with it, so its flat pattern, DXF and bend table were
refused ("`Body1 (2)` is not a sheet metal body"). Each copy now gets its
source's sheet model moved with it (SheetBody::moved). A reflection turns
the frame's handedness over, so a mirrored copy's frame sits on the moved
top face, thickness growing back toward the moved bottom: the same blank,
each bend turning the other way, machining depths taken from the other
face. A sheet feature on the copy rebuilds it in place.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The countersink's cone started 1% of the drilled body's size above the
entry face, widening on to rc + top tan(half) there, so above the face
it cut beyond the countersink's rim: a wall 0.15 mm from the rim was
notched (#5041, 12 faces instead of 10). The tool is now one revolved
profile: the cone from its rim on the face to just inside the bore, and
above the face the rim's own cylinder, as the bore's cylinder runs on
above it. The cone stays the profile's first segment (csink:side:0).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A Combine took its tools in their own component's coordinates and joined
them to a target kept in another component's: a root body joined into a
body of a component turned a quarter turn landed where the turn carries
its root coordinates (#3383, 2 shells and 17000 mm3 instead of one solid
of 16750). Each tool is now moved through its own placement and the
inverse of the target's (component placement, then body offset), as
state_in_frame places bodies, and a combine's cache key follows the
placements so moving a component recomputes it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The case's base is a stadium ring (r 3..5 arcs, centres 20 apart, z 0..2),
as its description and expected volume (2 (80 + 16 pi) for the base) say,
but its extrude picked no profile, and "all profiles" is every drawn region:
the ring and the stadium island inside it. The base came out filled and the
join 1066 mm3. SMLib's union was exact for those operands. The extrude now
picks the ring (the region at (4, 0)).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The record's torus label is a ring torus (major > minor; the validator
refuses any other). SMLib keeps horn and spindle tori: a half disc revolved
about a line through its rim's end is a horn torus, its major radius a
rounding below its minor, and the export wrote it as a torus label the
record then refused (rig sweeps/4007, "ring torus radii"). Such a face is
now its NURBS form, unlabelled, which is its geometry in any case.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
solid.shell refused an empty face list ("select 1…1000 faces"), and the
shell feature found its body from the first face picked. SMLib shells with no
face opened (the void sealed inside, its own shell): the command now takes
`faces: []`, the feature shells the one body there is or the one named
(`body`), and an outside or both-ways shell pushes no open face out of the
cavity. Rig shell/4009 declares the void's shell (`shells: 2`), as its method
says.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Where a face's loops pass a vertex more than once (two holes touching at a
point, a loop through another hole's corner), each pass took its parameters
from the pcurve of the edge leaving it, and the two pcurves round them 1e-14
apart. The triangulation got two nodes 1e-14 apart with constraints crossing
between them and refused the face ("boundary crosses itself"): rig 307, 3092
and 3887. A pass whose parameters agree with the vertex's first pass to that
rounding (1e-9 relative) now takes the first's; farther apart they are two
places of the surface (a seam's two sides, a closed spline's ends).
Rig 494 now fails: its revolved face has a dead-end edge walked down and back
up in one loop, which met the triangulation as a needle 1e-14 wide and now as
the slit it is.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fillet_max bisected the radius down on any error of a candidate fillet,
not only KernelError::TooLarge, and fillet_max_each fell back to one
radius for all on any error of a group or of the combined fillet. A
kernel failure at the radius asked for (or at any candidate) came back
as a smaller radius reported as "the largest that fits".
fillet_max and fillet_max_each now return Result<Option<..>>: only
TooLarge narrows the search or falls back to one radius, and any other
error is passed on. Ok(None) is still "no radius fits". The doc's
FilletFit::Reduce passes the error on. The test of a point with no edge
near it now expects the fillet's own error, not "no radius".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The pieces of an IGES trim curve run on from one another, and the sewer made
each meeting point one vertex, but left a piece that missed it where it was,
the gap on the vertex's tolerance: an edge 0.01 mm short of its corner, 2000
times the file's resolution, which SMLib refuses (SmTrimmingTools #0, SmBrep
line or clamped B-spline piece that misses it is made to end there exactly;
any other piece that misses keeps its ends, and the gap is reported.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0113gDVmStT5tCGKW9Fxgc5K
A cone face bounded by its base circle and its seam out to the apex and back
turns once round the axis and closes at the apex. add_seams counted the turn
as a loop without a seam and added another, out and back from the same
vertex: the drill point of a turned pin read back from STEP had a sliver
whose boundary crossed itself, and the body would not tessellate (#5193). A
winding loop with a vertex on a pole is left as it is.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0113gDVmStT5tCGKW9Fxgc5K
form::extent took the box of 32 points per edge. A face's NURBS form is built
over that box, so where the boundary peaks between two samples the form
stopped short of its edge: the re-imported rod's bore ended at v = 9.99977
while its intersection with the rod reaches 10, and SMLib found the edge
1.9e-4 off the bore (SmBrep #4, SmEdge #13). Each sample higher or lower
than its neighbours now brackets the edge's extreme there, found by golden
section.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0113gDVmStT5tCGKW9Fxgc5K
form::extent unwrapped each loop of a face from its own first point, so a
hole beside the seam (a cross bore through a rod at the rod's seam angle)
came out a period away from the outer loop. The box over both was more than a
turn, cut to one turn from the hole: the face's NURBS form ran over quarters
3..7 while its seam uses lay at 4 and 8, and SMLib found the re-imported
rod's edge ends outside its face domain (SmBrep #70-#72). Each loop after the
first now goes the whole periods along that keep the face's box smallest.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0113gDVmStT5tCGKW9Fxgc5K
tangent_chain followed tangent edges without looking at their convexity: a boss's concave base
arc that runs on tangentially as the convex rim of the overhang above it (rig fillets/4174,
4369) and a convex outline running into the concave edges under an overhang (4365) were sent to
SMLib as one chain. No round can pass from one side of the faces to the other, the cases'
answers stop at that vertex, and SMLib rightly refused the request (1001 in its Nx2 corner).
A chain no longer runs on into an edge of the other convexity: convex(e, v) tells, from the
second face's natural normal, the loop's direction and the first face's normal, whether the
faces meet at less than 180 degrees inside the body. With SMLib's corner and end fixes this
makes rig 4174, 4365 and 4369 pass with their analytic answers; nothing else changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0113gDVmStT5tCGKW9Fxgc5K
Rig case shell/4711 (a 60 x 40 x 10 block, an r3 through hole at (22.5, 7), a 10 x 10 notch on
the front edge, shelled 2 open at the bottom) failed with "SMLib shell: operation: SMLib failed
(status 1027)". The hole is 4 = 2 x 2 from the front face: the hole wall's offset (r5) is tangent
to the front's at (22.5, 2), the walls meet along a line and the cavity would pinch to zero
thickness there, which is no solid. Refusing it is right; the refusal did not say why.
SMLib now refuses an offset whose walls would meet with its own status, SMLIB_ERR_WALLS_TOUCH
(-1110). The smlib crate reads it as Error::WallsTouch; the kernel's offset backend turns it into
a sentence with the wall, as blends turn NoRoom into one: "shell: the 2 mm walls would meet along
a line (between a hole and the side, for example), which leaves no solid there; use a thinner or
thicker wall, or move the feature". Thicker builds too (SMLib shells 4711 at 2.5, the walls
crossing and merging between the hole and the front), so the sentence offers both.
4711 expects that refusal and its message ("outcome": "refused", "refusal"), its "what" and
"method" saying why: the wall is 4 = 2 x 2, the walls meet along a line. Tests: the smlib crate
shells 4711 built from primitives (2 refused as WallsTouch, 1.5 and 2.5 valid), the kernel a
plate whose hole is 3 from its side at 1.5 (refused, with "the 1.5 mm walls would meet along a
line").
Results (run-sc.sh against SMLib a759622's code, which differs from the tree run only in
comments and docs; baseline 858f381fc-5448b9a): rig 410/447 (409
before; 4711 fail -> pass as the refusal, nothing else changed), ladder 109/116 (unchanged),
kernel lib 535 passed / 202 failed (one new test; the same failing set), smlib crate 53 / 2
(one new test; the same failing set).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0113gDVmStT5tCGKW9Fxgc5K
SMLib now builds a round or chamfer whose rail runs off its face across a cliff (an opening's
rim, the body's outline): the blend is made as if its faces continued and what it removes is
subtracted, so it never bridges an opening (owner decision for rig 5158/5181; fork commit
"SMLib, capi: a convex blend overflowing a cliff is subtracted from the body"). The tests that
expected a refusal for such blends now expect the body, each answer from the geometry:
- variable_radius_on_a_box_edge: 2 -> 50 on the 10x20x30 box's top edge builds at
3546.18767939 (circular sections in y = const, the r x r square less the quarter disc,
clipped to the box; rel 1e-4).
- variable_radius_from_a_spline: the "big" law builds at 6000 - 726.5334566720 (de Boor law,
split at the knot and where r crosses 10).
- a_chamfer_meets_sloped_end_planes: chamfer 25 removes 11500 - 4625 * slope
(integral of 40 - 2 * slope * y over z >= y - 5).
- g2_fillet_refuses_invalid_geometry: G2 at 25 builds at 6138.1606889 (the quintic section
with control points at thirds, clipped to 20x20, along 20).
- fillet_max_each_reduces_only_where_the_radius_does_not_fit: both edges take 4, 24438.7317761.
- hostile_inputs_are_errors: chamfer 30 on the cube builds at 1000 (y + z <= 10 along 20);
R50 still fails.
- kernel-cases fillets/1080: the 9 mm chamfer into the openings builds at 3350
(method in expected.json).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0113gDVmStT5tCGKW9Fxgc5K
Cases from the issues filed after #4160, each a synthetic recipe with its own answer
(analysis or invariants); every other issue in the range is listed with why it was skipped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Cases from the issues filed after #4160, each a synthetic recipe with its own answer
(analysis or invariants); every other issue in the range is listed with why it was skipped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Cases from the issues filed after #4160, each a synthetic recipe with its own answer
(analysis or invariants); every other issue in the range is listed with why it was skipped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Cases from the issues filed after #4160, each a synthetic recipe with its own answer
(analysis or invariants); every other issue in the range is listed with why it was skipped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Where a cut-out's corner touches the outline (a triangle on the sheet's edge in a bend), four
walls meet along the corner's line. The touching sew paired them round the empty wedges and
gave each fan at the corner its own vertex, so a cap kept its outline and the cut-out as two
loops with a vertex each at one point: SMLib refuses that face (coincident vertices, loops
touching: its face tests 32, 38 and 39), and the bend face's two corners, a hair apart in
(u, v), left a boundary no triangulation follows.
The walls are paired round the wedges of material, as SMLib keeps a solid touching itself along
a line: an edge per wedge, the two sharing their end vertices. A face's loops that meet at a
vertex are joined there into one loop through it twice, SMLib's form of such a face.
With SMLib's face builder taking the same profile (an extrusion of the plate and its
cut-out), fixes rig sheet-metal/1095.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit 0bf765aa815a5552e5b090fd8854b1dc7a981e3f)
A loft cut from a block's whole outline (flush with its sides) down to a pad cuts the slab
above it loose: the slab and the rest touch along the outline's four edges, round the pocket
the loft leaves. SMLib keeps the pocket as a void region of its own (its check finds the body
valid): its one runtime shell is the slab's underside and the loft's faces, so the slab's
material shell faces two empty runtime shells, the infinite region's and the pocket's, and so
does the rest's. The export made one record shell per pair of empty and material runtime
shells, so each solid's closed boundary came out as two open shells ("an open, wire or vertex
shell lies in one region").
A record shell is now one material runtime shell, whatever empty shells its faces face. Empty
regions one solid's boundary faces together are one record region: the infinite region where
it is among them, else the void whose outer shell bounds it alone. The pocket, which no record
shell bounds once the solids' shared edges are split, is part of the empty space round them.
Fixes rig sweeps/96-loft-cut-box-sections-pad-first and -top-first.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit 796920968265c57c59f11dd908ec022d53f5c647)
With touching: separate (the default) a join groups the joined body's
pieces by face contact. The groups that hold the material of the body
before the feature stay in it (each unless it touches one already kept;
the first group where the lineage doesn't say), and so do the groups that
touch nothing: a virtual grouping later operations take as one body. Every
other group that touches another along an edge or at a point becomes a
body of its own, and the feature's notice says so. The joined body keeps
its name and never holds an edge or point contact. A join that only
grouped pieces apart gives no notice, so KeptSeparate::Apart is gone.
one_body still keeps every piece in one body.
SMLib's export copies an edge or vertex that touching solids share once
per solid, so the record never listed those contacts:
Brep::region_contacts now takes such copies (the same curve and interval
between vertices at the same points, a vertex at the same point) as the
edge or vertex they share. A contact SMLib leaves out of the topology (the
3612 spoke tangent to its hub) is found on the exact geometry, the least
distance within the record tolerance, and reported as KeptSeparate::
Touching ("along a line or at a point").
The pending format change join-touching-bodies stays: older designs kept
pieces touching along an edge or at a point in one body too. Its notes say
so, and the reopened legacy emboss still expects touching: one_body.
Tests restored to their form before cc49081df and 733d3bd19: kernel-cases
2896-join-arm-end-near-ring-wall-gap (one body of two shells), 3379, 3455
and 3526 (no touching: one_body); the engine tests of posts joined while
apart (twice), two cubes for interference, a ring with a cylinder, the
buried tab and a_join_adds_only_its_sketch's 0.1 mm gap.
join_meeting_nothing_keeps_the_first_body expects one body and no notice.
The #3378 spoke stays 1 mm inside the hub.
New tests: blocks touching along an edge are split with the along-an-edge
notice; blocks apart stay one body without one; a ring with an arm on its
corner edge and a hub apart gives ring and hub in one body and the arm on
its own (doc); the engine's notice for an edge contact; the record's
contacts of cubes sharing an edge or a vertex (kernel).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each edge is sampled to the chord tolerance on its own, so where two curves of one face run
closer together than their chords sag (the two arcs of a thin lune, the sliver caps of a
joined ring of cylinder segments) the chords crossed although the curves do not, and no
triangulation follows such a boundary ("boundary crosses itself").
After the edges are sampled, every face's boundary is checked for chords that cross another
of the face, in its loop or another, and on a closed surface across whole periods (chords
ending on a pole are left out). The edge segments that cross are halved at their curve
parameter's middle, on the edge, so every face of the edge gets the same new nodes, and the
faces of the edges split are checked again until no chords cross. A boundary still crossing
after 16 rounds, or an edge past its node cap, is an error naming the face and the edges.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
(cherry picked from commit d2b2bdb8e2720690d79123a16bc18c768b1598e8)
Every edge segment now spans at most a quarter turn of its curve, whatever the chord
tolerance: a hole far below the tolerance (a 1 mm hole through a 990 mm plate at a 2 mm
chord) stays a polygon with area, where each half arc had been one chord and the two
chords lay on one line, a hole of no area.
The boundary check takes boundary nodes at the very same (u, v) as one vertex, as the
triangulation does: two loops touching at a vertex they share (a ring standing on a
perforated plate, tangent to the holes beside it) bound a face the triangles follow. A
seam's two uses, one point in space a period apart in (u, v), stay two.
Fixes rig booleans/3665 and sheet-metal/1040.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
(cherry picked from commit 53f4f9a2a991fcd6d1374ca2d647e3a1be36554f)
The r3 on the top edge at x = 60 takes tangent_chain: false, so it
rounds that edge alone and not the r5 round's tangent chain with it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
(cherry picked from commit cf46e2a713c45ae0427e831cae1dc1d3b858febc)
seam_flags matched only an edge between two faces on one curved surface.
A whole turn of a cylinder is one face whose seam it uses twice, and
history::edge_faces lists that face once, so the seam came out as [x]
and was never flagged: it was drawn and picked, and a blend pick at the
vertex where it meets a rim (ladder 07-fillets/017's (23, 15, 20) on a
boss's base circle) resolved to the seam, which SMLib then rejected. An
edge both of whose uses lie on one curved face is now a seam too, for
every consumer of the flag (display, picking, blend resolve, projection).
The blend's own edge selection took the plain nearest edge: the tangent
chain seeded from it and grew into seams (a pick at the top of a boss's
seam ran down it into the base round's seam, ladder 08-chamfers/008), and
the variable fillet's edge and the fillet's grouping of picks could
differ from the edge the blend resolved. They now take the edges the
blend's resolve picks, and a chain never grows into a seam or a smooth
join.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
(cherry picked from commit 709c19c7fd60db522d35e147f21821360bf34668)
Every feature command reads touching (separate | one_body) as it reads
participants; the box and extrude contracts declare it. A feature result
carries notice {message, contact, action} when its join kept solids
apart ("These bodies only touch along an edge, so they were kept as
separate bodies."), the action a timeline.edit that sets touching to
one_body; programmatic calls open nothing.
Tests adjusted: join_meeting_nothing_keeps_the_first_body now expects the
separate body and its notice by default and the old single body of two
lumps with touching one_body; the schema key lists and the
FeatureResult literal gain the new fields.
Engine tests whose subject is a body joined from pieces that share no
volume make it with touching: one_body (posts joined while apart, a
cube joined clear of another, a cylinder joined clear of a ring, a tab
joined in a component); a legacy design reopens with touching: one_body
on its emboss; the 0.1 mm gap of a_join_adds_only_its_sketch leaves the
block a body of its own; and the #3378 spoke starts 1 mm inside the hub,
so the join merges its targets instead of touching the hub along a line.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
(cherry picked from commit 7a2915fe32e6549cbf3170d69e0f5b6518e19da3)
A feature's touching parameter says what a join makes of solids that
share no volume and no area: separate (the default) splits the joined
body's pieces that share no face into bodies of their own, named, with
history and lineage, as a cut's pieces are; one_body keeps one body of
several pieces. The state records how the kept-apart pieces met
(KeptSeparate: along an edge, at a point, apart) for the feature's
notice.
Older builds kept every join's pieces in one body: the pending format
change join-touching-bodies gives each joining feature of an older design
touching: one_body, so existing designs keep their bodies.
kernel-cases 2896-join-arm-end-near-ring-wall-gap: the arm meets nothing,
so it now stays a body of its own (body_count 2, one shell each).
kernel-cases that join solids apart to make the body they test ask for
it with touching: one_body: 3379 (the hub joined to the ring before the
spokes reach it), 3455 (64 pegs in one body) and 3526 (a two-lump target
for an intersect).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
(cherry picked from commit 2c2aaed86f4be60f7f88e40bc24b60de60085148)
Brep::region_contacts lists each pair of material regions that shares a
face, an edge (used by faces of both) or a vertex, with the most it
shares; pairs not listed are apart. Body::region_contacts reports it for
a boolean's result (lump positions as in Body::lumps), and Body::parts
cuts a body into groups of lumps (lumps is parts of one each). Exact
facts from the record only: no geometry, no decision.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
(cherry picked from commit 64a1d688e68e5f0c5b1f9e4fe0d2facfaa4744be)
Every edge of the hand-written cube was a LINE along one filler direction,
trimmed by vertices up to 1 mm off it; the reader refuses such an edge (a
degenerate edge on a line). Each edge now has the line through its two
vertices, so the assembly, instancing and unit tests read a valid cube.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
of_faces gave every edge it was not told about the meeting of its two faces.
A builder (extrude, revolve, loft, a primitive) has no operands: every face it
makes is new, so its edges are new too (EdgeProv: empty when it is new and lies
between faces new to the operation), not where profile segments meet. With no
operands an edge's origins are only what the builder states.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
The display's smooth joins were certified only between ruled B-spline patches
along a Line record. SMLib gives an extruded profile's rulings as degree-1
B-splines of two control points, and sweeps a profile curve within tolerance of
an arc as a cylinder (a ring of 256 cubics is 512 cylinder faces), so no join
was certified. A straight edge is a line or such a B-spline; a plane, cylinder or
cone gives its outward normal along the edge in closed form at both ends (their
difference is the uncertainty, so an edge that is no ruling certifies nothing).
The closed-edge fixture cuts its hole with one circle: SMLib trims a loop of
eight cubics as eight edges.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
The kernel blends exactly the edges given; the Fillet and Chamfer features
expand each pick to its tangent chain (kernel::tangent_chain) before the call
(4593fd07). Six tests written before that passed one pick and expected the
chain: they now ask for it as the features do, with their analytic expectations
unchanged (10 faces along the rounded corner, the sphere corners of #1557, the
boss's half circle, the rounded rectangle's tiny end, the rounded box's loops).
blend_smlib's notes no longer say SMLib follows the chains.
fillet_tabbed_plate_outer_loop_and_partial_selection_after_hole is left as it
is: it fillets every exact subset of the outer edges, and SMLib still fails some.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
A degree-1 patch of four control points on a parallelogram (a plane as an IGES
or STEP B-spline writes it) is that plane at every parameter, so its pcurves
take the plane's closed forms instead of going to SMLib (which drew the rim
circle of 3806's base as a 385-point spline).
SMLib may run a one-face sheet's loop against ours. A use's pcurve is the same
either way, so the loop's way round is read from an edge the face uses once and
every use is matched with it (3806 failed with 'a use SMLib did not keep').
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
A use with no closed-form pcurve (an approximated intersection curve on a cone,
a spline edge on a spline surface) no longer refuses its solid. Its face goes to
SMLib alone, an open sheet with the trims it has and without the others, and
SMLib makes those as it makes every trim it needs (GetOrCreateUVTrimCurve on
export); they come back on the record's parameters and the face is closed and
checked as before. The STEP reader takes the trim maker from the kernel
(read_solids_with, read_solid_with); read_solids stays exact-only.
Kernel lib tests: 490 pass, 239 fail (were 468 / 261; pushed head 475 / 254),
none newly failing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
exact_pcurves gives each use's pcurve where it has a closed form (or is given)
and None where it has none, without closing the loops; pcurves_within closes
them once every use has its pcurve. TrimFace, TrimEdge, TrimUse and the Trims
trait carry a face to a kernel that makes the pcurves of the uses with none.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
along took u or v the edge's own parameter at slope ±1. A rim SMLib makes (an
extruded or revolved circle, four quarters with scaled knots) runs the face's
quarter grid at another steady pace, and a closed rim's two ends are one point,
so neither its ends nor slope ±1 give its line. The pace is read from the edge's
points a quarter and three quarters along, unwrapped from the middle; the line
is taken only where it lifts onto the edge within the tolerance.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
Validation refuses a circle or ellipse edge, or a cylinder, cone, sphere,
torus or revolution face, without its NURBS form. Test fixtures are built on
forms; seam and domain defects use the form's parameter (one turn = 4).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
`Assembly::finish` (lip, emboss, delete face, and the other operations that take a body apart
and rebuild it) puts on a NURBS form every cylinder, cone, sphere, torus or revolution face that
none of its uses brings a pcurve to (over its boundary's extent on its label), and every edge
whose uses bring none and whose faces are planes, splines or on forms (a circle about such a
face's axis labelled on its frame). Their pcurves are then exact on the forms. A face or edge
with a pcurve of its own keeps the parameters that pcurve is on.
Rig and ladder unchanged (333/372, 101/116; 529 and 3368 flip between pass and a 40 s timeout
with the machine's load). The kernel tests of lip, emboss, delete face and assemble fail as
before (none new).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
sew and the STEP reader build a revolved face's NURBS form over `form::extent` (its boundary on
its label) instead of a pass of pcurves on the label, and compute its pcurves within the form's
box. The STEP reader gives revolutions their form too, and a trimmed swept profile its file's
trim range exactly (points on the axis fit either side of it). The STEP seam tests read edge
angles through `Edge::curve_interval`, intervals and domains being in the forms' parameters.
step 37/37.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
- A meridian of a sphere or a tube circle of a torus (a circle in a plane through the axis) has
its exact pcurve on the label: u fixed at the arc's azimuth, v its angle from the outward
radial, alpha + s t.
- A placed revolution has a NURBS form: its profile's control points turned about the axis on
the quarter grid from angle zero, weights multiplied, then its matrix; the record takes forms on
revolutions.
- `form::extent`: a face's parameter box from its boundary projected on its label (the collapsed
parameter at a pole bounds nothing; a boundary that never goes round covers the turn), what its
form is built over with no pcurve involved.
- A face that goes all the way round gets a form of exactly one turn, closed and so periodic.
- `along` projects inside the face's box (a spline extended past its knots, or a profile turned
past the face, has other preimages), and takes a closed form within the record's tolerance, as
the closed forms on analytic surfaces are.
Tests: fits are gone, so curves with no closed form on their face (a tilted circle on a sphere,
an ellipse on a cylinder, an intersection record, a curve weaving across a thread flank) are
asserted refused; the closed-spline fixtures are built as forms. geom 166/166.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
`On::fit` and everything it used (samples, least squares, breaks, filled poles) are deleted, and
the STEP reader no longer retries a face at ten times its tolerance. A pcurve is exact (closed
form, an affine image, or a straight line along a form's parameters) or the face is refused:
"not supported yet: no exact pcurve of <curve> on <surface>".
The rig and ladder are as with the fit (333/372, 101/116; none lost; 3806, already failing, now
refused on import). 16 unit tests that build sphere meridians, torus seams, closed spline
bands, intersection curves and surfaces of revolution on their labels now refuse: those paths
need meridians labelled on their face's meridian plane and revolutions on the quarter grid.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
A surface of linear extrusion or of revolution over a circle or ellipse built its profile with the
reader's own conic spline, on a parameter of its own; the profile is now `form::curve`, the form
its edges carry, so the surface's parameter along the profile is theirs and their pcurves are
straight. The reader's conic spline is gone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
SMLib holds every ellipse as an exact rational NURBS; the fork now sends ellipse edges with that
form and takes them back with it. The export reads an ellipse with its form as it reads a circle
with one (its frame turned so its angle grows along the form, its trims carried over as they are);
the import writes an ellipse on its form as its conic entry plus the NURBS.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
`form::edge` is the one rule sew and the STEP reader share: a circle about a revolved face's
axis is labelled on that face's frame, and a circle or ellipse edge gets its form. The STEP
reader now gives every imported edge its label and form before any pcurve, and every
cylinder, cone, sphere and torus face its form over the extent its label's pcurves give; the
pcurves, the outer-loop choice and the turn-over test then run on the forms, and the record
carries them.
With the pcurve fit turned off, the STEP and IGES circle cases pass (8 of the 9 import cases
that needed it; 3208's ellipse on an extruded spline surface remains).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
A flat-pattern edge folded across a bend is a helix, which no NURBS is: sheet metal already
gives its 3D curve as degree-7 Hermite pieces of the fold, within 1e-7 mm. Its pcurve on the
bend is now built the same way, on the edge's own parameter: degree-7 Hermite pieces matching
the point and first three derivatives of its angle about the axis (through the form's quarter
parameter, by the chain rule) and of its height along it, broken where the angle crosses a
quarter (the form's parameter is only once differentiable there), halved until they lift onto
the edge within 1e-7 of its size. An approximation by construction from exact derivatives, like
the edge; nothing is fitted. B-spline and NURBS edges alike (a NURBS's derivatives from its
homogeneous form). Breaks within rounding of each other are one.
With the pcurve fit turned off, all 44 sew cases that needed it pass; with it on, all 44 pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
The bend's rim arcs carried frames of their own (x toward the arc's start), a fraction of a
quarter from the bend cylinder's, so their forms were off the cylinder's quarter grid and no
straight pcurve lay along them. `form::on_frame` relabels such a circle on the face's frame (the
same circle, its interval shifted); its form is then on the face's grid, its pcurve u = ±s + c.
`along` takes u or v as the edge's parameter either way along, offset. The volume test steps to
a pcurve's ends exactly (t0(1-f) + t1 f), as an interval's end may now carry rounding.
With the fit off, 38 of the 44 sew cases pass (from 29). With it on, the rig is unchanged
(333/372, ladder 101/116, no case changed).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
Every circle and ellipse edge sew builds now carries its NURBS form, its interval in the form's
parameter; every cylinder, cone, sphere and torus face its form over the extent its label's
pcurves give; and every pcurve is computed on the forms, so a bend's rim circle is a straight
line on the bend and an affine image on the flange, exactly. The volume test evaluates the faces
on their forms. A form interval's end within rounding past its last quarter stays on it.
With the pcurve fit turned off, 29 of the 44 sew cases that needed it pass (from none); with it
on, all 44 pass as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
On a plane, a B-spline or NURBS edge's control points go through the plane's affine parameters
and the result is cut exactly (bspline::segment) to the edge's interval, so a circle's form gives
its exact pcurve on any plane. On a spline face (a NURBS form), an edge that runs along the face's
parameters at its own pace gets the straight (u, v) line it is: between its ends' parameters, or
with u or v its own parameter (a rim on the same quarter grid). Each is taken only where it lifts
onto the edge to within 1e-9 of the geometry's size; nothing is fitted.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
A form's parameter is now the angle's quarter on its frame plus the arc's parameter in it, and a
form spans the quarters its interval needs (up to five for a turn). Every form on one frame has
the same parameter at the same angle, so a circle edge and the cylinder it bounds agree along it,
and an arc across the frame's x axis, or starting anywhere, has a form. Traces its label to 1e-14
on such arcs, negative angles and a turn from 30 degrees.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
`brep::form` builds each analytic kind's exact B-spline the way SMLib holds its own: a circle
or ellipse as four rational quadratic quarter arcs from its frame's x axis on [0, 4] (angle and
parameter related in closed form), a cylinder or cone as that circle times a line along the axis,
a sphere as the circle times a half circle of latitude, a torus as the circle times its tube's
circle, a line as the degree-1 B-spline on its own parameter. A circle and every face it bounds
then share one parameterisation. Each traces its label to 1e-14.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BnQawySUFx57iaJZq7z1Fm
The assembly kept a face's domain and pcurves, which are in its NURBS form's parameters, but
dropped the form and computed new pcurves and the domain on the analytic surface. The form now
travels with the face, and its pcurves and domain are computed on it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A shell's `forward` is about the surface the face's (u, v) are on, its NURBS form when it has
one, which may run against its cylinder, cone, sphere or torus. The kernel read it against the
analytic surface, so a fillet's bore whose form runs the other way was a shaft and an M10 thread
was refused in its 8.5 mm hole. `Face::form_turns` says when the form turns the normal.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
No floor of ours on tolerances (1e-6 of the body's size, 30 times SMLib's 1e-5 default): the
record's tolerances go back into SMLib, which approximates to them, so the floor coarsened every
later operation. No recognition of NURBS edges as lines or circles: an edge is a circle where
SMLib holds an SmCircle (with its NURBS form), and a NURBS edge keeps SMLib's trims on its own
parameter. crashes/2992 and sweeps/1853 pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Edge::curve_interval (and brep::curve_interval for assemble's edges) gives the interval on the
edge's curve itself: a form circle's angles of its ends, its frame turned by the SMLib conversion
so the angle grows along the NURBS. lip and the ruled surfaces read arcs by it; the hull, the
blend tangents, the patch pcurves, point paths, IGES output and the prism runs evaluate on
Edge::parameters. assemble carries an edge's form through.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The SMLib conversion reads and writes EDGE_NURBS_FORM: a circle edge's record keeps the circle
as what it is and the NURBS SMLib holds it as as its parameter, so the trims on it carry over
exactly (TrimMap identity).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The edge mirror of a face's NURBS form: Edge::nurbs keeps the NURBS SMLib holds a conic as, and
the edge's interval and its pcurves' parameter are its. Edge::parameters is the curve every
evaluation at a parameter uses; curve says what the edge is. Nothing writes the form yet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
prism.rs's sheet regions import Surface and Vec3 only where they are built (the smlib feature),
so the kernel without it has no unused imports; the cli's tests that measure meshes at their own
tolerance depend on solidcraft-kernel.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
An edge whose two faces meet tangentially has no corner to round, and SMLib refuses a blend along
it. The chain now runs on sharp edges only, as a tangent chain does (fillets/3369: a rim edge that
ends on the tangent seam of a stadium).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A whole circle is one closed edge now, so the outer wall and the hole wall each have a seam; the
rig counts seam edges (4005, 4006), and 1c490ddd's count without them broke those cases.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
doc: world_state, world_state_reusing, state_in_frame_at, state_in_frame, face_point_state,
face_point_names and construction_state return Result. A body that could not be placed, a
frame that could not be inverted, a sketch plane or thread axis that did not survive the
placement, or a failed placing worker used to leave things where they were without a word;
now the call fails with what could not be placed. is_placement is the one test of an
occurrence's placement (finite, a rotation that does not mirror, a translation within
1e9 mm). Opening a design names every occurrence that fails it; add_occurrence refuses
one; a joint solution that would place an occurrence outside one is not applied and the
joints placing it report it.
engine: Session::world_state returns Result and keeps a failure in its cache as it keeps a
placed model. bounds, home_camera, open_sketch_plane, the scene builders, the world
previews, frames::to_local and auto_operation pass the error on; helpers answering "not
found" take the placed model from their caller. Commands setting a placement check it
(valid_position, exploded view, fastener at a point, STEP insert matrix).
ui: panels, picks and hovers read the session's placed model through
open_meshing::session_world; a model that cannot be placed shows no placed bodies, and the
viewport says why across its bottom. Pictures and the control channel report the error.
cli, mcp: snapshots, renders, benches and the oracle report the error; the kernel-case rig
records it as the case's failure. Tests read the new Results with unwrap.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Extent.corner is optional: unset, each profile preserves curvature where
every curve is a line or an arc and takes the tangent rule where another
curve has a corner, with a note in the feature's warning; a rule the user
gave is kept. draft_corners reports not_exact as the kernel builds (any
free curve in a cornered profile). Builds (release CLI); cases not yet run.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The kernel blends exactly the edges given; the feature's tangent_chain (default true, saved
only when false: pending format note blend-tangent-chain) expands each pick to its tangent
chain (kernel::tangent_chain) before the kernel call. A variable or spline round takes its
edge alone. Scripts and MCP passing one pick get the chain again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Every SMLib analytic surface is a B-spline surface, and its trims are in that NURBS's
parameters. Face::nurbs keeps that NURBS: the face's domain and pcurves are its (u, v), and
`surface` says what the face is. Face::parameters is the surface every (u, v) evaluation uses.
The SMLib conversion reads and writes it (fork FACE_NURBS_FORM, SMLIB_REV c4483f6); the STEP
writer writes the analytic surface and the face's edges without pcurves.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
SMLib (fork 68e3086) refuses a blend whose rails leave the faces they roll on with
SMLIB_ERR_NO_ROOM, which the kernel reports as KernelError::TooLarge. The kernel's own
pre-checks go: blend_fit's fit check and the 256-station band check of the variable fillet.
The shim no longer follows tangent chains; the Fillet/Chamfer dialog already expands a pick
to its chain (its Tangent Chain option) and stores every edge in the feature.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Exporters write a plane as a CONICAL_SURFACE whose semi-angle falls a whisker
short of 90 degrees (#4006: pi/2 - 7e-12): its height over the face spans
4e-11, a parameterisation ill-conditioned for everything downstream. The
reader now measures the cone's deviation from its limit plane (the plane
through the placement across the axis) over the face's boundary points and,
should the face hold the axis, the apex; within the file's distance
uncertainty the face's surface is that plane, oriented as STEP's normal.
One criterion, the deviation over the extent against the tolerance; no
semi-angle threshold.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Pinned at the fork's draft-corners head: a drafted rectangle (or any
line-line corner) has its walls meet at a corner edge under the tangent
rule as well, 6 faces rather than 14. A curvature refusal for a curve
that is neither a line nor an arc names that curve's start.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Extent.corner (preserve_curvature by default, tangent or round), saved
only when not the default (one pending format change), and the
solid.extrude `corner` parameter. draft_corner_options reports what each
rule gives at each corner of a feature's profiles, for a dialog to offer
the others where the chosen one is a poor fit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
extrude_tapered(_between) take a DraftCorner: PreserveCurvature (each wall
continues as itself, a line's plane or an arc's cone, until it meets its
neighbour; SMLib's SM_OC_CURVE_EXTENSION), Tangent (the offsets continued
along their end tangents) or Round (an arc about the corner). A corner
the rule cannot close is refused with its point and why
(solidcraft_smlib::Error::Corner). draft_corners reports, per sharp
corner of a profile, what each rule gives: how far each offset curve runs
on or is trimmed, whether the corner cuts inside the exact offset, or why
the rule refuses it. SMLib pinned at the fork's draft-corners head.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The angles a cylindrical wall spans came from its arc's two ends about its middle; a wall in
one piece has a whole circle for an edge, whose ends are one point. The circle's parameter is
its angle, so the span is the edge's interval about the angle of its middle: half the turn
either side for a whole wall, as for any arc.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A rim's arc was its two ends and the point halfway, its circle found through the three: a
whole circle, one closed edge now, has one end and no circle through it. Each arc piece carries
its circle's centre and the angle it turns about the face's normal, read off the record circle
(its parameter is the angle), and offsets, profiles, edges and walls are made from those; the
loops' turning is read through points a third and two thirds along each arc. assemble's
unused edge_point goes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Loop2::circle and circle_from gave two half arcs, so everything built from a circle came out in
two: an extruded wall in two faces, a rim in two edges, and so every round, chamfer or sweep on
it in two B-spline halves that SMLib cannot join (two separately parameterised arcs meet at
different curve parameters, SmVertex::IsTopologicalVertex; different free-form surfaces are not
topological, SmEdge::IsTopological). A circle is now one arc of a whole turn from its start
angle: one closed edge, one wall with a seam, one round. Sketch circles, the cylinder builder
(circle_from at its seam angle) and every other builder take it through Loop2; SMLib's arc
curves take a whole turn (smlib_sweeps clamps to one) and ArcSweepCylinder makes the whole
cylinder.
Cases, singly on 9f5ad20c: fillets/1191, 2797, 3753 and sweeps/1269, 2295 fail before and pass
now; eight hole and six thread cases give what the base gives (12 pass, booleans/1279,
booleans/307 and threads/2038 fail as on the base).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Seven sheet-metal cases failed in smlib_planar_faces (status 1001/1006). Every refused
curve list, in the rig's dumps, has curves of which more than two meet at one end point:
flange regions of one flat pattern touching at their corners (405, 349 x2, 436, 1481, 725,
723) and an outline passing a point twice where a cut-out's corner lies on the sheet's edge
(1095). SmBrep::CreatePlanarFacesWith3DCurves, behind SmApiCreatePlanarFaces, documents
that it fails for such "spine points (where more than two curves share a common
end-point)" (SmBrep.cpp); the sheet handed it all of an operand's loops at once.
A sheet is now built one simple loop at a time: each loop (outer counter-clockwise, holes
clockwise) split where it passes a point twice, each simple loop its own planar face, the
counter-clockwise ones united and the clockwise ones taken away with SMLib's 2D boolean,
which takes faces touching at a point. extrude_planar_regions uses the same sheet.
The read-back keeps the contract the regions had: a face's loop that passes a point twice
(a cut-out touching the outline) comes back as simple loops, counter-clockwise pieces
material and clockwise ones holes of the piece that holds them.
Release CLI, SOLIDCRAFT_KERNEL=smlib: 405, 349 (72 and 90 degrees), 436, 1481, 723 and 725
pass. 1095 now gets past the planar regions; its extrusion of a plate with a cut-out
touching its edge at one point fails in SMLib's export ("a closed shell bounds two regions
and encloses one"). Every other sheet-metal case fails or passes as in rig-1, and 529
(a hang in rig-1) passes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The five "boundary winds around the surface" refusals of the first SMLib rig run (solid.hole
1355, solid.revolve 1853, 1866 at 180°, 3191, 466) are one mechanism: a revolved face whose
profile ends on the axis (a drill point, a cone tip, a half slot about its centreline) has a
pole, and its two meridian edges meet there with pcurves half or a whole period apart in the
angle, the parameter the pole leaves free. The boundary unwrapped every step to the nearest
copy, so at the pole it moved the next meridian a period along (1866's quarter sphere: from π
to 2π instead of 0) or, where the pole closed the loop (3191: v from 1 back to 0), the closing
step did; the loop then drifted a whole period and the face was refused as winding.
The pcurves are exact and run on continuously, so nothing within an edge is unwrapped now.
Where one edge meets the next, a pcurve starting a whole period from where the previous one
ended is moved back by it, except in a direction the meeting point is a pole of: there the
two pcurves' own values are the ends of the boundary's run along the pole, which the split
poles then carry. Each loop starts at a node off every pole whose predecessor is off them too,
so seam::drift's nearest-copy closing step never crosses a pole. All five cases pass;
tessellation/ and sweeps/ otherwise as in the rig run (1856 and 3006 pass, slow under a
machine load of 190; 2552 hangs as before).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
SMLib's mass properties no longer refuse bodies that are not one manifold
solid: several lumps and lumps touching along edges or points are the sum
over their material regions, and a sheet is its faces' area and centre with
volume 0. Docs of Brep::mass_properties and kernel::measure say so.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The ruled surface's far-curve solve kept its own copy of NURBS Book A2.3. The
geom crate now exposes bspline::basis_derivatives (the basis functions alive at
a parameter and their derivatives up to an order, each from its local knots via
local_basis, on the span tess::span_of picks); tess's second-derivative surface
evaluation and the ruled surface both use it, and the copy is gone. smlib's
mod list back in order.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Owner's decision for strips whose rulings turn freely along their edge (a face's
normal along a free-form face, which was refused): the far edge is not the
offset curve but a B-spline the kernel defines, and the strip is exactly the
ruled surface between the edge and it.
The far curve has the edge's own degree, knots and weights (so the ruled NURBS
needs no knot merging); it runs from the strip's start corner to its end corner
(where the rulings at the edge's ends reach); its interior control points are
given (handles) or the least-squares solution of weighted terms, the
construction's parameters: closeness to the rulings' ends at the edge's
Greville abscissae (W_CLOSE), fairness by arc length (W_FAIR), each end leaving
along the chord to the other end (W_TANGENT, one small function), and at a joint
where the chain runs on smoothly with shared sides the first and second
derivatives by arc length matched to the neighbour's (W_JOIN), in one solve
along the chain, an exact neighbour's far edge matched as it is. Exact strips
are unchanged.
ruled_far_handles gives each free strip's default handles; ruled_surface_with
takes moved ones. The Ruled Surface feature saves them as `far` (per chain
edge, null for the default; pending format note ruled-far-handles), and
surface.ruled takes them as `far`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
No caller is left for: Assembly::sheet and assemble::outward_normal, Body::shell_nesting,
regularize_generated_planar_faces, with_shell_planar_sources, boolean_record's Traced and
of_pair, offset's normal, plane_fit's plane_through, EdgeSampler, planar_face and
plane_frame (fill has its own on the record), polyhedron's solve3, prism's wall_along,
record::regions/region, splitface's side_of_point, Analytic, analytic_face, face_sides and
canonical, surfaces::picked_loop, sweep_path's closed_path, turn and stations,
shell_groups' groups and parents with their limits, and the whole of spline_conv. Their
tests go with them: prism's curved_prism_wall_requires_positive_separable_rational_weights,
body_lump's shell_grouping_preserves_large_disconnected_patterns_and_bounds_hostile_work,
and regularization protection's check of the generated regularizer. Three names follow
clippy's conventions: into_face -> direction_into_face, from_to -> ends_along, and Params::
to_eval takes its Copy self by value; fill's clash loop walks its slots.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Machine-applied by cargo clippy --fix: borrows the compiler dereferences at once, an
unused import, a manual += , an entry() for contains_key+insert, a needless Ok(?),
contains() for iter().any(). No behaviour changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
mesh memo (one remembered mesh per tolerance), history (a bare body from a record), tests.rs
(no coarse-missing field, lumps for touching shells), outline (the foot found by the planar
faces in its plane), chamfer sides (the rim from the record's edges), seam faces, coplanar
and issue 679 and pinch tests (the boolean is SMLib's: the diagnostics assertions go, every
volume, shell and validity check stays), boolean history (a cut of two pieces through the
boolean), integration tests naming solidcraft_kernel::MeshTolerance.
Deleted, testing internals that are gone: boolean_history_tests' glue_merges_across_the_seam
and faces_pushed_apart (coplanar::glue, boolean_apart), planar_fallback (polybool's
planar_boolean) and bore_union_of_a_bore_in_two_arcs_has_no_history (coplanar::bore_union);
issue_851's the_sphere_at_a_box_corner_is_the_planar_corner_builders (blend_vertex); pinch's
prism_boolean shell check.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The truck reads become record reads: an intersection edge of a moved cut, the fillet
mitre edges sampled with CurveEval, the cache estimate from the record's faces and edge
uses, holed faces by their loops. The drain pads and the coincident-faces booleans keep
their volumes and faces without the boolean diagnostics and the polygon boolean.
Deleted, testing internals that are gone: a_tapered_s_curve_is_cut_into_pieces_that_follow_its_offset
(build::offset_error/offset_splits), planar_fallback_names_the_curved_faces (the polygon
fallback), mixed_loop_blend_guards_selection_and_construction and
loop_chamfer_mixed_arc_and_line_uses_offset_intersections (loopblend),
an_attempt_budget_stops_the_boolean_and_is_shared_by_nesting (with_attempt_budget),
placing_a_body_does_not_turn_its_revolved_faces_again (REPHASED, truck's revolution phase),
and the truck debug runs debug_coplanar_union_rate, debug_truck_sphere, debug_sphere_patches,
debug_sphere_cases, debug_sphere_rotations, debug_cube_sphere_measure, debug_heal_sphere_cut,
debug_coplanar_cylinder_union, debug_drilled_hole, debug_face_param_areas and
debug_revolved_tube_cut (truck_shapeops booleans, build::sphere_patches).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
native.rs's test module keeps its encode/decode, wire-record, hostile-input and appearance
tests. Deleted with the adapter they tested: pcurves_on_planes_are_the_curves_mapped_exactly,
pcurves_fit_curved_faces_within_tolerance, pcurves_of_seams_lie_a_period_apart and
pcurve_pictures, with their helpers (pcurves_of, check_pcurves, check_faces, curved_bodies,
seam_faces, draw). smlib_import_tests count shells on the record; their notes no longer
speak of truck.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
- thin faces: the face pinched against its own edge is a record sheet (sew::assemble) meshed
by the body's mesher; its triangles still cover its boundary polygon.
- trimmed loft faces, loft rails: distances to faces through the record's SurfaceEval.
- untouched faces: edges and in-line stations from the record's edges and vertices.
- intersection curves: the #1252 regression on a record sheet whose edge is a traced
intersection of parallel planes: meshing neither panics nor goes non-finite.
- lumps: shells counted on the record; the certified witness is Brep::compound of its three
bodies, faces known by surface and box.
- Deleted, testing internals that are gone: intersection_curve_tests'
degenerate_intersection_polylines_return_none_without_panicking,
closed_intersection_retains_the_closure_and_spherical_supports and
open_intersection_keeps_endpoints_on_both_planes (truck's IntersectionCurveWithParameters)
and a_result_with_an_edge_that_does_not_evaluate_is_refused (boolean_record::edges_evaluate);
body_lump_tests' unowned_overlapping_and_touching_cavities_are_errors_not_dropped_geometry
and shortcut_clearance_refuses_curved_contact_hidden_between_mesh_vertices (truck shells
joined without regions, separated_lumps) — kept as spheres_apart_join_as_two_lumps — and the
shell-order permutation of nested_cavities (the record names each cavity's region).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
- placement: a rigid placement moves the record and nothing else (regions, shells, loops
with their pcurves, edges and their uses equal; every vertex moved; the identity's
history), on a cylinder whose wall's loop uses its seam twice.
- issue 2005, 616, 674, 1038, 1510, join between caps, regularization protection, history,
mesh memo, display mesh: the truck reads become record reads (shells closed by edge
uses, faces by the record's surfaces, vertices by the record's points, the mesher's own
boundary budget); the nearly-in-plane slab of 1510 is an extruded record body.
- Deleted, testing internals that are gone: issue 616's glue_joins_every_face_the_pieces_share
(coplanar::glue) and the_retry_guard_refuses_changes_away_from_the_tool
(boolean_guard::check_retry); issue 674's partial_contact_does_not_join_tangent_or_separated_faces
(coplanar::imprint_union); issue 1038's hinted_division_matches_the_search_from_nothing
(truck's vendored parameter_division); issue 1510's a_face_close_to_the_plane_but_off_it_is_not_moved
(ops::off_plane). The boolean-diagnostics assertions (no retry, no tool nudge) went with
capture_boolean_diagnostics; the rest of each test is kept.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
solidcraft-smlib is a dependency of every build (it answers NotBuilt without
SMLib), so the SMLib sides of surfaces.rs and blend_smlib.rs no longer sit
behind cfg(feature = "smlib"), and their stub modules go; blend_smlib's tests
run in every build like the rest. This also uses offset_backend::thicken and
the smlib_sweeps curve conversions in builds without the feature, which left
them unused there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The three refusals of the cornered controlled sweep become the owner's rules:
- an orientation direction turns the section about the corner's binormal like
any sweep, then turns it back about its normal to the direction's frame
evenly along the next run, as a twist (direction_offset: the turn between
the carried section and the direction law's section there);
- guide rails follow a path with corners when each rail has a corner of its
own at every corner of the path, in the plane halving it and by the same
turn about the same axis; the rails are cut there into each run's
(run_rails), anything else is refused with the corner named;
- round a closed path with corners the first run starts from the closing
mitre and the last ends on it; a twist is allowed where the profile maps
onto itself (closing_offset), an end scale is refused.
Each run's station count now follows its own turn, twist included.
solidcraft-smlib is a dependency in every build (it answers NotBuilt without
SMLib), so smlib_sweeps' imp is no longer behind the smlib feature and its
stubs and not_built go; polyhedron calls offset_backend in every build.
Deleted as unused: sweep_path::{closed_path, turn, stations}.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A sweep that twists or scales refused any path whose direction jumps by more
than 1e-3 rad, a branch on the path's shape. It now takes the corner model of
the plain sweep: each smooth run is swept through stations by the laws a smooth
path has (twist and scale running with the length along the whole path, each
run going on from the section the last ended with), and the runs meet at each
corner in a mitre. Turning with the path, each run goes on past its corners as
a prism of its section there (the laws hold still across the mitre: the two
runs' sections meet on the plane halving the corner only when both are the
corner's section) and is cut back to that plane, then turned through the
corner onto the next run; only moving along the path, the runs meet at the
corner's section. The mitre plane is one function for both (mitre_of).
Refused, with the reason, where a control has no meaning at a sharp corner: an
orientation direction (it turns the section by its own turn, not the corner's,
so the two runs' sections don't meet), guide rails (their sections jump where
the path turns), and a closed path with corners (a twist or scale keeps the
last section off the first).
(not built: base in progress; no error in these files)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Intersection::point started its Newton solve from the last point found (a Cell holding both
surfaces' parameters) when that landed near the leader, as truck took a hinted point. The
answer then depended on what was evaluated before, so CurveEval was not Sync, and the
clearance check, which shares a body's DistanceShape (its edges' CurveEvals) across worker
threads, did not compile; with threads it would also have made the result depend on their
interleaving. Each point is now solved from the leader's own point, the path the hint fell
back to, and the unused trace tolerance goes with the hint.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
sheet_hole_cut chose by failure: a hole's drills were tried all at once and, if that was
refused, one by one; each drill was tried as a flat cut-out and, if that was refused (or
machined nothing), as a bore. Now the drill's geometry decides, one path each:
- SheetBody::drilled_panels: the panels a straight drill meets and how square it is to each
(|panel normal · axis|); a drill into a bend is refused, as before. add_bore uses it.
- sheet_drill: square across every panel it meets is Drill::Across (sheet_primitive_cut, a
cut-out or pocket of the flat pattern), along every one into a rim is Drill::Along
(sheet_bore_cut, a bore), meeting no sheet is None, and anything else (square to one panel
and along another, or at an angle) is refused with the angle message.
- The all-at-once pass (sheet_drill_cuts) passes its errors on; its None (the drills not all
square along one axis) still puts the model back and drills them one by one.
- sheet_bore_cut passes add_bore's refusal on instead of answering None.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The boss's post and ribs, the boss's hole, the snap fit's arm and the rest started a few
hundredths of their height inside the material (and the hole and the groove ran on as far
past the face), and the groove's cutter reached past the rim edge into the free side, so no
face of a tool lay on a face of the part. SMLib joins and cuts touching faces exactly: each
tool is now the feature's own shape, from the face (a drafted post or rest sized at the
face, the groove's band the rim's own, free reach 0). The material each adds or removes is
the same.
kernel::rim_band's `free` is now always 0 from the doc.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The kernel has no boolean retries now, nor their diagnostics
(capture_boolean_diagnostics, BooleanDiagnostics, BooleanStrategy), nor the near-coaxial
crescent (coaxial_overlap, coaxial_candidate, with_attempt_budget). What read them goes:
- doc: FeatureResult and the evaluation cache carry no boolean diagnostics, and no "Boolean
retry" warning; the saved model no longer writes them (an older file's are ignored);
- engine: timeline.inspect / feature inspection drop `boolean_diagnostics`; fuzz runs and
the CLI fuzz summary drop their retry counts;
- inspect.interference: each pair is the bodies' Intersect, measured; the near-coaxial
crescent, the by-piece intersection of bodies in several pieces and the unverified notes
go (the command's doc no longer lists near_coaxial);
- inspect.clearance: an overlap is the bodies' Intersect, measured near its region; the pin
tries, the cut-down copies of large bodies (clipped, narrowed), the union cross-check
(union_disagrees, contradicted) and the malformed-part check (plausible) go, and the
measure near the origin passes its move's error on.
Tests: the retry test and the unit tests of the deleted checks are removed; the barrel join
test (tests/boolean_diagnostics.rs, now tests/barrel_join.rs) keeps its geometry checks;
the pin-in-bushing test expects the filleted case measured; assertions on empty diagnostics
and on near_coaxial are dropped; the corpus fixtures lose the diagnostics key and the one
retry warning.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Every lib warning under --features smlib outside the files agents are editing
(assemble.rs, sweep_path.rs) pointed at code nothing calls any more:
- body: shell_nesting;
- boolean_record: of_pair and its Traced alias;
- planar_regularize: the generated-wall regularization (and its branches in the
certified one, which was its only other path); planar_sources:
with_shell_planar_sources;
- plane_fit: plane_through, EdgeSampler, planar_face, plane_frame (the patch is
fill's); only loop_area is left;
- polyhedron: solve3; prism: wall_along; offset: Surf::normal;
- record: regions, region; shell_groups: groups, parents and their limits;
- splitface: side_of_point, Analytic, analytic_face, face_sides, canonical;
- spline_conv.rs, all of it (surfaces builds NURBS records); an unused import in
tess/uv.rs.
The tests of deleted items go with them: prism's wall_along test, the shell
grouping test in body_lump_tests, and the generated-regularization assertion in
regularization_protection_tests.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
split_band_edges_are_seams tested split.rs's splitting of a face that winds around its
surface into two (deleted with the truck reader; the record keeps the face with a seam):
deleted. ImportedBody::file_faces says each body face is a file face.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Each body's record is written directly: a manifold solid B-rep (186) per material region
with its voids, open shells as 514 form 2, faces on the record's exact surfaces and edges
on its exact curves over their intervals, in the STEP form solidcraft_step::write::topology
gives them (one geometry model for both exports). The detour through STEP text and its
re-parse is gone; the maximum coordinate is the largest coordinate written.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
write::topology::layout says which shells make each solid (a material region's outer shell
and its voids) and which are sheets, and outer_loop which loop of a face is its outer one;
the STEP writer uses both, and the IGES writer can too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Import: every solid is read by solidcraft_step::topology into its record and taken into
SMLib as given (step_in/parallel.rs smlib_part), the one path; the truck readers
(step_in/brep.rs, split.rs, geom.rs) are deleted. Placements are column-major 4x4 matrices
from solidcraft_step::read placements; an instance is its part's record moved there
(ops::transform_matrix), or the very part at the identity. Assemblies, names, colours,
shared displays and the public API are unchanged. step_orientation_errors lists the faces
the STEP reader turns over (their loops disagree with their same_sense).
Export: each body's record is written by solidcraft_step::write::topology as the exact
B-rep (solids per material region, open shells as a surface model, the record's pcurves);
step_out writes products, units, colours and assemblies around it with Writer. The truck
writer and every post-processing pass (split-face merging, sheet trimming, analytic
recognition, short-edge collapse, the re-read fallback) are deleted. The context's
uncertainty is the loosest record tolerance exported.
Tests: ported to the record API with their assertions. Deleted, as tests of deleted
internals: conic_tests extruded_ellipse_seam_uses_profile_angles_for_oblique_sweeps and
conic_arcs_are_exact (truck geometry reader), tests offset_surface_analytic_geometry_and_
orientation, offset_surfaces_nested_and_hostile and periodic_knot_bsplines_are_read_on_
their_domain (truck geometry reader), corpus_truck_and_schema_4_counts (truck comparison);
the orientation counter asserts (crate::body::ORIENTED, gone with orienting by mesh).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A placed revolution is written exactly whenever its placement maps revolutions about the
axis onto revolutions (one scale across the axis, another along it, then a rotation or
reflection), its profile placed with it; a conic profile stays a conic under a similarity.
A void's CLOSED_SHELL keeps the record's sides (normals out of the material, into the void)
through an ORIENTED_CLOSED_SHELL .T., as SolidCraft has written voids. Reader::frame reads an
assembly placement (AXIS2_PLACEMENT_3D, or _2D in the XY plane).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
write::topology writes each material region as a MANIFOLD_SOLID_BREP (BREP_WITH_VOIDS
with its voids as ORIENTED_CLOSED_SHELLs), open shells as a SHELL_BASED_SURFACE_MODEL,
faces as ADVANCED_FACEs on their exact surface with FACE_OUTER_BOUND/FACE_BOUND of
EDGE_LOOPs and VERTEX_LOOPs, edges as EDGE_CURVEs on their exact curve over the record's
interval, and the record's exact pcurves as PCURVEs of SURFACE_CURVE/SEAM_CURVE. A record
surface's (u, v) map linearly onto STEP's (planes on their orthonormal frame, narrowing
cones turned over, revolutions angle first), and pcurves go through the same map.
Writer::add_distinct writes topology entities unshared. topology::Solid lists the faces
the reader turned over (their loops disagree with their same_sense).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
doc's FeatureResult has no boolean_diagnostics any more (truck's Boolean
retry reports, gone with truck; SMLib's boolean has no retries). The
timeline.inspect and doc.inspect replies drop the key.
Tests of the removed reports: tests/boolean_diagnostics.rs is deleted (it
checked truck's retry events and their survival through the cache, undo and
reopen); the assertions that a feature's report had no events go from
features_more_tests, patterns_tests, hole_face_tests and tests/rim_cut (the
rest of each test stays). The CLI corpus's expected replies
(move-bodies, issue-489-stacked-join, issue-616-lid-ribs) drop the key; the
issue-616 one listed a planar_fallback event. docs/robustness.md: its
section on retries becomes one on SMLib's booleans, which have none.
Left to the blends work: engine fuzz.rs (its retry-event counts) and
clearance/interference (capture_boolean_diagnostics, coaxial_overlap,
with_attempt_budget).
Not built: base in progress (kernel-smlib/no-truck 66ab6b7f does not build).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The kernel's BooleanDiagnostics and capture_boolean_diagnostics are gone
with truck: they reported the retries and fallbacks truck's booleans took
(shifted operands, refined tolerances, a planar fallback), and SMLib's
boolean has none. Model::evaluate ran every feature inside
capture_boolean_diagnostics, warned with its summary and kept the report
in FeatureResult, its cache entry and the saved model ("diagnostics").
The feature runs as it is, FeatureResult and CachedEval lose the field, and
the saved model neither writes nor reads "diagnostics" (a model saved with
it still opens: the key is not looked at).
Callers outside doc that read FeatureResult::boolean_diagnostics change
with it: engine's timeline.inspect reply here; engine's fuzz report and
clearance/interference belong to the blends work.
Not built: base in progress (kernel-smlib/no-truck 66ab6b7f does not build).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The vendored truck-geometry they pinned is gone; the same finite-difference checks, degrees
2 to 5 and a rational surface, now pin SurfaceEval::second, which every curvature the kernel
reads comes from.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The helpers read every face of the fill as a record surface over its box, the loops'
(u, v) polygons from their exact pcurves, the rim from the free edges and curvature from
SurfaceEval::second; check_face takes the fill's faces (one, or the patches about a centre)
whose free edges lie on the body's edge curves. Expectations changed by the coordinator's
decision on loops of five or more edges: the pentagon fills as five patches (each rim edge
in halves, five spokes: at least 15 edges), the hexagon through a point as six patches with
18 edges. Every other assertion is kept.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
finish lays out the faces left and added (each stored relative to its surface's natural
normal with its pcurves and box, now one function, stored()), the source's shells and
regions (a region whose outer shell lost every face gone) and the edges and vertices as
they are, then calls Brep::canonical; its own ordering walk is deleted. check_geometry
and the face and edge maps are kept.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
sweeps_along_3d_paths asserted at most 8 faces for a 41-point path: one wall
per profile segment fitted through the points, which the exact construction
no longer makes (no curve is fitted through samples). A point path is the
polyline its points make, swept line by line and mitred at each point, so the
circle's two arcs give two walls along each of the 40 lines, and two caps:
82 faces. The volume check stands (a mitred sweep of a profile centred on its
path keeps area × length).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
An Iso link along an explicit direction on a solid now takes
kernel::face_section_along, the face cut by the plane holding its exact normal
and the direction, instead of the triangles' section. With the projection, the
u/v iso curves and the base's exact intersection edges, every linked 3D curve
on a solid keeps its exact curves (Wire::exact); curves that meet end to end
become one wire (joined_wires, shared by the projection and the iso curves).
Mesh bodies, which have only their triangles, keep the triangle-based curves.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
face_section_along cuts a face through a point by the plane that holds the
face's exact normal there and a direction (Brep::face_section): the iso curve
"along a direction" of a link, as record curves on their intervals, the
B-splines SMLib made. The projection and the section share the picking reach,
the curve tolerance, the error words and the conversion to record curves.
(Intersection links take the base's meeting_edges, SMLib's imprint.)
lib.rs (coordinator): export point_path::face_section_along.
(not built: base in progress; no error in these files)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Brep::face_section (smlib_face_section: one face cut by a plane) hands back
the NURBS SMLib made with their intervals, through one reader of smlib_curves
(take_curves) shared with project_curves. The face-pick errors map to
NoMatch/Ambiguous in one place (picked). SMLIB_REV at fork eedfa51.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
LinkSource::Iso with `u` or `v` on a solid now takes kernel::iso_curves: the
face's isoparametric curve through the point, exactly, cut by the face's
loops, each piece a wire with its exact curve. An explicit direction (the face
cut by a plane: an intersection curve, waiting for SMLib's exact SSI curves)
and a mesh body still use the triangles' section.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
iso_curves gives the isoparametric curve of a body's face through a point, along
its first or second parameter, as record curves: on a B-spline surface the
surface's own row at the point's other parameter (each control point its
column evaluated there, homogeneous, so exact for rational nets), on an
analytic surface its line or circle. It is cut where it crosses the face's
loops: each crossing solved on the loop's pcurve itself (bracketed on samples,
then bisected on the pcurve), the pieces between kept where their middle lies
inside the face (even-odd over the loops), neighbouring inside pieces joined.
A revolved surface record is not supported yet.
lib.rs (coordinator): export point_path::iso_curves.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A sketch curve linked onto a face (LinkSource::OnSurface) was projected onto
the face's display triangles, so the wire was a polyline on the mesh. On a
solid it is now kernel::project_onto_face of the curve's exact segments, along
the sketch normal turned toward the face: the wire's exact curves are the
B-splines SMLib made, pieces meeting end to end one wire, the points sampled
from them for display. A mesh body, which has only its triangles, is still
projected onto them.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
project_onto_face projects sketch segments along a direction onto the face of a
body at a point through Brep::project_curves (smlib_project_curves), each
projected piece the B-spline SMLib made, returned as a record curve
(RationalSpline) on its interval, within a millionth of the body's size.
lib.rs (coordinator): export point_path::project_onto_face.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The base no longer links truck, so build.rs and sweep_path.rs drop what was
kept only for other modules: chain and profile wires, the profile face,
loft wires and their ruled and smooth faces, across/section_parameters, the
cubic interpolations and NetFit, the uniform wires, the six sphere patch
faces, and sweep_path's ruled_sheet and smooth_sheet with the station helpers
only they used. Their callers switch to planar_face_record,
smlib_sweeps::{chain_curves, body_from_record, sphere}, sketch_record_curve,
the SMLib sweeps with open ends, and bspline::CurveN::interpolate.
offset_backend and polyhedron pick faces through provenance::closest_on_tri,
boolean_guard being gone.
Callers left to switch (other ports): surfaces.rs, loft_rails.rs,
surface_offset.rs, surface_ruled.rs, step_in/brep.rs, trimmed_face_mesh_tests.rs.
(not built: base in progress; no error in these files)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The FFI for smlib_project_curves and the smlib_curves result (fork 5f9af50),
and the safe call: a curve list projected along a direction onto the picked
face, back as Curve::Nurbs with each curve's interval. SMLIB_REV at 5f9af50
(fork integration head 1934b57 + the untrim status fix + the projection shim).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Rulings that are constant, or keep their angle to an arc's axis, move the edge by
an affine map: the far edge is the edge's exact image and the strip exact. Any
other ruling field (a normal turning along a free-form face, a tangent carried
round a free-form curve) ends on an offset curve, which has no exact NURBS form
in general: SMLib holds one only implicitly (SmOffsetCurve) or approximates it
(NL_CrvOffset.h, N_CrvOffsetApprox), and its ruled surface (SmApiCreateRuledSurface)
takes two B-spline curves it is given. Such a strip was a B-spline fitted within
1e-3; it is now refused by name. The fitted-strip test checks the refusal.
The round-corner cases count one face per corner: SMLib's revolve makes a quarter
turn one face, where truck's cone builder made two (12 faces and joins, now 8).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
SMLib is the only kernel now. cargo xtask kernel-cases ran every case once
per kernel asked for (--kernel default|smlib|both), setting
SOLIDCRAFT_KERNEL per run, and wrote a column per kernel; solidcraft-cli no
longer reads that variable.
The rig and the ladder run each case once, through solidcraft-cli built with
--features smlib (cargo run passes it), and report one result per case:
--kernel and the env var are gone. Each run has an output folder (--out,
default target/kernel-cases, target/kernel-ladder for the ladder) holding
results.json (every case's full result, which --results used to write on
request) and dumps/<area>-<case>/: every case runs with
solidcraft-cli kernel-cases --dumps=<out>/dumps, so a failed SMLib call
leaves its inputs for replay.
The ladder's golden holds one entry per case: golden.json is the last
recording's smlib entries, which predate SMLib running every case (its
reason says so); record it again after the first SMLib-only run.
docs/kernel-cases.md (until the next run regenerates it),
tests/kernel-cases/README.md, docs/kernel-smlib-sweeps-shell.md (its truck
column and comparisons dropped), xtask's and solidcraft-cli's usage text:
one kernel. Nothing was run.
clippy -p xtask --all-targets: clean.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The helpers that read a face's surface, its edges and its seam shape read the record
(SurfaceEval, CurveEval, matchsurf's Patch) instead of truck faces; every assertion is
kept.
(not compiled: the kernel's test target waits on the base's step_*, build, sweep_path and
surface_ruled ports)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Before sewing, each edge of one input is split exactly (Brep::split_edges) at every vertex
of another input that lies on it within the stitch tolerance away from its own ends, so a
patch whose corner meets a body's edge part way (the five-plus-sided fill's midpoints)
sews with no T-junction. The pieces carry their edge's history. The search is the nearest
of 64 samples refined by golden section, over the vertices within a chord's reach of the
edge's samples. Nothing else in stitch changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Each piece keeps the edge's curve on a sub-interval, a new vertex sits on the curve at each
split, every use becomes uses of the pieces with its pcurve cut at the same parameters
(bspline::segment), wire edges likewise; the result is in canonical order with each
edge's source edge. What a sheet needs so that another's vertex on one of its edges is a
vertex of both.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
sew::assemble and surfaces::Parts::canonical each lay out their record as its parts stand
(regions, shells, faces, edges with no uses, vertices) and end in Brep::canonical; their own
ordering walks are deleted. spline_conv::tensor_face (no callers left) is deleted in favour
of surfaces::tensor_sheet, and fill's copy of the spline surface record in favour of
surfaces::spline_surface.
(not type-checked: the kernel stops at name resolution in step_*, build, sweep_path and
surface_ruled)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
part()'s walk becomes renumbered(): the record of given source regions in canonical order.
part() hands it the regions it cuts; canonical() hands it every region as it is, open,
wire and vertex shells included, for records builders put together in their own order.
Each edge's uses come from the walk; a source list naming exactly those uses keeps its
radial order (turned to start at the primary use), as part() did.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The strips keep their model (exact where the rulings move the edge by an affine
map, a ruled B-spline fitted within 1e-3 otherwise; mitred or rounded corners;
an inside corner cut back). Each strip is built on its own on the record: the
chain edge as its exact NURBS (a circle or ellipse arc rational quadratic), the
far edge its image (its control points mapped) or the fit's, the surface a plane
or the ruled NURBS between them with straight pcurves along its parameter lines.
A mitre's triangles are flat polygons; a round corner's sector is SMLib's revolve
of the incoming side about the corner's axis. The faces are stitched by SMLib
(surfaces::sheet_of, which now takes the operation's operands), each strip made
in place of its edge, each corner face in place of the two edges there. Tests
read the record; the check that the strips stand on the box's own edges now
looks for edges lying along them, as the result is a body of its own.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A surface sweep along a path with corners was refused. Each smooth run is now
swept on its own with open ends by the solid sweep's corner model: carried on
straight past each corner, imprinted with the planes halving the corners and the
pieces between them kept (a point inside each face, smlib_face_points), the
profile turned through each corner onto the next run; the runs are stitched.
Only moving along the path, the runs meet as they are. Each wall keeps the
segments its uncut face came from through SMLib's tags. Sweep options along a
path with corners are refused as the solid sweep refuses them. The sheet
builders use smlib_sweeps' curve and path_curve; their own copies are gone.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
smlib_sweeps' loft_rails kept its own copy of the rail law with the anchoring
4056 was refused by (a rail lying in the last section's plane met at its first
crossing there). It now builds the sections with loft_rails::Section and takes
the sections and stations loft_rails::railed gives, so the law and its fix live
in one place. path_curve is exported beside curve for the sheet builders.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The doc puts exact tools into SMLib's boolean as drawn, so nothing asks whether a touching tool
can be joined, batches extrusions, or probes for room to grow a profile.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
machine_bores cut with the drill clipped to the panel and, refused (coincident faces at the
panel's rims), cut again with a longer cylinder a hair past the panel. SMLib's boolean takes
the flush ends exactly, so the clipped drill is the one cut; its refusal is passed on.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A lip tried kernel::raise_band and, refused, joined an extruded band in its place (whose
walls coincide with the walls below); a groove tried kernel::rim_band and, refused, cut a
band offset from the rim face's display-mesh loop (a sampled outline). Both second paths are
gone: a refusal of the exact band is the feature's refusal. offset_left, which only the
sampled band used, goes too. The rim and opening checks are unchanged.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
SMLib's boolean takes touching and coplanar faces exactly, so the tools no longer bend
around them, and no tool is tried a second way after a refusal:
- extend_for_coplanar / grow_profile_for_coplanar (a cut or join tool run on past faces it
lies on, its profile grown out of walls it touches), Surroundings and its clearance probe,
fill_void_holes (holes over empty space taken into the profile, #1360),
trim_embedded_join_ends (a join's buried ends cut off, #2059), the touching-face checks
(kernel::can_join_touching_faces / can_join_all_touching_faces) that decided when the
exact tool could go in, the profile that "grows wherever the samples find room" when the
build failed, region_samples, and the revolve's grown profiles;
- a tapered tool's run-on into the face it stands on (taper_run_on_fits, face_level_at) and
its straight TAPER_FOOT: the tapered tool is the profile drafted over its length;
- apply_traced's batched boolean that, when refused, put the model back and ran the tools
one by one (and common_join_target, which only it used): the tools run as the operation
always runs them;
- a pattern's copy of a reduced fillet tried again as a new fit when the copied radii were
refused: the copy uses the radii the original was built with;
- union_apart / join_apart errors are passed on instead of leaving the tools apart.
The command behaviour stays: the same tools, the same lineage, the same at-once boolean of
tools apart. The tests of the deleted growth and trimming (coplanar_growth_tests) go with
them.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
thread_groove built the faceted mesh groove truck cut; nothing calls it since the thread
cut is SMLib's screw sweep (kernel::sweep_helix) and a B-rep boolean. With it go
GrooveShape's facet count and the refusal it fed ("the modelled thread is too fine for this
diameter"): an exact groove has no chord error, so a fine thread on a large diameter is cut
as drawn. Dimensions that are not finite are still refused.
kernel::sweep_helix_mesh is now called only by crates/engine/src/thread_tests.rs.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
loft_blend hands its net to surfaces::tensor_sheet (the bridge's own one-face sheet);
section_frames picks the face that bounds the whole chain and faces the run best from the
record's edge uses and the face's SurfaceEval normal, and reads its frames with
bridge::frames. Behaviour and messages unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
match_surface and seam_continuity read faces as record surfaces (SurfaceEval over the
face's domain), pick the face and its kept side's edges from the record, chain and frame
the target through bridge::chain/frames, keep the crown whose bridge strays least from the
original, and turn the bridge face to the target's sense by building it again turned
(surfaces::Parts). The history is over [body, target]: the face modified from the body's,
its edges new. Behaviour, constants and messages unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
general_fill takes loops of three to sixteen edges. quilt_body assembles fill_iso's
patches into one sheet: each loop edge whole, or for five or more its two halves at the
middle of its interval, both on the edge's exact curve record and sharing a midpoint
vertex; each spoke a new edge from that vertex to the centre; every pcurve the straight
line in (u, v) of its side. The history marks every face generated, each edge or half a
piece of the body's own edge, each spoke new. A reversed loop edge's knots keep the
interval's ends exactly, so pieces name the edge's own vertices.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A loop of N >= 5 edges is N four-sided patches about a centre point: each edge split at
the middle of its parameter interval, a spoke from each midpoint to the centre, patch k
on the second half of edge k-1, the first half of edge k and the two spokes beside them.
All patches share one compatible space, so each spoke is the same control points in both
patches it divides and the centre one point; the least-squares solve takes them once
and adds rows opposing the two patches' derivatives across each spoke (tangent there).
The one- and three-/four-sided fill is the one-patch case of the same model. The result
lists each patch's sides (edge pieces with their parameter ranges, or spokes) and the
spoke curves; every pcurve is a straight line in (u, v).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
sketch.intersection_curve (LinkSource::Intersection, crates/doc/src/
project.rs) intersected the two bodies' display meshes triangle by triangle
and chained the segments into polylines: a curve known only as points, off
the true intersection by the meshes' chord.
kernel::meeting_edges(a, a_face, b, b_face, chord) takes each side whole or
as one face (that face alone as a surface body, surfaces::taken), tags both
and imprints the first by the second in SMLib (smlib_imprint: "each of its
faces that `tool` meets cut along the intersection", capi/include/smlib.h).
SMLib intersects the faces' surfaces and trims each curve to both faces, so
the edges the imprint makes are exactly where they meet; its own history
says which they are (an edge no edge of either operand became). Each comes
back as its record curve on its interval, from SMLib's export of the
imprinted body, with its points from that body's mesh at `chord`. Sides that
do not meet give none.
The intersection link now stores each edge as a wire with its exact curve
(Wire::exact, the existing WireCurve; no new type or format change), its
points at 1/1000 of the larger body. project3d::mesh_intersection, its
triangle/triangle test and segment chaining go; project3d::mesh_of becomes
part_of (the body and, for a face reference, the face index).
Needs in crates/kernel/src/lib.rs (the coordinator's): `mod intersection;`
and `pub use intersection::{MeetingEdge, meeting_edges};`.
Not built: base in progress (kernel-smlib/no-truck 513ebe47 does not build).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A rig case whose SMLib call fails left nothing to replay it from: nothing called
solidcraft_smlib::set_dump_dir. `solidcraft-cli kernel-cases --dumps=DIR` now points SMLib's
dump folder at DIR/<area>-<case> before each case runs, so a failing call leaves its
arguments and B-reps there (smlib_brep_read reads them back into a scratch C program), and
turns it off after the run. Off without the flag. The kernel exposes it as
smlib_dump_dir, beside smlib_built (the CLI reaches SMLib only through the kernel).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
truck-modeling, -topology, -shapeops, -stepio and the vendored truck-geometry, -meshalgo,
-shapeops, -stepio and matext4cgmath (with their attribution rows and vendor notes) are removed.
The kernel-cases runner has one kernel, SMLib, and no build-mismatch check; a modelled thread is
cut by SMLib's exact screw-swept groove (the faceted groove path is gone).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Edge chains are the body's record edges, walked along their curves the way the
chain runs; G1/G2 conditions read the open edge's face on the record with its
exact first and second derivatives (SurfaceEval::derivs, ::second). The face is
built on the record: each side the network surface's boundary row on its knots,
split at the chains' inner vertices into edges over parts of the side's interval
(no knot insertion), each with its straight pcurve; a collapsed side is left out.
boundary_join counts the record's edges. Tests ported.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
ssi::intersect took a sample count and gave each intersection curve as
samples: its 3D points, its UV points on each surface and the surfaces at
them. The fork's 947eedc hands out the curves SMLib's intersectors return,
exactly (smlib_ssi_curve_nurbs: degree, knots, control points, weights), and
drops the samples; SMLIB_ABI_VERSION 4.
SsiCurve now holds the 3D curve as a schema 4 Curve::Spline or
Curve::RationalSpline (rational where a weight is not 1) and each UV curve
the method returned as a Curve2, with SMLib's type, deviation estimate and
the 3D curve's natural interval; ssi::intersect loses its sample count. A
curve SMLib does not hold as a B-spline is Error::Unsupported (it is never
approximated); a B-spline whose knot count is not points + degree + 1 is an
argument error.
The SSI corpus scorer (tests/ssi_corpus.rs) samples the exact curves itself
where the corpus runner samples them: 4001 parameters spread evenly over the
3D curve's natural interval and over each UV curve's interval, the surfaces
evaluated at the UV samples, samples that cannot be evaluated dropped. The
unit test plane_through_sphere_is_a_circle evaluates the exact curve.
Nothing else used the samples. SMLIB_REV: the fork at 947eedc.
clippy -p solidcraft-smlib --all-targets, with and without native: clean.
The corpus gate (smlib_global_intersect_at_1e_6) is not run here.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
bridge::{chain, frames, frames_across, chain_closes} and surfaces take
body.record() instead of encoding one; a surface body is Body::from_record of the
canonical record, a stitched one Body::from_smlib_record of SMLib's export, its
history matched to the input record when SMLib's import numbers it alike.
record::attach takes the new form.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The bridge core was already its own geometry; its readings went through truck.
Chains are edges of the body's record (index, direction), measured along each
edge's record curve the way the chain runs; a support face is a record face whose
surface is evaluated with its exact first derivatives (SurfaceEval::derivs) where
truck's were finite differences; the station's parameters come from SurfaceEval's
projection. The bridge face is surfaces::tensor_sheet, and blend_between records
it as new over [a, b]. frames, frames_across and chain_closes take the body and
face index. Boundary patches go through the same bridge. Tests ported.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
surfaces.rs gains the record pieces NURBS faces are built from: a NURBS surface
and curve as record geometry, a pcurve along a parameter line, a whole tensor face
bounded by its net's boundary rows and columns (exact for a clamped surface), and
all_new for a face new to an operation. tspline_sheet makes a regular T-spline
that tensor face and an irregular one its Bézier elements, each side split at the
T-points, each edge an exact Bézier piece of its side with a straight pcurve on
every patch it bounds, shared between neighbours. Tests ported to the record.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Copy, unstitch, reverse and remove faces, Patch (a sketch region, or a planar loop
of a body's edges) and Extend work on the body's record: Parts puts any faces,
edges and vertices into canonical order, and each face taken is the body's face
(Kept), its edges the body's. Sheets are SMLib's sweeps with no caps (extrude,
revolve, curve sweep, stations for twist/scale/orientation/rails), a chain at a
time; a loft lofts each column of matched segments and stitches the columns.
History: a wall comes from the segments its edges in the start plane lie along.
Stitch joins the bodies' records and sews them with smlib_stitch, faces and edges
carried back to the bodies through SMLib's tags. Offset Surface offsets each face
on its own with smlib_offset_sheet and stitches the offsets within the tolerance:
smooth neighbours stay joined, sharp ones come apart, as before; the fitted
offset surfaces and their fold check are gone. Thicken is SMLib's thicken.
Not yet: a surface sweep along a path with corners (refused). Tests ported;
the fold-radius message test now checks only the refusal.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
loft_rails.rs placed stations between truck wires. Its law now works on the
sections' exact NURBS curves (Section, from the sketch segments' record curves)
and returns the sections and stations, each a curve list a loft is skinned
through; nothing of truck is left.
Case 4056 (a keel rail lying in the bow's sketch plane): rail() anchored every
section at the rail's first crossing of the section's plane after the section
before. A rail lying in the last section's plane crosses it all along, first at
the far section, 100 mm from the bow's curve, and the loft was refused ("a guide
rail crosses section 2's plane"). A rail runs from the first section to the
last, so the last section is met where the rail ends; the sections between are
met at their first crossing, as before.
smlib_sweeps.rs's solid loft_rails keeps its own copy of the anchoring and has
the same cause; it should call loft_rails::railed.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The safe wrappers of the fork's smlib_stitch (a body's faces sewn at a tolerance,
oriented, with whether it closed) and smlib_offset_sheet (a sheet's faces onto
their offsets), for the kernel's Stitch and Offset Surface.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
freeform.rs turned each free-form sketch segment into a truck edge. It now gives
every segment (lines, arcs, Béziers, conics) as a record curve and its pcurve on
the sketch plane's own surface, on one parameter with the same knots and weights,
so the pcurve lifts onto the edge exactly.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
trim_keep told each piece's side of the tool from the tool's field at the
centres of its mesh triangles. Each piece now takes one point on its own
surface inside its trims (SMLib's CalculateAnInternalPoint, through
smlib_face_points), mapped to the record's face order through SMLib's
numbering of the split body; the tool's field there says the side. A piece
whose side can't be told is an error, never guessed.
(not built: base in progress; no errors in this file)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Each face's point from SmFace::CalculateAnInternalPoint, in SMLib's face
order; the fork at 89a1933 (also: smlib_extend_sheet extends every surface
as SMLib's B-spline, its analytic kind given back).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
planar_regularize now calls the fork's smlib_merge_faces as it is
(SmBrep::RemoveTopologicalEdgesAndVertices with the edges under the kept points
staying): every edge the lineage does not choose is kept by a point at its
middle, so SMLib joins the faces across the chosen ones only. A body with more
kept edges than one call takes stays as it is, as the other budgets leave it.
splitface, planar_sources and planar_regularize read the body's record in place
(Body::record) and a face's use from Body::face_forward; the two-lump test body
is Body::from_record of the two records side by side.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
planar_sources reads each face's oriented plane from the body's record
(actual_planes: the surface's plane, turned with the face's use) instead of
truck faces; the lineage rules are unchanged. planar_regularize picks the edges
to remove as before (two faces of one shell on one certified plane, from one
source unless generated walls, off every split line) from the record's edge
uses, and SMLib joins the faces across them (Brep::merge_faces: each edge
removed where SmEdge::IsTopological finds the faces' surfaces coincide), so the
merged face is SMLib's and the history its own tags (Merged), composed through
the body's. SmBrep::Simplify is not this operation: it refits curves and
surfaces by knot removal and sampling.
The planar_sources tests read face points, edges and planes from the record;
the two-lump body is built from the two records side by side, each face given
the source of its own operand's face on that plane. Deleted:
planar_regularize's alternate_adjacency_path_cannot_remove_protected_internal_edge,
a test of the truck face builder merged_face (only listed edges are removed now,
so no other path can remove a protected one).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
coplanar.rs (coincident faces pushed apart, faces glued back to back, bores and
insets joined, imprint unions), seam_faces.rs (round faces halved before a
retried boolean) and planar_faces.rs (flat B-spline faces made planes so the
coplanar strategies see them) existed to get round the mesh-based boolean's
coincident faces and seam loops. SMLib's boolean classifies coincident faces
itself (SmMerge::ManifoldBoolean, step 2), so these are retries and second
paths around the operation, not operations: deleted, not ported.
Their tests of what the boolean gives keep their volume, lump and validity
assertions and now run against the boolean itself (coplanar_tests.rs and
seam_faces_tests.rs need their own mod lines in lib.rs); the assertions on the
strategies used, and the tests of push_faces, grow_faces_exact, trim_exact,
planarized, coincident_planes and halved, go with them.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A tool is now a set of schema 4 record surfaces, each over a parameter box,
unbounded where the tool is: a plane; a face's surface over its own box; a
sketch curve as one patch per segment swept along the direction (a line a plane
strip, an arc swept square to its sketch a cylinder strip, any other segment its
exact rational Bezier extruded), an open curve continued along its end tangents
past the body. SMLib cuts the faces where the patches meet them
(Brep::split_faces: its intersection curves trimmed to each face by
MergeCurvesOnSurface) and slices a body with one patch (Brep::split_body); the
history is SMLib's own tags. Nothing is traced on a mesh, sampled or fitted.
Which side of the tool a point is on is its offset along the normal of the
nearest patch, each patch's surface prepared once with its projection grid
(ToolSide), never per point. ToolSide::of and side_of_point take the body the
tool must reach past; face_side becomes face_sides (every face of a body, one
mesh); analytic_field becomes analytic_face (a face of a body, on its record);
split_any returns the face map instead of the truck imprint. split_edges_at and
fit (a Catmull-Rom fit through samples) are deleted with the truck code.
ssi::Surface::from_record turns a record surface over a box into SMLib's.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
split_faces cuts faces (SMLib face order) where ssi::Surface tools meet them and
says per face how many faces it became; split_body cuts a solid in two by a
surface, the part behind it and the part in front. SMLIB_REV moves to the fork
commit that adds smlib_split_faces and smlib_split_body.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
face_params and outline are gone (no callers on the record). The fork head adds the 2D planar
boolean (smlib_planar_faces, smlib_boolean2d), the grazing-curve crossing rule (solidcore#480) and
untrim's parameter-box status.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
It imprinted two bodies touching back to back on each other along the outline of their
contact and glued them, for the old boolean's coplanar unions that could not join the pieces
its own imprint left. SMLib's boolean is the one path, and nothing calls it any more. Its tests
of the union itself (a lip swept on a round rim joins it, the lip's foot is a plane) stay as
boolean tests, without the old retry diagnostics.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Its sampled parameter loops (points projected back onto the surface, the outer loop guessed by
area) and its iso-curves of truck surfaces served Untrim and Extend, which no longer use them;
on the schema 4 record a face's loops in (u, v) are its exact pcurves, the outer loop the one
running counter-clockwise about the natural normal. The only use left is a truck test in
native.rs's seam fixtures.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
geom's NURBS curves and surfaces become the record's rational splines, weights kept.
tensor_face builds a whole tensor surface as a one-face sheet (assemble's new Assembly::sheet:
one open shell in the infinite region): the net's first and last rows and columns as edges on
the surface's own knots, sharing the corners, each with its exact pcurve (an iso-line, the
edge's parameter being the surface's along it), and returns the body.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
plane_through gives a plane record (its natural normal turning with the loop); EdgeSampler
samples a record edge through CurveEval in the direction a loop runs along it (an intersection
curve traced as the record holds it, with the evaluator's own hint); planar_face certifies a
loop of record edges from their records (lines by their ends, circles and ellipses by their
frames, polylines and splines by their points with positive weights, intersection curves by a
plane among their surfaces) and returns the plane, the caller building the face on the record.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A face's reach into a region comes from its record surface: planes (as offset::surf_of reads
them) lie in their triangles, a spline face's control net bounds it (its hull, read off the
record's points and positive weights), walls along the direction are prism::wall_along's, and
round faces keep their span and side. Truck's surfaces and the guard around them are gone.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The exact step of the distance between bodies works on the record: analytic faces in closed
form as before, other faces through SurfaceEval::project, edges by a bracketed golden-section
search of the record curve's own points from the nearest of evenly spaced ones (one method for
every curve kind, traced intersections included), the direction there from the curve's
derivative. Truck's point inversion is gone with it.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
face_surface is the record face's analytic surface (offset::surf_of); edge_circle takes a
circle record's centre, axis and angle exactly (its parameter is the angle about the frame's
z), and for any other curve keeps the test that its points all lie on one circle (SMLib holds
its primitives' circles as rational splines). The spindle torus's signed major radius is gone
with the old kernel: the record's tori are ring tori.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Seams and merged topology tell faces on one surface apart by the record's own surfaces
(offset::surf_of: plane, cylinder, sphere, cone, torus) and, for free-form faces, by the very
same surface record, never by fitting circles, spheres and cones to mesh vertices or by
comparing mesh normals along an edge. Edges are welded by their record vertices and middle
point, split lines and pinches read the record edge's points, a cone's apex is checked against
its faces' record vertices. The mesh still gives faces of no area and a cylinder face's extent
along its axis; a cylinder's side (hole or boss) is the record's.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
planar_regions_boolean replaced each free-form loop (a font's curves, an ellipse) with arcs
fitted within PLANAR_FIT_TOL before its own line/arc arrangement, and a loop the operation
cut kept those arcs: a cut free-form outline was extruded inexactly. Now each operand is a
sheet of planar faces built from its loops' own curves (Brep::planar_faces; lines, arcs and
the Béziers and conics as NURBS, exactly) and SM_API's 2D boolean combines them
(Brep::boolean2d): SMLib's topology intersector, the one the solid booleans use, splits the
operands' edges where they cross and the result's edges are copies of those curves. The
faces read back exactly (prism_curve::projected: lines, arcs, each Bézier span as the
segment it is). The Rust arrangement (crossings, ray parity, chaining, the fit and the
"exact again" restoration) is deleted: one source, no fitting.
extrude_planar_regions unites its regions the same way (balanced unions of one-region
sheets) and refuses pieces of the union that share a vertex, read off the result's
topology instead of a line/arc crossing test on fitted arcs.
Signatures stay (doc/sheet.rs and mesh_convert call them as before); `tolerance` is the gap
a loop may close within and the least tolerance edges are read back at. PLANAR_FIT_TOL
stays for doc/engine's flat patterns, which fit their own loops. smlib_sweeps::curve (the
Seg2 to SMLib curve conversion) is made pub(crate) for the sheets.
Tests: those of the deleted helpers (crossings, ray parity, closest_param, segment boxes)
are gone; the free-form test now asserts what the exact boolean gives (the half-ellipse
notch is the ellipse's own conics, area exact to 1e-6; the joined ellipses are conics), and
the fitted-arc budget assertion is dropped. Without the smlib feature the booleans are
refused ("not supported yet").
(not built: base in progress; cargo check shows no errors in prism.rs or smlib_sweeps.rs
with and without the smlib feature)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Wrappers of the fork's smlib_planar_faces (a sheet of planar faces from closed curve loops in
one plane, nested by SMLib) and smlib_boolean2d (two such sheets combined as regions of the
plane by SM_API's 2D boolean, the inputs' own curves split where they cross; an empty result
is None). SMLIB_REV moves to the fork commit that adds them (78617dd).
clippy -p solidcraft-smlib --features native --all-targets -D warnings: clean.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The general fill (G1/G2 patches, patches through points, loops that bound no height
field) is solidcraft_geom::fill_iso's patch on the loop's own edges: each line or spline
edge, on its interval and run the loop's way, is a side, so its pcurve is a straight line
in (u, v), exact. The frames across a G1/G2 edge are read off the one face beside it in
the record: its outward normal, the way it leaves the edge, and for G2 its normal
curvature from the surface's exact second derivatives. Corners whose neighbours' tangent
planes disagree are exempt from the angle tolerance. Loops of five or more edges, and
conic, rational or traced edges, are refused by name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The loop's edges become the sides of the unit square (three sides: the fourth collapses
to the first corner, a pole), each mapped affinely onto [0, 1]. Opposite sides are made
compatible exactly (split into Bezier pieces at every break, raised to the higher degree,
rejoined), the bilinearly blended Coons patch is the start, and with the boundary rows
held the interior control points solve the G1/G2 rows, the points to pass through and a
vanishing multiple of the thin-plate energy by least squares, refined by halving spans.
Every edge lies on the patch exactly and its pcurve is a straight line in (u, v), affine
in the edge's own parameter: no pcurve is fitted.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A solid body's volume, area and centre are SMLib's mass properties at its documented 1e-4 and its
box SMLib's exact one; a body SMLib cannot measure is an error. The three-mesh measure and the
mesh cross-check beside SMLib's numbers are gone, and with them the kernel switch
(SOLIDCRAFT_KERNEL, Backend): SMLib is the kernel. The smlib feature gates in the crate root, the
validity check and the offset backend go, since SMLib's crate is always linked (answering not
built without the feature).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Encoding writes the body's own record and decoding makes the body from it: the adapter to and
from the old kernel is gone. A record goes into SMLib as given (SMLib's own form); one SMLib
refuses is an error that names why, never sent through SMLib's healer. A body made by SMLib keeps
the SMLib body. The bounded preview of a decoded body is a coarse mesh from the planner, charged
to its triangle budget.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A boolean is SMLib's boolean, its history SMLib's tags: the jittered retries, tool nudges, prism
and coaxial routes, the retry result guard and their diagnostics are gone. A rigid placement moves
the record; any other matrix goes through SMLib's transform (new Brep::transformed over
smlib_brep_transform). Bodies apart join as their records side by side (brep part::compound),
with each piece's faces, sources and history; apart means their exact boxes do not meet. A split
is the intersection with a half-space box, with no faces pushed off the plane.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The fork head carries untrim and extend shims with their history tags, the local solvers that
take no value from a failed evaluation (solidcore#481), the B-spline curve radius check
(#475), the torus minor-interval check (#482), the blend rail along its face's rim (#485) and the
intersection path's statuses.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
What stays is the composition of a multi-step operation's histories (composed, placed, kept,
through, operand, of_pair, Traced). healed(body) is crate::heal::heal with the body's split
lines kept, its history over [body] SMLib's, each face keeping the construction source its
faces shared and the body its regularization protection (as the prism result's healing did).
Deleted with the old kernel: heal_build, truck_result, heal_body and heal_prism_result (heal
and record::after replace them), Shapeops and shapeops (truck's boolean history), edges_evaluate
(sampling truck's intersection curves), copied and void (truck copies and the void boolean,
no callers left), join (records apart join through brep::part::compound) and parts_of (a lump's
history is history::of_part). Tests: the truck-retry tests (shifted attempts, the nudged tool)
deleted; shell counts read the record.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Body::from_record takes the finished record as the body, so its faces and edges are the
record's and the history needs no round trip through SMLib's numbering. The edge origins
(kept edges, pieces) go to record::attach, which gives every other edge its faces' meeting.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
An operation that builds its result says, per result face, what it is made from (an operand
face it is the whole of or a piece of, profile segments, caps, operand edges a blend replaces),
and per result edge the operand edges it is or is a piece of (pieces_of); of_faces and attach
make the history of the result's record from that, every other edge the meeting of its two
faces, operand faces under several result faces marked Split. That is the one model in place
of the identity-based Operand, Rebuild/Was, traced/build/build_placed, edge_index/face_index/
edge_origins/edge_provs/edges_by_identity, from_faces, solid_counts and rebuilt/rebuilt_body,
and of the truck sweep followers (ruled, follow, columns, Column, swept_face), all deleted.
The face constructors, regions, counts, after and meeting stay. Tests read edge_faces of the
body's record.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The SMLib point sweep refused a polyline whose points turn gently (a smooth run
of more than two points), a branch on the shape of the input: a polyline is
the path its points make, and SMLib's cornered sweep mitres it at each point.
Curves with an exact form no longer reach the kernel as points (previous
commit); what still does (curves projected onto a face, intersection and iso
curves, sweep_path's point lists) is followed as those points.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A sweep path through 3D sketch curves sampled every planar sketch curve in it
into a 0.01 mm polyline, and took every linked 3D curve as the points its link
stored, so SMLib got a many-cornered polyline (and refused a smooth one). Now:
- a planar sketch curve in such a path is its exact segments
(PathPiece::Segs of PathSeg::of_sketch over exact_segs);
- a linked curve whose source has an exact curve keeps it: Wire::exact holds
the record curves (WireCurve: schema 4 curve + interval) end to end along the
wire, set with the wire when the link is resolved: an included body edge its
record edge (kernel::brep_record), an included sketch curve leaving the plane
its segments (kernel::sketch_record_curve). Wire::map carries them with a
placement. The sweep follows them through kernel::curve_path (line, arc,
ellipse as rational quadratic spans, polyline, B-spline as Bézier spans).
- curves known only as points (projected onto a face, intersection and iso
curves: the link resolves them on the mesh) stay points.
Kernel, for the sheet and blend ports: planar_face_record (a profile face as a
schema 4 sheet record with exact pcurves), smlib_sweeps::chain_curves (a
segment chain as SMLib curves), body_from_record (any record into SMLib,
checked), sketch_record_curve; record curves from design files are checked in
curve_path. Format: pending/wire-exact-curves.rs.
lib.rs (coordinator): export SpaceBezier, curve_path, edge_path,
sketch_record_curve and planar_face_record from the kernel.
(not built: base in progress; kernel and sketch check clean, doc not reachable)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
PathSeg::Bezier(SpaceBezier) replaces PathSeg::Cubic: a rational Bézier span in
space of degree 1 to 9 (homogeneous control points), so a drawn spline's cubic
spans and an edge's own curve are one kind of segment, swept by SMLib as the
NURBS it is. SpaceBezier evaluates, differentiates (quotient rule on the
hodograph), measures and walks by length (Gauss-Legendre), reverses and maps.
edge_path gives a body edge's record curve as path segments exactly: a line,
an arc, an ellipse as rational quadratic spans, a polyline's lines, a B-spline
as its Bézier spans between knots (bspline::segment); an intersection curve
(no exact curve in the record) is refused. PathPiece::Segs carries such exact
curves in a point path; PathSeg gains reversed, mapped (a placement) and
of_sketch (a sketch segment on its plane). The cubic_* helpers are deleted.
surfaces.rs: its one PathSeg::Cubic arm follows the rename.
lib.rs (coordinator): export SpaceBezier and edge_path from point_path.
(not built: base in progress; cargo check shows no error in these files)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
extend_surface keeps its signature, options and error messages. One model:
SMLib extends the surface past a side of the face's parameter box
(smlib_extend_sheet, CreateExtendedSurface: natural continues the end spans,
tangent G1, curvature G2) and the face stays one face on it, tagged so the
face is Modified and its sides keep their edges. Several picked sides grow
the box on each. Up to a target, the face is extended 50 body sizes past
its edge and trimmed back with trim_keep (SMLib's exact intersection),
keeping the piece the face lies in. Back to the bounds is untrim of the
picked edge's face, in place, with its holes; one already there is
refused. A plane's straight edges with neighbours lengthened still go
through surfaces::extend. Edges are picked on the schema 4 record (samples
of each edge's curve), an edge two faces share or of a hole refused as before.
Not ported (every far edge there was a cubic fitted through marched
samples, which the exact-B-splines rule forbids): an edge that is not a
whole side of its surface's box (trimmed edges, curved planar edges such as
a disc's, a notch), mitred corners on such faces, sides run square to the
edge on a non-box face, and bodies of more than one face. These now answer
"not supported yet: ...". Tangent and curvature no longer add a separate
run-off face: SMLib's G1/G2 extension is one face (the engine's
surface.extend description says so).
Tests ported keeping their assertions: the planar extension, the cylinder
strip along its axis, around and both (now an extruded arc, not a truck
revolution), the extruded strip, the parabola extended naturally (now a
conic extruded), back to bounds (the saddle from loft_sheet and trim),
hostile requests. Deleted, as what they test no longer exists:
neighbours_lengthened_or_square, self_intersecting_extensions_are_refused,
a_disc_grows_by_its_curved_edges, neighbouring_faces_extend_together, the
tangent/curvature run-off half of the B-spline test, and
a_continued_bspline_is_unchanged_on_its_range (the deleted continue_span).
SMLIB_REV: fork 9531966 (tags carried through untrim and extend).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
untrim keeps its signature and messages. The face is picked in SMLib by a
point inside it (the centre of its largest mesh triangle), holes by their
points; SMLib deletes the holes or rebuilds the face over its surface's
parameter box with the holes kept (fork c6c7fb4). The body goes tagged as
operand 0, so faces and edges come back with their origins; the untrimmed
face is Modified. A face whose span goes all the way round its surface now
closes on the seam instead of being refused.
solidcraft-smlib: Brep::untrim_face (ffi/trim.rs, trim.rs) and
Error::NotExact for SMLIB_ERR_NOT_EXACT.
Tests: ported to the record (loops counted on brep_record; the saddle made by
loft_sheet and trim instead of truck types; the torus face copied out by a
point). Deleted: the_outer_loop_is_found_whatever_order_the_loops_are_in: it
built a truck face listing its hole first; SMLib's import always puts the
outer loop first, so the case no longer exists.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
trim / trim_keep / mutual_trim keep their signatures and behaviour; the
truck walk over the split solid is gone. The body is split where the tool
meets it (splitface::split_any), each piece's side is the tool's field at
the centre of its mesh triangle where the field is largest (the record
mesher), regions come from splitface::regions, and the pieces not kept go
through remove_faces, whose history record::after carries back to the
body. A mutual trim's nearest boundary edge is found on the schema 4
record: edges one face uses once, their middles through CurveEval.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A solid body holds its record (canonical faces, edges and vertices are the body's everywhere) and
builds its SMLib body from it on first use, or keeps the one the operation made. Queries read the
record: planar faces, outline and exact bounds, counts, closed shells, lumps from its material
regions (record parts, with their history), shell nesting from its regions. The orientation mesh,
the revolved-face rephasing, kept tolerances and the old kernel's mesh caches are gone. History's
edge faces come from the record's edge uses. The kernel no longer depends on the old kernel's
crates; SMLib's crate is a plain dependency (without the smlib feature it answers not built).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
compound concatenates the records and takes the part of all their material, so the result is
in canonical order with each face, edge and vertex numbered from its input. part now keeps an
island inside a void when both its region and the void's material are taken.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
patch_loop picks the closed loop among the record's edges, certifies a plane (lines by
their ends, splines by their control points, conics by centre and tilt, traced curves by
a planar support) or fits the height field, and assembles one face on the body's own edge
records; the height field's parameters are plane coordinates, so each line or spline
edge's pcurve is its exact projection (segmented to the edge's interval). The history
marks the face generated and each edge the body's own. A fit on circular, elliptic or
traced edges, whose pcurves on the spline have no exact form, is refused by name, as is
the general fill (G1/G2, points, loops bounding no height field) until it is ported.
Tests ported to the record; the general-fill ones stay and fail until then.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
corner_g2_patch builds the same tensor patch and returns CornerPatch: the B-spline
surface record, its v = 0 boundary line (the net's first column on the same u knots)
and each side's interval of it, so the seams lie on the patch exactly without cutting
the curve. The fold check and seam_error read the record surface through SurfaceEval.
Tests ported (the rolling ball is a sphere record).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Edges, the faces each bounds and each face's outward side come from the schema 4
record; stations on the edge's curve (CurveEval::derivs for its tangent) are projected
onto each face's SurfaceEval, and the normal curvature across the seam is read from the
surface's exact second derivatives (SurfaceEval::second). Behaviour, budgets and
messages unchanged.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Closed forms on cylinders, cones, spheres and tori; on a spline the basis functions' own
second derivatives (local_basis) through the quotient rule for the weights; zero on a
plane. A revolution has none (its profile evaluator carries first derivatives only).
The seam continuity measure reads normal curvatures from them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
fair_bspline_curve/_surface and fair_nurbs_curve/_surface take and give the schema 4
record's Spline and RationalSpline kinds (the weights kept, the knots unchanged).
fair_faces fairs the picked faces' surfaces in the body's record, keeps its loops,
edges, vertices and pcurves, and has SMLib build the body; its history marks the
faired faces modified and every edge kept when SMLib keeps the record's order, and is
left off otherwise. Tests ported to the record types with their assertions.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The faces' descriptions become a record with exactly the topology they give: vertices
and edges shared by key (lines, circles or flat arcs as their exact rational quadratic,
Bezier chains as B-splines), planes about the boundary's mean, revolved lines as
cylinders, cones or planes, extruded arcs as cylinders, ruled walls as B-splines with
exact iso-line pcurves. assemble() groups faces into shells, puts closed shells in
regions (a cavity where the outward normals enclose negative volume, in the solid whose
winding number about it is one), orders everything canonically and validates it.
SMLib imports the record as trusted; the body and the face each body face came from
follow SMLib's order. The edge-sheet and vertex-fan logic for touching bodies is kept.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
assemble (records rebuilt after removing and adding faces) joins; the boolean guards that re-ran
a failed boolean with a changed tool (bore_guard, candidate_reach, sphere_lens, tangent_cylinders,
touching_walls) are gone; the lens-cut result stays as a boolean test. SMLIB_REV at the fork's
integration head with smlib_merge_faces.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
heal: a body's faces on one surface merged and its edges on one curve joined by SMLib
(Brep::merge_faces, SmBrep::RemoveTopologicalEdgesAndVertices), the edges along its split lines
kept, with SMLib's history over the body. Its booleans already end the same way
(SmMerge::BooleanPostProcess). The truck healing is deleted with what only it did: coplanar and
collinear merges by hand, flat surfaces rebuilt as planes, and approximate edges refitted as
lines or arcs from samples. on_split_line stays; edge_points reads a record edge.
Deleted, with no port: the retries around the old boolean's failures (touching_walls,
bore_guard, tangent_cylinders, sphere_lens, tool_reseam, candidate_reach). Each re-ran a failed
boolean with a changed tool or split faces; SMLib's boolean is the one path, and a case it
fails is its defect to fix there. Their own tests go with them (the helpers' unit tests in
issue_788_tests, bore_union_tests, candidate_reach_tests, and sphere_lens' relaxed-tool test);
the #2750 lens cut's result test stays, as a boolean test, without the retry diagnostics.
Heal's tests built truck prisms and a truck spline edge: deleted.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The wrapper of the fork's smlib_merge_faces (SmBrep::RemoveTopologicalEdgesAndVertices): faces
on one surface meeting along an edge become one, edges on one curve meeting at a vertex nothing
else uses become one, and the edges under the given points stay. SMLIB_REV moves to the fork
commit that adds it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The neighbours of the deleted faces are rebuilt from the body's record: each keeps its own
plane (every moved corner lies on it) and its edges away from the gap; corners on the gap move
to where the neighbouring planes meet, and the gap closes with straight edges shared between
the rebuilt faces. Moved and restored edges are recorded as pieces of the edges they replace
(assemble.rs: Assembly::piece), so the history is as before: untouched faces kept, neighbours
modified, edges kept, pieces of their edges, or where their faces meet.
Test deleted: delete_pocket_keeps_an_imported_face_around_it bowed the edges of a truck body in
place to show that a face losing only whole loops keeps its surface; a record body has no such
in-place edit, and every rebuilt face now keeps its own surface.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The band is built on the body's record: the face keeps its other loop and the band's edge, the
walls rise along the face's loop (continuing the walls below) and along the band's edge
(planes along lines, cylinders along arcs), and the band's top is the face's plane moved up.
Each face is given with its outside, so no orientation is tried: the eight flips the old
construction tried in turn are gone. rim_band reads the loop's lines and arcs from the record.
History as before (the trimmed face, each wall from its loop edge, the top an offset).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The emboss is built from the body's record: the wall's face loses the profiles' footprints,
each profile adds planar walls and its cap on the wall's cylinder at the new radius, and SMLib
builds the body from the finished record (assemble.rs: a record taken apart and finished again
in canonical order, pcurves from the record's routine for new uses, the record's own for kept
ones). surf_of and with_same_surface read the record's analytic surfaces; the topological
offset and draft that only the old kernel used are deleted (SMLib's offset faces and draft in
offset_backend replace them).
#3782: the saved pick had to lie on the half of the wall the sketch is tangent to (its angle
inside that face's span). The wall is every exterior cylindrical face of the body (a stepped or
interrupted cylinder is refused before), so the pick is checked on the cylinder between the
endcaps. The sleeve test's far-side pick now succeeds; a pick on the bore is still refused.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A lump of a body is the part of one of its material regions: its outer shell, its open and wire
shells and a void for each closed inner shell, renumbered in the record's canonical order, with
the source index of each face, edge and vertex for history. Islands inside those voids are parts
of their own. Read from the regions the record already holds, with no mesh or ray test.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The fork's shims now hold an operation SMLib reports done to having done
something (SMLIB_ERR_NO_EFFECT, -1107: a blend that made no face, a face
move or draft that moved no face). It maps to Error::NoEffect, and the
kernel says "SMLib made no change: ..." for blends and for offset, shell
and draft. Case 1080 (chamfer over openings) returned an unchanged body
as success.
The fork pin moves to 6bb2938: the chamfer side fix of the piecewise
merge, the no-effect contract, and SmStatus marked must-use with the
dropped statuses of the shims and of the fillet, boolean, offset and sweep code passed on.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
faceted_boolean (a body with a closed triangle mesh) faceted the body's curved faces and ran
its own BSP polygon boolean, then rebuilt a truck solid. It now builds the mesh into a
solid: each run of coplanar triangles sharing edges (corners within `tol` of each other's
planes) one planar face, its boundary chained into loops (split where a loop passes a vertex
twice, holes given to the innermost outer loop), sewn with sew_touching; SMLib's boolean
takes it with the body, whose faces stay exact. The history over [a, mesh] is the boolean's
over [a, tool] with the tool's own over the mesh substituted: each tool face from its
triangles, each tool edge the meeting of the runs either side.
Deleted with the BSP and the rebuild: planar_boolean and sliced_planar_boolean (the truck
boolean's exact fallback for touching planar faces: on SMLib the boolean is one path),
reassembled and split_beside (prism_local's), components_in_order (truck shell order) and
the tests of what no longer exists (face_planes diagnostics, the planar fallback beside a
plate's small rounds).
lib.rs (coordinator): `pub use polybool::faceted_boolean;` (planar_boolean is gone).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The prism module kept two things: the 2D arrangement of planar regions
(planar_regions_boolean, and extrude_planar_regions, which normalises the material with it,
proves the pieces apart and extrudes them with crate::extrude, SMLib's extrusion with its
face sources), and a second route to the solid boolean: prism operands read section by
section, the result sewn and its history rebuilt from section tags (prism_boolean,
prism_union_between_caps, the stacked and free-form variants, prism_lineage.rs) and the
column-local variant (prism_local.rs). On SMLib a boolean is smlib_boolean, with SMLib's
own history: the second route is deleted, with its tests (prism_spline_tests.rs, the
stack/prism tests in planar_tests) of what no longer exists.
The section tags shrink to what the arrangement uses: the free-form segment a fitted arc
stands for. wall_along reads a record surface (cylinder, revolved line, spline net) for
clearance.rs. The 2D tests and the wall test keep their assertions.
lib.rs (coordinator): drop `mod prism_local;`.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
certified_prism_equivalent proves two bodies the same uniform prism from their records
(crate::native::brep_record): caps are the planar faces square to a candidate direction,
their outward side from the shell's oriented faces, their loops run about the outward
normal and read by prism_curve; walls are planes, cylinders (their axial extent from the
face's domain), revolutions of a line or ruled spline nets, each bounded as before. The
interval arithmetic, loop matching and separation proof are unchanged. Tests build their
altered bodies by editing the record (a wall's surface swapped or moved, a cap plane's
origin moved far away) instead of truck faces, with the same assertions.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
prism_curve::segment and ::projected take a Run (a record edge's curve over its
interval, run one way, with the vertex points it must reach) instead of a truck edge.
Lines and circles are read exactly as the record holds them; a polyline as its lines; a
B-spline or NURBS curve span by span (clamped to the interval with Boehm insertion, then
split into Bezier spans), certified by the existing whole-span Bernstein proofs; an
ellipse as its exact rational quadratic spans of at most a quarter turn. segment_fitted
(arcs fitted to a spline, only the prism boolean's free-form route used it) is deleted.
The existing tests run on record curves; the part of the aliasing test that asked the
deleted prism boolean to decline is dropped.
Callers in body.rs, bore_guard.rs, sphere_lens.rs (other owners) build a Run with
prism_curve::Run::of(record, edge, forward) or Run { curve, interval, forward, ends }.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
measure_backend asked SMLib for mass properties at 1e-8, the finest its
integrator allows and below the range SmApiBrepComputeMassProperties
documents ("Relative accuracy, clamped to [1e-4, 1e-1]", SmApiQueries.h,
whose cost is "governed by dRelativeAccuracy"). The tight request was there
because the documented range did not hold on faces bounded by circles
(KNOWN_ISSUES.md); the fork's dcbb07d integrates each trim-curve knot span
on its own, and such faces now meet it (a cylinder's volume 1.8e-6 off at
1e-4). At 1e-8 SMLib's measure was most of the regression rig's time on a
modelled thread (threads/602).
The measure now asks 1e-4: ten times inside the 1e-3 it is held to (AGREE
against the mesh measure, and the rig's volume and area tolerance). NURBS
spheres and tori can still miss 1e-4 by a small factor (tangency-labs/
solidcore#384; a torus's volume 2.5e-4 at 1e-4), inside that 1e-3.
Brep::mass_properties keeps MEASURE_ACCURACY (1e-8) as its tests' default.
SMLIB_REV: the fork at dcbb07d.
Not built: base in progress (kernel-smlib/no-truck 52f25ca3 does not build).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A rotation and a translation move every point, axis and control point and leave parameters,
pcurves, tolerances, orientation and the canonical order as they are: the same B-rep in other
coordinates. Placing a body (assembly instances, patterns) no longer needs a kernel round trip.
Reflections and scalings change parameters or geometry kinds and go to the kernel.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
SMLib's blend builds rolling-ball rounds, corners, loops and vertex spheres in one call; the
modules that built them on the old kernel were deleted by the blends port.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A join is hidden when both faces are clamped Bezier B-spline surfaces swept by one
linear translation (read from the record's spline net, the face's side from its shell)
and the edge is a straight line on a ruling of both, their normals agreeing within the
certified angle, as before. The edges are the record's, in the order the mesh has them.
Tests take the record (a closed edge's equal end vertices, the ruled face by its spline
net, ruled() on record control nets); the case that bowed a truck edge in place is
removed (a record edge cannot be changed under a body).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
curveblend.rs (rolling-ball rounds and chamfers of curved edges, their corners, the law
and chord variants), curvecorner.rs, loopblend.rs (blends round whole face loops) and
blend_vertex.rs (a sphere at three rounded edges) built blends on truck; SMLib's blend
makes all of these in its one call, with its own corners. curvecorner_tests.rs tests
fillet and chamfer through the public functions only, so it stays, mounted from blend.rs.
curveblend_tests.rs and curveblend_law_tests.rs go with the builder: they test its
internals (fit_sections, curve_blend_with_limit, ROLLS, its NURBS surfaces and contacts).
lib.rs needs `mod curveblend;`, `mod loopblend;` and `mod blend_vertex;` removed.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
fillet_variable, fillet_variable_points and fillet_variable_spline round the edge with
SMLib's variable-radius solver (smlib_fillet_variable) on any edge; fillet_chord is
SMLib's constant-chord round (smlib_blend, SMLIB_BLEND_CHORD). On a straight edge
between planar faces a radius law whose band runs past the faces at any of 256 stations
is refused as too large before SMLib runs, as before. The edge setup, the band check and
the taper's room read the body's schema 4 record: planar faces' outward normals, the
directions into them, side edges' exact tangents at the ends; a curved boundary is
divided to within the clearance for the containment test. RadiusSpline evaluates with
the geometry crate's B-spline basis.
The truck variable-radius surface builder and the rolling-ball law and chord blends are
gone. taper_mesh_tests keeps its mesh-accuracy test; its three tests of truck's mesher
on a surface with an arbitrary pole are removed (that mesher no longer exists).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The corner builder (one, two or three rounds at a vertex of planar faces, taking earlier
rounds off and blending them again) is gone: SMLib's blend makes those corners itself,
with each edge's radius. The tests check through the record (curved faces' normals along
their shared edges, vertex points, planar faces' vertices on their planes). Removed: the
assertion that a concave edge meeting a convex one is refused, a limit of the deleted
builder.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
fillet, fillet_each, fillet_g2, chamfer and chamfer_sides blend every picked edge in one
SMLib call (smlib_blend / smlib_chamfer_sides), which builds the rounds, their tangent
chains and the corners where they meet. The #1080 fit check (blend_fit) reads the body's
schema 4 record: planar faces' outward normals, the direction into a face across a
straight edge, edge tangents at the ends. It now runs before SMLib, so a blend that runs
past its faces is TooLarge with the largest size that fits (fillet_max_each, the taper
and the doc's reduce path rely on that), and the volume and wrong-way checks after it.
fillet_max_each groups the edges by their faces and ends in the record. fillet_g2_corners
is refused: SMLib's corners are a sphere or a Coons patch, not G2.
The local truck blends, their reorder search and every truck helper they used are gone.
Tests: blend.rs's tests call fillet/chamfer instead of the deleted internal `blend`; the
assertions on the deleted quintic construction's surface and curve parameters and on
extension_fits (deleted) are removed, the analytic volumes kept.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Every builder in build.rs keeps its signature and is SMLib's: extrude and
extrude_whole_sides (smlib_extrude), revolve and revolve_whole (smlib_revolve;
SMLib makes one face per profile segment and cap, so the two agree), the
tapered extrude (smlib_extrude's draft), loft, loft_with_rails, loft_regions
(smlib_loft, smlib_sweep_stations for rails), sweep and sweep_with
(smlib_sweep / smlib_sweep_placed, controlled ones smlib_sweep_stations). Box,
cylinder and torus are built through those, keeping their face sources and
history; sphere is SMLib's own (smlib_make_sphere), each face from the half
disc's arc as before.
Deleted: every face-by-face build (tsweep/rsweep shells, the revolve-on-axis and
touching-axis builders, poles, reseams, the healed whole revolve, the six-patch
sphere and its rotated variant, the tapered side faces, cap_and_close,
close_sweep, the full-circle sweep), and the taper's crossing pre-check and curve
splitting that only that path used (offset_crosses, offset_splits,
offset_error, offset_tolerance). Wire, loft-face and interpolation helpers other
modules still build on stay until those are ported.
(not built: base in progress; cargo check shows no error in this file)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
sweep_path_with is SMLib's sweep along the path's lines (smlib_sweep_points
through sweep_with, corners mitred). Deleted: the B-spline walls fitted through
sampled stations (smooth_walls, smooth_closed_walls), the ruled walls between
stations (ruled), and the helpers only they used (rotated_wire, start_cos,
swept_volume). The open walls of chains (ruled_sheet, smooth_sheet) stay for
surfaces.rs's sheet sweeps until those are ported.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Brep::sweep_placed (smlib_sweep_placed, fork f6f428b) returns the rigid motions
carrying the profile onto the sweep's first, middle and last sections. Each run
of a path with corners now takes its end pose from the last motion (less the
straight lead-out along the end tangent) and its inside seeds from the middle
one, for every path segment kind: the 3D sketch spline next to a corner is no
longer refused. Deleted: the predicted carry along a run (carried) and
point_path::cubic_split, which only it used. SMLIB_REV at f6f428b.
Fork: build + ctest, only smlib_flat_roundtrip's known incline failure.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
convex_polyhedron builds its schema 4 record straight from the planes (corners
at triple-plane intersections, line edges, planar faces with exact line pcurves,
canonical order) and SMLib takes it (Brep::from_record, trusted) and checks it.
shell, draft and offset_faces are SMLib's local operations (smlib_shell,
smlib_draft, smlib_offset_faces through offset_backend).
Deleted: the half-space shell and draft of convex bodies, the planar-offset
cavity shells (barrel, opening overreach, opening sweep), the non-convex draft
walls, the split-piece press after a failed offset, round_all_edges and the
helpers only they used, and the test of the deleted rim pre-check
(rim_preflight_tests).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
sweep_helix is SMLib's screw sweep (smlib_screw_sweep) and loft_to_point SMLib's
loft to a point (smlib_loft_to_point), with their checks and history as before.
Deleted: the B-spline walls fitted through sampled stations (smooth_helix), the
ruled walls between stations and the face-by-face pyramid, and the Pappus
estimate only they used.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The case expected two bodies for a combine join of an arm 0.01 mm clear of a
ring. A join's tool that meets no target goes into the target as a separate
lump and the tool body is consumed (doc eval Operation::Join; the engine test
join_meeting_nothing_keeps_the_first_body): one body, Ring, of two shells,
with the volumes' sum. The case's point, nothing fuses, is kept: shells 2 and
the summed volume.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
ladder/05/011 expected 8 faces (the stock's 6 and one per hole). With equal
radii the two walls cross along two ellipses that meet at the two points
where the walls are tangent (x = 25 and 35, y = 20, z = 10), so each wall is
left in two pieces that touch only at those points, with no edge between
them: separate faces, 10 in all. SMLib makes those 10 (volume exact); truck's
8 came with an invalid record. Derivation in the case's method.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
With no targets picked, an extrude's Join, Cut or Intersect changed every body
of the active component it met, shown or hidden: a hidden coincident copy was
joined in and disappeared. Where some bodies of the active component are
hidden, the shown ones become the feature's participants, kept with it, so
showing or hiding bodies later doesn't change the model. Explicit targets and
participants, and a face's or a face sketch's own body, still win.
Test: the issue's plate with a hidden copy and a rim joined on.
Main reached the same fix for Cut and Intersect in #3961 (its targets-based
shown_scope in extrude); this one, which also covers Join and keeps the shown
bodies as participants, replaces it. Its test stays.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A closed path was always followed from the start of its first picked curve,
wherever the profile stood on it. A profile drawn at the middle of a line was
then carried round the path standing off it by the distance between the two
(walls offset, the rounds turned the wrong way, on both kernels). Now the
closed chain begins where the profile's plane crosses it nearest the profile,
the curve there split in two (or at the joint there); open paths and paths
that already start on the profile's plane are unchanged.
Test: a 4 x 3 profile at the middle of the bottom line of a rounded rectangle
swept round it: Pappus volume and the tube within 2 mm of the path.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Replaying a Mirror or Pattern feature asked it for its tools, which only primitives
and sketch features have: 'Mirror1 cannot be patterned or mirrored yet' (#3652). A
copy of a pattern or mirror of features now replays that feature's own features with
its moves followed by these (a mirror of Mirror1 reflects the hole Mirror1 made),
nested at most 8 deep, never through itself, and to at most MAX_INSTANCES copies in all.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Face capture estimated its work as every face times every triangle and refused a
body past 2M (#3914, a threaded block: 'face capture work limit exceeded'), though
only planar faces through the point scan the mesh. Each such scan is now counted
when made.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
solid.revolve and surface.revolve took any other axis string as a sketch line id,
so a construction axis by name failed with 'unknown sketch line' (#2516). A name
that is no line of the profile's sketch and names a construction axis is that
axis where it is now, as picking it in the view already passes it ({origin, dir}).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Three defects in restating trimmed surfaces (144/143) as STEP, kernel-agnostic:
- A boundary loop is one closed curve (IGES 5.3 composite curve: each piece ends
where the next starts), but each piece's ends were sewn into vertices by
distance alone, so pieces that miss by more than the sewing tolerance left the
loop open and the face was dropped (#3845). Surfaces are now read first and sewn
after: every meeting point of two pieces is one vertex (union of the two), the
STEP reader then carries the distance as the vertex's tolerance.
- With no parameter-space curve to give the start, a composite whose first piece is
written end to start kept that direction, so the second piece did not connect
(#3913). The first piece now runs from its end away from the second.
- All sewn surfaces were written as one shell, so two separate closed boxes came in
as one body of two shells that SMLib's check refuses (#3972). Each connected set
of faces is now its own shell and body (Body1, Body2, ...); a shell writes only
the vertices its edges use, in the sewer's order.
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
EvalSurface::wrap asked uv::closed whether a spline closes on itself every
time it evaluated a point outside the knot interval, and that test walks the
whole control net (its extent, then the first and last rows). A tessellation
of a closed spline unwraps its parameters across the seam, so about half its
evaluations are outside: on a coil's helical tube (thousands of control
points) the test was 80% of the tessellation's time, 30 s of the coil
command on SMLib (sweeps/3832, sweeps/793).
EvalSurface::Nurbs now carries `closed`, set by EvalSurface::nurbs from the
same test; wrap and uv::domain read it. An affine map of the surface keeps it
(closedness does not change under one). The answers are the same as before. The constructor's callers in
crates/kernel/src/tess (size.rs, surface.rs) change with it.
Not built: base in progress (rebased onto kernel-smlib/no-truck 41adb372).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The tests in tess/ build their bodies as before and plan crate::brep_record of them; the
assertions stay. Changed where what they compared is gone:
- surface.rs: each face evaluates through eval_surface and its record-parameter map as the
record's own evaluator (SurfaceEval) evaluates it, point and natural normal over the face's
box (was: as truck evaluates it); the kinds counted are the record's (planes and analytic
surfaces; SMLib's primitives have no spline faces, which truck's sphere was).
- edges.rs: edges are the record's; the intersection-node test becomes: every edge between two
curved faces puts each node, through each face's pcurve at the node's curve parameter, on that
face's surface at the node, within the edge's tolerance (was: Newton's per-node parameters
within 1e-7, which tess/ssi.rs no longer computes).
- tests.rs: "reference mesher" comparisons are against Body::mesh (the same mesher on the same
record, at a finer tolerance for the cross-hole volume). Removed the assertion that
seam::turns_more_than_once finds the thread's winding walls: that test of which mesher to use
no longer exists.
(not built: base in progress) tess/ has no errors under cargo check --tests.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Each edge node keeps its curve parameter; each face reads its nodes' parameters from its own
pcurve of the edge at that parameter, so nothing on the boundary is projected and a seam's two
uses (each with its pcurve) are its two sides: the cases the projection could not tell apart
("boundary winds around the surface", sides that cross where projection snapped them, poles).
The record's (u, v) go to the evaluator's in closed form (tess::surface::Params): the same for
planes, splines and revolutions; (θ, v) to (−v, θ) on a cylinder's or cone's generator; on a
sphere or torus (θ, φ) to (1 − s(φ), θ), s the rational meridian's parameter, from
tan(φ'/2) running linearly over each span (φ' from the span's middle).
A pcurve that does not cover its edge's interval, or places a node farther from the edge than
ten times the chord tolerance plus the edge's own tolerance, is an error naming the face and the
edge: the record is wrong there (for an analytic face, a trim left in SMLib's internal NURBS
parameters rather than the record's would show here).
(not built: base in progress) tess/ has no errors under cargo check.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
On kernel-smlib/no-truck: sweep_point_path_with goes straight to SMLib's sweep (no truck
branch), and PathSeg::flattened (truck's polyline of a path) takes no Cubic arm (the wildcard
keeps the segment).
(not built: base in progress)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A drawn 3D spline reached the kernel only as its sampled polyline (Wire.pts), so SMLib refused
to sweep along it ("not supported yet: SMLib sweep along a smooth path of points"). The doc
holds the spline's fit points; the curve is the centripetal Catmull-Rom spline that
fit_polyline samples. Each span of it is a cubic (Barry-Goldman), and its derivative at a fit
point depends only on that point, its neighbours and the two knot intervals, so neighbouring
spans share it: the spline is exactly one cubic Bézier per span, C1 across the spans.
- geom: catmull_rom holds the curve: samples() (fit_polyline's arithmetic, moved verbatim;
sketch::fit_polyline now calls it) and beziers(), the exact spans; ArcLength on CurveN
(Gauss-Legendre per knot span, Newton for the parameter at a length).
- kernel: PathSeg::Cubic, a cubic Bézier in space (the 3D counterpart of PathSeg::Curve, which
is planar), with its arms wherever PathSeg is matched (start/end, tangents, length, point at
a length, turn, SMLib curve as a degree-3 NURBS, halving). PointPath / PathPiece carry a path
through 3D sketch curves: polylines as points, drawn splines as their fit points (reversed
when the path runs them backwards); PointPath::segs gives SMLib the spans and the lines;
sweep_point_path_with sweeps it (SMLib through sweep_with, so corners, twist, scale, rails and
orientation take the existing routes). A smooth polyline piece (a projected or linked curve,
which the kernel only has as points) is still refused on SMLib.
- doc: path_points builds the PointPath (a wire with three or more fit points is a spline
piece); solid sweep and pipe call sweep_point_path_with; rails, surface sweeps and curve
networks take its points as before.
Checked in scratch (not committed): the spans pass through every fit point and every
fit_polyline sample at its parameter (1e-12), C1 at the joints and round a closed spline;
samples() is bit-identical to the old fit_polyline over 1180 random open and closed inputs;
SMLib sweeps of a radius-2 circle along an open spline and round a closed one match
pi r^2 L (L by independent quadrature of the curve's definition) to 1e-7 and 3e-8.
sketch/wire.rs: the existing closed-spline test reads SPAN_SAMPLES from its new home.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A Join with one target takes every tool, whichever it meets, but it grew the
target by one tool at a time: each union took the whole result so far, and
each step first meshed it to see whether the tool met it (which with one
target decides nothing). n tools cost n^2 in every per-body step of a
boolean (converting the operands, checking the result, exporting it back),
whichever kernel runs it: a combine of 64 separate pegs (booleans/3455) took
400 s on SMLib, of 36 lofted blades and a disk (crashes/2992, sweeps/1856)
485 s and 97 s.
With one target, the target and the tools are now unioned pairwise, level by
level, so each body takes part in about log2(n) unions. The result is the
same union under the target's name, with the lineage of every union. Several
targets keep the absorption order they need. The doc and engine join and
combine tests passed and failed on SMLib as before (measured before the
rebase).
Not built: base in progress (rebased onto kernel-smlib/no-truck 41adb372).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
uv::project, given no hint that converges within `accept`, evaluated a grid
of up to 16384 points of the whole surface for every point it placed, then
sorted all of them by distance. The grid is the same for every point of a
surface, and a face's pcurves place hundreds of points on one surface: on a
modelled thread's helical faces that grid was most of a boolean's time on
SMLib, in the operands' records, SMLib's export of each body and the
result's (threads/2526, the ladder's threads).
uv::Starts keeps the grid, filled on first use; uv::project_with takes it,
and a face's pcurve fit (pcurve::On) keeps one per face. The 32 nearest
starts are picked with a partial select instead of a full sort, in the same
order (nearest first, grid order where equally near), so every point gets
the parameters it got before. (The tessellation's boundary projection can
take the same grid; crates/kernel/src/tess is the mesh agent's.)
Not built: base in progress (rebased onto kernel-smlib/no-truck 41adb372).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
On a B-spline face the export now takes each edgeuse's UV trim exactly as
SMLib holds it (EDGEUSE_NURBS_*): the face's parameters are SMLib's, so the
trim is the pcurve, reversed exactly (knots t -> a + b - t) where the
edgeuse runs against its edge, its knots mapped onto the edge record's
parameter where that is an affine image of SMLib's (identity for spline
edges, as_line's map for NURBS lines, the C side's map for lines), and its
end knots set to the edge interval where they differ by rounding (within
the schema's 1e-9 slack, else an error). Changed: export.rs `curves`
(per-edge TrimMap), new `trims`, `mapped`, `Face::given`,
`Edge::trim_map`, the face loop in `native_ordered`, `pcurves`,
`outer_area`.
Analytic faces and circle/ellipse edges keep brep::pcurve: SMLib keeps an
analytic face's trims on its internal rational B-spline (SmCylinder.h:
"the map from one to the other is not linear") and a circle's on its NURBS
parameter, neither an exact B-spline in the record's angle parameters.
brep::pcurve::EdgeUse gains `given`, a producer's exact pcurve taken as it
is (structure and interval checked, moved by whole periods with its loop,
never refitted). Existing construction sites (STEP reader, tests, truck
adapter) pass `given: None`; the existing tests change only for the new
field.
Built, clippy-clean and tested on 4b2c0eaf before the rebase onto
kernel-smlib/no-truck (not built: base in progress).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
tess::plan takes a brep::Brep: faces, loops, shells' face sides and edges in the record's
order. Surfaces go to the evaluator's form from the record (planes, splines and revolutions
through SurfaceEval; a cylinder, cone, sphere or torus as the exact revolution of its line or
rational-arc generator about its axis). Edges are sampled on the record's own curve evaluator
(CurveEval), an intersection curve included, so tess/ssi.rs and the per-node parameters it
carried are gone; every boundary node's parameters come from projection.
No fallback is left in the planner's path: a curve that does not evaluate on its interval is
an error naming the edge (it was a two-point chord, or truck's evaluation); a surface the
evaluator cannot express an error naming the face and its kind. Errors name the face, its
surface kind and, for a node off the surface, the loop, node and point. The module doc states
the order: a cheaper layout where the face suits one, else constrained Delaunay in (u, v),
else an error.
MAX_ANGLE_BOUNDARY and MAX_RELAX are tess::size's own. Removed with truck: the winding test
and pole-to-pole seam test that chose between this planner and truck's mesher, and
boundaries_on_surfaces (truck blends). The tests in tess/ still build truck bodies and are
left for the record-based Body.
(not built: base in progress) tess/ has no errors under cargo check; the rest of the crate is
the no-truck branch's.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Body::mesh / mesh_plan / mesh_of and MeshTolerance (chord, display, display_of, export_of)
replace tessellate, tessellate_with, display_mesh*, display_plan*, reference_display_mesh_angular;
every caller in kernel, doc, engine, io, ui-egui, gpu and the CLI renamed mechanically; the
truck-mesher fallback, mesh_shells*, refined* removed; Body::new orients from the planner;
truck-meshalgo dropped from Cargo.toml with the tests of its hooks. Still to fix: trait imports
in tess/*, splitface/lumps/preview/step_in/boolean_record meshalgo uses, measure.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Over the fork's new smlib_brep_bounds (SmBrep::CalculateTightBoundingBox): the box of a body
from its exact curves and surfaces, for measuring without a mesh on the SMLib kernel.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Through SMLib, in place of "not supported yet":
- A sweep along a path with corners: each smooth run swept on its own and
the runs joined at each corner in a mitre (each run swept on straight past
the corner and cut back to the plane halving it, where the next meets it
face to face), including the corner where a closed path closes; only
moving along the path, the runs meet on the profile. History: walls from
their segments (found half-way along each run), caps where the path ends.
- A free-form path curve (a spline's Bézier pieces, conics) reaches the
kernel as PathSeg::Curve, its exact piece, instead of the sketch's
polyline of lines: SMLib sweeps along the curve itself; truck's sweeps
flatten it as before (now the piece's own polyline within 0.01 mm, which
turns ladder 04/004 into a pass for truck too). Patterns along a path
flatten it and measure as before. A sweep that twists, scales or follows
rails places its stations by length along the curve.
(Such a controlled sweep still refuses a path with corners.)
- A loft through sections with holes: each section's outline and holes are
one SMLib profile, the holes matched as for truck.
- A loft to a point: SMLib's pyramid (smlib_loft_to_point).
- A draft extrusion with a straight foot: the foot's prism and the drafted
part (sloping from the foot's top to the far end the taper gives) joined
face to face; each segment names its foot and its sloping wall.
Cases now passing on SMLib: booleans/3633, sweeps/1184, 2435, 3111, 3006,
3505, 561, 96, ladder 04/003 and 004.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The fork's new smlib_loft_to_point (a pyramid from a planar section of lines
to an apex) as Brep::loft_to_point. The pin brings the sweeps' other
changes: a loft's section may hold holes (loops lofted loop to loop), a
sweep along a straight path is an extrusion and one turning with arcs about
one axis a revolution (exact walls), and a sweep round a closed path closes
its walls on each other, without caps.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The case's own method says one body with an inner shell (the void of
radius 18.5 inside the ball of radius 40), but without a "shells" entry
the rig asks one shell per body. It now asks two, as the method derives.
SMLib still fails the case on its volume: 294558 mm3, near the sum of
the two balls (294604) rather than their difference (241561), so the
inner shell is not taken as a void.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Two SMLib fixes (tangency-labs/smlib r-shell-fillet):
- b0e84f4: faces an extended offset already mated are not intersected
again in the offset's merge (rig shell/3103, a two-half cone frustum
shelled open at the top, now one valid solid at the closed-form
volume).
- ffaf4e8: an intersection edge's end goes to its own vertex when its
ProtoVertex resolved into several (rig fillets/1488, the slab joined
with a nudged pleated loft, valid again; booleans/2801 passes too).
Every rig case that passed on SMLib at 2d7c7ed still passes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
On SMLib a sweep with a twist, an end scale, an orientation direction
or guide rails, and a loft with guide rails, were refused. They now
place the profile at stations by the laws the kernel already defines
for them (sweep_path::controlled_stations / railed_stations, and the
loft_rails law: plain loft between the sections moved onto the rails),
each section the profile's exact image, and SMLib skins the walls
through them (Brep::sweep_stations, Brep::loft). The path is sampled
exactly on its lines and arcs, a station per degree of turn: SMLib's
skin interpolates with free ends, its error falling as the square of
the step (a quarter-twisted square is 2e-7 off in volume at 1.4
degrees, 2e-6 at 2.8). Round a closed path the last section has to
land on the first (closing_offset, as for truck) and the ends are
joined. A railed loft needs each section's curves to pair with the
others' (same kind), else "not supported yet".
A path of points (3D sketch curves) was refused outright on SMLib; a
line or a polyline with corners is now swept as its lines (the
corner refusal still applies), and only a smooth polyline of more
points (a drawn 3D spline) stays "not supported yet".
History in the builders' vocabulary: caps from the start and end
section planes, walls from the profile segments (round a closed path,
from points inside the walls at the middle station).
Kernel cases now passing on SMLib: sweeps/1269-sweep-twisted-square-
ninety-degrees, 2005-closed-sweep-twisted-cut-inside-a-torus,
2005-closed-sweep-twisted-new-body, 201-loft-one-guide-rail-cone-
frustum, 2862-three-pipes-meeting-at-one-point, 2891-twisted-sweep-
cut-through-a-tube, 771-sweep-end-scale-cone-frustum, 771-sweep-one-
and -two-guide-rail(s)-cone-frustum.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The wrapper of the fork's new shim (61ee7f5): a planar profile placed
by an affine map at each station, the walls skinned through the
sections, the ends capped, open, or joined round a closed path
(Ends). Tests: a twisted square (V = A L), a scaled disc (frustum), a
square carried round a circle with its ends joined (Pappus), and
singular or too few stations refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
SMLIB_REV moves from 276b509 to capi-solidcraft 2d7c7ed: a fillet crash at an
unsupported corner (#453), rounds ending on a tangent seam (#418), radius laws
with a corner (#430) or that cannot be evaluated (#458), a round over a hole
(#422), a sphere's own patches kept apart (#423), offset surfaces extended the
same way whatever was evaluated first (#419), swept walls sharing their planes
(#434), the internal thread's last vertex gap (#410), shells across rounds
thinner than the wall (#397), nested shells imported in any order (#388) and
the tessellator held to its chord tolerance (#390).
Three tests follow: a radius law with a corner is now built, so the SMLib
tests check its volume (8000 - (1 - pi/4) 20 28/3) and that a law pinching to
zero inside the edge is refused; the split coil's halves now measure within
1e-5 of the closed form, so they are held to it rather than to the whole
coil's mesh, which is only within 3e-3.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
The kernel accepts `truck` (as well as unset) for its truck backend, and
`boolean_kernel()` reports that backend as `default`. The rig compared
the raw setting with that name, so a run with SOLIDCRAFT_KERNEL=truck
reported every case as "this build runs booleans on the default kernel,
not truck" instead of grading it. The rig now names unset, empty and
`truck` alike as `default`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A face's grid lines are thinned to MAX_FACE_GRID points (and now to at most that, where the
rounding of thin() could leave it a little over), but the cut grid adds the face's motorcycle
traces to them and then refused when its cells came out over the cap. The face fell through to
Delaunay, whose refinement budget a face of that size is far past: a split coil's wall at the
measuring tolerance got 90 000 triangles where it needs 320 000, and the halves of #793's coil
measured 1.5% short of the whole. The cut grid now thins its full lines (the same evenly spaced
choice as the face's grid, moved into grid.rs) until the cells fit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
SMLib writes a modelled thread's groove walls as splines winding once per pitch, and splits the
bore or shaft into one face per turn, so no face of revolution has a loop turning more than once
and turns_more_than_once missed the body. Truck's mesher then took the winding walls: meshes
reaching 1.4 mm past the block (#1995, #2526, #3311, ladder 06/004), volumes of -1.8e7 and boxes
120 m wide (#2038 join, ladder 06/003), or spade's "!previous_region.is_empty()" assertion inside
its constrained triangulation (#3952, ladder 06/001, 06/002). A spline face now has an axis too:
the mean binormal of an iso-line that winds more than once about it, through the iso-line's
centre; a boundary loop turning more than once about that axis routes the body to the display
plan, as a helical trim on a cylinder already did.
Truck's own coils (sweep_helix's spline walls) go the same way; their walls are longer than a
grid's 1024 lines resolve at the measuring tolerance (three turns of a 20 mm coil need ~1200),
so a direction may now have 8192, the grid as a whole still capped at MAX_FACE_GRID.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A thread's root band that runs out through an end face (a strip a quarter pitch wide winding
ten turns, ending obliquely, so it has no four corners to blend) fell through to the butterfly
of a round face. Its cuts ran lengthwise through the strip, seeded the whole way: ~200 000
points whose triangles zigzagged along the band, six times its area and up to 1.6x the chord
tolerance off the surface. A curved face whose division across one parameter has no inner line
(straight and within a side there) and with no room for a lattice seed now skips the patch and
cut-grid layouts and goes straight to Delaunay of its boundary nodes with sag refinement:
11 600 triangles for the whole part instead of 438 000, within the tolerance, the triangles'
area matching the surface's. A fillet narrower than a side but curved across keeps its layouts
(Delaunay from its boundary alone runs out of refinement budget at fine tolerances).
cdt::has_room tells "no seed fits" apart from "the lattice is too large to look through".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Each loop of a face is projected on its own, so along a closed direction a
hole can come out a whole period away from the outer loop. A revolved
dome (rational spline closed in v, bounded by its rim, its seam both ways
and the pole) has its outer loop opened at the pole into v in [-1, 0],
while a hole cut by a ball next to the seam projected into [0, 1]: the
triangulation covered the hole as well, and the cut measured 48118.36 mm³
instead of SMLib's 46424.50 (tangency-labs/solidcore#433, kernel case
sweeps/3528-revolved-dome-cut-by-a-ball-on-its-seam). Every other loop is
now moved by whole periods to the copy nearest the outer loop.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
On a thread's root band (1.4 mm tall at a pitch of 1.5) a point on the band's lower rim lies
0.1 mm from the rim of the turn below. The search's starts are dense across the band and
sparse along it, so its four nearest starts all lay on the turn below, Gauss-Newton settled
there, and the pcurve fit refused the edge ("the curve at 0 lies 0.1008 mm off the surface",
ladder/06-threads/007). The search now goes on through the nearest 32 starts until one
converges within the accepted distance, and stops at the first that does.
The test strips now put their knots at other angles on each turn (7.3 control points a
turn), as SMLib's screw sweeps do.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
SMLib's intersection curves carry knots of full multiplicity where the curve is in fact
differentiable (its derivative the same either side). The pcurve fit took every such knot
for a corner and fitted the two sides apart; on a thread's flank, whose u runs hundreds of
times as fast as its v, the pieces met at up to 1.4 deg in (u, v), and SMLib's check refused
the record read back (Test #7 Bad CurveKink, offset 0.9997 against cos 1 deg).
A knot that leaves a spline only continuous by its multiplicity now gets a cubic with a
tangent there when the spline's one-sided derivatives (EvalCurve::sides) agree within 1e-9
of their size. Knots where the curve does turn are corners as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
A thread cut by SMLib's screw sweep leaves flanks that are cubic splines of hundreds of knot
spans along the helix (an M6 through 20 mm: about twenty turns over u in [0, 1]). With no
hint (an edge's first sample), the projection started from a 17 x 17 grid over the parameter
box: one start every turn and a quarter, so the nearest starts lay on the turns beside the
point's own, Gauss-Newton settled there, and the pcurve fit gave up with "the curve at 0
lies 0.87 mm off the surface" (pitch x cos 30 deg, the next turn's flank). Every modelled
thread failed its export that way.
The search now starts, along each direction where the surface has its own knots (a
spline's, a revolution's profile's), from each knot span's start and two points inside it as
well as the 16 even steps, capped at 16384 points of the surface, so every turn is searched.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
Without the native feature nothing reads a sweep's curve list, and clippy -D warnings
refused its two fields as dead code. The list is built through List::new in both builds;
only the native build keeps the slices the C API reads.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
SMLIB_REV moves from e1ac3c3 to capi-solidcraft 276b509: thread groove
booleans (#435, #410), a holed profile swept along a bent path (#411), a
revolution a turn past its range (#428), shell and offset faces (#402, #380,
#403), the check linear in the body (#431), a far box that leaked (#377),
fillets and chamfers where a neighbouring round collapses (#416, #417, #420,
#379), tangent faces round a non-manifold edge (#383) and split faces of an
exact NURBS surface (#387).
The last one is why the SMLib block-and-cylinder decode test no longer finds a
spline edge on part of its curve: SMLib keeps a split face's parameterization
now. The test keeps its volume check; the decode of such edges stays covered
by spline_edges_on_part_of_their_curve_load_as_that_piece.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
An SMLib result where one shell uses an edge four or more times (a hole tangent to the outline,
touching holes, stepped cuts touching along a line) was refused as "not supported yet: a solid
touching itself along an edge". The export now reads SMLib's radial order round each edge
(EDGEUSE_NEXT_RADIAL, EDGEUSE_RADIAL_ENTRY) and the side of each face its material is on, and
gives each pair of faces across one sector its own copy of the edge: the pairs across material,
so the surface is cut apart round each wedge of material, unless one face would get two copies
(a face on both sides of the line), which SMLib's check refuses as coincident edges (tests 33,
34); then the pairs across the empty sectors, so that face keeps one edge on both sides, as
SMLib's crack edges do. The vertices at the ends stay shared: copies would give a face passing
the point twice two coincident vertices, which SMLib refuses too (test 32). Geometry unchanged;
each copy keeps its edge's tolerance and history (Order.pieces). The radial arrays are checked
as hostile input (in range, cycles on their own edge).
Tests: a hole tangent to a block's outline exports with the line copied once per wedge (each
copy joining faces on one side of it), every edge two uses, and goes back into SMLib valid at
the same volume; a blind hole tangent to a side leaves the side a crack; through SMLib both come
back as one solid with verified history and empty validity.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
After a parameter edit, a feature's stored face point is moved onto the
face its name finds. That used the face's nearest point to the old pick,
which lies on the face's boundary whenever the pick no longer lies on the
face: a shell opened at the end of a bar picked (150,15,10), on the edge
it shares with the bottom face, so truck opened the wrong face and SMLib
refused the pick (#2531).
FaceHit::pick_point keeps the point where it still lies on the face clear
of its edges, and otherwise takes the face's nearest point among points at
least a margin inside its triangles (a few pick tolerances, capped at half
the roomiest triangle's inradius on narrow faces).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017itGh5fpMEousnZ91cYcVt
With SOLIDCRAFT_KERNEL=smlib, kernel::draft hands the picked faces (a point inside each,
from the face under each pick) to Brep::draft, with SMLib's own history: the tilted faces are
the faces themselves (Modified). Faces SMLib can't draft (a round wall, a face not parallel to
the pull) are "not supported yet", never truck. Tested in modify_history_tests: a 40 x 30 x 20
block's side at 5 degrees removes 20 x 20 tan 5 deg / 2 x 30, with its history checked against
the geometry.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Flat faces tilted by an angle about the line where each meets a neutral plane, leaning in along
the pull, and the faces around them extended or trimmed to meet them
(tangency-labs/smlib@e1ac3c3, SMLib's local operation). A face that isn't flat or isn't parallel
to the pull is Error::Unsupported. The test drafts a cube side by +-5 degrees, removing or adding
the analytic wedge to 1e-6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
cargo xtask kernel-cases --kernel both, solidcraft-cli built with smlib
at c65fc91a (fork bf7c272), SMLib through SOLIDCRAFT_KERNEL=smlib. 372
cases: default 235 pass, smlib 269. SMLib's gains are booleans (50 to
66), fillets (32 to 41), shell (5 to 11) and STEP import (13 to 17, its
solids now SMLib bodies); its losses are sweeps (54 to 51 pass, 15 not
supported yet on SMLib) and threads (17 fail, of which the default kernel
hangs on 4).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face on a B-spline surface whose first and last control columns are the
same (closed around u), bounded by loops that go once around without a seam
edge, was refused ("a loop across a closed spline surface's seam"). It now
gets a seam as cylinders and cones do, along the surface's own boundary
column (u at the start of its knots, an exact iso-curve of the net), each
loop split where it crosses that column when it has no vertex on it; the
period is the knots' width. At a pole a loop may jump across u, where the
seam's two uses meet a period apart.
The step-conformance NURBS cylinder, dome and torus ring now read. With
SOLIDCRAFT_KERNEL=smlib all 17 parts import into SMLib with their exact
volumes and face counts. A spline closed in v only, or both ways, is still
refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A gap the boolean fixed on one side only left the freed edge in its common-edge map, and the next
pass read it: a square-wire coil cut into the wall of a bore crashed the process on SMLib
(sweeps/2552). It now fails with an error.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib leaves a corner patch where blends meet untagged, and blends() matches no single edge to
it, so it had no origin, and the edges between two blend faces (two chamfers' mitre) none
either. The history checks refused both (modify_history_tests' corner, whole-box and mitre
cases on SMLib). Now a patch without origin is made in place of all the edges of the blend
faces beside it, as truck records a corner, and every edge of a face new to the blend is where
its two faces meet.
modify_history_tests with SOLIDCRAFT_KERNEL=smlib: the three blend cases pass (28 of 30; the
painted shell and the painted box cut's pieces are the shell's and boolean's). Truck: 30 of 30.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With SOLIDCRAFT_KERNEL=smlib, chamfer_sides (two distances, or a distance and an angle) runs
SMLib's new unequal chamfer instead of being refused: `d1` back along the face facing up most
(down most with `flip`, as truck's chamfer_sides picks its first face), the other side by the
distance or angle, with the body's history as for the other blends. A radius law SMLib refuses
as having a corner is "not supported yet", never truck's round.
Tests (with the feature): 5 and 2 on a cube edge either way round (the top face keeps 15 or 18
of its 20, the chamfer face comes from the edge), 5 at 45 degrees, and the corner law refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Brep::chamfer_sides reaches the fork's smlib_chamfer_sides: `distance` back along the face
whose outward normal leans most along a given direction, the other side by a second distance
(ChamferSecond::Distance) or at an angle to the first face (ChamferSecond::Angle). The fork's
smlib_fillet_variable refuses a radius law with a corner with SMLIB_ERR_UNSUPPORTED, which the
crate reads as Error::Unsupported ("not supported yet"): SMLib's section fit ran smoothly
across the corner and missed the law (4 -> 2 -> 4 on a 20 mm edge took 35.1 mm3 for 40.1;
tangency-labs/solidcore#430). Both are in the fork at the pinned bf7c272.
Tests (ops_tests.rs), against analytic values: 5 and 2 either way round (the centroid tells
which face took the 5), 5 at 30 degrees, rejected arguments; a quadratic law 4, 1, 4 and a law
down to zero (1e-5), the corner law refused and the same points joined C1 built.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With SOLIDCRAFT_KERNEL=smlib, kernel::measure and measure_mass take volume, area and centroid
from SMLib's Brep::mass_properties (integrated on the exact surfaces, accuracy 1e-8). The mesh
measure is still made, as an independent check. Where SMLib disagrees with it beyond the mesh's
own accuracy (1e-3; tangency-labs/solidcore#384), or can't take the body, the mesh measure
stands and the measurement says why. BodyMeasure and MassMeasure carry `measured_by` (smlib |
mesh) and `notes`, and inspect.measure shows both. The box and topology counts stay the mesh's.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib's tolerance update now fixes the vertex gaps its check reports (gaps over half a vertex's
tolerance), and every boolean, blend, shell, offset and sweep result grows its vertex and edge
tolerances over gaps below SMLib's own rounding bound before it is handed out. A thread's groove
cut from a shaft no longer fails SMLib's check on vertex gaps of about 1e-5 mm, and an offset
sphere face no longer fails it on an edge gap of 5.015e-6 against 5e-6 (shell/2939).
Also carries the fork's blend commits since ca2559f (second offset per edge, two-setback
chamfers, variable radius laws with a corner refused) and the IGES knot check.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Where two edge uses of a loop meet at a pole (meridians of a sphere), the periodic parameter's
jump there is free, and the pcurves took whichever jump made the loop counter-clockwise. Two
meridians a three-quarter turn apart bound a quarter of the sphere or three quarters of it, and
the loop alone can't tell which: a half disc revolved 270 degrees on SMLib came back as a quarter
ball turned inside out (sweeps/2667, V = -226 against 678.6).
pcurve::pcurves_within takes the face's parameter box, as SMLib gives it, and picks the jumps
whose extent in that parameter is nearest the box's; pcurves is unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each solid item of a STEP file goes through the schema 4 reader
(solidcraft_step::topology, at the file's units and uncertainty) into
SMLib through the flat converter, healed where SMLib's check refuses it,
and comes back as a kernel body; its faces keep the file faces they came
from. A solid the reader or SMLib refuses is not imported, with the
reason: there is no truck fallback on this path. truck's own reader is
unchanged and stays the default.
Test: the 17 step-conformance parts read into SMLib with their exact
volumes and face counts, every face traced to an ADVANCED_FACE; the three
NURBS parts stay refused by the reader (a loop across a closed spline
surface's seam), as on schema 4 today.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
smlib_record_ordered is smlib_body_ordered for a record of any origin (a
STEP file's, not only a body's): SMLib's body of it, healed where SMLib's
check refuses it as given, with where each of SMLib's faces and edges is in
the record's order. A record SMLib refuses outright now goes through the
healer too before the import gives up.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Schema 4 lets a circle or ellipse edge run up to one turn past its seam
(a quarter arc after the seam is [2π, 2.5π]); SMLib's circles take angles
within [-2π, 2π], and the flat import refused the whole body as invalid
input. The mapping now gives SMLib the same arc a whole turn back, and moves
the edge's pcurves on B-spline faces by the same amount, since their
parameter is the edge's. step-conformance's plate_100_holes now imports.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Solid::face_origin lists the ADVANCED_FACE or FACE_SURFACE entity of each
of the record's faces, in the record's face order, so a body made from the
record (in another kernel) can trace its faces to the file, as the truck
import's face_origin does.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Imprint cuts each face of a body where a tool body meets it (the fork's
smlib_imprint, SMLib's imprinting boolean) and gives None where the tool
crosses no face. Trim builds on it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
169 rig and ladder cases on both kernels, graded against each case's own answer: where SMLib
fails, grouped by what goes wrong (wrong geometry, a crash, the thread groove's boolean failing
SMLib's check, shell refusals, sweeps with no SMLib path yet).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SolidCraft's own record of a body SMLib made (the rig's record check, the saved body, the next
SMLib import) re-fitted every pcurve at the kernel's working tolerance, 1e-5 mm. SMLib's fillet
and chamfer surfaces lie some 2e-5 to 2e-4 mm off their faces, so 15 fillet and chamfer cases
failed that check although SMLib's own passed. And SMLib reported one tolerance for a whole body,
the largest any part of it needed, so one loose edge (a pocket cut left one at 0.23 mm) made the
whole body loose on its way back into SMLib.
Now the tolerances go per edge and per vertex, both ways:
- solidcraft-smlib reads and writes the fork's new SMLIB_FLAT_EDGE_TOLERANCE and
VERTEX_TOLERANCE arrays. Export gives each edge and vertex record its own (never below the
floor, which becomes the body's). Where those understate what an approximated edge needs on
its face (SMLib holds a blend's end curve 2.3e-5 mm off its plane at 2e-5), the export falls
back to SMLib's largest everywhere, as before. Import writes each record edge's and vertex's
tolerance.
- `geom::brep::pcurve::EdgeUse` carries its edge record's tolerance, and a pcurve is fitted to
it where it is larger than the face's.
- The kernel's decoded body keeps its record's edge and vertex tolerances
(`Body::tolerances`, in its edge and vertex order) after an SMLib operation and a saved body's
decode, and keeps them through a rigid placement. Its record writes them again; anything that
builds a new solid drops them.
SMLIB_REV moves to the fork commit that adds the arrays (smlib-blends-abi2 4a67471, ABI 2).
Rig and ladder fillet and chamfer cases on SMLib: 43 pass before, 60 now, none lost. Tests: a
pcurve fitted to its use's own tolerance, and the ladder's 08/006 cylinder rim chamfer whose
record, saved body and placed copy all stay valid and SMLib's check takes back.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A pick at the vertex where a cylinder's seam meets its rim lies on both edges. The body's
nearest-edge order took the seam on SMLib-made bodies, and SMLib refused to chamfer it
(the ladder's part chamfer, step 22, SM_ERR_INVALID_INPUT). A seam or a tangent join is never a
blend's edge, so a pick as near to one as to a sharp edge takes the sharp edge; anywhere else the
nearest edge stands.
Test: every pick round a cylinder's top rim, the seam's end among them, takes the rim.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Every blend entry point asks `backend()` and, on SMLib, hands its arguments to blend_smlib.rs:
fillet, fillet_each and fillet_radii (a radius per edge), fillet_g2 (SMLib's G2 blend-curve
section between the round's contacts), fillet_chord (constant-width section), chamfer, and
fillet_variable_spline (the radius law as an SMLib B-spline law; fillet_variable and
fillet_variable_points reach it, and fillet_max, fillet_max_each and fillet_tapered run on these).
Options SMLib has no construction for are refused as "not supported yet", never built by truck:
G2 corner patches, and chamfers with two different setbacks (distance and distance, or distance
and angle; equal setbacks are the equal-distance chamfer).
Each pick takes the body's nearest edge in its own edge order, as truck's blends do, and SMLib is
given that edge's midpoint, so a pick at a vertex never leaves SMLib to choose between edges. The
body goes in tagged as operand 0, SMLib's check must pass, and the decoded result carries SMLib's
history in the body's orders, the blend faces given FaceOrigin::Edge of the edge each replaced
(the picked edges and the rest of their tangent chains). The truck path is unchanged.
Tests (with the feature): a round's volume and its face's origin, a size per edge for rounds and
chamfers, a linear law from either end, refused sizes and picks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The Rust side of the fork's smlib_blend and smlib_fillet_variable, which the kernel's fillets and
chamfers need on SMLib: `Brep::blend` takes a size per picked edge and a `Blend` (circular round,
curvature-continuous round between the round's contacts, constant-chord round, equal-distance
chamfer); `Brep::fillet_variable` rounds one edge with a clamped B-spline radius law from the end
nearest a start point. Selection errors map as for the other picking operations.
SMLIB_REV moves to the fork commit that adds them (smlib-blends, on capi-solidcraft ffcf56d).
Tests: per-edge rounds and chamfers and the chord round against analytic volumes, the linear law
from either end, and refused laws and picks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The fork's booleans now hand out closed non-manifold results (SMLIB_REV e87e9b3, ABI 3): solids
touching each other along an edge or at a vertex, which the C API refused as "not a closed
solid". The SMLib → schema 4 mapping now gives each closed boundary its own record shell: the
faces sharing one runtime shell on the empty side and one on the material side, so solids that
share their empty side's runtime shell get a shell each. An edge or vertex faces of more than
one shell use is split into one copy per shell (same curve, interval and point), so every
shell is a closed manifold boundary; regions list each shell as before. The history follows: an
Order's `pieces` give each record edge of a split SMLib edge with the faces it bounds, and each
piece keeps the edge's origins.
SMLib's own check refuses the radial order of faces round an edge solids share, so a
non-manifold result is checked as the separate solids it becomes (imported as given). A solid
touching itself along an edge (one shell using an edge more than twice) is still refused, now
as "not supported yet: a solid touching itself along an edge".
Tests: cubes sharing an edge or a vertex unite into two shells of 1000 mm³ each, every face and
edge's history verified against the geometry; two opposite quarters cut off a slab leave two
solids. Of the pinch tests run through SMLib, cut_leaving_pieces_meeting_along_a_line now
passes; the other nine are solids touching themselves.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The kernel runs exact helical sweeps and never resizes; the check that
refuses a thread off its hole's or shaft's size is the engine's sugar
(AGENTS.md). So the seven cases (995, 1443, 2152, 2969, 3366, 2369,
2705) go back to expecting that refusal and its message, as 1375e905 had
them, reversing 1588dd70.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
On SOLIDCRAFT_KERNEL=smlib the thread's groove (the ISO 68-1 trapezoid, as before) is a B-rep
screw sweep (kernel::sweep_helix, SMLib's screw sweep) cut from the body by the B-rep boolean,
whose history carries the threaded face's names on. Truck keeps the faceted groove and the
faceted boolean. thread_groove_shape holds what both build from; the size checks before it
(a hole or shaft off the thread's size is refused with the size to use) are unchanged, so
neither kernel fits a thread to a face that is too large or too small.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
With SOLIDCRAFT_KERNEL=smlib, extrude (and extrude_whole_sides, so boxes and cylinders too),
revolve and revolve_whole, extrude_tapered(_between), sweep and sweep_with, loft,
loft_with_rails, loft_regions and sweep_helix (coils, and the threads' grooves) build their
bodies with the fork's sweeps (smlib_sweeps.rs): the profile's segments go to SMLib as curves
(lines, arcs, Béziers and conics as NURBS), the result comes back through its schema 4 record,
and one SMLib's own check refuses is an error. What SMLib has no sweep for answers "not
supported yet", never truck's sweep: twists, end scales, guide rails and orientation
directions; paths with corners and paths of points; lofts with holes or to a point; draft
extrusions of profiles with holes or with a straight foot.
History is in the builders' vocabulary, found from the result's geometry as provenance.rs does
for operations without exact history: a wall comes from the segments its start edge runs along
(and, for extrusions and revolutions, whose middles it holds half-way), the face in the start
plane where the profile bounds it is the start cap, the one at the end the end cap. Straight
extrusions also carry their faces' construction sources (ExtrudeCap, ExtrudeSide), which the
box and cylinder certificates need. The loft's section matching is truck's own, now
loft_numbered_loops (split out of loft_numbered_wires, unchanged), so both kernels pair the
same vertices.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Brep::extrude (straight or drafted), Brep::revolve, Brep::sweep (turning with the path or only
moving along it), Brep::loft and Brep::screw_sweep wrap the fork's new smlib_sweeps.cpp shims.
Profiles, paths and sections are lists of sweeps::Curve (lines, circular arcs, NURBS), turned
into the C API's curve lists here; the FFI is ffi/sweeps.rs. SMLIB_REV moves to the fork head
that has the shims (c33df0f, on capi-solidcraft 1499681).
Tests check each sweep against closed-form volumes and SMLib's check, and hostile curve lists.
(The crate's test build currently stops at import/tests.rs, which calls import::surface without
its new analytic_planes argument; that is the import's, not this change's.)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The SMLib shell passed the user's pick points to SMLib, which refuses a point as near to two
faces; a pick on a hole's wall at the line between its two halves (kernel-cases shell/4010)
failed. The open faces are now picked as truck's shell picks them (the face of the nearest
triangle), grown by the faces on the same surface joined to them (offset::with_same_surface:
a round wall stored in pieces opens whole), and SMLib gets one point inside each.
Offset faces keeps its design on both kernels: when moving the faces fails on a body with
split lines, the chosen pieces are pressed (swept and joined or cut) instead, as truck's path
always did; the booleans of the press run on the selected kernel.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
smlib_body_ordered keeps SMLib's face and edge order for history only where SMLib's export
numbers the body as its record does, and compared each loop edge for edge in the same order
and direction. SMLib walks a face whose surface it turned (a hole's wall, a concave cylinder)
the other way round and may start the loop elsewhere, so a plate with a hole lost its history
through every SMLib boolean and shell although its faces and edges are numbered alike. A loop
now matches when it walks the same edges in the same cyclic order, or in the opposite order
with every edge reversed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
With SOLIDCRAFT_KERNEL=smlib (the one kernel switch), shell, offset_faces and thicken send
the body to SMLib tagged as operand 0 (offset_backend.rs) and bring the result back with
SMLib's own history, as booleans do. A result SMLib's check refuses is an error.
SMLib's tags travel through the copies its offset makes, so a shell's inner wall and a
thickened sheet's moved copy read as pieces (Split) of the face they came from. A result face
lying more than half the operation's distance from its tagged face is recorded as that face's
offset (FaceOrigin::Offset) and its edges as where its faces meet; a thickened sheet's walls,
new to SMLib, are recorded as made in place of the sheet edge their edges are tagged with, as
truck's thicken records them. Offset faces pick the faces truck's path picks (a round wall's
pieces together) and pass SMLib one point inside each; the moved faces are Modified.
The histories are checked against the geometry as truck's are (modify_history_tests), with
the analytic volumes, the thicken direction, and the refusal of walls past the L-shaped body's
inradius (8.79) that its box (half extent 10) let through before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Brep::offset_faces and Brep::thicken reach the fork's new smlib_offset_faces and
smlib_thicken; SMLIB_REV moves to the fork commit that adds them. The shim now refuses an
inward wall that leaves no room inside the body with its own status, Error::NoRoom, instead of
an argument error, so callers can say what is wrong; the test that a wall thicker than the box
fails now expects exactly that.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The analytic mapping refused every revolution that is none of SMLib's plane, cylinder, cone,
sphere or torus (a profile of any other curve, a placement that is no rigid motion), which
now also kept those faces from kernel-smlib's exact rational B-spline import (da7ccba9). Such a
face now stays a revolution, so the import sends it as its B-spline; the analytic surfaces are
still taken wherever they fit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
All operations move to SMLib at once, so the selector is no longer the booleans' own:
`SOLIDCRAFT_KERNEL=smlib|truck` (read once, unset is truck, the same refusals for an unknown
value or SMLib in a build without it) is answered by `solidcraft_kernel::backend()`
(crates/kernel/src/backend.rs), which every operation's dispatch calls. The boolean's SMLib path
moves to smlib_boolean.rs. The regression rig's kernel check (`boolean_kernel`) now answers from
the switch, and the rig and CLI set SOLIDCRAFT_KERNEL.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The record of a truck body is built to the body's working tolerance (1e-6 of its size, at
least 1e-5 mm), and truck's offsets and blends leave some edges a few times farther than that
from a face they bound: no pcurve fits, so the record and every SMLib boolean on the body
failed ("pcurve of loop 0 use 0 ...: the curve at 0 lies 3.2e-5 mm off the surface" in a
shelled revolve of the robustness generator). The record for SMLib now retries at ten and a
hundred times the working tolerance for the body, its edges and vertices before refusing; the
saved native body keeps the working tolerance. Robustness seed 64 (a filleted extrusion
joined to a cylinder), refused this way before, now builds as truck does to 1.6e-6 of its
volume.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
An SMLib cylinder, cone, sphere or torus decoded as a truck revolution with its parameters
swapped (angle, profile), which truck's own revolutions never have. The tessellator then put a
cone's apex at the wrong angle: a drill point through SMLib and back meshed 41 % too large and
oracle cases 18, 55 and 56 measured wrong areas and volumes. The profile now runs against the
record's v, unswapped, which keeps the record's natural normal and truck's (profile, angle).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Truck's bodies are trusted into SMLib as given, but some miss SMLib's rules by a little (an
edge's vertex 4.4e-5 mm off it against a tolerance of 3.3e-5 after a truck operation), and
every boolean on them was refused by the check of its result. An operand whose trusted import
fails SMLib's check is imported again through SMLib's healer, and the healed body is used only
where SMLib then finds it valid; its numbering is checked against the record as before, so a
heal that renumbers leaves the result without history rather than with a wrong one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A revolve's flat ends (a line square to the axis turned about it) were refused, so any
revolved profile off the axis could not reach SMLib (oracle case 29). Such a face now goes as
SMLib's plane square to the axis, its range the box round the sector the line sweeps.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
kernel-smlib already gives solidcraft-engine and solidcraft-cli an `smlib` feature (for the
regression rig); the note on SOLIDCRAFT_BOOLEANS now says they are how the oracle, the naming
corpus and the robustness goldens run their booleans through SMLib.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The export takes a closed boundary's faces from the side in the empty region and skipped
every use there that is not `same`, a rule meant only for a sheet (a shell that is its own
mate). It tested the chosen side against the mate instead of the shell against itself, so a
cavity's shell, taken from the void's side, lost its faces whose uses run opposite to their
surfaces: "a face no shell uses" for every cut that leaves a void.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Truck extrudes a circle into rational splines, so every truck cylinder reached SMLib as
splines, and a box cut by one came back with edges 0.78 mm off their faces (SMLib's own check
refused it). A rational spline that is a circular arc swept straight along its axis (one
direction linear, the second row the first moved by one vector square to the arc's circle,
weights alike) now goes as SMLib's cylinder with the face's range on it.
The revolution mapping becomes one module for both (import/analytic.rs): each candidate face's
analytic surface is checked against the record's at 25 points of its domain, each point placed
on the analytic surface by its own (u, v), and turned over where the normals disagree.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Truck's planes run through three points, their axes the face's sides, so the SMLib import sent
every truck planar face as a bilinear NURBS patch: SMLib then intersected splines where it has
planes. A plane on any axes now goes as SMLib's plane on unit axes along U and at right angles
in the plane, its range the record's domain on those axes (a face's range need not be tight;
SMLib rebuilds an analytic face's trims from its edges).
A revolution's analytic frame turns to the face's first angle so u starts at 0: SMLib's check
refuses a cylinder whose range ends past 360°. A failed SMLib check of a boolean's result now
says whether an operand already fails it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Truck holds every round surface as a placed revolution, and a body SMLib made comes back from
truck that way too, so the SMLib import refused spheres, tori and any second boolean on a
first one's result. The import now turns each revolution of a line along or through the axis,
or of a circle centred on it or in a plane through it, under a rigid placement, into SMLib's
analytic surface: the face's range mapped from the revolution's parameters (angles moved into
the first turn: SMLib's torus breaks on a tube range below zero), checked against the
revolution at 25 points, and the face turned over (loops reversed, shell uses flipped) where
the analytic normal runs the other way. Other revolutions are still refused.
A result body's order is checked by each face's boundary edges as well as its identity:
Body::new rephases revolved faces onto new faces with the same edges.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
With the `smlib` feature, SOLIDCRAFT_BOOLEANS=smlib (read once per process, default truck)
sends every kernel::boolean through SMLib: each operand's schema 4 record imported and tagged
as operand 0 or 1, SmBrep's boolean, SMLib's check at every level, the result exported and
decoded (with its geometry check). A result either check refuses is an error. The setting
naming SMLib in a build without it, or anything unknown, fails every boolean instead of
quietly running truck, so a comparison never compares truck with itself.
The history is SMLib's tags renumbered into Body::history's index spaces. The export now
reports where each SMLib face and edge lands in canonical order (its arrays follow
SmBrep::GetFaces/GetEdges, the order the tags use). The result's order is checked against the
decoded solid by face and edge identity, the operands' by re-exporting the imported body and
comparing its numbering with the record's; where either differs the result has no history
rather than a wrong one. The result gets the operands' regularization protection and planar
roles as truck's does.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The owner's rule for the kernel-smlib branch: the kernel runs exact
operations and refuses what it cannot do exactly; sizes, standards tables,
defaults, pre-checks and conveniences live in the engine and doc. Threads
as the example: a helical sweep in the kernel, while the size table, the
off-size refusal and multi-start threads are sugar.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Only the Delaunay layout checked its triangles against the surface. The
grid divides each parameter direction within the tolerance, but on a
doubly curved surface the two sags add up in a cell's middle, along its
diagonal (SMLib torus 1.16x at 0.01, truck sphere 1.23x at 0.1, truck torus
1.40x at 0.01); the patches round a pocket cut into a sphere at its pole
fan out from the cut with no check at all (4.34x at 0.1, 1.90x at 0.01).
Every curved face's finished triangulation now goes through tess::sag:
each triangle is measured at its centroid and at the middle of each side
inside the face, across the surface (along the normal at the point of the
same parameters, which near a pole need not be the nearest), and the side
that leaves the surface most is split, both triangles on it together, so
the face stays conforming. Boundary sides are never split: they carry the
edges' own nodes, shared with the neighbouring face. At a pole the free
parameter of a midpoint comes from the other corners.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A sphere from SMLib cut through its centre keeps one face bounded by its
rim and by its seam from the pole up to the rim, used both ways. Unwrapped
along the loop, both runs of the seam landed on one parameter line and the
loop came out one period short, so the planner took it for a face winding
round the sphere and cut it along a new line of its own. The two seam runs
stayed inside that strip as a slit of zero width, triangulated on each side
separately: open edges along the seam up to the rim (gaps up to 9.5e-3 mm
at chord 0.1), a zero-angle triangle and a non-manifold edge at the pole.
Such a loop is now unwrapped from the node after the pole round to it and
started at the node before it: the seam's uses lie a period apart on either
side of the face's parameter box, which then meshes like any sphere patch,
both uses on the seam's own nodes. It is kept as it was unless the opened
loop spans at most a period with a positive area and no longer winds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Eight are hand-written synthetic files (MIT OR Apache-2.0). The other eight are byte copies of
CC0 files from tests/step-conformance and say so. Each row names the issue its case comes from.
cargo xtask assets passes again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLIB_REV moves to tangency-labs/smlib capi-solidcraft 1499681, eight SMLib
commits for surface/surface intersection: an empty curve is no longer
reverse-traced (SM_ERR_NULL_POINTER), tangent curves are traced once, curves
split at singular crossings, interior tangential contact points, tangential
curves and branches through tangent nodes, small loops above tolerance, and
high-order contact curves covered across their band. The corpus test's pass
list grows from 81 to 111 cases; the 8 still failing are all below SMLib's
1e-5 zone tolerance (angles 1e-6..1e-8, offsets 1e-7, loops of 2e-6 or less).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With the native feature, smlib_global_intersect_at_1e_6 builds both
surfaces of every case through solidcraft_smlib::ssi, intersects them
at approximation tolerance 1e-6 with the corpus runner's 4001 samples
per curve, turns the answer into what the runner writes for gsi, and
scores it. It asserts the exact set of passing cases (81, as score.py
grades the runner: 81 PASS, 25 INCOMPLETE, 11 WRONG TOPOLOGY, 2 ERROR),
so a regression and an improvement both fail until the list changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ffi/ssi.rs declares the C API's SSI block; ssi.rs wraps it: Surface
(plane, cylinder, cone, sphere, torus, NURBS, with the SSI corpus's
fields, angles in degrees), Method (Global, Advanced, Basic) and
intersect(), which returns every curve's type, UV flags, deviation,
interval and samples (xyz, uv on each surface, each surface at its uv)
and the isolated points. Without the native feature every call returns
NotBuilt.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ssi_score::score_case grades an engine result (sampled curves with xyz,
uv1, uv2, both surfaces at the UV samples and the curve type; points;
status; or a crash) against a corpus case: CRASH, ERROR, INCOMPLETE,
WRONG TOPOLOGY, INACCURATE or PASS, with score.py's flags and numbers.
ssi_corpus.rs checks every case parses, every NURBS surface lies on its
truth, the truth itself passes (bar unsplit singular points), and the
verdict rules. The ignored parity test scores the corpus runner's raw
outputs and compares with score.py: 119/119 identical verdicts and
flags for SMLib GlobalSurfaceIntersect at 1e-6 (81 PASS).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Test-support port of the parts of the SSI corpus's ssi_geom.py and
score.py the scorer needs: exact distances to the implicit truths, the
Gauss-Newton closest point on traced param patches, B-spline/NURBS
evaluation of the case surfaces, and score.py's segment-hash polyline
set, polyline distance, components, free ends and topology signature,
following numpy's arithmetic order.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The 119 surface/surface intersection cases of jyviko/ssi-corpus (b04c8a6,
MIT OR Apache-2.0), gzip-compressed, with its licence texts and a README
on the case format and the scoring rules. The 25 y23 cases come from
Yang, Jia & Yan 2023 (ACM TOG 42(6) Art. 211, CC-BY 4.0); both are in
ATTRIBUTION.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A PlacedRevolution record goes to SMLib as its exact rational B-spline
surface: each profile control point swept round the axis by a full circle of
four rational quadratic quarters starting at the face's first angle, then
mapped by the placement. The circle's parameter is not the angle, so the
face's angle range is converted pointwise and its pcurves are not carried
(SMLib rebuilds the trims). Where the placement mirrors, the record's natural
normal is the opposite of the map's, so the directions swap.
The B-spline matches the record's surface to 1e-12 with the same normal,
forward or not, mirrored or not; truck's sphere and torus now reach SMLib
with their analytic volume (1e-6) as manifold solids.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
784, 1753 and the two 1666 lofts came from holes in exported meshes, so
each now also requires a mesh with no open or over-shared edge. On today's
kernel 784 and both lofts are watertight; 1753's bowl, its rim cut by a
cylinder, still has 3 open mesh edges.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two faults made simple curved solids look open (#818's blind hole: 80 open edges, a stadium
slot: 52):
- It meshed with `refine`, the measuring mesh, which splits triangles across curved faces on
their own and leaves T-junctions on the edges they share with neighbours. It now checks the
display mesh, whose faces share each edge's points.
- It matched vertices by rounding to a 1e-6 mm grid, which splits points a rounding error apart
across a cell boundary. It now joins vertices closer than `tolerance.weld_mm` (default 1e-4 mm).
Both cases now have 0 open edges, also with the weld at 1e-6 mm.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2822, 3952 and 4149 cut long modelled threads into filleted or revolved
bodies; each takes minutes on today's kernel, so they get timeout_s 600
instead of the default 180.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
For the history areas' cases (threads, tessellation):
- A count may be an inclusive range `[min, max]`, where the geometry only bounds it (a modelled
thread's faces).
- `"watertight": true` meshes each body (chord tolerance `tolerance.mesh_mm`, default 0.01 mm)
and requires no open and no over-shared edges, with vertices matched by position.
- `"timeout_s"` gives a case longer than the run's timeout, at most 30 min.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
372 cases. The default kernel passes 232. smlib reports "not supported yet" for every case
until the SMLib boolean switch lands.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- tests/kernel-cases/sources.json gives each looked-at number's kind: an open or closed issue, a
merged PR, one of the frozen truck-geometry PRs, a kernel-paused PR. The report's coverage by
source counts how many of each were looked at, became cases or were skipped.
- crashes/2357: one join of a ring of 90 overlapping cylinder segments. The right answer is one
closed ring body. The old kernel still panics in its intersection curves and runs past the
timeout.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2969's description named another product's behaviour; 1443's and 2152's
still described the refusal. Each now says the hole is drawn at the
thread's nominal size and gets the fitted thread.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Reverts 1375e905. A modelled thread fits by default: a hole too wide for it
is filled to the tap drill, a shaft too thick is turned down to the major
diameter, so the seven cases (995, 1443, 2152, 2969, 3366, 2369, 2705) go
back to expecting the fitted thread and its volume.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A modelled thread never changes the size of the hole or shaft it is on
(#1018, #2853): one that isn't the thread's size is refused, naming the
size to use. The seven cases that expected the hole filled to the tap
drill or the shaft turned down to the major diameter (995, 1443, 2152,
2969, 3366, 2369, 2705) now expect that refusal, with its numbers worked
out from ISO 68-1 and the message in `refusal`. An opt-in fit option, if
it comes, gets cases of its own.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Runs are compared with the golden's times, so a run sharing the machine between cases flags
slowdowns that are only contention (seen with --jobs 4: four cases two to ten times slower).
--jobs still overrides it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Tiers: primitives; booleans in the degeneracy classes (coincident and coplanar faces, edge-on-edge,
vertex-on-face, tangent along a line and a circle, near-tangent gaps and overlaps of 1e-3 and
1e-6 mm, seam and pole crossings) and cuts that end in a single line (a knife-edge wedge, a V
groove and a cylinder that pinch the part in two, a hole tangent to a face, a notch flush with an
end); line-and-arc extrudes and revolves (partial, touching the axis); spline profiles, sweeps
and lofts; holes and tapped holes (blind, drilled, counterbore, countersink, M5-M8 taps, into a
curved face, near and tangent to an edge, crossing); modelled threads (external, fine, partial,
to a shoulder, internal, into a chamfer, at a filleted shoulder); fillets (constant, tangent
chains, convex and concave, variable down to 1e-3 and to 0 at a vertex, corners with equal and
unequal radii, g2 corners); chamfers (two distances, angle, three- and two-edge corners, curved);
shells and offsets; patterns and mirrors (overlapping instances that cut a core free); and
cumulative parts.
A running part grows one tier at a time through `base` (back wall, revolved boss and slot,
spline window, tapped holes, rounds, chamfers, shell, vents), and the last case adds its thread:
it carries every tier. Answers are analytic wherever the geometry has them (Pappus, Steinmetz,
ISO 68-1 basic thread profiles, exact Gauss quadrature of Bezier profiles, inclusion-exclusion
for chamfered corners), otherwise invariants; each `method` gives the derivation, and
tools/kernel-ladder/cases.py (standard-library Python, like tools/oracle-analytic) writes them.
Capabilities SolidCraft lacks are `missing` cases: multi-start threads, thread runout lengths,
fillet corner setbacks.
The golden is the first recording on truck, each tier sorted by its time. SMLib booleans are
not on this branch yet, so the smlib column records "not supported yet" for every case.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The ladder orders kernel operations by complexity instead of by random goldens: a manifest
(tests/kernel-cases/ladder/ladder.json) lists tiers in order and each tier's cases in order;
every case is graded against its own answer like any rig case, and a golden
(tests/kernel-cases/ladder/golden.json) records per case and kernel the status, the commands'
time and what the kernel made. A run compares with the golden and flags slowdowns (over twice the
time and 0.25 s more), status changes and changed counts. `--record REASON` re-measures one case
at a time, rewrites the golden with its reason and sorts each tier by the first kernel's time.
The report, docs/kernel-ladder.md, gives per tier and kernel the counts per status and the
frontier: the first case in ladder order the kernel fails. It reuses the rig's process-per-case
runner, timeout and cells; results now carry their path under tests/kernel-cases/ so cases from
other areas can be listed in a tier. The README documents `base`, `missing` and the ladder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
What the complexity ladder needs from the rig's case runner, kept in the one runner:
- A script may name a `base` case directory: that case's commands (and its own base's, up to
32 deep) run first, so a ladder case starts from the previous tier's finished part without
copying its commands. `$CASE` is each recipe's own directory; `$OUT` is the running case's
scratch directory, in its bases too.
- `missing` in expected.json reports the case as not supported yet without running it: a
capability with no command or parameter (multi-start threads, fillet setbacks) stays visible
instead of being faked with a stand-in.
- Each result carries `seconds` (the commands' own time, without process start or grading) and
`measured` (body count, face/edge/vertex counts, volume, area), for the ladder's golden.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A revolved cone tip with an exact volume and area; auto-reports without
a design are listed in the skipped.json files.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A tray with seven rounded edges, both rim rounds of a tube, a chamfer on
a joined L-shaped body, and a circular pattern of a chamfered hole, each
with an analytic volume.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
An M8x1 thread in a 7 mm hole, and three booleans of finely faceted
prisms (1000 and 1200 flat faces) with planar and curved partners, each
with a closed-form volume. Three auto-reports without geometry are
listed in skipped.json.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
`"outcome": "solid_or_refused"` takes a clean refusal, or else grades the result as usual. Two
bodies that touch along a circle only (#168, #3841) have no one-solid union: a refusal or the two
bodies left apart are both right; fusing them is wrong. The two cases said opposite things
(two bodies, a refusal) and now agree.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- `"shells": N` in expected.json grades the total number of shells: a body with a sealed void
holds two, and some joins keep several lumps in one body. Those cases (#3175, #3455, #3526,
#3730) are graded on validity again instead of being marked not valid.
- When SMLib's record conversion says "not supported yet" (revolved surfaces, traced intersection
curves), its check gives no verdict. That's shown but not graded, since it isn't a fault of the
body. solidcraft_kernel::SmlibCheck carries the three outcomes.
- `$OUT` in a recipe is a scratch directory outside the repository, for files a case writes. The
two STEP-export crash cases write there instead of /tmp.
- cargo xtask kernel-cases --results FILE keeps every result as JSON.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every other sweep, loft, revolve, pipe, coil and rib issue and PR, one line each: auto-filed reports that carry only command parameters, reports cut off in the source, UI and dialog items, feature requests, and failures that belong to import, shell, fillets, booleans or tessellation.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Coils split and trimmed by a plane, cut with a negative pitch, joined flush round a cylinder, intersected with a cylinder, cut as a thread groove into a bore, and placed with a square wire outside the diameter (screw-motion Pappus); surface revolve, sweep and loft sheets with areas worked out by integration, and a rail-guided surface loft in either section order.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
End faces of a loft, a sweep and a coil extruded, closed twisted sweeps round a torus, a twisted square cut through a tube, a ring swept along a short arc and joined to a collar, two overlapping swept arc walls, a rib grown along the sketch normal, and three pipes meeting at a point.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
End scale, parallel orientation and guide rails on straight paths (frustums and a sheared cylinder), a rectangle swept round a mitred corner, a profile mid-line on a closed rounded rectangle, second sweeps from a plane along a closed circle, hollow and cutting pipes, and a flat wall of a swept corner extruded.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Refusals for a profile across the axis and an axis outside the sketch plane, a construction axis picked by name, a cone left by a revolve cut, revolve joins and cuts whose end faces lie in a face of the target, a dome cut by a ball on its seam, a horn-torus turn, and half-disc revolves with a refused hostile sweep scale.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Spherical shell sketched off the origin, adjacent profiles on the axis, whole planar faces of full and partial turns and a face pick on a fresh revolve, a half slot turned about its centreline (alone and as a cut), a torus cut by an overlapping ring, and a 300 degree revolve crossed by a plate.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Added: 451 (block inside / crossing a hollow spherical-shell tool), 1107 (hollow hub joined
with nine blades), 1140 (spiral hollow lofted tube cut in half and joined), 2928 (cap-touching
lips with 0.0001 mm offset walls), 3850 with 3741 (cross cut touching a round side, and the
hole after it), 1984 (near-coaxial pin and rivet overlaps). #3573 is now a source of
shell/3576. Their lines are gone from the skipped files; #2259, #3518 and #3602 stay skipped
with corrected reasons.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Hollow elbows lofted along a bend (Pappus), hollow bands halved and joined, a flange touching the end of a bent tapered loft, tapered and twisted lofted blades patterned round a disk, and a loft joined into a window that it overhangs, alone and mirrored on a hollow case.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A shallow loft cut from a slab top face, a loft cut whose sections touch the block sides given in either order, circles lofted into a cone on a turned-over plane, a loft guided by a rail, a circle lofted to a comb outline, a square lofted to a circle, and a circle-to-rectangle loft joined to a cylinder.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Ruled lofts of zig-zag outlines worked out with the prismatoid formula, rotated rectangle sections turning through half a turn (Pappus), ring sections with round and rectangular holes, a cone loft meeting a cylinder wall near its top, a loft standing on a box face, a loft cut limited by targets, and a refusal for sections in one plane.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The rest of the thread cases (holes at the nominal size, shafts and holes
off the thread's size, the crash and sketch-attachment repros) and the
triage of the sources that are not kernel geometry.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
External and internal modelled threads, tapped holes and threads beside
other features, each graded against the ISO 68-1 groove volume (Pappus)
and the part's own closed-form volume.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
SMLib bounds an intersection edge on a piece of its spline (seed 32 intersect:
a block and a cylinder, the top disc's edges), and the truck decode refused
such an edge ("not supported yet: a spline or intersection edge on part of its
curve"), since truck bounds an edge by its curve's whole range.
bspline::segment cuts a B-spline to a sub-interval of its domain exactly:
Boehm insertion of both ends until each has multiplicity p, then the knots and
control points between them, clamped on the interval. The same points at the
same parameters, so the edge keeps its interval and the record its numbers.
load_edge_curve loads a spline or rational spline edge on part of its curve as
that piece; an intersection edge on part of its curve is still refused. Our
own records (interval = the curve's range) load as before.
Tests: segments of a rational cubic with a double knot, inside a span, across
knots, from or to a domain end, match the whole curve within 1e-12; both
spline kinds load on a piece with that range and the same points, and beyond
the domain are refused; SMLib's block ∩ cylinder decodes (it was refused) with
π r² h within 1e-3.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The adapter loads an analytic torus as truck's revolution of the whole profile
circle, a rational quadratic closed on itself, so an SMLib torus is one face
bounded by its two seam circles, each used both ways. The display evaluator's
domain gave a revolution a period around its axis only: projected along the
profile the loop jumped from one end of the knot interval to the other at its
start vertex, the plan failed ("boundary crosses itself") and the body fell
back to truck's mesher, which meshes a sphere bounded by its seam alone as
nothing (seed 60 union, a torus beside a sphere: 925.0 of 990.7 mm³).
tess::uv::closed_curve detects a closed profile the way closed does a closed
spline surface (clamped knots, first and last control point and weight the
same within 1e-9 of the polygon's extent; the shared checks are factored out),
the domain then has period = the profile's knot interval along it, and
EvalSurface::wrap/derivs wrap that parameter back into the interval, so the
plan's projection, unwrapping and both evaluators treat it as periodic. A
cylinder's or cone's line profile is unchanged.
Tests: a revolved whole circle, either parameter order, has both periods and
evaluates a period away as inside; SMLib's torus, alone and beside a sphere,
decodes without meshing to orient it, plans every face and measures within
1e-3 of 2π²Rr² (+ 4/3 πr³). With the closed profile undone the torus plan
fails as above.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Results of SMLib booleans on its own spheres came back wrong through
body_from_smlib (side-by-side runner: seed 13 cut −1611.9, seed 61 cut −252.4
against SMLib's 4056.9, seeds 46/61 union 80 % short). The schema 4 records were
right (check_geometry passes, outer loops counter-clockwise, shells out of the
material); three things on the truck side were not:
- Body::new orients a solid by the volume of truck's coarse mesh, and truck's
mesher meshes a sphere bounded by its seam alone as nothing: the rest of the
body set the sign and turned it inside out. A foreign record is oriented by
its shells, so body_from_smlib now builds the body with Body::oriented
(revolved faces turned as before, orientation kept, nothing meshed).
- rephase_revolved turns each revolved face so it sits around angle π. A face
that uses its seam both ways goes all the way round; turning its surface ran
the wrap through the face's hole, the display plan failed and the sphere
meshed as nothing. Such faces now keep their wrap on their seam (truck's own
primitives have none: their round faces are split).
- The display plan opened only a loop of one seam used both ways. A sphere face
whose meridian a trim splits uses each piece both ways, turning back at each
pole, and projected onto one parameter line ("boundary crosses itself"), so
the whole body fell back to truck's mesher and lost its sphere faces. A loop
through both poles that turns straight back at one now has its run between
the poles moved one period along, whichever way keeps it within one period
with a positive area (poles split as the planner splits them); the routing
check finds such loops too.
Test: SMLib's sphere less a box cap beside its seam, across it (seam split) and
a bar bored into its side away from the seam decode without meshing to orient,
oriented as recorded, every face meshed, within 1e-3 of the exact volume.
Before, the decode meshed the body to orient it; with only the plan's pole
handling undone the −x cap measures 50.3 of 469.1 mm³, with only the
rephase undone the bored ball 252.4 of 4059.5.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Every other fillet, chamfer and blend issue and PR, one line each: auto-filed reports from private designs (picks and error only), sketch fillets and chamfers, UI, previews and navigation, feature and fit-option requests, reference cases that already exist in tests/oracle, surface modelling, and failures that belong to booleans, shell, import, tessellation, sheet-metal or threads.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The graded volume was right; the derivation text quoted the union as 46476.5 instead of 45571.7.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Copies of filleted and chamfered features (a pasted block, a pattern of pockets with corner rounds, a pattern of chamfered bosses), a grille of rib roots, a join that needs a boolean retry followed by a fillet, and curvature-continuous fillets worked out from the documented quintic section.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Rounds along curved outlines, with the removed volume worked out by Pappus or by integration: merged cylinders, a D prism, a rounded rectangle with 0.02 mm end edges, a long box beside quarter turns, vertical corners after bottom rounds, a bent tube end, a boss's bottom arc, bars running into a boss, an edge ending on a curved wall, and the knife-edge arc corner.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Rounds with their own right answers: a radius larger than the ring it runs along (refused), a pick on a smooth seam (refused), a far edge pick (refused), a round that uses up an end face or a wall, a round over a hole, rounds on both sides of a thin wall and concentric with an inner one, a round on a sphere's flat cut, an edge ending on a tangent seam, and rims on B-spline ramps.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Chamfers on boxes, plates, discs and sectors with closed-form volumes: all twelve edges in the order that was refused, a concave chamfer that adds material, a chamfer next to square openings and one that runs into them, a hole rim tangent to the bottom face, a hex hole's corner bevels, a one-pick chain on a large part, a bottom loop as wide as the corner round, two distances on a circular rim, the end outlines of a rounded prism, a chamfer over a joined pin, and the equal chamfer of a slanted plug. Spline-sided pillar (invariants only) for chamfer and fillet.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Cases: mirror of a cut under a wall (#4011), coplanar intersect (#3671), multi-lump
intersect (#3526), split along a face plane (#3447), 64-peg join (#3455), boss on a rod
(#3485), collinear cylinders and pin in a bore (#3497), swept G2 cut (#3604), spokes
tangent to a hub (#3612), sphere and pyramid (#3633), column in a channel (#3665),
extrude to a point on an edge (#3729), emboss on a cylinder (#3782), circle-only join
(#3841), ring on a perforated plate (#3887), symmetric revolve cut (#4008), walls round a
floor (#4058), slab beside a bore (#4095), and others. The rest are skipped with a reason
in skipped-c.json.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every other sheet-metal-keyword issue and PR, one line each: UI, docs, file
format, naming and attachments, DXF output, feature requests, user designs
with no repro, and failures that belong to booleans, fillets, shell, sweeps
or threads.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Trays at 72 degrees and with four angles and heights, a tray shrunk by a
parameter and unfolded, an L bracket unfolded and refolded, opposite flanges
at 105 and 120 degrees, torn corners folded, and a rhombus whose square
reliefs would overlap.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Unfolded trays graded by the flat size W + 2(H - BD): torn, round and square
relieved corners, per-feature reliefs, rounded plate corners; a shelled tray
and a bracket converted at their bends; an L converted to the rule's radius.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A flange beside the leg of an L, along what a cut-out leaves of an edge,
beside a ledge as tall as its setback, a return on a flange's far edge, a
round hole, a slanted slot and an ellipse across a bend, a hundred slots with
four folds, and the refusal of rule values sheet.flange would ignore.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Cut-outs folded and refolded, a hole square to a 45 degree flange, a slot
that parts a flange from its base, an inside flange on a plate turned
62 degrees, cuts after cuts, a hole beside two adjacent flanges and a flange
on a flange's side edge; refusals for a flange on a bend line and a flat
pattern after a fillet; a mirrored sheet body that must stay sheet metal.
Volumes by hand (K = 0.5 folds: flat area * t; otherwise bend volumes).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Plates converted to sheet metal (holes, long thin strips, engraving, a
channel with modelled bends, position kept) and holes and cut-outs
through or into a sheet; volumes by hand from the plate and the holes.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Folds of tabs and plates, graded by closed form: a fold keeps the blank's
flat length, so V = flat area * t + theta * w * t^2 * (1/2 - K). Tabs
on one line (#264, #638), a tab on an L's inside edge (#639), a fold just
inside a tab root and a refusal at the root (#262), a fixed point on the
top face of a large plate (#685).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Cases for the near-contact gap join, a countersunk hole beside a distant
body, a perforated plate under a ring, a sealed void, a rim and tooth join,
a frustum cut, a line-to-line pin, a flush notch, spoke patterns and a join
into a rotated occurrence. The rest of the bundle is in skipped-b.json with
one line each.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Hand-derived cases for tapered and collinear extrudes, pattern and mirror
copies keeping their targets, a cut up to a face, a prism joined to a
round-cornered slab, a clip round a small hole, and near-contact joins.
2564, 2583, 2601, 2754, 2755 (pattern, mirror), 2801, 2810, 2826, 2580,
2896 (overlap)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Spheres cut twice (#1717), box/cylinder intersect (#1647), crossing cylinders
moved far (#1649), coaxial cut from a thin disc (#500), coincident cylinder
walls (#114, #2008, #1910, #1865), rim touching a coin (#168), holes along
walls and beside rounds (#559, #562, #1279, #1355, #2055), tangent holes
(#307), blind cut in a cylinder (#308), through-all cut (#423), shallow cuts
(#1858), skin left by a short cut (#1730), delete face (#1927), collinear
extrude sides (#1949), hidden bodies (#2464), mirror and patterns (#2003,
#2524, #2076), tapered extrudes (#831), text cut (#1195), split (#13).
Every other item in the bundle is in skipped-a.json with its reason.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every other tessellation-keyword issue and PR, with one line each:
rendering and GPU drawing, UI, performance, display-plan plumbing, and
failures that belong to booleans, fillets, threads, import, shell,
sheet metal or crashes.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
311 ramp cut across a round blank, 784/811 filleted and drilled bodies
for STL export, 818 blind hole in a filleted ring, 819/921 pocket that
splits a wall seam, 1263 concave arc with a knife-edge tip, 1615/1653
crossed lofted tubes, 1648/1850 sphere touching a box edge, 1666/1670
two twisted lofts, 1753 bowl rim cut by a cylinder, 1813/2712 lens face
thinner than the tolerance, 2069 notched ring moved far away, 2528
slot joined with an overlapping cylinder.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
140 join with a pad hanging past the edge, 1219 bore fillet then grooves,
1985/2243 STEP export of two enclosures, 3540 cut with a coincident cylinder
face, 3685/1559 delete a fillet face, 4092 loft + ribs + pad union (two
recipes), 2992 disk and 36 twisted loft blades. Everything else is skipped
with a reason (private designs, other areas, UI, perf only).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
4009 closed shell, 4010 shell open at a hole wall, 2531 point pick after an edit,
3468/3898 offset next to a fillet, 12 split face piece, 566 swept bar, 745/3893
freeform boss (two variants), 3103 tapered cylinder, 3669/2990 loft, 3283/3650
and 2939 curved offsets, 3506 and 1972 split then shell, 3576, 391, 1494, 1525.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Cases: #3702 (a board in the cavity of a case with a filleted rim, cut leaves the case
unchanged) and #258 (STEP cone to a point, with and without an apex vertex, and the 0.5 mm
pin). Everything else in the bundle is skipped with a reason in skipped-other.json.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- solidcraft-cli and solidcraft-engine forward an `smlib` feature to solidcraft-kernel/smlib, so
`cargo build -p solidcraft-cli --features smlib` gives the rig SMLib's check.
- solidcraft_kernel::boolean_kernel() says which kernel this process runs booleans on.
`default` until the SMLib switch (SOLIDCRAFT_BOOLEANS=smlib) lands and answers there. An smlib
run on a build without the switch reports "not supported yet" instead of grading the default
kernel under SMLib's name.
- An area may keep its skipped list in parts (skipped-<part>.json).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Rows without an analytic volume (the fuzz seeds) get one from the
primitives' definitions, by no kernel: along each vertical line both
operands are exact z intervals, combined exactly, integrated over a grid
whose cell edges include every straight face, with the half-resolution
grid bounding what error is left. Each kernel's volume is graded against
it (within 1e-3 relative or 3x that bound), in the JSON per row and in the
report's summary.
On 64 seeds truck is right on all 192 rows. SMLib's own measure is wrong on
13, all with cylinders or spheres, although truck's measure of the same
exported bodies is right on every one of them; 5 give no closed solid.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every geometry case the old kernel got wrong becomes tests/kernel-cases/<area>/<id>/: an
engine script or oracle recipe, and expected.json with the right answer, from analysis or a
hand calculation, or else the invariants. Never a kernel's output.
- solidcraft-cli kernel-cases <dir>...: runs a case and grades it. The checks are the outcome
(succeeds, or refuses cleanly), the body count, and one closed shell per body that passes the
record's structure and geometry checks and, with SMLib built, SMLib's check(All). Counts,
volume, area, centroid and box are graded where the case gives them. Each run is pass, fail,
not supported yet or crash.
- cargo xtask kernel-cases [--kernel default|smlib|both]: one process per case, so a native crash
or a hang (--timeout) is that case's result. It writes docs/kernel-cases.md: coverage per area
and source, results per kernel, and the sources looked at and skipped (<area>/skipped.json).
- solidcraft_kernel::validity and smlib_built, which the rig's checks use.
tests/kernel-cases/README.md gives the format.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When SMLib's volume of a boolean result differs from truck's, the report could
not tell a wrong body from a wrong SMLib measure. The runner now measures the
result it exports back to truck as well, keeps that volume in the JSON, and
shows it in the export column wherever it leaves SMLib's own measure or SMLib
could not measure the body at all.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The side-by-side runner now reports how long each boolean takes on truck and
on SMLib (the call alone), and apart from that the SMLib side's conversions:
the operands' truck→SMLib import and the result's SMLib→truck export, which
the app pays while truck still holds the bodies. The report gains a times
summary (median, worst, total), solid counts, the analytic volume of each named
case's union, intersection and cut, and how many results land on it.
It also runs the "each cut gets slower" case from TASK_kernel-smlib-booleans as
a named sequence: a 63.6 × 31.7 × 9.5 block, a Ø20 × 10 boss joined on top, a
slot front to back, a Ø6 cross hole through the boss and a slot that splits
the block, each step on the previous result. SMLib keeps its own results from
step to step (only the tools come from truck), in a child process like the
cases, and every step's volume is checked against its closed form.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The smlib_booleans example answered "not supported yet" for every SMLib case.
Now it runs solidcraft_smlib::Brep::boolean on the tagged operands (cut maps
to Subtract, None is an empty result), reads the result's history from the
tags, and measures its volume with mass_properties at MEASURE_ACCURACY, so the
report compares truck and SMLib on real results.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
truck lists some planar faces' hole loops before the outer loop, so the
rule that loops[0] is the outer loop never held for truck's records; moving
it in the adapter would renumber those faces' edges, which history and
names stand on. The outer loop is the one that runs counter-clockwise about
the natural normal (what check_geometry checks); the order is the
producer's, and SMLib's import puts the outer loop first, as SMLib needs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
check_geometry compared pcurve ends with edge points only, so a loop whose
pcurves enclose nothing passed whenever its uses lift onto the same 3D points:
SMLib's hole-rim fillet with both seam uses on u = 0, or a whole sphere whose
pole-to-pole seam has both uses on one line (the continuity check skips the
pole). Such a face has no surface and meshes to nothing.
Each edge loop's signed (u, v) area is now taken over 16 points of each use's
pcurve: a loop within 1e-9 of the square of its extent of zero is refused
(Problem::Empty), and each face must have exactly one counter-clockwise loop
about the natural normal, the rest clockwise (Problem::Turn). Outer and inner
are told apart by area rather than position, because the truck adapter writes a
face's wires in truck's order, which puts a hole first on some planar faces of
the sample designs (the model cache test caught it).
Fixtures: the pole-to-pole seam on one line, and a sheet whose plane axes are
swapped so its loop runs clockwise; a seamed sphere passes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A spline surface whose first and last control rows coincide in a direction
(points and weights, knots clamped there) closes on itself, but the display
evaluator's domain gave NURBS no period. The pcurve routine therefore never
moved a seam's second use by a period: SMLib's fillet round a hole's rim (a
rational spline closed in u) got both seam pcurves on u = 0, its outer loop
enclosed no (u, v) area, the decoded face tessellated to nothing and the body
measured 6839.73 mm3 against SMLib's 6989.00.
tess::uv::closed detects the closure from the record, within 1e-9 of the
control net's extent; the domain then has period = the active knot interval's
width, and the evaluator wraps a parameter outside that interval back inside
it, so pcurves, projection, sizing and the display plan all treat the direction
as periodic. The record stays as it is. The display plan hands both evaluators
wrapped parameters (the GPU does not wrap).
Tests: a rational torus band, a NURBS cylinder and a NURBS dome with a pole,
each bounded by a seam used twice, get seam pcurves one period apart; a record
of the band passes check_geometry and its seam uses on one line are refused;
SMLib's hole-rim fillet decodes with every face meshed and SMLib's volume
within 1e-3.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Brep::boolean (union, subtract, intersect; Ok(None) for an empty result), Brep::solid_count,
Brep::fillet, Brep::chamfer and Brep::shell, over the fork's smlib_ops.cpp shims (ffi/ops.rs).
Edges and faces are chosen by points, each the nearest within a tolerance; a point that
selects nothing or cannot decide is Error::NoMatch / Error::Ambiguous with its index, and a
solid input that gives anything but a closed solid is Error::NotSolid. Fillets and chamfers
take whole tangent chains; a chamfer distance is the setback along each face; a shell's
thickness is negative inward.
The C API with these shims is version 2 (the fork's capi-flat line), so ABI_VERSION is 2.
Tests (ops_tests.rs) against analytic volumes and counts, SMLib's own check at every level on
each result, and history through a tagged box minus box: union 9000 mm3 [11, 24, 16]; slot
6000 [10, 24, 16]; intersection 1000; empty results; a split body (2 solids); a cube with all
12 edges rounded r = 2 against the Minkowski formula; one chamfer (7910, [7, 15, 10]); shells
open at the top inward (3392), outward (4672) and closed; a tangent chain from one pick equal
to picking all three edges; hostile sizes, tolerances, points and counts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
build.rs asked Cargo to rerun only when capi/ changed, so a fix in the fork's source/ (SMLib
or SM_API) left the old objects linked: the tests kept crashing on a kernel bug already fixed
on disk. It now watches source/ too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The hyperbola and parabola cone pieces and the revolved cylinder, read by solidcraft-step into
schema 4 and built back into truck bodies, keep their closed-form volumes with their faces
pointing out.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SURFACE_OF_REVOLUTION reads as a placed revolution (u the angle, v the profile's parameter,
STEP's normal); SURFACE_OF_LINEAR_EXTRUSION as an exact spline surface, the profile as a
(rational) B-spline times degree 1 along the vector; a circle extruded along its axis or
revolved in a plane through the axis as the cylinder or ring torus it is. Line, conic and
open-conic profiles are bounded by the face (a closed conic's seam outside it, a revolved
profile stopped at the axis). OFFSET_SURFACE of the analytic kinds reads as the moved
surface, checked at each level. Hyperbola and parabola edges read as exact rational
quadratic pieces and a Bezier on STEP's parameter. Faces winding around a revolution get
seams along its profile. Every mapping is checked point by point and normal by normal
against STEP's own surface.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Schema 4 counts every pcurve's control points as items, which schema 3 had
none of, so a body that fitted the old 1,000,000 no longer does (a STEP
plate of 50k faces). The binary envelope's own caps on bytes, values and
entries per collection still bound a body.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A face winding around a closed (non-periodic) spline surface without a seam edge has loops
that jump across the surface's domain; it is refused for now instead of making a record whose
loops do not close. The step-conformance parts are read in a test, and the kernel's ignored
corpus comparison builds each record back into a truck solid to compare volumes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A face on a cylinder, cone or sphere bounded by loops that go round the axis (a band
between two circles, a dome on its equator, a cone on its base and apex) gets a seam:
a ruling or meridian from one loop to the other (split where it has no vertex at that
angle) or to the pole it covers, so every loop closes in (u, v) as schema 4 requires.
Pcurves may be fitted up to 1 mm off for edges that lie off their surfaces. The kernel
gets an ignored corpus test comparing truck's face and edge counts with schema 4's.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
read_solids walks MANIFOLD_SOLID_BREP, BREP_WITH_VOIDS and SHELL_BASED_SURFACE_MODEL
into solidcraft_geom::brep records: lines, conics, B-splines and polylines; planes, the
analytic surfaces and B-spline surfaces; edge intervals from the vertices' parameters;
loops relative to the surface's natural normal; outer loops, regions, voids and sheets;
pcurves from the 3D edges; the canonical order. Unsupported geometry refuses its solid
with a reason and the rest of the file is still read.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Boxes and cylinders go into SMLib as truck's own B-reps, through their
schema 4 records (smlib_body), so both kernels start from the same input
whatever the axis. Spheres and tori are SMLib's primitives about +Z, like
truck's, since the import doesn't take truck's revolved surfaces yet. All
41 cases' operands now build; only the boolean itself waits for its shim.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Cylinders (also against oracle 02 and 38's expected.json), spheres, cones, frustums and tori,
at the origin and 2e5 mm out: area, volume, centroid, moments and products within 1e-6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib B-reps leave the C API as its flat arrays, read through the crate's
one flat wrapper (crate::flat::Flat, copied out as import::Arrays), and
become schema 4 records: shell pairs merged, sheets collapsed, radial cycles
normalised, analytic surfaces as their kinds (narrowing cones kept), loops
reversed where SMLib's surface is inside out, every pcurve from
brep::pcurve, canonical order, then validate() and check_geometry(). Edges
arrive as their own kinds; NURBS edges that are lines or circular arcs (the
primitives' circles) are recognised as a fallback and counted. With the
kernel's `smlib` feature, body_from_smlib() decodes them into a Body.
Adds the cylinder, sphere, cone and torus primitives. Each primitive passes
SMLib's own check at every level, its volume from the exact mass properties
matches the closed form, and it goes round SMLib -> schema 4 -> SMLib with
its counts, volume and a clean check. Truck now meshes a whole sphere
bounded only by its seam (SMLib's sphere) through the display plan.
SMLIB_REV moves to the fork's capi-primitives (the primitives on capi-flat
40bb855).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
SMLIB_REV moves to the fork's capi-static 40bb855 (flat arrays, C ABI 2).
flat.rs wraps them (Flat: arrays by id, Brep::export, Flat::import,
Brep::is_manifold_solid; FFI in ffi/flat.rs). import.rs lays a schema 4 record
out as those arrays: a closed shell becomes one SMLib shell per region it
separates, a sheet's shell holds both sides of each face, the infinite region
is marked void as SMLib has it, pcurves become UV trims on B-spline faces only
(reversed for edgeuses against their edge), planes with non-unit axes go as
their exact bilinear patch. Brep::from_record builds the SMLib body.
kernel: brep_record(body) gives a body's schema 4 record, and smlib_body
(feature smlib) its SMLib body, trusted (no healer: healing moved a
cylinder's volume by 9e-7). Box, cylinder and tilted cylinder keep their
analytic volume in SMLib (1e-9, 1e-6) and are manifold solids; truck's
spheres and tori are revolutions, refused with "not supported yet" for now.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A whole sphere bounded only by its seam (SMLib's sphere: one meridian
used twice) came out with both uses on the same u line, a loop that
encloses nothing. Two causes in the per-loop closure:
- The pole-junction search took any positive signed area as the right
turn. With the two uses on one line the area is rounding, and in most
frames that rounding is a tiny positive number, so "no shift" won over
the period shift. A turn now needs more than 1e-9 of the loop's
period box.
- Junctions counted as poles only where the surface collapses exactly.
Ends within the tolerance of a pole (as a kernel's records have them)
were closed to the nearest period instead. A junction is now free
where moving its periodic parameter by half a period stays within the
tolerance, and fit samples within a quarter of it take their free
parameter from a neighbour, so the pcurve runs straight into the pole.
Tests: the sphere bounded by its seam in four frames, three meridian
angles, circle and NURBS meridians, both loop orders, ends at or a tenth
of the tolerance off the poles; and a torus bounded by its two seams.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Area, volume, centroid, and inertia and products about the centroid (with inertia_tensor()),
integrated by SMLib on the exact surfaces, through the fork's smlib_brep_mass_properties and
smlib_brep_area. MEASURE_ACCURACY = 1e-8, the finest SMLib allows. At that accuracy the fork's
own test puts boxes, cylinders, spheres, cones and tori within 2e-7 of their closed forms.
Tests: tests/oracle/01-box against its expected.json (to 1e-12), a box 2e5 mm out keeping its
inertia, and accuracy arguments. SMLIB_REV moves to the fork's measure commit (3e95b79).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ValidatePointers, then AssertValid over the whole topology graph (the fork's
smlib_brep_check), at CheckLevel::Basic, Geometry or All. A body that fails is still Ok, with
the failed tests counted and one line each in the CheckReport; Err means the check could not
run. Tests run it on every SMLib result and import.
SMLIB_REV moves to the fork's check commit (39f398e), which also fixes a seam-curve leak that
AssertValid hit on every periodic face.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
decode builds topology with Edge::new_unchecked and Face::new_unchecked,
so a record whose curves miss their vertices or whose pcurves miss their
edges loaded and failed later inside truck. It now runs check_geometry
after validate and refuses such a record (a cache miss at the IO
boundary). Only our own cache feeds it today; STEP and SMLib exports will.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
validate() is structural only, and a kernel adapter that builds topology
from the records without its own checks (truck's new_unchecked) would take
an edge whose curve misses its vertices and fail later inside the kernel.
check_geometry spot-checks what it assumes, each against the record's own
tolerance: every edge's curve meets its vertices at its interval's ends;
every oriented edge's pcurve, lifted through its face's surface, lies on
the edge's curve at the interval's ends and middle and continues the use
before it on the same period; every vertex loop's vertex lies on its
surface. A fixed number of evaluations per record, so linear in the item
budget; a point that cannot be evaluated is an error (Mismatch, naming the
record), never a panic.
The schema fixtures now carry real planes and pcurves (from
pcurve::pcurves), so every form passes; one-defect fixtures and hostile
records cover the refusals.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
SMLib makes a fillet or chamfer face with no tag, so history::blends gives
each Generated face FaceOrigin::Edge for every blended edge whose two faces
both lie next to it and no longer meet directly. For the fillet shim's
history once it lands; a corner patch between blends keeps no origin.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
examples/smlib_booleans runs truck and SMLib on the same cases (named pairs
whose results are known by hand, then fuzz::boolean_shapes for N seeds),
each as union, intersection and cut, and compares volume, face and edge
counts and the shape of the history (how many faces Kept, Modified, Split,
Merged, Generated; edges from an operand edge, new, or without origin). It
writes a JSON and a Markdown report and never changes the default kernel.
SMLib runs in a child process per case, so a crash inside it is reported as
that case's result. Its primitives and booleans arrive with the operation
shims; until a call exists the case is reported as not run on SMLib.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Brep::tagged gives a copy whose faces and edges are tagged as one operand's;
after an operation on tagged copies, Brep::history reads back each result
face's origins and change (Kept, Modified, Split, Merged, Generated) and each
edge's faces and origin edges, as SMLib itself carried them through the
operation's splits, merges, copies and coincidences. Raw::history turns that
into solidcraft_geom's History, renumbered into canonical order by each
body's Order, with a new edge recorded as where its two faces meet.
The FFI declarations are in src/ffi/history.rs, beside ffi.rs. Tests cover
the C layout, hostile records, the meeting edges, renumbering and bad
orders, and (with SMLib built) a tagged box.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A new crate, solidcraft-step (L1, no modelling backend), holds the Part 21
parser (moved from the kernel, which now reads through it) and STEP geometry
as plain data: lines, conics, B-splines (rational and implied knots),
polylines, trimmed curves, surface curves with their pcurves, 3D offset
curves; planes, cylinders, cones, spheres, tori (degenerate too), B-spline
surfaces, revolutions, linear extrusions, offset and rectangular trimmed
surfaces. Each keeps STEP's own definition and parameterisation, in mm and
radians, with pcurves scaled by their surface's parameter units. Shared
entities stay shared. A writer emits the same entities back, deduplicated.
Tests per entity check values against analytic points, units, round trips
and hostile input; an ignored test reads and round-trips every edge curve
and face surface of the tangency-labs/models STEP files.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ffi::line was a safe public function reading any raw pointer. The log
callback now lives in ffi.rs, reads the C string there, and hands the
crate a String through a safe sink, so no safe function takes a pointer.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A new crate (L1, only the kernel uses it) builds tangency-labs/smlib's SMLib,
SM_API and C API with the cc crate when its `native` feature is on, from the
commit in crates/smlib/SMLIB_REV (scripts/fetch-smlib.sh fetches just those
sources). The kernel's `smlib` feature turns it on. Without it nothing C++
is built and every call returns NotBuilt, so workspace builds need no C++
toolchain.
unsafe is confined to src/ffi.rs. The wrapper owns handles, maps statuses to
errors, checks the C API version, and sends SMLib's per-call log to the log
crate. Crash reports note the native kernel call in flight.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SMLib makes narrowing cones with a negative half angle; schema 4 now takes
them as they are (r(v) = r0 + v tan a already covers them), with the
natural v range below the apex. The check that no region lists a shell
twice was quadratic per region (about 1e9 comparisons at the caps); it now
remembers the last region that listed each shell.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The adapter now saves each edge's curve record and interval when the edge first appears and
computes every face's pcurves from those records and the surface record it writes, with the
shared plain-record routine; the truck-face version in native/pcurve.rs is deleted. Loops go
in as the record keeps them (relative to the surface), so a reversed face no longer has its
wires reversed and back again.
The native tests are ported (adapter::face_alone gives one face's record for them). The cut
body's intersection pcurves now fit at 1e-6 mm too: the routine follows the converged
intersection, and truck's own points are within its 1e-6 mm of it, which the lift check
allows for on intersection edges. docs/native-brep.md says where pcurves come from and what
a surface intersection's point at t is.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
solidcraft_geom::brep::pcurve::pcurves(surface, loops, tolerance) computes every edge use's
pcurve from our own records: the surface record and, per loop, each use's curve record,
interval and direction. No backend's parameter space is involved, so the truck adapter and an
SMLib conversion can share it (owner decision 2026-10-21).
brep::eval evaluates every schema 4 kind: planes, splines and revolutions through the display
evaluator (EvalSurface, EvalCurve, tess::uv), cylinders, cones, spheres and tori in closed
form in angle and height, and a surface intersection as the point of both surfaces in the
leader's normal plane (what truck's subs converges to), by Newton's method.
Exact in our parameters: lines, polylines and whole splines on planes; a circle about the axis
of a cylinder, cone, sphere or torus (a u-line); a line along a cylinder's or cone's ruling (a
v-line). Each exact pcurve is checked by lifting it; everything else is the cubic B-spline fit
the kernel had, with its span and round caps. Curve2::eval moves here from the kernel.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The parameter box, the projection of a point onto an evaluator surface (closed form on a
plane, damped Gauss-Newton elsewhere), unwrapping and the pole test work on geom's
EvalSurface alone. They move from the kernel's tessellator to solidcraft_geom::tess::uv so a
pcurve routine on plain B-rep records can use them without truck; the kernel re-exports them
and its tessellation is unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A test over every design the repository keeps (the action matrix's
corpus: old design files, examples, regressions, demo and typical
designs, the oracle recipes; plus the sample designs, the engine's
design fixtures and the older designs of the hole matrix, taper and
projection tests: 154 designs). Each is rebuilt and recorded in
tests/naming/<group>/<design>.txt: per body the names of its faces and
edges with whether each follows exact history, and every stored name
(feature face picks, fillet/chamfer/bend edges, sheet metal edges, the
face a sketch is attached to) with the face or edge it finds now. A
difference fails with the design and the changed lines; record an
intended change with NAMING_CORPUS_BLESS=1, as FEATURE_WALK_BLESS does.
Face and edge indices are left out on purpose. They are the kernel's
order: truck already gives a fillet's or a boolean's faces in another
order from one process to the next (oracle 07 and 37 changed on every
run), and SMLib will order them differently again, while what must not
change is the names and what references find. Names are listed sorted.
One design's names themselves change between runs on truck
(closed_sweep_torus_cut: level wall halves numbered #1/#2 in kernel
face order); it is listed in UNSTABLE with the reason and its
difference printed instead of failed on.
Records made on this branch (truck), macOS. The whole corpus builds in
about 20 s on 14 cores (designs run side by side).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The corpus naming test (next commit) records what every stored face and
edge name resolves to, and must find it exactly as a rebuild does. Three
lookups were crate-private: `locate_face_on_body` (a feature's face
points, tried on the body the feature names first), `blend_target` (the
body a fillet or chamfer rounds) and `Document::anchored_face_indices`
(the faces a face-attached sketch is on). Visibility only; no behaviour
changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
docs/native-brep.md describes schema 4: the records with their SMLib and
STEP names, regions and shells, the loop convention, every edge's uses,
the canonical order (a reversed face walked backwards), each curve and
surface kind's parameterisation, pcurve rules, validation, history, the
SCBREP05 envelope and what the truck adapter writes and refuses.
A test lifts every pcurve of a box, a filleted box with a bore and a
sphere onto its face and checks it is its own edge at both ends and the
middle; with a reversed face's pcurves left in truck's oriented order it
fails by 30 mm.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Edges, vertices and the body carry the face tolerance (10^-6 of the
body's size, at least 10^-5 mm): truck's traced intersection edges sit
about 1e-6 mm off their surfaces, so the linear tolerance is too tight for
a pcurve to lift onto them. The box-and-fillet golden fixture is the
schema 4 envelope now (NATIVE_BREP_BLESS=1 rewrites it), and the adapter
tests cover a pair of lumps. The model cache's producer revision becomes
solidcraft-truck-adapter/4.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The plain B-rep follows SMLib's flat form with STEP's names: regions
(infinite first), face/wire/vertex shells, oriented faces, faces with a
domain and edge or vertex loops, oriented edges with a pcurve each, edges
with an interval, a tolerance and their list of uses, vertices with a
tolerance. Curves and surfaces are unbounded; their bounds are the edge
interval and the face domain. Loops are relative to the surface's natural
normal. Schema 3's read-only intersection and revolution records go.
The validator keeps schema 3's caps and adds the structural rules:
references, loop shape and connectivity, regions and closed shells, the
pcurve interval, natural parameter ranges, every edge's uses (primary
first, upper = !forward) and the canonical order of shells, faces, edges
and vertices. A reversed face is walked backwards, as truck's edge_iter
walks it, so truck's face and edge indices are the canonical ones.
The truck adapter writes truck's absolute wires as the loops, numbers
everything as the kernel does, and builds regions from the shells (a
tessellated nesting test only when there are several closed shells).
Pcurves come from pcurve::pcurves only; until it supports a face, encode
fails and the body is a cache miss. The envelope signature becomes
SCBREP05.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
FaceOrigin, Change, FaceProv, Side, EdgeOrigin, EdgeProv and History move
unchanged (fields, serde derives) to solidcraft-geom, so a backend can
report history without depending on solidcraft-kernel. The pure methods
(composition, renumbering, count checks, mark_shared_split) move with
them and return brep::history::Failed, which converts to
KernelError::Failed with the same "history: …" message.
What needs a kernel body (History::identity, check, face_sources) stays
in the kernel as the BodyHistory trait; solidcraft_kernel::history
re-exports the types, so existing paths keep working.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A band of a cylinder bounded by two full circles and no seam cannot close
in (u, v): each loop's pcurve runs one period. The test records that such
loops come back continuous and a period open (the caller needs a seam
there), and that a hostile tolerance is refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Each sample of an intersection curve is one of truck's double projections
(about 0.13 ms in a release build), and a fit evaluates the curve at every
sample and halfway between. Starting from four samples a span instead of
eight halves that; the checks stay at eight points a span, and the knots
at the curve's own breaks already carry the shape. The cylinder cut by a
tilted box now takes 70 ms at 1e-3 mm and 149 ms at 1e-5 mm for all its
faces (release), from 183 and 638 ms.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Faces with seams as imported faces have them (the kernel's own
revolutions come in pieces): a cylinder whose seam line is used twice
between two full circles, each a closed edge on one vertex, and a sphere
bounded by one meridian used twice from pole to pole. Their seam uses come
out a period apart, in the base period. The checks now also require every
face's outer loop to turn counter-clockwise in (u, v) for a face along its
surface's normal and clockwise against it, holes the other way: the rule
the pole junctions rely on, which every test body follows.
With SOLIDCRAFT_PCURVE_SVG=<dir>, a test writes each test body's face
loops in (u, v) as SVG for review.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Every edge not exact on a plane gets a cubic 2D B-spline on its curve's own
parameter: through the parameters of the curve's ends, least squares
through samples between (banded normal equations), with spans halved and
sampled twice as densely wherever the lift misses the curve by more than
half the tolerance, at the samples or halfway between them. Knots start
where the curve's parameterisation is less smooth: a spline's own knots
and the vertices of an intersection curve's leader polyline, with the
multiplicity a cubic needs there (without them the cylinder-minus-box
intersection curves took 400 spans at 1e-3 mm). Spans and rounds are
capped; a fit that does not converge, or a curve farther from the surface
than the tolerance allows, fails naming the wire, edge and distance.
Samples are projected with the kernel's own surface parameter search (the
evaluator's form, parameterised as truck parameterises it; truck's search
for surfaces it does not take), each from the one before, unwrapped along
periodic directions. Where the surface collapses (a pole) the undefined
parameter comes from the nearest sample that defines it, so a pcurve runs
straight into the pole. Each wire's uses are then moved by whole periods so
each starts where the one before it ends, starting after a pole junction
if there is one; at further pole junctions the offset that gives the loop
a boundary's turn is taken. The wire then moves into the base period. A
seam edge used twice is fitted once.
Truck places intersection curves on their surfaces to about 1e-6 mm, so
their pcurves need a tolerance of 1e-5 mm or more; the test records that.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A plane's parameters are affine in space, so the pcurve of a line or a
(rational) B-spline on a plane is the curve's own knots and weights with
its control points mapped into (u, v): no fit. Curved surfaces and
intersection curves still return "not supported yet". A test helper checks
every pcurve the way the schema validator does, its interval against the
curve's range, the lift error and the wire's closure in (u, v); the box's
line pcurves lift onto their lines' ends bit for bit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The shared base for the schema 4 types and the truck pcurves: 2D B-spline
pcurves as SMLib keeps its UV trim curves, and the truck adapter's pcurve
function, not supported yet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Arabic and Hebrew are complete catalogs since #3768, so
every_complete_catalog_carries_every_english_message fails a pull request
that adds an English message without them. The localization guide's list of
languages a new label needs now says so.
Co-authored-by: rootarchitect <7781866+rootarchitect@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4771
* Move/Copy takes a browser body the same way Combine does (#4771)
A body row feeds the dialog's active input through the same pick as a
click in the view, and does not replace the selection (#4452). A group,
the Bodies folder, or a component is that pick for every body in the
set. M and the Move/Copy button also start from a selected body or from
a component picked in the tree.
* claim: #3457
* Hebrew interface language with right-to-left text (#3457)
Adds a complete Hebrew catalog (assets/locales/he.json, original
translations) and registers `he`.
egui 0.36 shapes each font run with harfrust, so a run of Hebrew letters
is drawn right to left, but it places the runs of a line left to right in
stored order. The new `bidi` module runs the Unicode Bidirectional
Algorithm (unicode-bidi, MIT OR Apache-2.0) over each message of a
right-to-left catalog once, giving egui the runs in display order with
brackets mirrored and placeholders kept whole, while leaving each Hebrew
run's letters in reading order for the shaper. A test lays messages out
with a real Hebrew font and checks where the glyphs land.
Desktop builds append a system Hebrew font (Arial Hebrew / SF Hebrew,
Segoe UI / Arial, Noto Sans Hebrew / DejaVu Sans). Layout mirroring,
wrapped multi-line RTL paragraphs and a web Hebrew font are follow-ups.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hebrew: translate the new Describe Command label; list he in the control protocol
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hebrew: catch up with main's catalog and the new ribbon (#3457)
- Translate the 258 messages main added since the branch was cut, drop
the 38 it removed (10 renamed ones keep their translation) and
retranslate the two whose English changed (toolbar → application bar).
- The timeline title is ציר הזמן, matching the glossary.
- Ribbon buttons split their labels into two lines from the start of
the string; Hebrew labels come in display order, so the first line
showed the end of the phrase ("חדשה / סקיצה"). For right-to-left
labels the first line now takes the words at the end of the string,
and a shortened line loses characters, and shows its ellipsis, on the
left. The word-boundary sweep checks Hebrew from the lines the galley
is built from, since epaint's right-to-left rows carry zero-width
glyphs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hebrew: after rebasing onto main, load the Hebrew face with the cached system fonts
main now searches the system fallback faces once (SystemFonts, on a thread);
the Hebrew face joins that search with its coverage check. Drop the three
camera-animation messages main removed (#4296) from the Hebrew catalog.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hebrew: translate the messages main added after the catch-up (#3457)
Translates the messages main added since the last catch-up: the history
bar's buttons and name toggles, the 3D mouse driver and axis mapping, the
thread size columns, the File view (home page), the redesigned Preferences, the new issue report
dialog, free orbit and the assistant's setup card. Drops the one message
main removed. The catalog has every English message again, so the
complete-catalog checks pass for he.
* Arabic interface language (right-to-left, on the Hebrew bidi path)
Adds an original Arabic translation of the whole English catalog
(assets/locales/ar.json, every message) and registers it. `ar` joins the
complete catalogs, so a message added to en.json later fails the build
until it is translated. It goes through the right-to-left path of #3458
unchanged: crate::bidi puts each message in display order and leaves
every Arabic word's letters in reading order for harfrust, which joins
them with the font's own tables. Ribbon labels use the right-to-left line
split from the same stack.
fonts.rs adds a system Arabic fallback after the Latin interface font,
which keeps spaces, digits and Latin text so each Arabic word stays one
shaped run: Segoe UI/Tahoma/Arial on Windows, SF Arabic/Geeza Pro/
Damascus/Tahoma on macOS, Noto Sans or Naskh Arabic/KACST One/DejaVu Sans
on Linux, or NotoSansArabic-Regular.ttf in CRAFT_FONTS_DIR. It is the first
fallback for an Arabic UI (Arial Unicode, a CJK candidate, has Arabic
letters without joining data) and the last otherwise; like Thai and
Hebrew it is never promoted for a CJK interface.
Tests: the catalog comes back in display order with no directional
controls, and Arabic laid out with a real font lands on screen word by word
in reading order.
* Hebrew: review the catch-up translations; restore the camera-animation messages (#3457)
- Review of the 533 messages translated in the previous commit: 21
corrected. Keys in the motion dialog are keyframes (נקודות מפתח, as
in Add Key), the Design workspace is תכן as in its label, a dropped
"(Direct Driver)" is restored, terms aligned with the rest of the
catalog (פורמט, סוכן קוד, רצועת הכלים), and a few calques reworded.
- Translate ui.animate_view_changes, ui.animation_speed and
ui.look_at_anim_desc, which main has again, so he stays complete.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Arabic and Hebrew: translate the messages main added since (#3457, #3762)
Both catalogs again carry every English message. The 760 messages main added
since the last rebase (render workspace, material editor, standard views,
sheet metal Path Flange, AI app connections, document-on-disk notices and
more) come from A-AL-ANAZI's translations on #3768; the five Perforate dialog
messages are new here. The earlier Hebrew review is kept on top of them, and
33 of the new Hebrew messages are corrected: "completed" was "perfect"
(מושלם → גמור), Regrade read as ranking (now תיקון צבע מחדש), shaders,
guarantee, the Mirror reflection mode, simulations and displacement.
Co-authored-by: Abdullah <alksa2025@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Arabic and Hebrew: attribution rows as a pending file
main now takes new ATTRIBUTION.md rows as one file per change
(attribution/pending/), so the he.json and ar.json rows move there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Arabic and Hebrew: translate the 23 messages main added since (#3457, #3762)
Both catalogs again carry every English message; the two theme-page messages
main removed are dropped. Twelve of the new messages (link updates while
loading, the agent chat's model choice, plugin search, render quality) come
from A-AL-ANAZI's translations on #3768; the sketch plane descriptions, the
rigid group and empty-boolean messages, free sketch points and the web update
note are new here.
Co-authored-by: Abdullah <alksa2025@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Arabic and Hebrew: translate the 12 messages main added since (#3457, #3762)
Eleven come from A-AL-ANAZI's #3768; "Turn Up Axis" in Hebrew follows the
catalog's existing up-axis wording, and the agent chat's no-tool-call message
is new in both.
Co-authored-by: Abdullah <alksa2025@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Arabic and Hebrew: translate the 22 messages main added since (#3457, #3762)
* Arabic and Hebrew: a label found by its English text comes back in display order too (#3457, #3762)
* Arabic and Hebrew: translate the 32 messages main added since (#3457, #3762)
* Arabic and Hebrew: translate the 34 messages main added since (#3457, #3762)
* Arabic and Hebrew: translate the 30 messages main added since (#3457, #3762)
* Arabic and Hebrew: translate the 7 messages main added since (#3457, #3762)
---------
Co-authored-by: rootarchitect <7781866+rootarchitect@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Abdullah alanazi <x1700@MacBook-Air-Abdullah-2.local>
* claim: #6943 free orbit Shift snapping
* fix(ui): snap free orbit direction and roll with Shift
* test(ui): align free orbit coverage and complete snap help translations
Evaluation and placement never read appearances, yet an appearance edit
re-evaluated the model and placed every occurrence again. A whole-document
proof that only appearances changed now takes the existing render-only refresh,
which keeps the model and retargets the world-placement cache.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Sketch Transform: scale, turn and move sketch entities in one command and one undo
sketch.transform scales about a centre, turns about it, then moves; the window and its
handles for existing sketch geometry build on it.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Sketch Transform window: one box with handles to scale, move and turn what is picked
Select curves and points in a sketch and press G (or Transform in the Modify panel or the
context menu): the picks get a box with a handle at each corner to scale (the opposite corner
stays), the box to move and a knob above it to turn (Shift snaps to 15 degrees). Position,
Width, Height, Scale and Angle in the window are the same values, to type exactly; OK applies
it as one step.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Move the new message to a pending locale file
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch Transform: follow the sketch field of Sel::SketchCurve
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Nathalie Schönwetter <nathalie@cfw-ag.ch>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #5542
* Reopen the designs that were open at the last quit (#5542)
A launch with no file to open reopens the saved designs that were open
when SolidCraft was last closed, in tab order. Designs never saved and
files that are gone are skipped. Preferences > General > Reopen designs
turns it off; a launch driven over --control or with --sample starts fresh.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Move the new messages to a pending locale file
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Each command that names a body (appearance, material, move, ...) mapped it to
its component's own name by working out every placed body's world name and
frame, so a command on a large assembly cost a walk of the whole design. The
list is now kept with the world-placement cache while it is current.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Naming a placed body scanned every occurrence and body of the design, so the
world state and every caller naming all placed bodies grew with bodies times
design size. WorldNames indexes the design once and names each body in time
proportional to its occurrence path; one naming routine serves both it and the
single-body scan. Names are unchanged (a test compares them with the scanning
implementation).
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: palette-indexed scene and incremental appearance updates
- Move triangle, line and instance colours and opacity into a per-scene
palette (an R32Uint texture, WebGL2-safe); appearance edits update
palette entries instead of vertex bytes.
- Build CPU body triangles as one canonical vertex stream per block with
opaque/glass/faded index lists; an opacity class change rewrites only
the index list.
- Cache resolved body appearances by a per-body key, so an edit resolves
only the bodies whose look changed.
- Split the model scene key from the session revision: selection and
view-only commands no longer rebuild or upload the model scene.
- Replace the O(n^2) section-index lookup and per-placement instance scan
with hash lookups.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Viewport: skip unchanged palette work when rebuilding
- Reuse each body's appearance key while the document is unchanged.
- Leave a palette slot alone when its value did not change, and allocate
every body's slots once before the build loop.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Viewport: measure scene rebuild cost by edit kind
An ignored scale test times the scene build after selection, appearance,
theme and one-body geometry edits at 200, 1,000 and 5,000 bodies.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Viewport: read faded grid alphas through the scene palette
The grid-fade tests now resolve line palette indices instead of reading the
line record's old colour byte.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Viewport: measure rebuild cost on few large meshes
Adds the 20/100-torus case where a block rebuild used to re-emit a large
part of the model.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Viewport: measure rebuild cost on an assembly
Unique detailed parts plus one component placed many times, timed for
selection, part and component colour, opacity, theme, hide and move.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: draw block classes as vertex runs, not index lists
A body's vertices are three per triangle in mesh order, so each draw class
is a few runs of the block's vertex buffer: all-opaque blocks need no index
buffer, and an opacity class change uploads nothing. A fragmented class
(more than 64 runs) falls back to one index buffer.
Each triangle is written as one record into a block reserved up front,
which takes the cold scene build of a 7.1M-triangle assembly from 97 ms
to 55 ms (main: 83 ms).
Also fix reindexing a block whose bodies are drawn elsewhere: it walked
the mesh's triangles from that body's start, past its own vertices.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: draw the highlight's own block, and light faded bodies without copies
The highlight scene keeps its feedback (a hovered face, a body copied on
the CPU) in a block of its own; since blocks became vertex runs, only the
model's blocks were drawn, so that feedback was uploaded and never shown.
Draw the highlight's blocks with its triangles.
Whole faded and see-through bodies (another component active, or an
appearance below full opacity) were lit by copying every triangle on the
CPU at each hover change: 85 ms for a design row of 812 bodies. They are
now drawn again from the model's own block vertices, in the highlight
palette with their slots set to the highlight colour, or as placements of
the model's shared triangles. The picture is the same as the copy's.
Adds a GPU frame benchmark for document-tree hover (ignored test).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: count scene uploads, and keep the sorted glass stream across colour edits
Renderer::uploads reports the bytes the model scene's last change uploaded
(geometry, placements, palette rows, planned colours, re-evaluated vertices).
Without order-independent transparency the global glass stream is kept while
the blocks' see-through triangles are the same, so a colour edit on a glass
body uploads palette rows only. Tests assert both on a GPU.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: keep palette slots and looks without rehashing every body
A rebuild found every body's palette slots by name and wrote them back,
allocating and hashing a new key per body, and copied every shown name into a
set to check the arena still belongs to the design. Slots are written back only
when a body's faces grew, and the looks cache is pruned only when it holds more
bodies than are shown. Adds the STEP-file benchmark (SOLIDCRAFT_SCALE_STEP) and
the colour command's own time to the assembly benchmark.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: round-pick construction lines through the scene palette
The axis and centre feedback for round picks (#4610) drew with a raw colour;
on the palette scene it is interned once like the other construction geometry.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: draw faded and see-through bodies from their plans with OIT
A body faded outside the active component, or see-through, left the GPU
tessellation path and was built triangle by triangle on the CPU, so working in
a component of a large assembly rebuilt every other body. Where the renderer
blends see-through surfaces order-independently (GpuTarget::oit, set by the
renderer), a body whose faces share one class is drawn from its plan: chunks
are split by class, opaque passes, AO and the section draw only opaque chunks,
and the OIT pass draws the rest. The sorted fallback keeps them on the CPU.
The model and its highlights decide through one rule (model_plan_class).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Code <noreply@anthropic.com>
* Move the design-link path helpers into one module
`component_link.rs`'s path rules for links to other design files (base_dir,
absolute, relative, stored, resolve, rebased) move, unchanged, into
`cmd/design_link.rs` with their test, so other kinds of link (#3891's drawing
files, linked parameters #5354, #5113's extract) share them instead of copying
or re-exporting them. No change in behaviour.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* design_link: keep base_dir and relative private
Only design_link.rs uses them; the module's surface is absolute, stored, resolve and rebased (review on #5420).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* component_link: import Document after the rebase
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Carel Theron <careltheron@Carels-MacBook-Pro.local>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #6861 eval scripts on input designs
* fix(cli): evaluate scripts on the input design (#6861)
* fix(cli): validate eval options before opening inputs
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6608
* Canvas: keep value boxes clear of the command window (#6608)
* Canvas: describe how value boxes clear the command window (#6608)
* Canvas: keep the value boxes clear of the gizmo handles (#6608)
* ui-egui: format the value box regression test (#6608)
* gpu: the startup MSAA check skips when there is no GPU adapter
The check needs a real adapter to draw the multisampled view and to
resolve it, so on a machine with no software Vulkan adapter it is
skipped, with the reason printed, instead of failing the build.
SOLIDCRAFT_REQUIRE_GPU=1 still makes a missing adapter a failure, as in
the CI job that has a software adapter.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A cut (or intersect) that names no bodies went through every body it met,
hidden ones too: a body hidden to keep it out of the cut was cut anyway. When
some of the active component's bodies are hidden, the cut now stores the shown
ones as its targets, so the hidden bodies stay as they are, and stay so once
shown again (the model never depends on what is shown). With nothing hidden it
goes through every body it meets, as before.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Until a connection had shown the token, its first line (up to 16 MiB)
was parsed whole into a JSON value. A line of [0,0,...] that size is
hundreds of MB once parsed, so 16 connections without a token could
make the app run out of memory and lose unsaved work. The token is now
read from the request object without building the rest of it: other
fields are only skipped, and anything but an object with a string
token is refused as before.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Canvas to Sketch: trace a canvas image into a sketch on it (#3931)
New command canvas.to_sketch, offered as Canvas to Sketch in the canvas
context menu. It traces the canvas image (automatic Otsu threshold, light
drawings on dark backgrounds inverted) with interpolated marching squares
and Douglas-Peucker, and adds lines and cubic Bezier splines through Insert
SVG's add_geometry to a new sketch on the canvas plane, with the canvas'
width, centre, rotation and flip. Holes stay holes; one undo step.
The image's declared width x height is checked before decoding (at most
40 megapixels) and the decoder runs under image::Limits, since canvases
are capped by file size, not pixel count.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Move the new message to a pending locale file
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #2605
* Insert SVG / DXF: place and scale the drawing before it is added (#2605, #3443)
After the file is picked, an Insert SVG (or DXF) window sets the
drawing's X, Y offset and scale, with a live preview; OK adds it in one
step and Cancel adds nothing. The insert commands already took `at`
and `scale`; the window is what was missing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Insert scenario: the picked drawing goes through the placement window (#2605)
`sketch_toolbar_imports_use_file_picker` expected the SVG added as soon as
it was picked; it now checks the placement window first and adds it with
OK.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Insert SVG / DXF: a size in millimetres, an angle, and placement by the drawing's centre
sketch.insert_svg and sketch.insert_dxf take width or height (mm, aspect kept, winning over
scale), angle (degrees) and align (origin or center). The new sketch.drawing_bounds reads a
file's own size without inserting it.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Insert SVG / DXF window: Width, Height and Angle, placed by the drawing's middle
Width and Height in millimetres follow the scale (aspect kept); Angle turns the drawing
about its middle, where X and Y now put it. Uses sketch.drawing_bounds for its own size.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Insert SVG / DXF: handles in the viewport to scale, move and turn the drawing
The preview gets its box with a handle at each corner (scales about the opposite corner,
aspect kept), the box itself (moves) and a knob above it (turns, Shift snaps to 15 degrees).
They edit the same X, Y, Scale and Angle the window shows.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Insert handles: the move that comes with the release counts
A quick drag could stop short of the pointer's last position because the final move arrives
in the same frame as the release.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Insert SVG / DXF: the Drawing Size command's label in the catalogs
`sketch.drawing_bounds` came without a catalog entry, which the i18n
test requires (and the complete ja and zh catalogs with it).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Move the new messages to a pending locale file
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Nathalie Schönwetter <nathalie@cfw-ag.ch>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #2113
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Crash reports: capture native crashes on macOS (#2113)
A signal handler for SIGSEGV, SIGBUS, SIGILL, SIGFPE, SIGTRAP and SIGABRT
writes the same <journal>.native.json as the Windows exception filter: the
signal, the faulting module and offset, the thread and a frame-pointer walk.
The next start reports it instead of "an unclean exit". The signal then goes
on to the previous handler or the default action, so macOS still writes its
own crash report.
native_crash_macos.rs joins the crash-safety skill's list of files that may
use unsafe (no safe API can catch these signals).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5641
* Cone primitive: solid.cone with base and top radius (#5641)
A cone or truncated cone placed like the cylinder: base centre, axis,
base radius (or diameter), top radius (0, the default, is a full cone)
and height, with the new body / join / cut / intersect choice. It is
built from what the kernel already exposes, a revolve of its half
section about +Z through the origin, moved into place. Toolbar entry
after Torus, a Cone dialog that also places it on a picked face, Edit
Feature, a timeline icon, pending UI strings and attribution, and one
pending design-format change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #1932
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Press Pull on a fillet's round face edits the fillet (#1932)
Q on a fillet's round face opened Offset Face, which refuses a face
between two others it is tangent to ("a corner whose faces don't meet
the same way after the move"), so OK stayed disabled. The engine's
solid.press_pull already changes the fillet's size on such a face; the
window now does the same and reopens the fillet (or chamfer) to edit.
Other curved faces still open Offset Face, planar faces Extrude.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6718
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Claim bot: let an unassigned claimant /unclaim (#6718)
GitHub doesn't assign people who aren't collaborators, so many claims
have the `in progress` label and the bot's "Claimed by @user" reply but
no assignee, and `/unclaim` refused the claimant. When nobody is
assigned, `unclaim()` now also accepts the user named in the bot's
latest "Claimed by" reply on the timeline, unless the label was removed
after it. Maintainers and the owner rule are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Add Open in Slicer for the selected bodies (#2595)
One command saves the selected bodies as a 3MF in millimetres (an STL if the 3MF can not be written) and opens it in the last slicer used. It is on the File menu, the Mesh tab and the body right-click menu. The first time, or when that slicer is gone, it asks which one with the same list Save as mesh uses. The Save as mesh dialog is unchanged.
* Send slicer files to a private per-user folder (#2595)
Files sent to a slicer used to go to a shared folder in the temp directory. They now go to the user cache folder (mode 0700 on Unix), under a name that stays the same for the same design and bodies, so sending again replaces the file and the slicer can reload it. The file is written to a temporary name and renamed into place. If the name can not be replaced, as on Windows while the slicer holds the file open, the copy gets a number and the status line says so. Files older than seven days are deleted at startup. The first send tells the user that sending again replaces the open file, and so does the tooltip of Open in Slicer. Send to slicer in Save as mesh uses the same path.
* Open files in a running slicer (#2595)
PrusaSlicer, SuperSlicer, Bambu Studio and Cura are started with --single-instance, so a slicer that is already open loads the file instead of opening a second window. OrcaSlicer has no such flag and is left alone, and macOS bundles are single-instance already. PrusaSlicer 3.0 ships a launcher entry that already carries --single-instance-on-url; it keeps that one and never gets a second flag. It is a preference, on by default, and scripts can pass single_instance. Preferences also gets a button to forget the slicer Open in Slicer remembers.
* List slicers the desktop entries say open 3MF or STL (#2595)
On Linux the scan of desktop entries now also lists a program that claims 3MF or STL in MimeType, even when it is not in our table of known slicers. The table still supplies names, ids, icons and flags for the ones it knows. Entries marked NoDisplay or Hidden are skipped, and so are terminal programs and SolidCraft itself. The existing size and count limits stay. The Windows scan now looks inside a Prusa3D folder, where PrusaSlicer installs itself.
* Only start slicers the user can see or chose (#2595)
file.save_mesh took a slicer parameter that could be any executable path, so a script or an MCP client could run any program. The slicer now has to be one of the detected slicers (by id or name) or a program the user already added under Custom, which the app hands to the engine from its settings. The same rule holds for file.open_in_slicer. The agent chat already asked before any file command; its question now also names the slicer.
* Review fixes for Open in Slicer (#2595)
The file name hash keeps all 64 bits instead of 32. When Open in Slicer has to fall back to an STL, the status line now says so. The engine test that sets the chosen programs clears them again. The unused "could not make the folder" message is gone from the catalogs, and the docs say plainly that finding slicers by file type is Linux only.
* Keep unsaved designs apart and hide Open in Slicer in the browser (#2595)
Two unsaved designs with the same name and body names got the same file in the slicer folder. Every document now has a uid that is new for each New, Open and Recover and stays the same for the life of the document, and the file name uses it (with the process id) until the design has a path. At the first Save As the name changes to the path-based one, so the slicer gets one new file; the old one goes with the seven-day sweep.
The browser can not start a program, so Open in Slicer is now disabled there: the engine command reports why, the File menu item and the body and component right-click items are greyed, and starting it anyway says so on the status line instead of opening a dialog. The Save as mesh dialog keeps its browser guard for the open flow too.
* Put the Session doc comment back on Session (#2595)
The new NEXT_UID static had landed between the doc comment and the struct it describes.
* List every install of a slicer, and test the first Save As (#2595)
Detection kept one entry per known slicer, so a system PrusaSlicer 2.9 and a Flatpak 3.0 beta showed up as one. Every distinct install is listed now. The same program found twice, through a desktop entry and through PATH, is still one. When a slicer has several installs, each gets an id made from where it was found (prusa#comprusa3dprusaslicer) that stays the same on every run, and a name worked out from its command and location without running anything: PrusaSlicer (system), PrusaSlicer (Flatpak, beta), PrusaSlicer (AppImage). A lone install keeps its plain id and name. The plain id or name still picks the first install, so a remembered "prusa" keeps working. Single-instance flags and the other per-slicer behaviour go by the id without the install part.
Also a test that the first Save As changes the file name once and later sends keep the name that comes from the saved path.
* Start PrusaSlicer 3.0 first and send the file once it is ready (#2595)
PrusaSlicer 3.0 starts through slic3r-app-launcher, which ignores a file given at a cold start whatever the flag, and loads it over D-Bus when an instance is already running. A small table in slicers.rs names such slicers by their command; today it holds that launcher, so PrusaSlicer 2.9 and OrcaSlicer are untouched. Ready means a name starting with com.prusa3d.prusaslicer.BroadcastListener. is on the session bus, read from busctl --user list with no new dependency. If busctl is missing or fails, nobody waits and the file is sent as before.
When the slicer is not ready, Open in Slicer starts it without the file and looks every 200 ms for up to 10 seconds. The wait runs on another thread, so the window stays live, and the status line says Starting PrusaSlicer. Once the name appears the file is sent with the same arguments as before. If the limit passes, the status line says the slicer did not respond, that the file is already saved, and to send again. A second click during the wait only repeats the Starting line, so there is one launch and one send. Scripts calling file.open_in_slicer wait inside the command by default, or pass wait: return or no.
* Always show the slicer picker for Open in Slicer (#2595)
File > Open in Slicer, the Mesh tab entry and the right-click entry now always open the picker, with the slicer used last time pre-selected, so the usual case is one click or Enter. Scripts and file.open_in_slicer with a slicer still send without a dialog. The preference button that forgot the remembered slicer is gone: it could no longer make anything ask again, and it did not clear the pre-selection either. The guide says how it works now.
* Leave Bambu Studio out of the single-instance list (#2595)
Open files in a running slicer now adds --single-instance for PrusaSlicer, SuperSlicer and Cura only. Bambu Studio is started as it is, so it may open a second window. A test pins that, for the detected slicer and for one the user picked as a custom program. The guide and the preference tooltip no longer name Bambu Studio as getting the flag.
* Sort the English catalog and leave features.md to the generator (#2595)
cargo xtask lists wanted one message moved in en.json after the rebase. docs/features.md is written by the Generated files workflow on main, so the row I had added by hand is gone from this branch; the workflow adds Open in Slicer when it lands.
* Drop the unused folder message from the Vietnamese catalog (#2595)
Vietnamese arrived on main after I removed ui.save_mesh_temp_failed from the other catalogs, so the check that every translated id exists in English failed on it. Nothing uses the message any more.
* Say so when a started slicer can not be found again (#2595)
After Open in Slicer had started a slicer on its own, a failure to look it up again to send the file just returned, so the status line stayed on Starting and then nothing. It now says the slicer was started, that it can not be found to send the file, and where the file is saved. A test covers it.
* Say why a failed busctl ends the wait (#2595)
If busctl fails partway through a wait, the wait stops and the file is sent, the same as when busctl can not be run at all. That is on purpose: a file sent too early is dropped quietly and costs one more send. A comment in wait_for_ready says so.
* Keep the readiness wait inside its limit (#2595)
wait_until asked the probe before checking the deadline, including right after a sleep, so a yes could be accepted after the limit, and busctl ran with no timeout, so one look could outlast the whole wait. The deadline is checked first now, the probe is told how much time is left, and busctl is given at most two seconds (less near the end of a wait) and is killed after that. A hung busctl counts as not telling, which sends the file as before. This covers the first look on the UI thread too. Tests cover a slow probe, the shrinking time left, and a fake busctl that hangs, answers and fails.
* Send the file that was already written once the slicer is ready (#2595)
When the wait for a started slicer ended, the UI ran file.open_in_slicer a second time. That exported the design again on the UI thread, from whatever the session looked like by then, and could write a numbered copy while the timeout message still named the first file. The wait now keeps the slicer it found and the answer of the first call, and opens the file that was written. A test changes the design during the wait and shows the same file is sent, is not written again, no command runs, and no second file appears.
* Send with the single-instance flag after starting a slicer (#2595)
After Open in Slicer has started a launcher that needs a start first, the file has to reach the instance that is now running, so the send that follows a start carries --single-instance whatever the preference says. A normal send, to a slicer that was already up, still follows the preference. A comment in both places says why. The test uses a plain launcher command with the preference off: the normal send has no flag and the send after a start has it.
* Find the remembered slicer again after an uninstall (#2595)
With two installs the remembered key is prusa#prusaslicer. Remove the other install and the lone one is plain prusa, which matched neither the exact key nor the plain-key fallback, so the picker fell back to the first slicer in the list, maybe a different one. The key is now also matched by the part before the #, so the same slicer is pre-selected. A test covers the lone install, a plain key with two installs, an exact key, and that another slicer is never picked.
* Say a second click did not send anything while a slicer starts (#2595)
While a started slicer is being waited for, another click showed Starting <the first slicer> even when it had picked another slicer or other bodies. It now says it is still waiting for that slicer and that nothing else was sent. The test checks the wording and that it names the slicer being waited for.
* Label macOS installs by their application (#2595)
An open -a command resolved its program to /usr/bin/open, so every application on a Mac was labelled system. Labels for open commands now come from the .app path: /Applications is system and a folder under home is shown as ~/Applications. A test with two PrusaSlicer.app bundles and a made-up home checks both labels and that the ids differ.
* Pick the same place for a program found twice every time (#2595)
When two desktop files, or a desktop file and PATH, pointed at one program, the first one read was kept and its file name went into the id, so the ids could change with the order the folders were read in even though nothing was installed or removed. The place that sorts first now speaks for the program. A test with two aliases and a second install lists them with the folders in both orders and gets the same ids and commands.
* Treat flatpak run spelled two ways as one install (#2595)
flatpak run ... and /path/to/flatpak run ... with the same arguments had different install keys, so one invocation could be listed twice. The launcher is now compared by its file name and the arguments are kept, so both give the same key, while a different branch is still its own install. The test uses a real launcher file so the full-path entry is not skipped as stale.
* Say plainly that a late yes is accepted on purpose (#2595)
The comment on wait_until said a slow probe could not accept a yes that came too late. That was not what the code does: no probe is started after the limit, but one already running may finish and its yes is taken, because the file should still go. The comment now says so, with the bound: the limit plus that last probe, which in production is given the time left and so ends by the limit. A test has a probe that runs past the limit and says yes, and checks it is accepted and the wait stays within limit plus the probe time, and that a probe that says no is not followed by another.
* Give reading busctl the same deadline as running it (#2595)
After busctl exited, the output was read with a join that could wait forever if a process it left behind kept the pipe open. The reader now hands the text back over a channel and the wait uses the time that is left of the same limit, and gives up with nothing otherwise. The polling sleep no longer overshoots the time left either. A test uses a fake busctl that backgrounds a sleep with its output inherited and exits at once: the call returns within its limit.
* Tell two different flatpak launchers apart (#2595)
Comparing only the file name of the launcher made /opt/tool-a/flatpak run ... and /usr/bin/flatpak run ... with the same arguments one install even when they are different programs. The launcher is now found the way any program is, through PATH and with links followed, and its name stands in only where it can not be found. The test has flatpak on PATH and the same file by its path as one install, a different executable of the same name as a second, a different branch as a third, and the fallback when there is no PATH.
* Keep waiting when busctl is slow, send only when it is missing or fails (#2595)
Since the busctl timeout, a slow busctl came back as "can not tell", and that was read as send now. One slow look, likely just when a cold PrusaSlicer 3.0 is starting under load, ended the wait and sent the file early, which gets dropped. The look now says why it has nothing: slow means not up yet, so the wait goes on to its limit and then reports that the slicer did not respond; a missing or failing busctl still means it can not be told and the file is sent. The comments name only those two cases. Tests use fake busctl programs: one that hangs keeps the wait going to the limit and times out, a missing one sends at once, one that says not up and then fails sends, and one that names the slicer is ready.
* Do not open an empty path when the answer had no file (#2595)
If the answer kept for a started slicer had no path, the send ran open with an empty string. The tick now says there is no file to send and starts nothing. The test gives a wait with no path, checks the error line, and that the fake slicer was never run.
* Drop the unused folder message from the Danish catalog (#2595)
Danish arrived on main with ui.save_mesh_temp_failed, which this branch removed from English because nothing uses it any more, so the check that every translated id exists in English failed. The entry is gone from the Danish catalog too.
* Always start the 3.0 launcher with the single-instance flag (#2595)
A hand run showed that starting the PrusaSlicer 3.0 launcher with no file while a 3.0 is already running forwards to it and exits when the flag is there, but opens a second full window when it is not. A slow first look at the bus can start the launcher twice, and with the preference off the flag was stripped. A slicer that has to be started first now keeps the flag on that start whatever the preference says, the same reasoning as the send after a start; when nothing runs the flag does no harm. A normal send with the preference off still has none. The test covers the launcher entry, a bare launcher command, a normal send and a slicer that is not on the table.
* Swap --single-instance-on-url where it does not hand the file over (#2595)
A hand run on the PrusaSlicer 2.9.6 Flatpak showed that its entry's --single-instance-on-url does not forward a file to the running instance: every send opened a new one, three after a cold start and two sends. With plain --single-instance the second process finds the first, forwards, and exits. So when a command carries --single-instance-on-url and the slicer is not on the wait table, that is, not the 3.0 launcher, it is replaced by --single-instance, with the file markers handled as before, when the preference is on, and stripped when it is off. On the 3.0 launcher the flag does forward and stays. The comment on the rule says why. A test uses the real 2.9.6 and 3.0 Flatpak command lines.
* Cut long slicer names to fit the picker (#2595)
With installs listed on their own, a name like PrusaSlicer (Flatpak, master) is wider than its row. The list painted it with no limit, so it ran out of the popup, and the closed selector wrapped it onto a second line. The list now lays the name out on one line that ends in an ellipsis when it does not fit, the selector truncates, and the full name is in the tooltip when it was cut (for a program the user added, ahead of the delete hint). A test checks that a long name is cut to one line within the width and a short one is left alone.
* Keep the 3.0 launcher's flag on a send whatever the preference says (#2595)
A hand run with the single-instance preference off showed the 3.0 launcher needs its flag on a send too, not only on a start: a cold start with a file and no flag loaded nothing, and a send to a running 3.0 left a launcher process that never forwarded. A slicer on the wait table now gets the flag on every launch, as it already did when started alone. The doc comment says why. The preference text, in English, Japanese and Chinese, and the guide now say that PrusaSlicer 3.0 is the exception to turning this off. The tests that took the flag away with the preference off now expect it, for the launcher entry, a bare launcher command and the 3.0 Flatpak line; the test of the send in the UI checks the first send carries it.
* Keep the install-flag swap to PrusaSlicer, found by what it is (#2595)
The swap of --single-instance-on-url for --single-instance ran for any program whose command had the flag, so another program with that flag would have been rewritten. It is now a PrusaSlicer rule: it applies to a slicer recognised as PrusaSlicer by its identity, its id or, for a program the user picked, the name it goes by, never by a version number. The single-instance table uses the same identity check. A test keeps the flag of a program that is not PrusaSlicer, strips it when the preference is off, and swaps it for a picked program named PrusaSlicer.
* Do not promise a hard bound in the wait comment (#2595)
The comment on wait_until said the production probe ends by the limit. Its clock starts after the process is spawned, and killing and reaping the process and scheduling are not bounded by it, so it is a polling and read budget with process overhead on top. The comment says that. Comment only.
* Tell PrusaSlicer 2 from 3 by the installed version (#2595)
PrusaSlicer 2 and 3 need different handling, and the rules had to guess from the command line. Each install of PrusaSlicer now carries its major version, read when it is detected from the release at the top of its appstream metadata and never by running the program: a Flatpak's own files (the branch the command names, else the current one), or the metainfo file beside a desktop entry. PrusaSlicer older than 3 is sent the file with --single-instance in place of --single-instance-on-url, which does not hand a file over there, and is not waited for. PrusaSlicer 3 and newer keeps the launcher's flag, is started on its own, waited for, and sent the file. When the version can not be read, the command decides as before: the slic3r-app-launcher launcher counts as 3, and any other entry is left as it is. A program that is not PrusaSlicer is untouched whatever its flags. Tests use metadata fixtures, no Flatpak: 2.9.6, 3.0.0-alpha10, a branch that is not installed, a system package, an unreadable version with and without the launcher, and a non-PrusaSlicer program with the same flag.
* Make the busctl wait tests steady under load (#2595)
One of them failed once in a full run at its first assertion and passed alone. A script that has only just been written can fail to start (text file busy) while another test forks, and that reads as a busctl that is missing. The fake programs are now started once, patiently, with a flag they answer at once, before a test relies on them; a hanging one uses exec so it is killed cleanly. The time bounds that could trip on a busy machine are several seconds now, and the checks are on outcomes: not ready and keeps waiting, send now, and the number of looks. Thirty runs alone, thirty of the whole slicers filter and thirty of the busctl tests under eight busy loops all pass.
* Say what the busctl reader thread leaves behind (#2595)
The thread that reads busctl output is left to end when the pipe closes. A busctl that answered but left a process holding its stdout reads as slow, and every poll of such a bus can leave one blocked reader behind until that process lets go. A comment next to the thread says so. Comment only.
* Make Save as mesh a plain export (#2595)
Save as mesh asked for the bodies and the format, and could also send the file to a slicer with a checkbox and a slicer list. Sending now has its own command, so the checkbox, the list and the Send label are gone from this dialog: bodies, format, a file dialog, Save. The two settings that only the checkbox used, save_mesh_send and save_mesh_slicer, are no longer kept; a settings file that still has them loads, and they are ignored. The strings only they used are gone from the catalogs. The slicer list stays in the dialog of Open in Slicer. file.save_mesh keeps its slicer parameter for scripts. The guide describes the two as separate.
* Rename Open in Slicer to Send to Slicer (#2595)
The command sends a file to a slicer rather than opening one, and a later Send to App can share the name. File menu, Mesh tab, right-click menus, the dialog button, status lines, the guide and the English, Japanese and Chinese strings now say Send to Slicer, and the command id file.open_in_slicer is file.send_to_slicer. It is new in this change, so there is no alias. The remembered slicer is stored as send_slicer, and the question the agent chat asks says send to. Module and type names stay as they are.
* Put the send targets in one table and one send pipeline (#2595)
The known slicers, the list of slicers with a single-instance flag and the rule for slicers that must be started and waited for were three separate things. They are one table of targets now: id, label, kind, names and program files, accepted formats and the default, how the file is handed over, the single-instance flag, the version from which PrusaSlicer's on-url flag hands a file over, and the wait-for-ready rule. Programs that are not in the table are slicers with the same defaults. The kind is slicer, the only one so far; its comment says it is where a later Send to App would plug in. The send itself is one function: write the bodies, find the target, start it and wait if its rule says so, hand the file over, report. Detection and resolution take a kind. Nothing changes for what already worked.
file.save_mesh keeps its slicer parameter, now through the same function, with its export arguments and result unchanged: a cold PrusaSlicer 3 is started and waited for before the file goes, and it accepts the program path of a detected slicer as file.slicers lists it (file.slicers gives it as program). A path is accepted only when it is the program of exactly one detected target, and that target's own command and arguments run; a launcher several entries share, such as /usr/bin/flatpak, is refused as ambiguous, and any path nobody listed or chose is refused. The parameter and the code say it may be deprecated later in favour of file.send_to_slicer or a Send to App workflow, which is a possibility and not a plan.
Tests go through file.save_mesh itself for the two reported failures, a listed AppImage path and a cold PrusaSlicer 3 that is waited for, with the slicers and the busctl made up; and they check that the result and the file are what an export without a slicer gives, and that a path that is not listed, or is shared by two entries, writes nothing. The pipeline has its own tests: a cold launcher waited for, one already up, wait return, and one that never comes up. A test of save_mesh that shared a file name with another is given its own.
* Keep an AppImage's id through an upgrade (#2595)
When a slicer has several installs, each gets an id made from where it was found. For an AppImage that was its file name, which carries the version, so upgrading it changed the id and the remembered choice missed. The id is now made of the slicer's id, the fact that it is an AppImage and a short hash of its folder, so a new version that replaces the old one in the same folder keeps the id, and the same AppImage in another folder is its own install. Two versions kept side by side in one folder are two installs: while both are there, their file names are added to their ids to tell them apart.
A remembered id is found again only when it names exactly one install. An id that was one AppImage and is now two versions, or a plain id that is now several installs, is treated as not remembered, so the picker starts from the first entry as on a first run and does not quietly pick another install; a single install left of the same slicer is still found, as before. Installs that are not AppImages are told apart as before. An id from before this change no longer matches. Tests cover an upgrade, two versions side by side, a second folder, and the picker's side of each case.
* Add Open with EasyPrint to Send to Slicer
EasyPrint is Printables' slicer in the browser. It's a plain text row in
the Send to Slicer dialog, backed by a new file.open_in_easyprint
command. The command exports the bodies as a 3MF (or STL), hands them to
the host to upload to slicer.printables.com/import, and the app opens the
link that comes back. The desktop app uploads with ureq and the browser
build with fetch, so this is the one send target that works on the web.
The model leaves the machine, so the command refuses without
consent: true, the dialog asks once and remembers, and an AI agent's
request asks first. A model over 100 MB is refused before anything is
sent, and a failed upload shows as an error with the reason.
The chat's privacy note said SolidCraft sends nothing anywhere, which
isn't true any more, so it now says that about the chat. New strings are
in every catalog.
* Place Send to Slicer on the Insert Files panel
* claim: #6170
* Move: after picking a body, go on to From, Point or Axis (#6170)
In Point to Point, Point to Position and Rotate, once a body is picked
and the next input is empty, the dialog goes on to it, so the next click
picks the point or axis instead of toggling another body. Free Move
stays on Bodies: its Pivot is optional.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6070
* Sketch: a changed length dimension keeps its start point, so a side grows from its first corner (#6070)
Changing a dimension's value re-solved its sketch as the smallest move from
the stored positions, which splits the change between both ends: a
rectangle side set from 40 to 60 mm grew 10 mm each way. parameters.change
now first solves a sketch whose length dimension (distance, horizontal or
vertical distance, line length) the parameter drives with that
dimension's first point held, and stores the result, so evaluation has
nothing left to move. Where holding it can't solve (a symmetry moves both
ends), evaluation solves as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5099
* Browser: a feature's eye shows or hides the bodies it made (#5099)
A row in the Browser's Features folder has an eye when the feature made
bodies that are in the design now (a patch, a box, a pattern's copies;
not a fillet, which changes bodies). It runs browser.visibility with the
new features: [id or name] (and path: for one placement of a component
placed more than once), which hides or shows those bodies.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A path a design names (a linked design, a sketch text's font file) now
goes through one rule before any file call: a network location is
reached only once its server is allowed, a device path never. Update
Links and Open Linked Design ask first (Allow once / Always allow this
server / Cancel); Preferences > Design lists the allowed servers and
can allow every network location. component.links reports such a
link's found as unknown with a reason, without looking for it.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A homemade 3D input device's driver app (MAKMOUSE) listens on a loopback
port; with Preferences > Navigation > "Navigate with a device app on this
computer" on (off by default), the desktop app connects, says hello and
reads JSON lines: axis motion (orbit, pan or zoom by mode), zoom steps and
standard views. The motion goes through navigation::device like a 3D
mouse's, only in the focused window; a HID device the app names is left
to it. Documented in docs/3d-mouse.md.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* feat: add personal Hole parameter presets
* feat(hole): prompt for preset names and recall changed selections
Ask for a name only on Save new or Rename in the kit confirmation. Keep Enter and Escape inside the prompt, preserve cancelled drafts, recall only changed selections, and retain independent recessed-hole bottom choices.
* feat(hole): preserve thread direction and partial depth in presets
Carry Left/Partial, explicit source units and Custom draft guards through personal preset capture and recall. Preserve legacy Right/Full/flat defaults and reject unknown options before changing the form. This production bridge depends on the handedness and graphical-depth changes inherited below it.
* docs(hole): describe supported preset thread options
* feat(ui): save modified hole presets in place
* fix(presets): retain reference placement and current locale contracts
* test(ui): borrow restored preset library during serialization
* test(presets): preserve modal focus with viewport focus restoration
* Use UI kit controls for hole preset selection
* Themes: Clay (light) and Clay Dark, warm paper and charcoal with a terracotta accent
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Clay: on-accent text a touch lighter for 4.5:1 on the accent
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6724
* Extrude: the picked region is the one extruded again (#6724)
Since #5530 (#5302) every index or point pick grew to the whole drawn
profile that reference geometry cuts, so a lens or crescent clicked in
Extrude, where the face's projected outline cuts a circle, extruded the
whole circle while the dialog highlighted and counted one region.
Dialogs now send exact_regions: true with their profile picks, and the
command keeps each pick to the region it names. Picks without the flag
(scripts, agents, the CLI) still take the whole drawn profile (#5302).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5345
* Select Other: list what is under the pointer front to back, one body at a time (#5345)
The list was sorted by kind first, so on stacked bodies it showed every
body's face, then every body, mixed across depth. Now each body's rows
stay together (vertices, edges, face, then the body), bodies ordered by
their nearest hit along the pick ray; an edge counts at its point under
the pointer. Sketch overlays stay first, profiles, planes and axes last.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4784
* Part properties: one table of keys, each with its scope, check and label (#4784)
SolidCraft's part-property keys (project, part number, revision, author,
organization, description, licence) were declared once but listed again
by hand in each command, error, dialog and window. They are now one
table, metadata::KEYS: each row has the key, where it may be set
(Scope: the design, a component or both), how its value is checked and
the message id of its label. A new key is one row.
Metadata::set is told whose metadata it changes (Owner::Design or
Owner::Component), and a key is taken where its scope covers that
owner. is_well_known, both set commands, document.metadata,
component.metadata, the refusals, the Design Properties dialog and the
Bill of Materials window read the table; the hand-written lists are
gone (WELL_KNOWN, the label match whose fallback named any unknown key
"License", the window's three labels). metadata::BOM_KEYS names, in
column order, the keys a bill of materials lists for each part:
component.metadata returns them, component.metadata.set takes each as a
parameter of its own and the window edits them; a test keeps each one a
row of the table that a component takes. Free properties use the keys'
text rule (metadata::check_text) instead of a copy of it. The params
docs point to document.metadata's properties instead of repeating the
keys.
Every key accepted before still is, on the design and on a component.
Two things change:
- a refused key or a full metadata on a component says "part property"
("`bogus` is not a part property: …", "part properties are limited
to …"), not "design property";
- changing the design's own properties no longer evaluates the model.
The fields that change no geometry are named once,
Document::with_part_properties_of: the design's metadata, and each
component's metadata and free properties. It compares them first and
copies the design only when one differs, as before.
The shortcut is for a session's own edits only. A command that gives
the session another design (doc.open, file.new, doc.recover: through
Session::replace_design, which now counts them) is refreshed in full,
however little the new design differs from the one before, so its
joints are solved and its occurrence state pruned. Before, opening a
design that differed from the open one only in render settings or
component properties skipped that; with the design's properties in the
list it would have for two revisions of one part too.
Tests: the table (one row per key, the dialog's order); the bill of
materials' keys (rows a component takes, in column order); each scope
on each owner, and refusals that name the owner (the empty key, an
unknown key, a 256-byte key shown cut); the text check at 4096 bytes,
line breaks, tabs and control characters; hostile keys and values
refused without a change; only-part-property changes in each scope,
and changes beside them (units, a component's name, an added, removed
or swapped component); document.metadata.set, remove, undo and redo
evaluate nothing; opening a revision that differs only in its author
solves its joint; the component wording, the shortcuts and the limits;
the dialog's and the window's labels from the table.
What the tests can't reach: every row is Scope::Both, so a known key
refused for its owner (the wrong-owner branch of check) can't happen
yet. It becomes live with the first single-scope row, bom_type, a
component-only key.
Left for later (not this topic):
- Metadata::others() ignores scope, so a component-only key stored on
the design would be neither a dialog field nor listed under other
tools' data. It matters from the first single-scope key (bom_type).
- assembly.bom's rows, its CSV columns and the window's grid headers
still name the three keys by hand; the bill of materials module's
rewrite will read BOM_KEYS.
- The bill of materials types (P1, #5107, into drawings) still use
WELL_KNOWN and the two-argument set: once this is on main, bom_type
is rebuilt on it as one KEYS row with Scope::Component, set through
Owner::Component.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5497
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Sketch pattern: a spacing arrow dragged past zero gives a negative spacing (#5497)
The drag arrow floored an unsigned value at a small positive number, and the
rectangular sketch pattern's two spacing arrows were unsigned, although the
command takes a negative spacing and typing one works. Both arrows are signed
now, like the Extrude, Move and offset ones.
The test drags each arrow back past its start: without the change the first
one stops at 0.001.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Drag arrows: the solid pattern spacing and one-sided sketch offset go negative too (#5497)
The solid rectangular pattern's Spacing arrow and the sketch Offset distance
arrow were unsigned, so a drag back past the start stopped at a small positive
value, although both commands take a negative value and typing one works. Both
are signed now. A symmetric offset uses only the distance's size, so its arrow
still stops at the curve.
Tests: the solid pattern arrow dragged 10 mm back gives Spacing -10 and the
copies on the negative side (without the change: 0.001); the offset arrow
dragged across its circle gives -1 mm and the same circle as typing -1 mm
(without the change: 0.001), and a symmetric offset stays positive.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* Release: sign the Windows binaries and MSI through SignPath Foundation
When the SIGNPATH_API_TOKEN secret exists, each Windows job sends the
three .exe files to SignPath before packaging, then the MSI built from
them, using the GitHub artifact flow from SignPath's docs. The policy
is test-signing unless the signpath_signing_policy dispatch input or
the SIGNPATH_SIGNING_POLICY repository variable says otherwise. Without
the secret the jobs build unsigned exactly as before.
package.ps1 gains -Phase stage|package so signing can sit between the
build and the MSI; the artifact configuration to paste into SignPath is
packaging/windows/signpath-artifact-configuration.xml.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* SignPath: drop the description attributes the Foundation plan doesn't allow
SignPath Foundation projects get an automatic signature description (its watermark);
the artifact configuration may not override it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Double-click a shown sketch dimension in the model view to edit its value (#7079)
A finished sketch's dimensions shown in the model view (Show Dimension) now take a
double-click: the same inline value box as sketch editing opens at the text, Enter
changes the dimension's own expression through parameters.change (one undo step), Esc
cancels. The sketch is not edited and the view does not move. Driven dimensions stay
read-only with a status hint.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Control channel: ui.selection lists the shown dimensions of finished sketches
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Test: a shown dimension that breaks a later fillet reports it, and Undo restores both
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #1457
* Agent chat: tool calls read as summaries, raw JSON behind Details (#1457)
Each tool-call card now says what the call did, built from the command's
translated label, its main parameters (lengths in the design's units) and the
bodies the result names: "Extrude Sketch1 by 10 mm", "Box 40 x 30 x 20 mm,
then Fillet 4 edges (r 4 mm)", "Measured Block: ...", "Saved block.solidcraft".
Calls it can't put in words keep the agent's title. The raw arguments (now
kept from the in-app client and ACP's rawInput) and the result sit under a
folded "Details".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat cards: show file names, never paths; temp files not named
ui.screenshot reads "Screenshot"; save/export/open and other tools' titles
show only the file name. The full text stays under Details and in the tooltip.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #5681
* sketch: exact circular arcs on drawn 3D wires (#5681)
Add WireArc (centre, unit normal, radius, start/end) as an optional analytic
form on drawn 3D sketch wires, with arc_from_3_points, arc_sweep and
sagitta-bounded arc_polyline; add_drawn_arc_in stores the arc, its three
defining fit points and its samples; move_drawn_point re-derives the arc
atomically and refuses collinear points; validate_wires checks saved arcs.
* engine,ui: sketch.arc3d exact 3D arc command (#5681)
sketch.arc3d draws an exact circular arc through three world points in a 3D
sketch, stored analytically beside its fit points and samples; sketch.move3d
rebuilds it, sketch.inspect reports centre/radius/sweep; placement mapping
(map_wire, now shared with the engine's world state) transforms the arc.
One pending design-format file records the new field. The viewport's 3-Point
Arc tool draws a 3D arc when clicked off the sketch plane, with a live arc
preview; the SKETCH panel offers sketch.arc3d.
* ui,engine: 3D Arc button starts an interactive tool (#5681)
sketch.arc3d now maps to a Draw(3) tool handled by sketch3d: three world clicks (start, end, through - same order as the 2D arc tool) run the command, with the live arc preview and snapping. The command is only enabled while a 3D sketch is open, so the toolbar never offers a tool that could only fail. Locale labels for the command.
* ui: cargo fmt
---------
Co-authored-by: RayGalt <RayGalt@pm.me>
* claim: #1383
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Viewport tests: the hover pixel test keeps its pictures when it fails (#1383)
The Metal hover pixel test failed once in a loaded full-suite run and has
not been reproduced since. On failure it now saves the plain and hovered
GPU and CPU frames and a mask of the pixels that do not light, and names
the folder in its message. The assertion is unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4384
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: shared module, with Rest as the first tool
The Rest handles spike becomes a shared module, place_gizmo, so other features placed on a face can
use the same marker and knob. The mechanics (crosshair with its three states, the grab offset, the
footprint while dragging, the turn knob) live in place_gizmo.rs. The handles own the pointer through
canvas::owns_pointer, as the move triad does, so navigation and Select Other's press-and-hold leave
them alone. Each tool is one file under place_gizmo/ that gives a Spec: which dialogs it serves,
which commands' Position input also takes sketch points, where the place is, how the pointer
re-places it, its footprint and its turn, if any. The module lists the tools one per line, so tools
added on separate branches each add one `mod` line and one list line. Rest is the first.
The handles are always on; the preference the Rest spike had is gone. The POINTS accept flag on a
Position input is granted from the tools' Specs instead of a Rest-only arm in the plastic dialog's
start. Handles are published as `rest:position` and `rest:rotate` as before.
Seen straight down, the dialog's height arrow collapses onto the place and its value box sits just
right of it, where it took the press meant for a knob pointing that way. The knob now slides out
along its ray until it is clear of the box.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Placement handles: Boss
The marker moves a boss across the model, also from Edit Feature, and its Position input takes
sketch points while the handles are on, so a screw boss can be dragged onto the centre of a
circle drawn for it. A boss is round, so it gets no knob; the footprint while dragging is its
outer circle. Published as `boss:position`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Placement handles: Snap Fit
The marker moves a snap fit across the model and its Position input takes sketch points while
the handles are on, so the arm can be put where a sketch says (#3423). The knob points the
catch: it turns between the four directions the Catch Direction combo offers for the face, so
the knob and the combo always agree; the command keeps only those four, so there is no free
turn. The footprint while dragging is the arm's foot, its thickness behind the place and its
width across. Published as `snap_fit:position` and `snap_fit:rotate`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Placement handles: the look in one file
Everything that draws the handles (the crosshair and its three eased states, the knob's ball, the
cased lines, the angle label, the colours) moves from place_gizmo.rs into place_gizmo/look.rs, so a
restyle touches one file and the mechanics (hit radius, drags, knob geometry) stay where they are.
No change in behaviour.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: behind a Design preference, off by default
The handles sit behind Preferences > Design > Placement > Placement handles (`placement_handles`,
also through `ui.prefs`), off by default as the hole placement work in #2912 is: a Rest, Boss or
Snap Fit is then placed by its click alone, its Position input takes faces only, and no handle is
drawn or published. The tests switch the preference on; one checks the default.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: the tests hide the origin through set_origin_shown, as main does since #1849
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: the move marker as designed in the prototype, drawn with theme tokens
The marker is a dot with four arms to the screen cardinals: short and dotted at rest, longer with open arrowheads when pointed at, solid while dragging, each state growing out of the previous one (360 ms). Sizes are the prototype's. Its core is the icon highlight and its casing the marker halo, the tokens main's hole handle uses, so the literal allowance goes. The tip shows while pointing, not during a drag.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: the dialog's drag arrow stands above the move marker
With the handles on, the Height arrow started at the marker's centre and ran through its upper arm, head on arm tip. The arrow now keeps the marker's full reach and a gap free at the base, its shortest length grows by as much, and the base dot is left to the marker. Other dialogs are unchanged (clearance zero).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: a turn ring on the face replaces the knob
The ring lies on the face about the move marker's centre, drawn in perspective, a fixed 65 px on screen like the move triad's arrows (it is a control of something on the screen, not a measure of the model). At rest it is a short dotted arc with an open chevron grip just outside; pointed at, the dotted circle opens from the grip both ways round and the grip slides in and fills; while turning, a solid arc around the grip. Dragging the grip turns the feature as the knob did, through the same handle name, so the harness tests stand.
The floating value box yields: with a handle's centre under it, it moves to the other side of the arrow's tip or above it, since seen straight down the box sat exactly where the grip is. A handle the arrow's stem crosses keeps the pointer.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: a compass while turning, and the turn pulled onto the face's compass points
Four ticks just outside the ring at the face's in-face world directions (the ones the Snap Fit combo names, +X, −Y, …) appear while turning and stay put, so you can see how far you have gone. A Rest's turn is pulled onto one of them within 8°, otherwise it steps by 5° as before; Alt or Ctrl frees it. The readout beside the grip names the point it sits on ("90° · +X") and keeps clear of the ring.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: the footprint in the preview's blue, the idle handle steps back
The footprint drawn while the marker drags is a filled shape in the preview's own blue with a thin darker edge, not a white outline, so white stays reserved for what can be grabbed. While one handle is dragged the other fades to a quarter and comes back when the drag ends.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: the ring in the preference text, the docs and the tests' wording
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: the look as tuned in the app
Marker: drawn to the redrawn design (dot r 3.9 pt, two dashes per arm, a filled head 8.7 × 7.1 pt whose base is where the outer dash began); pointed at, the outer dash becomes the head (its far end narrows to a point, its near end widens); dragging runs the inner dash on into the head as a shaft. Every casing is 0.4 pt outside the edge, for lines, heads and the dot alike.
Ring: the grip and the compass lie in the ring's plane like the circle, so the grip is never skewed against it; the grip is a thick chevron with a hole that flattens into the base as the grip slides onto the ring, and fills; the dotted circle stays while turning, with the solid arc laid over it; opening eases in and takes 0.3 s, leaving is a 0.15 s fade over the resting ring, not the sweep backwards; a drag keeps the angular gap of the press, so the grip does not jump onto the pointer; the compass ticks and the readout's point name are in their axis colours, and the grip takes the axis colour (eased) while aligned with one, with that point's tick left out; free turning keeps a hair of pull (0.5°) so an axis can still be hit; the readout wraps to a half turn.
Value box: with a ring it keeps out of the circle the ring, grip, compass and readout sweep, one place per view, nothing moves when a turn starts; also when the value does not evaluate and there is no arrow.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: the readout shows a half turn as 180°, not −180°
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: the motions in one place
The animations the handles grew (a factor eased in and out each way on its own time and easing, a handle stepping back while the other works, a colour that comes and goes from the last one it was on, the gap a press keeps for its drag, an offset that slides instead of jumping) move out of place_gizmo into motion.rs, so the next manipulator reuses them. The look's timings become Transitions in look.rs. No change in behaviour: the contact sheet before and after is identical, the harness tests stand.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Placement handles: the axis colours as main names them since #4430
The compass and the aligned grip take the one set of X, Y and Z colours (axis_x, axis_y, axis_z); the Move gizmo's own array is gone.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* claim: #3133
* Apply complete feature transactions off the window thread
Evaluate dependent history and prepare reference and GPU display caches before accepting a guarded result. Preserve the evaluated original model for safe editor cancellation and journal only accepted commands.
* fix(ui): preserve original marker across sketch-to-feature edits
* test(ui): verify edit-sketch cleanup and guard Unix link fixture
* test(ui): fix texture disposal and unreachable assertion in feature tests
* fix(build): compile CLI link inspection only on Unix
While an agent's batch computes off the window thread (#1989), the chip reads
N/M · <command>. It is centred on the navigation bar, just above it, and moves up
past any toast or panel it would cover.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Axes and points from circular edges and round faces
In 3D, a round edge or face stands for its circle wherever a command
takes an axis or a point.
Revolve, Surface Revolve, both Circular Pattern dialogs, Plane at Angle
and Move -> Rotate take a circular edge (through its centre, along its
normal) or a cylinder, cone or torus face (its axis). The features keep
the pick as a model reference, so they follow the model when a parameter
moves it. Straight edges work as before; flat faces are refused.
Every 3D point input takes a circle's centre: construction points (which
follow the circle), To Object, patch points, hole placement, and Move's
pivot and points. Hovering shows the axis as a dashed construction line
or the centre as a star, and the dialog's chip names what the pick
stands for ("centre of edge 10", "axis of face 3").
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Format the round-pick hover check
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Selection chips: name a round pick as a circle centre or axis
The chip has room for little more than the body and a word or two:
"Shaft · circle centre", "Shaft · circle axis", "Shaft · face 2 axis",
and "Shaft · point" for a point of the model.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Move: a circle's centre in a placed component is where the component shows it
Move's pivot and points read a picked circle from the model in its
component's own frame, while the pick is in the world. They now read it
from the world state (and, editing a Move, from the state before it,
placed the same way), as the viewport and the other round picks do.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Plane at Angle on a planar face about a model edge: test that it follows both
Upstream #1956 made a planar face a base for Plane at Angle; this branch makes a model edge an
axis. The rebase keeps both in one PlaneRef::AtAngle.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #1397
* External STEP corpus: nightly import checks on real-world files (#1397)
tests/external-corpus.toml pins 20 real-world NIST PMI STEP files (AP203
geometry-only and AP242 editions from several vendor exporters, public
domain) by archive and per-file SHA-256 with the licence recorded.
cargo xtask external-corpus downloads each source once into the gitignored
corpus/external/, verifies every hash, imports each file headless
(file.insert_step + inspect.measure) and checks body count, total volume
(0.1 %) and area (0.5 %) within tolerances and time and memory budgets.
A file whose upstream hash changed is reported as 'source changed', never
as our failure. --bless "why" re-records expectations in the manifest with
a dated reason. The External corpus workflow runs it nightly.
assets: skip corpus/ when scanning for attribution (downloaded case data
is never committed).
---------
Co-authored-by: Dahak Builds <dahakbuilds@users.noreply.github.com>
The blue, green, red scale's lightness rose and then fell (L* 44 up to
72 and back to 50), so a flat face and a tight radius looked alike with
hue lost, and its upper part merged for a red-green loss. The stops are
now dark blue, a middle grey-gold and light yellow in one straight
blend, as measured in the issue: L* rises the whole way and every
quarter step stays clear for normal, protan, deutan and tritan vision.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: a Weldments ribbon tab (#4003)
- WELDMENTS joins the ribbon after SHEET METAL and is built the same way:
Sketch (New Sketch), Create (Frame Member) and Configure (Parameters,
Configurations). Frame Member's home moves there from MODEL > Create.
- Always shown, as a frame starts from a sketch of its lines. It selects
itself and takes the contextual style when the selected body (or the last
body of the active component) is a frame member; sheet metal comes first.
- It uses the default accent until the palette has a colour for it.
- The tab label is in every catalog; docs/ribbon-layout.md, docs/frames.md
and the tab list in docs/control-protocol.md follow.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: Weldments keeps the ordinary tab look (#4003)
Weldments still selects itself when a frame member is the active body, but
no longer takes the solid contextual fill or the 2 px accent line under the
tab row: selected, it looks like a selected Model tab. Only Sketch and Sheet
Metal (on a sheet body) are contextual.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Expressions: a number and its unit are one value (10 mm / 2 mm is 5)
A unit written after a number was multiplied in at the precedence of `*` and
`/`, so `10 mm / 2 mm` was (10 mm / 2) mm = 5 mm² and `90 deg / 1 rad` a value
in angle², reported as a unit error. The number now takes the unit after it
before `*` and `/` apply; a fraction of numbers before a unit stays one value
(`1/2 in` is half an inch), and an exponent stays a plain number (`2^3 mm`).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Expressions: after a division, a fraction before a unit only when closed up
`width / 10 / 2 mm` read the `10 / 2 mm` as the fraction 5 mm and gave
width / (5 mm) = 8 instead of (width / 10) / 2 mm = 2. After a `/` the
fraction rule now needs the fraction written without spaces (`1 in / 1/2 in`
is still 2); elsewhere `1/2 in` and `10 / 2 mm` are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
?open= now also accepts an absolute path on the page's own origin, fetched
with the page's scheme, so a self-hosted setup that serves SolidCraft and
other tools under one origin can link a design on a LAN without a public
HTTPS address. Absolute URLs keep the HTTPS-only rule (a URL on the page's
own origin is also accepted), redirects must end on HTTPS or the same
origin, and protocol-relative, dot-segment, backslash and control-character
values are refused.
* Patterns: one before the feature it copies follows that feature's parameters (#6726)
A pattern or mirror of features made with the timeline rolled back comes
before its sources and replays them from their definitions. Its
fingerprint chained only the features before it and its own inputs, so a
change of a source's parameter left its result in place: the box grew,
its copy kept the old length, also after rolling forward.
The chain fingerprint of a pattern or mirror of features now also takes
the fingerprint of each feature it replays (and of their sources, for a
pattern of a pattern), wherever they are in the timeline.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Patterns: a test for a later feature taking a deleted source's name (#6726)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Patterns: fingerprint the replayed sources with the document, as fingerprint now takes it (#6726)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3538
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Updated AppImage takes the release's file name (#3538)
A versioned AppImage name (solidcraft-<version>-linux-<arch>.AppImage) is
replaced by the new release's name, so zsync-based tools keep finding the
update data. Other names are kept.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
A sketch dimension read Ø40.00, with nothing to tell millimetres from
inches, while the value inputs and the pointer's coordinates show the
unit. Length labels now end with the design's unit (Ø40.00 mm, 1.000 in,
(25.00 mm) for a driven one); angles keep their degree sign.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
AGENTS.md drops from 16.8k to 3.7k characters: the hard rules, each naming the
check that enforces it, and a table of what to read for each kind of work.
The rest moves into eleven new skills; the five existing ones stay.
cargo xtask skills fails when AGENTS.md or a SKILL.md cites a path, an xtask
command or a skill that does not exist, when the router table and skills/
disagree, or when the router or a new skill grows past its size cap. It runs
in cargo xtask ci and the repository checks job.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With a section shown, the marking pass decided where the cap shows (the stencil) and wrote the
depth everything after it tests against. #1768 gave it the faces' FACE_BIAS for the second job,
so edges draw over their own faces (#1434). For the first job its slope part is unfair: next to an
edge where a back face seen through the cut meets a front face, the front face won a band about a
pixel wide although the back face is nearer. The cap skipped that band, and the edge line showed
through the cut.
#6363 marked those samples again from the back faces, pulled toward the eye as far as FACE_BIAS
pushes. That closes the gaps beside the edges, but the pull also lets back faces win wherever they
lie within the push of a front face, along every body's far silhouettes, and the cap hatches that
band. On bherbruck's block, the band away from the cut went from 58 px orthographic and 534 px
perspective before #6363 to 618 and 780 with it. In the app, his filleted hub showed it in every
frame of a 36-step orbit in both projections.
This replaces that pass. While a section is shown the opaque model now draws in four steps (`Cut`):
1. mark: the faces by their true depth (no push): colour, the stencil and depth;
2. a triangle over the viewport sets the depth back to the far plane, colour and stencil kept;
3. push: the faces again with FACE_BIAS, depth only, so the depth is what unsectioned frames have
and edges draw over their faces as before;
4. cap: depth only, as #6728 made it; the mark pass has drawn the cap's colour and hatch, as fs_solid
draws a back face seen through the cut.
The window pass's depth reset and this one are one helper, `depth_reset_pipeline`, for either
depth direction, and its passes are built by #6728's `solid_writes`. Unsectioned frames are unchanged. The CPU rasterizer already chose the cap by
true depth.
Tests (crates/gpu/tests/depth.rs), with the device's float depth and the 24-bit fallback:
- a_cut_hides_the_edges_behind_it (new): bherbruck's block cut at y = -16, home view, orthographic
and perspective. No edge shows through the cap, and away from the cut the picture is a box of
just the kept part's: 0 px orthographic and 1 perspective (main: 618 and 780).
- bherbruck's pocket and hub tests from #6363 are kept and pass (pocket 0 px; hub 1 and 2 px over
its twelve views, as with #6363's pass).
- an_edge_on_a_face_draws_whole_at_any_angle and
an_edge_just_behind_a_face_stays_hidden_on_a_large_design also run with a section shown.
In the app (safe mode, WARP), with 625 patterned cylinders cut by a section, the frame time is
that of main (#6363): 33.5 against 34.3 ms at the home view, 34.1 against 35.2 ms with the model
filling the view.
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6720
* Face pick by point: linear work budget, so bodies with many faces take a sketch (#6720)
A face picked by its point alone charged the whole mesh once per face, and
face_contains scanned every triangle for each candidate. A block with a
modelled thread (about 5 000 faces) went past the 2 000 000 limit. The
triangles are grouped by face once, each face's test reads only its own,
and the budget charges triangles + faces.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3812
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
* Sketch Dimension: value box in the design's units (#3812)
A dimension added without a value restated what it measured as
"<value> mm", whatever the design's units. In an inch design the
Dimension tool's value box therefore opened at "25.4 mm" next to a
dimension whose text read 1.000 (inches).
Document::design_length writes a measured length in the design's units
("1 in"), with enough decimals to keep it to the micrometre, so a
dimension that restates the geometry still moves nothing. The Dimension
tool, AutoConstrain's dimensions and Toggle Driven (making a driven
dimension driving) use it. Millimetre designs are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
* Move the design-units dimension tests into sketch_ux_tests (review)
The #3812 tests go in the existing sketch dimension test file rather than
a new file of their own (AGENTS.md, #3226).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The evaluation cache's 64 MB budget charged every cached feature result the whole model, though a
state shares every body it did not change with the state before it. A design whose model is a few
MB kept only its last few dozen results, so setting a parameter back recomputed what had dropped
out. Each result is now charged for the bodies and sketches it does not share with the state it
started from.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6550
* Crash reports: keep journaling through file.new and doc.open (#6550)
`Session::replace_design` swapped in the fresh session without its
`crash_journal` flag, and the app turns it back on only at its next frame.
So `file.new`/`doc.open` never recorded how they ended, the commands sent
after them in the same frame (a control-channel or MCP batch) were not
journaled, and a crash report then named `file.new` as the command in
flight. `replace_design` now keeps the flag, like the other session-wide
fields it keeps; only the design on screen has it set.
Test: `commands_right_after_a_new_design_are_journaled` (crash_tests.rs).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6917
* Tests: the six timing assertions do not depend on the wall clock (#6917)
Six tests asserted on the wall clock. Their result depended on how loaded
the machine was and not only on the code under test: they passed on an idle
machine and failed when the test binaries ran beside each other, which is how
the suite runs (`cargo xtask test` starts every test binary at once).
CONTRIBUTING.md:196 says it: "Tests run beside each other on a busy machine,
so a test must not depend on wall-clock timing."
The change is confined to the six test files the issue names:
- geom: the tangent chain of a helix is a ratio (750 pieces against 3 000)
instead of a limit of 200 ms.
- io: the placement budget is spent inside the reader, so the clock is gone.
- ui-egui: the classification of many faces is a ratio (900 against 3 600)
instead of a limit of 10 s.
- engine: the clearance measurement drops its 30 s limit; the boolean that
gives up reports itself.
- engine: the fillet beside a modelled thread is a ratio (a 1 mm thread
against an 8 mm one) instead of a limit of 30 s.
- ui-egui: the window frame budget goes from 250 ms to 20 s, the limit #6072
gave its own probe.
The helper is the one #649 wrote, copied into the crates that need it, with
the noise of a run as a parameter.
* claim: #6558
* Shortcuts: Ctrl+Shift+S opens Save As in the workspace (#6558)
The Home page already opened Save As with Ctrl+Shift+S, but the design workspace matched Ctrl+S, so the chord saved over the current file instead of offering another one. The workspace now routes the chord to Save As, and the File menu shows that shortcut next to Save As, taken from the command registry.
* claim: #6600
Co-authored-by: Cursor <cursoragent@cursor.com>
* Save an imported mesh that only fits after compression (#6600)
The design file limit was applied to the mesh before Deflate, so a mesh
that had already loaded could not be saved or autosaved.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix: validate and version large mesh saves (#6600)
---------
Co-authored-by: Colin Faulkingham <colinfaulkingham@Colins-Mac-mini.local>
Co-authored-by: Cursor <cursoragent@cursor.com>
* claim: #5679
* Top-level cut and join reach bodies in sub-components (#5679)
An extrude, revolve or press-pull from a sketch on a face of a body in a
component inside the active one now cuts or joins that body, moved or
turned, in its own component, so every placement shows the change. The
automatic operation probes that body where it is shown. A linked part, or
a face on a further placement of a part placed more than once, is refused
with the reason instead of a silent no-op (#5678).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6423
* test: a copied body must keep the look of the body it came from
solid.move with copy: true leaves the copy without the appearance, the
material or the face looks of the source body, so it renders with the
default look. The tests fail for that reason: the copy has no entry in
appearances.bodies, in materials or in appearances.faces.
* fix: a copied body keeps the look of the body it came from (#6423)
A move with `copy: true` pushed a new body but left it with the default look, so the copy came out a different colour than the part it came from. The copy now carries the body's appearance and its material, as the component copy already did.
* test: keep the copy regression to the body look and the material (#6423)
* test: each copied body keeps the look and the material of its own source (#6423)
Two bodies with different appearances and different materials, copied in one solid.move: each copy must keep the look and the material of the body it came from.
Scope of the fix: the look of the BODY and its material are carried over to the copy. The per-face looks are NOT carried over: a face look is keyed by the point on the face, so the copy's faces would have to be matched and their points translated, which is a separate feature.
A drag that isn't live walks to the cursor in steps, each a held solve
on its own budget seeded on the last solved state. Past a failed step
the walk keeps trying every step (a stretch the geometry can't follow
can be followed by one it can), and each failing solve runs the
numerical solver to its limit: one drag in a 65-command sketch took
7.7-9 s, most of it in failing solves.
A failed step leaves the sketch as it was, so the steps after it all
start from the same state and none changes what the next one sees.
They are now solved eight at once on all cores; those that fail in a
row are taken as tried and the walk goes on at the first that doesn't,
as it would one by one. The drag above takes 3.7-4.3 s and every
command's reply is unchanged. Live frames, whose steps share one
frame's budget, are left as they are.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6569
* Sketch: an offset slot can still be moved by the dimension that places it (#6569)
An offset arc is drawn round its source's own centre, and its ends are
held while the sketch solves (the offset is rebuilt from its source
afterwards). Its arc condition, both ends as far from the centre, then
only tied that shared centre, and the source arc with it, to the middle
of the held ends: a slot offset by 1 mm lost a degree of freedom, and
the length that places it was refused as conflicting.
The solver and the constructive reading now leave out the arc condition
of an offset arc round its source's centre (Sketch::offset_arc_follows_source),
as #4466 did for the Parallel between an offset line and its source. The
source's own arc keeps the centre; nothing changes without an offset.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch: a test for a rounded rectangle with an offset round it (#6569)
kaievns's case from the review: the offset's corner arcs tied each corner
centre to a 45 degree line, so the height followed the width and every
width change was refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6841
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Sheet metal: a pattern of a cut made in the flat is kept like the cut (#6841)
A rectangular, circular or path pattern of an extrude cut through sheet metal was refused
("this sheet machining cannot be preserved in its flat material"), though the cut itself is
kept as a cut-out of the flat pattern. The pattern replayed its copies through the generic
boolean, whose sheet check refuses any change the sheet model can't keep; the cut goes through
the sheet's flat machining first.
Each copy of an untapered extrude cut is now machined into the sheets the same way, in its
moved sketch plane (a moved copy keeps the extrude's prism certificate), and then cut from the
other targets. A copy that meets a folded bend is refused as the cut is there. Mirrored copies
(the sketch plane turns over) and patterns of an intersect (which cut with the union of their
copies) keep the generic path.
Fixes#6841.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5098
* Sketch: deleting part of a link breaks off the rest instead of rebuilding it (#5098)
Deleting some of a link's curves left the rest linked. The next refresh saw a
different layout and rebuilt the whole link as normal geometry: the deleted
curve came back and the others lost their construction flag (a projected arc
returning as a solid arc, a projected face's lines turning solid).
sketch.delete now breaks the link for what remains, as Break Link does: those
curves stay where they are and keep their flags. Deleting a whole link, and
text links, are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FN5eEMrqWWXqi65FcK8GXq
* Move the #5098 tests up in sketch_project_tests.rs
They sat at the end of the file, where main has since added other
tests; moving them next to the break-link test avoids the merge conflict.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: sascha2101 <221609842+sascha2101@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5853
* Batch: refuse a step that would fail at its turn before the first step runs (#5853)
engine.batch and the MCP batch tool checked only a batch's shape up
front. A misspelled command id in a late step was found at its turn,
after every earlier step had run, and then all of it was rolled back: a
160-step batch with a typo as its last step ran for a minute and
changed nothing.
solidcraft_engine::batch_preflight::check finds, before anything runs,
the first step the batch would certainly refuse at its turn: a step
without a command, an unknown command id (the error names the closest
ids, as at the step's turn) or, with strict parameter checking, a call
that breaks its command's typed contract. Anything that depends on the
design is left to the step's turn.
- engine.batch (control channel) checks against the app's registry and
its session's level; view.pick, which the app answers itself, passes.
- The MCP batch tool checks when the commands run in its own process
(headless, with --param-check from the backend). A running app may be
another version with other commands and checks the batch itself; its
reply's not_run is passed on.
- The refusal is the failing step's reply with completed 0, undone 0,
not_run listing every step and a note.
Tests that used an unknown command only to make a late step fail now
use a step that fails at its turn (a bad extrude distance), so they
still test rollback, timings and file.new as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FN5eEMrqWWXqi65FcK8GXq
---------
Co-authored-by: sascha2101 <221609842+sascha2101@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6211
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Joints: a loop whose axes are a hair out of parallel still moves (#6211)
Holes picked on parts are rarely exactly in line. A closed loop through
axes a few hundredths of a microradian apart (a lever, link and a rod on
a cylindrical axle) can't close exactly: it closes to nanometres, and the
gap grows as it turns. The loop solver wanted 1e-7 mm, so it threw such a
closure away (the loop stayed open, its parts at 0 dof) or stopped a drive
or drag after a fraction of a degree.
The conflict check is now 1e-4 mm / 1e-6 rad (it was below what its
message can show), and joint.drive / motion and joint.reach close the
loop at each step as at the start: to the tight tolerance where they get
there, otherwise within the conflict check.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2175
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Body rename: an imported assembly's other bodies keep their names and components (#2175)
An Import keeps its body names by the file's body index, and builds no body for the
file's later instances of a repeated part (`skip`). Renaming a body stored the names by
the body's place in the model instead, so every name after the first skipped instance
landed one slot early (and the new name could land on a skipped slot). The rename looked
right until the design was rebuilt; then those bodies took other names, dropped out of
their components and showed up loose in the import's top component.
`Feature::body_name_slots` maps the bodies a feature builds to their stored slots, and
`Feature::pin_body_names` stores current names through it. The rename, the pinning of
later features' names (rename and suppress) and `Document::body_rename_from` use them.
Fixes#2175
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6348
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch: constrain to an edge of any body, a mirrored copy too (#6348)
A constraint tool ignored a click on a body edge, so only edges already in
the sketch (the face it sits on) could be constrained to; a mirrored copy's
edges never were. Constraint commands now take a model edge or vertex
({edge|vertex: [x,y,z], body?}) for an operand and project it first, in the
same undo step, and the pick tools pass a clicked edge or vertex that way.
The link keeps the edge's persistent name, so the constraint follows rebuilds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6646
* Control channel: document.revision, a design revision and a document id (#6646)
A program that follows the open design asks document.revision and exports
only when the answer moves: document_id names the open design (kept through
edits, saves and renames, another for each design started or opened, never
shared between two tabs or two runs) and revision moves whenever the design,
what is hidden or an uncaptured component position changes. It stays put on
reads, selections and view changes, unlike the session's view revision.
The revision is read lazily: the session remembers the design it last
reported (held weakly, so Arc::make_mut can never change it in place and its
address can't be reused) and takes a new number from one counter for the
whole run when the design is another one. The control method answers at once,
also while a slow command computes; engine.execute reaches the same command
for scripts and MCP.
* document.revision: hash the run's id prefix with the process's random keys
A clock set before 1970 gave every run the prefix 0, so a restarted app could
answer a new design with the id and revision of the old one. The docs no
longer promise an answer at once: a screenshot waiting for an opened design to
be meshed still holds the requests behind it.
* document.revision: its label in the complete catalogs (ja, zh)
* claim: #6985
* Control channel: control.subscribe and design.changed events (#6985)
A connection that sends control.subscribe gets the active design's id and
revision in the reply (as document.revision answers), then a
{"event":"design.changed","params":{document_id, revision}} line whenever
either changes, so a live link need not poll.
The window publishes the design at the end of each frame's logic and
drawing while a connection is subscribed (one atomic load otherwise). Each
subscribed connection waits on that from a thread of its own and writes
the latest after a 100 ms gap, so a burst of changes is one line. Replies
and events share one lock per connection, each written whole (#6390).
* control.subscribe: the reply before any event, a second one refused
The events thread started before the subscribe reply was written, so a
change right after the window's answer could reach the client before the
reply that named the older revision. The subscription now starts holding
the connection's line lock and writes the reply under it.
A second control.subscribe on a subscribed connection is refused on the
connection itself: its answer could arrive behind a newer event. When no
thread or socket is left for the events, the reply says it did not
subscribe instead of ok.
* control.subscribe: a second one does not lose a finished late reply
The already-subscribed answer came before the busy check, so a request that timed out and had finished since lost its reply: control.pending then said the app ended it without one. It now comes after the busy check and drops a finished late reply as every other request does.
* control_server: declare the subscribe guard at the top of the loop
Clear of the lines #6643 changes (its token check moves to where the declaration was).
* control_server: the subscription's items beside serve
Away from the lines next to send, where #6643 adds token_of.
* control test: New on a blank design reuses its tab, so redo the box first
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #617
* fix(sketch): manual Horizontal/Vertical between a circle centre and an edge (#617)
The manual H/V tool silently did nothing when a point and a straight line
were picked or selected together (only two points or a lone line worked),
while auto-constrain inference applied the alignment. A point picked or
selected with a straight line now aligns the point with the line's
midpoint, in the mode the line's dominant axis suggests, through the
existing midpoint alignment in sketch.constraint.horizontal_vertical.
- tools: constraint_preselection consumes [point, line] selections; the
interactive tool applies the alignment on the second pick (a line).
- context menu: Horizontal/Vertical Constraint items for point + line.
- engine + ui regression tests pin centre-edge alignment and the
edge-to-centre dimension from #617.
Written by Qwen.
* style: formatting cleanup for #617 files
---------
Co-authored-by: Dahak Builds <dahakbuilds@users.noreply.github.com>
* Extend: join a stop that is already another curve's end
sketch.extend refused a target that coincided with another curve's end
("the extension target coincides with existing geometry") instead of
extending onto that point. The stop is now joined: a new line ends on the
existing point, and a free end grows onto it held by a Coincident. The
preview shows that stop too, so the line it draws is the line the click adds.
A stop at the end of a spline or conic joins that end the same way, and
extending an infinite line now refuses instead of moving one of its two
points.
🤖 Generated with Codebuff
Co-Authored-By: Codebuff <noreply@codebuff.com>
* Extend: close an arc, take an equal stop, preview with the command's search
An arc with nothing on its path now grows all the way to its own far end and
closes into a whole circle instead of reporting that there is nothing to
extend to.
Two curves stopping the extension equally far are both on the way, so the first
stop is taken instead of refusing to choose between them.
The line preview and the command now share one stop search (`line_aim`), so the
preview can no longer draw a line the click refuses.
---------
Co-authored-by: n0tfr3shm1lk <n0tfr3shm1lk@ukr.net>
Co-authored-by: Codebuff <noreply@codebuff.com>
* claim: #1994
* plastic.snap_fit: base_height and new_body options (#1994)
Base Height extends the snap fit's base below the picked face; New Body
makes the snap fit its own body so it can be patterned or moved. Both
optional, serde-defaulted, old designs replay identically.
* ui: snap_fit length-field defaults list gains the optional base height
---------
Co-authored-by: Dahak Builds <dahakbuilds@users.noreply.github.com>
* Browser: a feature row edits its profile sketch, suppresses and rolls back
A sketch button on feature rows with a profile sketch, and the feature menu gains
Edit Profile Sketch, Suppress/Unsuppress and Roll Back Here. Features behind the
history marker or suppressed are drawn faint, as in the timeline.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Browser: the sketch button sits right after the feature's name
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6545
* cli_link: the Unix-only link code and its test build on Unix only, so the UI crate builds on Windows (#6545)
slot() is used only by link_in's Unix version, and links_only_where_it_may
makes its own link with std::os::unix::fs::symlink; gate both like link_in.
On Windows the test failed to compile (E0433) and clippy -D warnings flagged
slot as dead code.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6755
* Frame rate readout stays beside a right-side pane, not under it (#6755)
The readout and its details sat in the viewport's bottom-right corner even under the
floating Assistant. They now move just left of any pane over that corner and come back
when it closes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Every view attached to an API-agent message stayed in the history as a
base64 data URL, and the whole history goes with every request (up to 24
per message). Requests grew by one screenshot per message until the
endpoint refused them with HTTP 413, after which every message failed.
Keep the newest two views as pictures and swap older ones for a short note;
the text sent with them and every other message stay. HTTP 413 now says the
conversation is too large for the endpoint and to start a new chat.
Fixes#2298
Main now speaks the Anthropic protocol too, where a view is an `image`
block rather than an OpenAI `image_url` part: drop_old_views trims both,
before trim_history. ui.ai_http_too_large is in every catalog that has the
other ai_http_* messages.
Co-authored-by: Framewrk CI <ci@framewrklabs.ai>
* claim: #6396
* Sketch: a regular line on a construction line is picked before it (#6396)
Coincident sketch curves tied on pick distance and the one drawn first
won, so a construction line drawn first hid the regular line on part of
it. Construction and centre lines now get half a screen point added to
their pick distance: on a tie the regular curve is hovered and picked,
and a construction line clearly nearer the cursor still wins. Clicking
again and Select Other still reach the construction line.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FN5eEMrqWWXqi65FcK8GXq
---------
Co-authored-by: sascha2101 <221609842+sascha2101@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
wgpu's default uncaptured-error handler panics ("wgpu error: Out of Memory").
The headless picture devices already report instead; the window's device had
no handler. Errors the code expects stay under error scopes.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
A line tangent to an arc at the arc's end had no sense: the tangency held
with the line on either side of its touch point, so pinning a line's far
end across the corner left the line cutting back through its fillet, and
when the solver found the mirrored corner the arc kept its stored
direction and became its three-quarter complement, inside out.
The solver now keeps each fillet an arc of its corner. A line tangent at an
arc's end keeps its sense there (it leaves the arc as it did) and the arc
its side of its chord, compared with the positions before an interactive
move or the stored ones. A corner that mirrored gets its arc on the
mirrored side: its ends swapped where the solve ran the arc round the
other way with the lines following, or the arc reflected across the line
that stayed and solved again where a line cut back through. A solve that
would loop the arc the other way round, or fold a line onto it, is refused
as a fold is. A drag's cursor-following steps keep the fillets too, so a
line dragged through its own fillet stops at the arc.
The kit slider adds a real `egui::Slider`, lets egui paint it, then paints an
opaque rectangle over it and draws the kit's track and thumb on top. That mask
was exactly `slider_width` wide, and egui places its handle so the circle's
right edge lands on `rect.right()` at the maximum, so the fill stopped flush
with the handle and its last column survived as a light sliver beside the
value box. The kit's own thumb cannot hide it either: its radius is
`t.space_8` while egui's handle radius is `input_radius`, about two pixels
larger.
Extend the mask one pixel past its right edge. The gap before the editor is
`item_spacing.x`, so nothing else is covered, and only the maximum was
affected: at the minimum the handle's left edge coincides with the mask's left
edge, which a fill does cover.
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
* claim: #6524
* Sketch display: each sketch samples within its own budget (#6524)
After a large DXF was inserted, lines drawn in a later sketch vanished
once finished: only their end points showed and could be picked, and
hiding the DXF sketch brought them back.
The display cache had one sampling budget for every shown sketch (4096
curves, 131 072 vertices per model state and view). Sketches are drawn
and picked in timeline order, so a large imported drawing used it all
and every later curve came back empty (BudgetLimited): not drawn, not
pickable, while the points, drawn separately, stayed.
Each sketch now has its own budget with the same limits (curves,
profiles, vertices and nodes). The cache as a whole is capped at 65 536
curves and 16 384 profiles, so memory stays bounded however many
sketches are shown.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FN5eEMrqWWXqi65FcK8GXq
---------
Co-authored-by: sascha2101 <221609842+sascha2101@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6657
* Context menu: the Appearance flyout closes when the pointer moves to another row (#6657)
Hovering Appearance opened its quick-choice flyout, but resting on a plain
row of the same menu never closed it: only another submenu row replaced it.
Now a flyout closes once the pointer moves onto another row and rests there
for 0.3 s, so a diagonal path to the flyout across a neighbour keeps it open.
A flyout opened with the keys stays open while the pointer is still.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A feature's face points are moved onto their named faces before it is
evaluated, so a To Object extrude follows its target face when an earlier
feature moves it. A pattern's or mirror's copies were built from the
feature as stored, with the face where it was picked: after a gear's disc
was made thicker, the tooth reached the new top but its pattern copies
kept the old height. The copies now use the same moved points; the move
is one helper used in both places.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Test: sketch Intersect keeps its exact cuts, crosses curved faces on the surface, cuts on a top face
Intersect's cut of a box, a cylinder, a sphere and a tilted cylinder, as it is on main, pinned.
Two cases where it is not exact: a plane through a cylinder's rim cuts the mesh's chord (its sides
come out at 6.3159 and 6.3085, not the square root of 40; a hole's walls at 25.0122), and a plane
on a body's top face finds nothing ("projects to nothing on this sketch plane").
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Sketch Intersect: crossings on curved faces lie on the surface; a plane on a top face cuts it
Problem: the plane cut that sketch Intersect makes of a body's mesh (project::section) took a
crossing on an edge where a curved face meets another (a cylinder's rim on its cap) from the
merged normals, which are undefined at the crease, so it fell on the chord: a cylinder's sides came
out 0.01 mm off and uneven. A plane through a vertex or on a face counted the vertices on it as
below, so a plane on a body's top face crossed no edge and found nothing.
Solution: the crossings are found by section_segments, which section fits as before: an edge where
a curved face meets another takes the curved face's normals, so the crossing lies on its surface;
a vertex within 1e-9 x the mesh's size of the plane is on it, on one side for every edge (below
when the mesh lies on or above the plane, else above), and an edge ending there crosses at that
vertex, which is an exact point for the fit. Each segment keeps its B-rep face, for a section that
groups them by face. Cuts that were exact are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The first test in each GPU path shard compiles the path pipeline cold
on lavapipe (335-502 s on 16 vCPU, 665 s on 8, against the 600 s setup
budget); the later ones read Mesa's disk cache in about 30 s. Put that
cache on a sticky disk that main commits, so the first test reads it
too until the shaders change.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A change computed off the window thread ran here with only the results of the evaluation
ahead's last evaluate primed. A parameter change evaluates twice (before and after its sketch
stores the new solution), so its first evaluate here found nothing to take, and on a design whose
states pass the evaluation cache's budget it recomputed the features on the window thread: the
window froze for seconds after Computing ended. The run now also takes what the evaluation ahead
kept (KeptResults, as an agent's control requests already do).
The catalogue entry Quantity served both the pattern dialogs' number of
instances and the motion study's plotted quantity. Nine locales had
translated the second meaning, so the rectangular pattern dialog showed
e.g. "Größe" above "Anzahl 2" in German.
The motion study's row now reads motion.plot_quantity (a pending file
that keeps every locale's previous word there). Quantity takes each
catalogue's own ui.quantity text, which the "Quantity 2" row already
shows; nothing is newly translated. The glossary lists Quantity under
words with more than one meaning.
Fixes#6538
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
On 8 vCPU the first merge-queue run after #7066 failed: one gpu path
test went past the 600 s pipeline-setup budget (665 s; 204 s for the
same test on the PR run) and shard 2 took 35 minutes. lavapipe compiles
on every core, so the shards move to 16 vCPU, and four count partitions
(27 tests each) replace three hash ones (41/40/27).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The 108 gpu path:: tests run one at a time (gpu-path test group) and took
about 36 of Native Linux tests' 45 minutes in a merge-queue run. They now
run in a GPU path tests job, three hash partitions on 8 vCPU, beside the
other tests; Native Linux tests leaves exactly them out. A `gpu` output of
Select checks drives both the job's `if` and the `ci` check.
The Windows jobs now run in a merge group only when its diff reaches what
they exercise (the pull request patterns and the Cargo.lock graphics-stack
check), not on pushes to main, and still nightly and by hand.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The README and docs/downloads.md ship a universal macOS disk image, but the
contributing guide listed only Windows and Linux.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Docs: on Windows, give claude mcp add the CLI's full path (#5895)
The Windows installer puts solidcraft-cli.exe in C:\Program Files\SolidCraft but not on PATH,
so the README's and docs/mcp.md's `claude mcp add solidcraft -- solidcraft-cli mcp` registers a
server that can't start there. Say so next to the command and give the full-path form.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Docs: word the Windows full-path note so it holds once the installer sets PATH (#5895)
#5898 makes the MSI add its folder to PATH. Say "if a terminal doesn't find solidcraft-cli
(older Windows installers don't put it on PATH)" instead of stating that the installer never
does, so the note stays true whichever change lands first.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Docs: the Windows CLI path is the default, not the only place (#5895)
The MSI installs to ProgramFiles6432Folder and lets people choose the folder
(WixUI_InstallDir), so C:\Program Files\SolidCraft\ is only the x64 installer's default; the
x86 MSI on 64-bit Windows lands in C:\Program Files (x86)\SolidCraft\. Say "by default" in
the README and docs/mcp.md (review on #5896).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
SolidCraft runs through XWayland whenever an X display is available. A
compositor that doesn't let X11 apps scale themselves draws the window at
100 % and stretches it. docs/nix.md said so only for fractional scaling,
but GNOME 47 (RHEL and Rocky Linux 10) does it at any scale: the report
shows 1.05 physical pixels per point at 200 %.
The troubleshooting page gets a section with the cause, the Window system
preference, SOLIDCRAFT_WAYLAND=1 for 0.21.0 and earlier, and the file drop
trade-off; docs/nix.md links to it.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The component colour tests no longer assume a design always carries a list of
pending format changes; once the release folds them, there is none.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4214
* Viewport: the grid fades with the distance instead of cutting off (#4214)
The ground grid stopped at a fixed distance from the view, so its edge showed
as a cut across the picture. It now fades out with the distance from the
middle of the view, and it is gone before either of the two lines that used to
cut it: its own edge, and the picture's far plane.
Each grid line is drawn in pieces, and each piece is as faint as the fade
makes it at the distance where it lies. The 'under' batch is drawn by the
overlay pipeline with alpha blending, so the colour's alpha reaches the
frame.
* The headless grid fades with the distance too (#4214)
The headless path (solidcraft-cli snapshot, the home-page thumbnail, the MCP
screenshot) drew the grid from the engine's palette, where a line had no
alpha: the grid still ended in a cut, and at a low camera angle the cut was
in the picture.
A SceneLine now carries an alpha, the CPU rasterizer blends with it and the
GPU capture passes it on, so the engine's grid can fade. The fade itself
moves to the engine, next to the grid's other geometry, and both paths use
it: the headless grid now reaches as far on the ground as the view is wide
(the ground stretches as it is seen at a slant) and dissolves into the
distance instead of ending in a line.
* Fix the headless/viewport grid fade divergence (#4214)
The engine stretched the grid's size with the slant of the plane it is seen at, the viewport stretched it only on the ground, so on a sketch's plane the picture's grid reached further than the window's (153.42 mm against 120.00 mm). Both now share one grid_ext, which the picture and the window draw the same grid from.
* The grid fades out before the picture's far plane cuts the ground (#4214)
The window's grid faded against a far plane of its own: the scene's radius
clipped by the projection's camera (260 mm away at 400 mm of distance), while
the picture itself cuts the ground at the far plane of its own projection
(223.7 mm at the same distance, because the projection frames the scene's
bounds and not its radius). The grid was still a third of its strength where
the picture cut the ground, so the cut showed as a straight line across the
view.
The fade now ends at the far plane the picture's projection uses, and the
grid is gone before that plane cuts the ground.
The test that watched the fade now measures the ink of each row of the
picture, and checks that the grid is already gone in the ten rows above the
row where the far plane cuts the ground: the point it compared before lay
beyond that plane, where both renders draw nothing at all, so the comparison
was always true.
* claim: #5734
* Docs: don't offer winget install before the package is in winget-pkgs (#5734)
The README and docs/downloads.md said Windows users could run
`winget install bherbruck.SolidCraft`, but winget-pkgs has no manifests for
bherbruck.SolidCraft (manifests/b/bherbruck doesn't exist), so winget finds
no package. winget.yml only updates a package that is already listed, and
does nothing until then. Drop the command from both pages, and say in
docs/releasing.md that the first version has to be submitted once by hand
and the command added back when it's listed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Releasing: a first WinGet submission is already open; take it up (#5734)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(ui): finish the open sketch when a modeling feature starts
* fix(ui): wait for the marker roll before starting a feature from a sketch
* fix(ui): start the feature once the marker is back
* test(ui): keep the sketch open in the palette clearance test
Saving into Documents fails with the Windows dialog's "File not found" when
Controlled folder access blocks the app, and the message gives no hint.
Refs #4479.
* claim: #4640
* Docs: AppImageLauncher 2.x can't start the AppImage; update to 3.0 (#4640)
AppImageLauncher 2.x intercepts every AppImage through binfmt_misc and
can't start static-runtime AppImages such as SolidCraft's: it fails with
"execv error: No such file or directory" after a FUSE message. 3.0 starts
it. One paragraph in the Linux AppImage section says so.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Docs: AppImageLauncher 3.0 is so far only published as betas (#4640)
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
GitHub Pages reports the base URL as http:// for the custom domain, so the
.flatpakrepo file, the remote-add line on the repository page and the mirror
of the previous repository all used plain HTTP. Force https:// once where the
base URL is read.
* docs: two notes for agents and contributors (the gate's working tree, and linking a picture)
* docs: how to show a picture without attaching it
The reviewer asked for the pictures to be attached to the PR (or in a gist),
not hosted on a branch of a fork, so the note no longer teaches the branch
trick.
* docs: say how to attach a picture, and what a gist is for
The review of this pull request pointed at two things in the note on
evidence:
- Attaching works in the browser for anyone who can comment: drag the
file into the comment. The limit is the command-line agent, whose
upload needs write access to this repository. Say that, instead of
claiming that a contributor without write access cannot attach a file.
- A gist suits a log or another text file; it does not render a picture
inline. Say what it is for, so nobody hits that dead end.
The note in AGENTS.md about the gate reading the working tree is
unchanged.
* claim: #4851
* User guide: local models with the Assistant (#4851)
A page on choosing and setting up a local model (Ollama, LM Studio, or any
OpenAI-compatible server) for the Assistant: tool support, VRAM, permissions,
the measured results from the standard test in #4851, and troubleshooting
for the common failures.
* Guide: local models table with the full set of measured results
Point to the `<file>.bak` copy every save keeps, the design.solidcraft in crash
reports, the recovery folder for unsaved designs, and how to attach the file
to an issue.
"Start every session here" sent readers to plan/STATUS.md,
plan/execution-plan.md, plan/architecture.md and
plan/adr/0001-geometry-kernel.md. plan/ is gitignored, so nobody who
clones the repository has those files.
Point the steps at what ships instead: ROADMAP.md, CONTRIBUTING.md's
"Where help is wanted most", "How the code is laid out" and "Kernel
work is paused" sections, docs/architecture.md, the help wanted, good
first issue and needs-triage labels, and the licence rules with
deny.toml and cargo xtask licences --check.
Signed-off-by: Ray <raykholo@users.noreply.github.com>
* Roadmap: a short user-facing ROADMAP.md, developer detail moves to docs/roadmap-detail.md
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Roadmap: feature milestones with progress counts, daily releases through the in-app updater
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Roadmap: count issues as pieces, show work built on the drawings and render branches
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The Joint dialog kept a round edge (a hole's rim), a straight edge and a
corner as plain points where they were picked. When the part got thicker,
a hole moved or the part got longer, the origin stayed where it was and
the jointed part overlapped the other one. An edit that left the picked
spot where it was (lengthening an arm away from the jointed hole) looked
fine, which hid it. Faces and hole walls were already found by name.
The dialog now sends a round edge as its circle's centre with its z, and a
straight edge or corner as an edge pick with its z. The engine keeps them
by name (the round end by its wall and the face across its rim, the edge
and corner by their edge and face names), with the frame the pick showed;
where it can't (no planar face beside the edge, no names), it keeps the
point as before. Edge origins get an optional axis (along the edge, or a
fixed direction) so a hinge on an edge keeps turning about it.
Tests: every joint type and as-built, for each way the dialog picks an
origin, with the arm made thicker, its hole moved and the arm lengthened,
inside one design and through Update Links.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Move and Align on placed components: picks and vectors where they are shown
A body in a component placed away from its own frame (an imported part
set upright) moved the wrong way: Move's vector, axis and pivot were
taken in the root's frame, so moving the upright plate up moved it along
its own frame's z. They now turn with the frame of the bodies moved.
Align mapped each pick into the frame of the body under it, so aligning
two separately placed parts mixed two frames (the plate landed at right
angles), and its snap points were never mapped at all. Align now finds
its picks, face normals and snaps where the view shows them and takes
both ends into the moved bodies' frame.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Preview: a feature's tool is drawn where its component is shown
Extrude, revolve and the primitives preview their tool on its own. That
preview was built in the model's frame, so on a component placed away
from it (an imported part set upright, edited with its component
active) the tool showed at right angles to the picked face, though the
feature itself came out right.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Follow main's lazy body frames and first-part grounding in the placed-component work
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Cache assembly browser data between unchanged frames
Extend #1762's cached component membership, placements and viewport hover
feedback with cached occurrence labels, folder entries, row keys and
tooltips. Skip offscreen row work and reuse browser hover buffers while
reading selection, visibility and locking live. Cache #2681's standard-part
configuration labels with the document snapshot instead of formatting
them on every redraw. Preserve #3009's live selection and hover marks
with highlight keys owned once by cached entries, and retain #2431's
feature rows in the reusable drawing order.
Related to:
- tangency-labs/solidcraft#662
- tangency-labs/solidcraft#1762
- tangency-labs/solidcraft#2681
- tangency-labs/solidcraft#3009
- tangency-labs/solidcraft#2431
Co-authored-by: Codex <noreply@openai.com>
* Browser cache: lock state moved to the session visibility
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Browser cache: reveal a placement's body by its row, and drop a needless borrow
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Pipe: an Operation row, so a pipe can join, cut or intersect (#2820)
The engine's solid.pipe takes an operation, but the Pipe window had no way to
choose one, so a pipe always made a new body. The window now has the same
Operation row as Loft and the primitives, and editing a pipe keeps its
operation.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Fit the Pipe operation row to main: sketch-qualified pick, length_input pattern
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #4846
* Assistant: say so when a turn ends without using SolidCraft's tools
When tools were offered and a completed turn made no tool call and no
change, the chat panel shows a short notice pointing to tool-capable
models: always for an empty reply, and for a text-only reply to an
explicit edit request. Plain answers to questions stay quiet. The
control channel exposes it as turnWarning.
* Fix the Anthropic loop test for the tools event; add ja and zh for the no-tool-call notice
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Every new command comes with input and output schemas
A test fails for a command without a contract or an output type, unless it is a planned!
placeholder or listed in schema/unconverted.txt, which holds the 400 commands that predate
the rule and only shrinks: the test also fails when a listed command has its schemas, when
the list names a command that doesn't exist, or when it isn't sorted.
`checked!` now takes only the command's id and checks the contract that id publishes
(`run_checked_as`), so the contract enforced is always the one described. Another test runs
every command that has a contract in strict with an unknown key, so a contract is never
published without `checked!`, under the command's own id, enforcing it.
A new book page, Command schemas, explains how to write a contract and a reply type, the
tests to add, the shared field types and how to add one, with a checklist; the failure
messages point to it. "Adding a command", CONTRIBUTING.md and AGENTS.md now include the
schemas, and the MCP tool description and docs/mcp.md no longer list the converted commands.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* List the commands added on main since: view.layout, view.history_layout, view.normal_to
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Command schemas guide: what participants and targets do, checking the evaluation, warning-free replies
- `participants` changes the result for some features and nothing for others; where it and
another key override each other, or where it has no effect, say so in that key's
description or in the summary.
- The example primitive's `targets` default to every body in the feature's component, not the
bodies it meets; they're ignored when `participants` lists bodies.
- Check what each value does where the feature is evaluated, not only how the handler reads it.
- A reply test runs every form in Warn and checks for no `param_warnings` (Strict for a reply
that isn't an object).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* List the 43 commands added on main since: frame, component, render and others
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* List the 11 commands added on main since: browser folding, standard views, parameter sweep, path flange
browser.collapse_all, browser.expand_all, inspect.parameter_sweep, sheet.path_flange and the
seven view.* standard views (back, bottom, front, isometric, left, right, top) landed on main
without schemas; the gate flags them, as intended. 457 commands listed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Schema rule warns instead of failing for now
A command without input and output schemas prints a warning listing it
(test output and a CI annotation). SCHEMA_RULE_STRICT in coverage_tests.rs
(or SOLIDCRAFT_SCHEMA_RULE_STRICT=1) makes it fail again. Docs say "should,
and will become required".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #3965
* Parameters: a click can correct an invalid name (#3965)
Refusing the new row was putting focus back on the field being left, so the correction was typed into the expression.
* Parameters: the invalid-name test types its expression draft (#3965)
A new row starts with a blank expression, so the regression test types 10 mm and checks that correcting the name leaves that draft in place.
* claim: #4146
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Viewport theme: a light 3D view in a dark window, and the reverse (#4146)
Preferences › Interface › Theme gets a Viewport row under the theme tiles:
Follow theme (the default, as before), Light or Dark. Light or Dark draws
the 3D view in the built-in Light or Dark theme's colours whatever the
window uses, so a dark window can show the model on a light view and the
other way round; it works with every theme, theme files included.
The view takes the background and grid, bodies, selection and hover, the
origin, previews, sketches with their dimensions and glyphs, the navigation
cube and triads, box selection and the analysis colours. Panels, dialogs,
menus and the chips over the view keep the window's. Text and warnings drawn
straight onto the view (the sketch status line, the loose-end and midpoint
markers, the section's cut outlines, the 3D sketch tool and the sketch tools'
previews) take the view's side, so they read on its background.
theme::ViewportTheme is the setting (saved as `viewport_theme`: theme, light
or dark; an unknown saved value reads as Follow theme, as ColorVision's
does). Tokens::get() takes the view's colours from the other side only when
the view's dark or light differs from the window's; Tokens::view() is the
view's side for colours drawn onto it. The ground shadow follows the view
(the scene and highlight caches already key on the colours themselves),
ui.render draws in the view's colours, and ui.prefs takes `viewport_theme`
and reports `viewport_dark`. An older release ignores the setting and draws
everything in the theme.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Viewport theme: follow the single set of axis colours
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #754
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
* Sketch Dimension: arc length from an arc and its end point (#754)
The arc-length dimension existed (sketch.dimension type arc_length) but
the Dimension tool could only give an arc its radius. An arc and one of
its own end points, in either order, now dimension the arc's length:
click the arc, then its end. The pair was refused before ("cannot
dimension that combination"), so no existing result changes. A point
where the arc's end is (a coincident line end) counts as that end.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch: hovering a circle's edge wakes its centre
A circle's or arc's centre was hard to take with Line or Dimension: it is a small
point in the middle of empty space, taken only within 8 points of the pointer, and
nothing showed where it was. Hovering a circle or arc of the sketch being edited
now wakes its centre: drawn (a cross in a ring) and taken from 16 points away, so
the pointer can go from the edge across the inside to it. Hovering another curve
or point puts it back to sleep, as does leaving the sketch.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* A woken centre's extra reach yields to a curve under the pointer
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Browser: surface bodies get the surface icon in the Bodies folder (#4165)
Every row in the Bodies folder had the solid body icon, so a surface (an
open shell, such as a Surface Revolve of an open curve) looked like a
solid. A body whose shell does not close now shows the existing surface
icon; solids keep the body icon.
Body::is_surface walks every shell's edges, so ModelBody caches it once per
body (shared by clones, like outline_bounds) instead of the browser
checking it every frame.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Add the surface icon to the requested-icons manifest
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Look At in the viewport context menu for a planar face or plane
Fixes#4105
* Look At tests: turn the camera without the animation
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4846
* Warn when Ollama models lack tool support (#4846)
Check Ollama model capabilities in the background and cache results per
endpoint and model. Show an advisory warning for models without tools,
expose it through ui.agent_chat, and document the behavior.
Cover capability parsing, warning selection, background probing, caching,
and selection changes with synthetic regression tests.
* Keep the Ollama capability probe out of the web build
The panel never calls it there, so it would only be dead code.
* Wrap the model warning instead of cutting it off
In a narrow panel the end of the sentence, the part that says what to do, was truncated.
* Fix AgentConfig default after rebase onto main
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Move the Ollama no-tools warning into a pending locale file with ja and zh
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
solid.loft already closes to a point (a section {sketch, point}), but the Loft
window's Profiles input took profiles only, so a pyramid or cone tip couldn't
be made from the UI. The input now also takes a finished sketch's point, sent
as an apex section in pick order; profiles of one sketch still make one
section.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2348
* Open linked designs from assembly browser
* claim: #2348
* Test linked design action with multiple components selected
* Fix linked design lookup in browser storage and translate its menu action
* Add Turn Up Axis to complete UI catalogs
* Keep browser DXF download after linked-design checks
* feat: add colored components inspection toggle
* feat: persist component colors and add cycling controls
* feat: assign component colors in creation order
* fix: update component color menu integration and translations
* fix: proper description for pending format change.
* fix: deleted duplicated format change description from docs/pending.
* Compare the component colour counter in the features-only check
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Vadim Kalinsky <vadim@kalinsky.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch: a click picks the sketch's own curve over a projected one under it
* Select Other tests: follow origin visibility rename
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Own-curve pick test: click away from the lines' midpoints, which are picks of their own now
* Use the select-other test helpers main already has
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5543
* Extrude: a profile drawn twice is cut once, not as two coincident tools (#5543)
A circle drawn twice at one place is two profiles of one region. Extruded
together they were two coincident tools: a cut through a 121-hole grid with
four such circles retried the boolean with the tool nudged (four times) and
took 25 s instead of 0.5 s.
merge_regions, which every extrude, revolve and surface feature runs on the
regions it builds from, now leaves out a region whose boundary repeats an
earlier one (within 1e-7 of its size; corners and segment middles on the
other loop both ways, holes alike). The selection itself keeps its
multiplicities, as profile_refs' tests require.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6186
* Timeline: a reorder evaluates the design once, from the cache it already has (#6186)
timeline.reorder checked the move on a copy of the model, then dropped it,
so setting the reordered design evaluated the moved part of the timeline
again; reorder_many (and moving features into a group) checked on an empty
model, recomputing the whole design even for a move near the end. The
check now starts from the session's own model, and the session keeps the
model it was checked on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Marker is Copy: drop the clone
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Search Console lists docs.solidcraft.io/, /introduction.html (the same page) and /guide/mcp.html as duplicates without a user-selected canonical. The Pages build now adds a canonical link to every book page (index.html and introduction.html both name /), skips the redirect stubs that already have one, marks 404 and the toc frame noindex, and writes a robots.txt pointing at a generated sitemap.xml.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5849
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01G1b41qKhyhmZoiAFYt9YwX
* Context menu: a right-click on a face of a selected body keeps the selection (#5849)
A right-click selects what is under the cursor first, so the menu is about it. The check
for "already selected" compared the pick with each selected item by identity, so a face
of a selected body never matched: right-clicking a face of a selected body replaced the
whole selection with that one face, and the face menu opened instead of the body menu.
A face, edge or vertex of a selected body now counts as part of the selection, whether the
body is selected on its own or through an occurrence. The right-click then leaves the
selection alone, as it already did for a face in a multi-selection of faces (#3733).
selection::same_item is unchanged: dialogs and Select Other compare picks with it, and
there a face of a selected body is a different pick.
Fixes#5849.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01G1b41qKhyhmZoiAFYt9YwX
* Pass the held keys in the selected-body context menu test
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Move: placements of one part moved together each go the way asked (#4768)
A part placed more than once (an imported assembly's corners, screws or
fan blades) is one body in its component. Moving several of its
placements together moved that body once per placement listed, each
placement turning the move its own way, so the copies flew apart.
When solid.move names one part through two or more placements, it now
moves those placements in the world, as occurrence.move does with space
world: a placement held by a rigid group (an imported assembly) moves with
the outermost placement holding it, one inside another being moved isn't
moved again, and root-component bodies listed are moved as bodies, all in
one undo step. Other moves are unchanged, and so is a move with copy.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Free the grounded first part before the placement-move tests move it
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Test: a text's words follow its text parameters, and so does its cut
A sketch text whose words are an expression of text parameters (`Label + '-L'`): a parameter edit
and a configuration row that sets the parameter relabel the text, the text's cut follows, and the
expression is saved with the design.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Text parameters: a sketch text's words built from them follow every change
A design's parameters are numbers; a text's words (a part label, a size code) could only be typed.
Text parameters hold words: `parameters.add` or `parameters.change` with `unit: "text"` (the
expression is the text, quotes optional), and `sketch.text` and `sketch.edit_text` take
`text_expression` (quoted texts and text parameters joined by `+`, `Label + '-L'`) instead of
`text`. The text is laid out anew from its expression on every evaluation, so a parameter edit,
a configuration row or a script relabels it and the features built on it (a cut, an emboss)
follow in the same rebuild. The design file keeps the text parameters and each text's expression
(one pending format change); a configuration table's parameter column can be a text parameter;
the Parameters dialog lists them under the unit `text`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Text parameters: the joints' same-but-features check compares the text parameters
The joints are solved once per change unless the design differs only in its features
(same_but_features, #6372), which names every Document field. The text parameters are a
field of their own and change what a text's words evaluate to, so two designs that differ
in them are not the same but for their features.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
`cargo clippy --workspace --all-targets -- -D warnings` fails on Rust 1.97: the
reflection radiance bound in crates/render/src/raytrace/image.rs (#5494) trips
`clippy::manual_range_contains` (`v < 0.0 || v > 1e12`).
It now reads `!(0.0..=1e12).contains(&v)`. The two differ only for NaN, which the
`is_finite` check in front of it already refuses, so nothing else changes.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Project: don't lay a second copy over geometry the sketch already shows
A sketch on a face already holds the face's border, linked. Projecting that face
again, or one of the border's edges, added a second set of curves and points over
the first: on a rounded corner that shows as extra points and lines. Project now
skips a reference whose curves are all linked into the sketch already (lines,
circles, arcs), reports it as `already_projected`, and refuses a pick of nothing
new so it adds no undo step. Points on their own (the origin, a vertex) still
project.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Projecting the face a sketch already shows adds no second copy in the face scenario
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #976
* One set of X, Y and Z colours for every axis (#976)
The origin axes, the navigation cube's triad, 3D sketch triads and joint
origins drew `axis_x/y/z` (255,0,0 / 0,255,0 / 2,2,248), the view triad and
the headless picture drew `triad` (220,60,60 / 60,170,60 / 60,100,230) and
the Move gizmo drew `gizmo` (225,70,60 / 80,190,80 / 70,130,240). The
`triad` and `gizmo` tokens are gone: every axis reads `axis_x/y/z`, now the
view triad's values (the engine's `view::colors::AXIS_*`, so headless
pictures are unchanged, and the set the red-green table was measured for).
Older theme files' `triad` and `gizmo` lists still read and set the axes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5860
* Doc: stop the names of split faces from growing with every split
* Joints: the long-piece-name test takes the longest name a piece now gets
Split-face names are bounded at 512 bytes since #5860, so the test cannot
build one over 512; it takes the longest the history gives.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Docs: refresh every UI screenshot, GIF and video from the current UI
Re-record the README demo, the rover build, the sample plate and the class-A surface pictures
with 0.22.1 in the default dark theme. The demo script is fixed for the current sketch dimension
boxes and MCP file limits, and the rover replay and class-A pictures now have scripts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Docs: cut the idle time out of the demo and rover recordings
The README demo plays at 6x and the rover replay at 15x with a constant-rate camera orbit, with the stretches where nothing on screen changes cut out by the new trim_idle.py.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#5187 fixed the frozen Render preview: product frames now set ss_ready
themselves instead of keeping the flag from the last Design frame, so
paint() no longer blits that stale offscreen picture. Nothing tested it.
The new test draws a see-through Design frame (offscreen, for
order-independent transparency) and then an opaque product frame on the
same renderer, and checks the middle pixel shows the product. With the
product branch's ss_ready update removed it fails with the stale red
Design colour, [255, 129, 129, 255].
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* i18n(uk): translate 188 pending messages
Bring the Ukrainian catalog up to date with en.json after rebasing onto the latest main. Covers the crash-report form, AI preferences, preferences pages, solid tool help, free orbit and history controls. Keys stay sorted via cargo xtask lists.
* i18n(uk): translate 114 pending messages
Rebase onto the latest main and bring the Ukrainian catalog up to date with
en.json. The new messages cover the mouse-gesture editor and navigation
preset help, the frames and sheet-metal dialogs, fillet control radii, the
timeline folder actions, the motion keyframe controls, library appearance and
library file actions, and the measure dialog.
Also correct "Add Key" to "Додати ключовий кадр": the only user of that
message is the motion keyframe button, and the catalog already translates
"Keys" as keyframes.
Keys stay sorted; cargo xtask lists --check and the i18n catalog tests pass.
* Sort the Ukrainian catalogue
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Home: a theme's File view follows its own colours where it sets none (#4868)
and no built-in theme sets them, so every theme kept Dark's or Light's blue rail and grey cards
under its own app bar and headings. A theme that leaves them unset now gets them from its own
palette: page and cards from its panel and dialog, the rail from its accent deepened as Dark's
and Light's are, the primary action from its soft accent, and each ink the one of its own colours
that reads best there. Dark and Light themselves are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Leave a theme that sets no colours as its base palette
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* i18n(da): complete upstream catalog and align CAD help terminology
Add 371 missing Danish translations, refresh changed source messages,
and independently rewrite six help descriptions from SolidCraft sources.
Preserve all 2397 keys, canonical ordering and placeholders.
* Sort the Danish catalogue
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* test: cover malformed and unbounded recovery metadata
Add regression tests for oversized JSON, mismatched entry IDs, and
non-regular recovery metadata. These expose failures in the baseline
recovery reader before the implementation fix.
Signed-off-by: Taha Zarif <tahazarif380@gmail.com>
* fix: bound recovery metadata reads and validate entry identity
Reject oversized and non-regular recovery metadata before reading it,
and bound the read if a file grows after inspection. Require each JSON
entry ID to match its filename for orphan listing and recovery loading,
preventing unrelated save paths from being attached to a recovered design.
Signed-off-by: Taha Zarif <tahazarif380@gmail.com>
* Format recovery.rs
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Signed-off-by: Taha Zarif <tahazarif380@gmail.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6226
* Offset Face: split a face off its seam and try again, as booleans do (#6226)
A hole's wall arrives from a STEP file as one whole turn bounded by one wire that
runs along its seam twice, where SolidCraft builds the same wall as two halves.
Offset Face heals the body before it moves anything, and healing such a face
fails inside truck with `This wire is not simple`, so the panic guard turned the
whole operation into a kernel internal error. Since importing is how a part with
a hole usually arrives, Offset Face did not work on imported parts with holes at
all.
`seam_faces` already answers this for booleans: split every such face in two
along a plane through its axis, run the operation, and compose the history back
over the body as it was given. This gives the offset path the same retry.
- `seam_faces::one_body` is `boolean`'s answer for an operation that takes one
body, with `needed_one` beside `needed`.
- `offset_faces` now does its healing, its face picking and its offset inside one
closure, so the retry runs the whole of it on the split body; the first attempt
is guarded, so the failure is an error the retry can act on rather than an
unwind. The faces are picked per body, because the pieces are numbered
differently once a face is split — the points do not move, so they pick the
same surface either way.
The regression test is in `seam_faces_tests.rs`, beside the boolean ones: a
20 x 20 x 10 block with a 6 mm hole, read back from STEP, whose top moves 2 mm
and lands on the same volume as moving the built body's top, with its history
intact.
Checked against `main` before and after: the three kernel tests that fail here
(`blend::…seam_twice`, `step_in::…parallel_body_preparation`,
`tests::a_box_cut_round_a_small_hole…`) fail the same way without this change.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: CarlosKeese <dudukeese@gmail.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
* claim: #4544
* Add Norwegian Bokmål translation
* Refresh Norwegian Bokmål translations against current main
* Drop Norwegian messages whose ids are no longer in the English catalogue
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5103
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Redefine Sketch Plane names the plane a sketch is on now (#5103)
For a sketch with no body-face attachment the panel only said so. It now shows
a Current plane row: the origin plane, a construction plane's name, or the
plane it is offset from, turned from or flipped from. English, Japanese and
Chinese messages added.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Drop needless to_string calls in the plane name format
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Themes: Okibi (dark) and Akari (light), from the Okibi design system
Two built-in themes mapped from the Okibi design system v1.8
(https://kaievns.github.io/okibi-akiri-design-system-spec/): cool charcoal
ash with a single ember accent, and its warm-paper light twin. Every one of
the 156 colours is set, so nothing falls back to the Dark or Light palette.
* Give the Okibi and Akari themes an icon palette
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6392
* perf: apply a joint solve with indexed joints and occurrences
Every change to a design re-solves its joints and applies the solution. Applying it searched
the joints for each value and, for each occurrence the solve places, searched the occurrences
for it and walked its parents' first occurrences by searching again.
Joints and occurrences are now indexed once, keeping the first of each id as the searches
found it, and the parents' transform is read through the index as the loop leaves it
(component_transform shares the walk, transform_through).
Synthetic design, headless (2000 components at the root with one component inside each, 4000
joints, a solution placing every occurrence): apply 14.1 -> 2.2 ms (median of 12).
The new test applies a solution that moves a component and the components inside it, and
checks each lands where the solution put it, read through its parent as the loop has just
moved it.
* claim: #6394
* perf: group bodies, features and occurrences once in component.list
component.list searched every body, every feature (three times) and every occurrence (twice)
for each component, so its cost grew with components times contents. Bodies, features,
occurrences and children are now grouped by component once, keeping their order, and each
component reads its own groups. The output is unchanged.
The new test lists nested components and checks each gets only its own bodies, features,
sketches, occurrences and children, in timeline order.
* Dialog curve picks keep their sketch (#3726)
* Format
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Format
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Adapt context-menu test to Sel::SketchCurve's sketch field
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Path Flange picks keep their sketch too
Path Flange (#4096) landed after this PR and still looked its path up in the newest sketch with those ids.
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Home: stop a recent card spinning forever when its picture can't be made (#4655)
A recent design whose thumbnail job returned no picture (empty or unreadable design) kept the loading placeholder, which draws a spinner, with no retry. Record such keys as a terminal 'failed' state and draw a box placeholder instead; opening or saving the design clears it and tries again. Sample previews are untouched: a failed sample build still keeps its shipped picture.
* Format the gallery sample row
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The row itself showed the key the command runs from (#1810), but its help
tooltip still named the default one.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With G bound to the centre rectangle, G, Escape, R started the centre
rectangle again, because R's family remembered the variant used last
(#4308). A remembered variant that has a key of its own no longer takes
over the family's key; one without a key still comes back on it.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
When an agent's models come from several providers (Hermes and OpenCode
name each one "Provider · model" or say "Provider: ..." in its
description), the chat header shows a provider drop-down before the model
drop-down, and the model drop-down lists only that provider's models
without the provider in front. It opens on the provider of the model in
use; picking a provider alone changes nothing. ui.agent_chat takes
provider and reports providers and provider.
* i18n(pt-BR): translate 319 missing messages
Catch up with en.json: the new Preferences pages, assistant settings,
navigation preset help, the bug report form, history names, feature
folders, component links and copies, joint keys and the reference
geometry setting. Messages added to en.json after this branch started
stay in English for now.
* i18n(pt-BR): translate Delete as Apagar, not Excluir
"Excluir" (Latin excludere, to shut out) means to leave out or exclude;
"apagar" is the Portuguese for deleting, wiping something out. Use
apagar for every Delete already in the catalog (22 messages: commands,
the delete confirmation, rules, rows, sets, keys, crash reports), with
its forms: apaga, apagados, apagado(s), apagá-lo. The Delete key's name
stays as it is printed on the keyboard.
* Revert "i18n(pt-BR): translate Delete as Apagar, not Excluir"
This reverts commit 3bc73c5930.
Keep the catalog's existing word for Delete until the term is agreed
for the whole catalog in its own issue; the change can follow as a
separate PR.
* i18n(pt-BR): use Excluir in the five new Delete messages
Apply the review's suggestions: the new messages keep the catalog's
existing word for Delete.
* claim: #3983
* Agent skill for reproducible bug reports (#3983)
skills/solidcraft-reporting has an agent gather the version, steps and
exact error, reproduce the problem in a fresh design (a command script or
one batch with rollback off), shrink it while it still fails, check the
numbers with measure and a screenshot, keep private designs out, search
for duplicates, and write the report in one shape: Reproduced on, Steps
with a minimal script, Expected, Actual, Attachments.
Listed in skills/README.md and docs/mcp.md; the bug template's steps
field points agents to it. The skill-name check now covers its files.
* Reporting skill: GitHub takes a design only as a .zip
* Show scroll arrows on the theme row when more themes are hidden
* Fade the theme row edges and use the chevron icon for its scroll arrows
* Remove the outline left on the theme row arrows after a click
* Trim the theme row scroll arrow tests
* Drop the earlier theme row arrows' test and labels
* claim: #6389
* perf: write each control reply to the socket in one piece
The control server wrote each reply with writeln!(out, "{reply}") on the unbuffered TcpStream.
Formatting a serde_json::Value hands the writer every string, number and punctuation mark
separately, so each became its own write system call. A reply of a few megabytes (component.list
on a design with thousands of components) took about half a second to send.
Every reply is now serialized to one buffer and written at once. Writing one large Value to a
loopback TcpStream: 1.79 MB took 1,052 ms with writeln! and 4.9 ms with one write_all.
The test counts the write calls for a 2000-component reply (one) and reads it back.
In an assembly the bottom timeline listed every feature of every component,
unreadable with a few dozen parts. While a component is active (or one is
picked in the browser), it shows only that component's features and those of
the components inside it. A toggle ahead of the features names the component
and switches to the whole design and back. The marker and drops map from the
shown features to the design's positions.
Tests: the_timeline_shows_only_the_active_components_features,
a_slot_among_the_shown_features_is_a_design_position.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4323
* Gruvbox Dark and Dark Soft: readable selected tab labels (#4323)
The Model tab and the shared tabs (Assemble, Insert, Inspect, Manage,
View) draw their selected label in selected_ink on accent_soft. Both
Gruvbox dark variants inherited the default dark selected_ink #9cc7f5,
which reaches only 4.48:1 on #48544f (Gruvbox Dark) and 4.13:1 on
Set selected_ink to the Gruvbox foreground #ebdbb2 in both files:
5.76:1 and 5.31:1. Tabs with their own accent keep their own ink.
A test checks the selected label of every workspace and contextual tab
against its soft background in Dark, Light and every built-in theme
file.
Every STL, OBJ and 3MF export meshed each body with a chord tolerance of
2e-4 of its size, with no way to ask for finer: a 3 mm hole in a 200 mm
part came out with about 14 sides. file.export and file.save_mesh take
`quality`: standard (as before, the default), fine (5e-5) or finest
(1e-5), and Save As Mesh has a Quality row, remembered like its format.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5024
* Hole: refuse a position that isn't on a face (#5024)
`solid.hole` took any `position` and drilled from the nearest face, so a
point 1 mm past a box's side was accepted and cut a hole across the edge.
A position given as a point (numbers or length expressions) must now lie on
a face of a body, within the same tolerance the plastic features use for
their placement (the check is shared with them). Holes at sketch points are
unchanged.
* Hole: a position over the mouth of a bore or opening of a planar face is on that face (#5024)
A counterbore is centred on an existing bore, in the plane of the face round it.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #5329
* Update Links: show which linked design it reads, off the window thread (#5329)
Update Links read and rebuilt every linked design on the window thread, so the
window froze, with nothing on screen, until it finished. From the toolbar it now
runs as a file operation: off the window thread, with the overlay naming the
linked design being read (cover (2 of 3)), its rebuild step and progress, and
Cancel. The assembly stays as it was until the update is in, and it lands as one
undo step. The command reports each design through doc::progress::item.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The cap pass redraws every back face behind a marked pixel with depth test
Always to put the section plane's depth there, and it also wrote colour.
The last back face drawn painted the cut, so a body behind the cut (one
inside another's hollow, an assembly's inner parts) showed through it in
its own colour and outline, like an X-ray. The pass now writes depth only;
the marking passes have already drawn the nearest back face's colour (the
same split as #4622).
Test: a_body_behind_the_cut_does_not_colour_the_cap in
crates/gpu/tests/capture.rs (a red body in a green box's hollow, 4 mm
behind the cut: the picture is the same with and without it, hatched and
plain, orthographic and perspective). On main 472 pixels differ.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The viewport's picker took the nearest face along the ray whatever the
section removed. Over a cut it hovered the cut-away face in front of the
plane (the outer wall of a sectioned hub, say); the shader clipped that
face's tint, but its outline, drawn dimmed where the model hides it, ran
behind the cap and showed through it. While orbiting with the pointer on
the model the hovered face changed from frame to frame, so lines and
patches of the far side came and went over the hatch.
The picker now does what view.pick already does: faces, edges and corners
on the removed side are skipped (the ray walks past them), and where the
plane crosses a body nearer than any kept face, the cap hides the faces,
edges and corners behind it and nothing is hovered there. The walk that
tells whether the plane is inside a body moves into
view_pick::body_cut_at so both pickers share it.
Test: view_pick_tests::the_mouse_picker_stops_at_a_section_cap (a box
with a pocket, cut through the pocket: over the cut material nothing is
picked, over the pocket its floor, and with the section cleared the top
face). It fails on main, which picks the removed top face.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5938
* Edge flange: height measured from the outer sharp, the inner sharp or the bend tangent (#5938)
An edge flange feature takes height_from (outer, inner or tangent) in sheet.flange and Height From in the Flange dialog, saved only when it is not the outer sharp. A height shorter than the bend takes is refused with the smallest height that fits.
* Edge flange: material outside and tangent bend positions (#5941)
An edge flange's position can now be material_outside (the inner virtual sharp on the edge) or tangent (the bend's outer face tangent to the edge's face), in sheet.flange and the Flange dialog. The bend starting at the edge was already the outside position, which now also answers to bend_outside.
* claim: #6252
* Open hosted designs in the web app and add an embed layout
Fixes#6252.
Fetch HTTPS design links with bounded reads and open them as unsaved copies.
Keep model panels and joint controls in embeds without browser persistence,
application chrome, or unload prompts. Add a full-app link and iframe docs.
Cover query parsing, hostile URLs, size limits, and the compact layout.
* Translate the linked-design messages into Japanese and Chinese
* claim: #4167
* Viewport: ambient occlusion no longer shows ghost copies of the model (#4167)
The AO shader used the same 32 sample offsets at every pixel, and nothing smoothed the
result. On a flat face, each sample darkened its own shifted copy of a nearby part's outline.
Where the radius is large on screen (a big design seen close up), those copies show as faint
separate ghosts of the part. Where it is small, they pack into stepped bands along contacts
(#3407).
The sample pattern is now turned by a different angle at each pixel, using interleaved
gradient noise of the pixel position (Jimenez 2014). The copies break up into an even fine
grain, which 2x supersampling smooths further. The cost is per pixel only (one noise value,
one sine and one cosine). The 32-sample loop, the depth prepass and AO off are unchanged.
Test: ambient_occlusion_draws_no_copies_of_a_part (viewport.rs) renders a block on a plate
from above with AO off and on. AO must still darken beside the block, the plate must stay
clear beyond the radius, and the profile beside the block's straight side must have no
copy-sized steps (0.0144 before, 0.0026 after; limit 0.005). The test-only offscreen helper
gains an AO radius, like render_offscreen_ssaa.
Fixes#4167
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
At launch the app adds the solidcraft entry to each installed AI client off the window thread,
names newly connected clients once, and refreshes a stale path. Settings > Assistant has
"Make SolidCraft available to AI apps on this computer" (on by default); off removes the entries it
added. Entries the user changed are left alone; automation runs never touch the files.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2926
* Edit Feature: a primitive in a moved component keeps its placement (#2926)
Edit Feature passes a primitive's placement (corner, base, axis, center)
through as saved, in its component's frame, as timeline.inspect reports it.
timeline.redefine read it as a world point and mapped it through the
component's placement, so in a turned component the edit moved the
primitive: in the Hinge Assembly, Leaf B's rod flipped and the revolute
joint put the leaf on the other side.
A placement equal to the saved one now stays as saved; a new one is still
mapped from the world.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TBKuBKHW3LfC7CMW1HYpwL
* Move the #2926 tests up in frames_tests.rs
They sat at the end of the file, where main has since added the #6000
test; moving them avoids the merge conflict.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: sascha2101 <221609842+sascha2101@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Every command in a design with placed components places their bodies
again on up to 16 threads, though nearly all of them are taken as they
were placed before (the same body by the same matrix). Those are now
taken on the calling thread and only the rest are placed on threads.
Warming the meshes (and display plans) of a state likewise starts
threads only for the bodies that have none yet.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
solidcraft-cli mcp setup (alias install) and Help > Connect AI assistants edit each installed
client's own configuration in place, with this program's absolute path: OpenCode, Claude Desktop,
Claude Code, Cursor, Codex, Gemini CLI, Copilot CLI, Windsurf, Zed and VS Code. Only the solidcraft
entries change; a stale path is refreshed and an entry the user changed is left alone.
Plain mcp keeps looking for the app while its headless session has no work, and says on every
tool result that it is headless when the app was not found.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4094
* Sheet metal: Path Flange, a chain of bends on an existing sheet edge (#4094)
Path Flange (`sheet.path_flange`): sketch an open chain of lines across a sheet edge, pick the
edge and one line; each line becomes a flange as long as the line and each corner a bend, one
feature and one body. The edge flange takes further bends (`then`), each on the far edge of the
one before, and the Flange dialog shows them as rows with a bin each.
* Release: sign and notarize the macOS app with a Developer ID (#5749)
package.sh --sign signs the app, the CLI and the DMG with the hardened runtime and a secure
timestamp, notarizes the DMG and the CLI zip, staples the DMG and checks it with stapler and
spctl. The app carries entitlements.plist: com.apple.security.cs.disable-library-validation
lets the hardened runtime load 3Dconnexion's client framework (crates/threedx-client).
signing.sh --setup imports the certificate into a temporary keychain and writes the API key;
--cleanup removes both. The release workflow builds both architectures first (macos-build, no
secrets), then signs and packages in the join job. Without the APPLE_* secrets (forks, dry runs)
the app keeps its ad-hoc signature and the job summary says so.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* macOS release: staple the app before the DMG, retry stapler, document renewal (#5749)
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #3271
* fix(ui): use document units across length fields (#3271)
* Format
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Adapt drag-extrude and joint tests to document-unit length helpers
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(ui): preserve blank parameter drafts and cover bounded lengths
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Selection fields: only the active one looks active
* Selection fields: a filled field keeps its tinted border where a theme fixes the control border's colour
* Select Other: Shift and Ctrl add the choice to the selection, as a click does
* Select Other: follow origin visibility rename in test
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Select Other test: click away from the lines' midpoints, which are picks of their own now
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2320
* Value fields step with the arrow keys and the mouse wheel (#2320)
Up/Down in a focused value field step a plain number by 1 in the unit it is written in
("20 mm" -> "21 mm", "2,5 mm" -> "3,5 mm"), 10 with Shift and 0.1 with Alt; the wheel over
the focused field does the same, a step a notch, instead of scrolling the panel under it.
Expressions such as `width / 2` are left as typed, and an open completion list keeps the
arrows. Every expression field gets this through `params_dialog::complete`: the value box
beside the geometry in every dialog, the dialog rows, sketch dimensions, the dimension
editor, Drive Joint and the Parameters window.
* The value box beside the geometry stays under the pointer
The box rides on the arrow's tip, so each wheel notch that stepped its value moved it, and
after a notch or two the box slid out from under the pointer and the wheel zoomed the view
instead. While the pointer is over it the box now keeps its place, and it goes back to the tip
when the pointer leaves.
* Value stepping is a preference, off by default
Preferences › General › "Step values with the arrow keys and the wheel" (`step_values`,
also through `ui.prefs`) turns on Up/Down and wheel stepping in value fields, and the value
box staying under the pointer while the wheel steps it. Off, as by default, value fields
behave as before: Up/Down move the text cursor, the wheel scrolls, and the box beside the
geometry follows its arrow.
* Value stepping test: Bend Radius is under Advanced now
* Value stepping test: the preference is on the Design page
* Value stepping test: draw the field straight into the test's frame (#2320)
The shared outer-host rule now covers tests too; the field needs no panel around it.
* claim: #4968
* Sheet metal rules: bend angles as the included angle (#4968)
A sheet metal rule's angle_display (bend, the default, or included) sets
how the DXF bend notes read each bend: a 58° bend is B1 UP 122° R1 under
an included rule. The bend table always gives included_angle_deg beside
angle_deg (CSV and JSON). Set in sheet.manage_rules and the Sheet Metal
Rules dialog, listed by sheet.rules; the flange angle stays the bend angle.
* Bend table: the included angle as the CSV's last column, so the columns before it keep their places (#4968)
A test also covers a body whose base flange names its rule, which keeps that rule's notes whichever rule is active, and a bend down.
* Sheet Metal Rules: Angle Display reads Bend or Included, so the choice fits its box (#4968)
The rule editor's row contract lists the new row and the key the dialog sends.
---------
Co-authored-by: begsaquib <saquibbeg4444@gmail.com>
* claim: #5580
* fix: follow named lid edges after parameter changes (#5580)
* fix: initialize viewport test frame tints
* Drop duplicate tint fields in gpu view tests (main already has them)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5907
* Fix Manage Units to open the shared unit menu (#5907)
Open the browser's existing unit choices for interactive command starts
instead of executing document.units without its required parameter. Anchor
the menu below toolbar buttons and menu rows. Choices still run the normal
command, while programmatic calls require an explicit unit.
Cover error-free opening, unit choices, button anchoring and dismissal,
the current-unit marker, one-step undo/redo, and programmatic calls.
Fixes#5907
* Sketch Dimension: point or circle to the edge of a circle or arc (#4064)
Adds PointCircleDistance and CircleCircleDistance (edge-based, like #1863):
solver, constructive seed, drawing, command wiring, tests, UI scenarios and docs.
Saved designs gain two constraint kinds, in one pending format file.
Fixes#4064.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Sketch Dimension scenario: pick the circle at its top edge (#4064)
The bottom edge sits under the Dimension tool's status hint in the scenario window, so the second click never reached the tool.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #5439
* Agent chat: each design tab keeps its own conversation (#5439)
* Adapt chat-per-tab changes to the new Pending context and cursor tests
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
solid.revolve already revolves a planar body face (`face`, `face_body`), but
the Revolve window's Profile input took sketch profiles only, so a face
couldn't be picked. The input now takes planar faces too, and a picked face
is sent as `face`; sketch profiles win when both are picked.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5467
* Warn when a design is open elsewhere or changed on disk (#5467)
The desktop app keeps a lock file next to each open design
(.~<file>.lock: computer, user, version, opened and refreshed times),
refreshes it every minute and removes it on close; a lock not
refreshed for three minutes is stale and ignored. Opening a design
that another SolidCraft has open asks: Open Read-Only, Open Anyway or
Cancel. A read-only design saves only to another file.
A design remembers its file's size and modification time when it is
opened or saved. Save over a file another program or computer changed
since asks: Save a Copy (the default), Overwrite or Cancel.
Programmatic calls never ask: doc.open reports open_elsewhere and takes
read_only, and file.save over a changed file returns an error until it
is given overwrite: true. ui.documents answers both questions.
* claim: #5336
* Editing an Extrude or Revolve shows its sketch (#5336)
A finished sketch is hidden once a feature uses it, and it stayed hidden
while that feature's dialog was open for editing, so no other profiles
could be picked. The session now holds the sketch of the feature being
edited; the edit dialog sets it while it is open and clears it however
it closes, and that sketch is drawn without changing stored visibility.
* claim: #2063
* Windows OpenGL: use the system title bar so the picture lines up with clicks (#2063)
On Windows with only OpenGL to start, the window opens with the system
title bar instead of using the application bar as its title bar. The
window without OS decorations keeps its caption style while having no
frame, and some OpenGL drivers (older Intel HD/UHD) show the picture a
caption's height lower than the window, so clicks land above what they
point at. Direct3D 12 and Vulkan keep the custom title bar.
The choice is made before the window opens, from the backends wgpu
starts, so the window never changes its frame while OpenGL draws into it.
* claim: #5377
* Report a bug: title from the workspace and the reporter's words (#5377)
With no dialog open the report was always labelled "Modeling" and its
summary pre-filled as "unexpected result", so every such report arrived
titled "Modeling: unexpected result" and the fallback to the first line
of "What happened?" never applied.
Label the report from the active workspace (Part, Sketch, Surface, Sheet
Metal, Plastic, Mesh, or Render), leave the summary empty so the title
becomes "<workspace>: <first line of the description>", and keep
"operation failed" when an error was recorded. The summary field now
hints that a short title is optional.
The triage script that asks for an empty in-app report to be pasted
recognises the new titles.
New UI strings go in assets/locales/pending/<slug>.json and new ATTRIBUTION.md rows in
attribution/pending/<slug>.md. The app build, xtask lists/i18n/assets read them as part of the
catalogs, and the Generated files workflow folds and deletes them on main with
cargo xtask fold-pending.
Refs #6439
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4031
* Triangulation: split a crossing boundary segment ourselves, never in spade's assert (#4031)
A trimmed face's boundary segment that crosses a constraint already in the
triangulation went to spade's add_constraint_and_split, which asserts
(cdt.rs:978, `!previous_region.is_empty() || from == final_vertex`) when a
split lands so near another vertex or constraint that the piece up to it is
blocked. The kernel guard caught it, but the whole operation failed and a
crash report was written; a STEP body imported empty or not at all (#4598,
also #4293, #4478, #5068, #5229).
PolyBoundary::add_constraints now adds every segment that can be added, then
splits each blocked one itself: a vertex at each crossing, inserted from the
crossed edge's own vertex, and the pieces between added where they can be. A
segment with crowded or very many crossings, or a face with more than four
blocked segments (a rim that runs back and forth across itself), leaves them
out, as truck did before #1653, so the face meshes slightly off instead of
not at all. Pinched faces (#1648) still split where they cross.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat: no per-message limit by default
The limit is optional now: an Unlimited checkbox on the AI page (checked by default), a
number field when unchecked. A saved 24, the old default, reads as unlimited. The API-key agent's
24-request and the gate's 400-request caps are gone; Stop still cancels.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat: add the new messages to the complete Japanese and Chinese catalogs
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Viewport: lift selected and hovered edge highlights toward the eye
A selected edge's highlight (3 px core, 6 px halo) is a screen-space band that
was depth-tested at the edge's own depth. A face next to the edge seen at a
grazing angle is pushed back by only about a pixel of its slope, so it buried
part of the band, unevenly along the edge: the highlight showed as a thin
wedge instead of a line.
The band is now drawn like a thin cylinder around the edge: its depth is
pulled toward the eye by half its width plus a pixel, measured on screen so
it holds at every zoom. Plain model edges keep their depth (#1434 still
holds); the part of a highlighted edge the model really hides stays dimmed
through hidden_line.
Not yet compiled or tested.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Viewport: lift an edge highlight by how steeply its faces are seen
The first lift (half the band's width plus a pixel, the same on every edge)
was not tied to the faces beside the edge. Each highlight segment now carries
the normals of the two faces it borders (octahedral, packed with its width in
the line's width slot, read as u32 by vs_line): the band is pulled toward the
eye by its half width times how steeply those faces are seen, so a grazing
face no longer buries it, and an edge seen square to its faces is not lifted
at all.
Test: a_selected_edge_on_a_grazing_face_draws_whole (tilts 60 to 88 degrees,
small and large designs, orthographic and perspective; the band 5 mm behind
a face stays hidden). Capturer::rgba_highlighted draws a highlight scene as
the viewport does, for that test.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5802
* fix: repaint on a theme change and key the frame cache on the theme tints (#5802)
theme::apply asks for a repaint, so the new visuals are painted on the next pass instead of waiting for another event to schedule one. The gpu frame cache key now includes the theme's hover and selection tints, so a picture cached under the previous theme's colours is not reused.
* test(ui): settle egui's startup repaints before asserting the theme's repaint (#5802)
* test(ui): name the call site that asks for the theme's repaint (#5802)
* test(ui): apply the theme outside a pass so its repaint request stands alone (#5802)
The AppImage writes a ~/.local/bin/solidcraft-cli launcher from its desktop-integration step
(same opt-outs, never over a file it did not write). The macOS app links the CLI into
/opt/homebrew/bin or /usr/local/bin at launch when one is writable, and Help > Install
solidcraft-cli command does it with an administrator prompt otherwise. Snap and Flatpak are documented.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5357
* A dragged command window keeps its place (#5357)
Command dialogs share one dragged offset per dock side for the session
instead of one per title, and closing a dialog no longer forgets it, so
the next command opens where the last one was left. The offset is still
kept inside the viewport, including after the window shrinks.
Choosing and swapping in a refinement rebuilds and uploads the whole model
scene. During an orbit of a large assembly that happened about ten times a
second, tens of milliseconds each. It now waits until the camera has stood
still for 200 ms.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
A Home picture still being drawn by the detached saved-thumbnail thread
could be on the GPU as the process exited, and the process died with
SIGSEGV inside the graphics driver. capture::close stops new pictures
and waits for those in flight (the app calls it in on_exit), and
dropping SolveApp waits for the Home picture threads it started.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: post-merge Render review corrections (#5493)
* fix(render): gate bundled studio tests in isolated engine builds (#5493)
Addresses jyviko review comments on #5494: compile the two positive payload-dependent regressions only with bundled-render-library. Keep all assertions and the explicit feature-absent refusals. Reproduced 5 pass / 2 fail before the fix; isolated default, CPU without pack, and CPU with pack tests plus strict scoped Clippy and repository checks pass. No production or CI configuration changes.
* Test: a text on a path follows the path when a parameter moves it
A construction line placed by a dimension that reads a parameter, and a text laid along it: after
the parameter changes the line has moved and the text has not. It follows one evaluation late.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Sketch text: a text on a path follows the path in the same evaluation
Problem: texts are laid out with the sketch's links, before the sketch is solved. A solve that
moves the curve a text runs along (a path placed by a parameter) left the text where the path had
been before the change, so it followed one evaluation late.
Solution: after a solve that keeps the sketch's orientation, every text that runs along a curve
of its sketch is laid out again along that curve as the solve left it; a text that no longer lays
out adds its warning.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Test: an offset plane on a sheet flange's face follows the flange
An offset plane on the inner face of an edge flange, picked by body and index as model.faces
lists it. When the base gets wider the flange moves along x and the plane stays where the face
was picked; a sketch on the same face follows it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Construction: a plane on a sheet's flat face follows the face
Problem: a construction plane on a flat face of a sheet body resolved the stored plane, which is
only where the face was when it was picked. When the sheet changed the plane stayed behind, while
a sketch on the same face follows it by its sheet face name.
Solution: a construction plane on a sheet's flat face resolves that face's plane as it is now, as a
sketch on the face does.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Test: a design keeps several hundred configuration rules
An assembly that switches a few hundred parts on and off, by two rules each, needs hundreds of
rules; the saved-rule budget refused more than 128.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Configurations: a design keeps up to 2048 rules
Problem: a configuration table kept at most 128 rules. A rule drives one feature's suppression,
so an assembly that switches a few hundred parts on and off (by two rules each) could not keep
its rules. Copying rules for an independent component copy looked up each new id by scanning all
rules, which a larger table makes slow.
Solution: MAX_RULES is 2048; a copy looks its new ids up in a set of the ids taken. The docs say
the new limit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The parts of one STEP assembly import share their face names, and each
part's body stays where the file has it, placed in its component by an
offset. A joint origin on a named face (a hole's centre, a face's
centre, an x reference by face, the end a round origin takes) looked the
name up on every body of the design and compared the pick, in the
joint's component coordinates, with each body as it lies. Another part's
face of that name could lie nearer, and the joint resolved on that
part's face: a pin jointed into one plate's hole landed where the other
plate's hole is relative to its plate.
named_face now looks only at the bodies of the joint's component, each
in its own coordinates, a face the name finds exactly before a piece of
it or a face it was merged into, then the nearest. The design-wide
lookup stays for a face whose body has gone to another component.
named_round, x_axis, pin_end and end_faces take the component, as
face_name_at and round_end_at already do.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A joint origin at a face's centre (`joint.origin {face}`, `Snap::FaceCenter`) was the face's
area centroid, holes included. A hole cut into that face, moved or suppressed therefore moved
every part held at the face's centre, though the face's outline stayed where it was: on a
200 x 100 plate with a 20 mm hole, 0.31 mm for each 20 mm the hole moves and 0.62 mm when the
hole, 40 mm off the middle, is suppressed.
On a planar face with holes the origin is now the middle of the area the face's outer boundary
encloses (`outline_centre`: the face's triangle sides that no other triangle of the face runs
back along, corners matched by place, chained into loops in the face's plane; the centroid of
the loop enclosing the most area). A named face (`NamedFace::plane`) and the planar face
nearest a point (`nearest_plane`) both use it. A face without holes, or one whose mesh gives no
clean outline, keeps its centroid as before.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A parameter change evaluates the model and solves the joints inside the
command (so that it can refuse a value that breaks a sketch), and
`Session::execute` refreshes again after the command and solves every joint a
second time, although nothing has moved since. On 40 small parts jointed
rigidly to a plate's holes (half of the joints lifted by an expression),
changing the lifts' parameter took 0.85 s on main and takes 0.45 s now;
changing one part's thickness, 1.96 s and 1.52 s (CLI, medians of 5 runs).
The session now keeps the design the joints were last solved for, as the
solve left it, and the model that solve read. A refresh that finds the same
model, and a design that differs from the solved one at most in its features,
leaves the joints as they are: the joints read features only through the
model, and a value edit stores the new solutions of the sketches it drove
after its solve, so its design is a new one by the time of the second
refresh. Placements and bounding boxes come out identical.
Test: `a_parameter_change_solves_the_joints_once` counts one solve per change
(two before), checks the jointed part still follows the hole, and that a
captured move of a jointed part is still solved back.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
An edge joint origin looked its edge and face up in the body's own coordinates with the pick in the component's,
and returned the frame in body coordinates. On a body placed by an offset (an imported part's bodies are) a pick on
the edge where it is shown found no edge ("joint origin: no edge there"), or the origin sat away from its edge by
the offset. Edge origins now take the picks into the body's coordinates through the inverse of its offset and place
the frame back by it, as named faces and round origins already do. A body without an offset is unchanged.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* CI: duplicate-PR check and a required 'Changes a default' line (Refs #6439)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* PR checks: enforce the defaults line only on PRs opened from 2026-10-25 (Refs #6439)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Since #1768 the section's stencil marking draws every face with FACE_BIAS.
Where a back face seen through the cut meets a front face along an edge (a
pocket's wall and its side wall or floor, a fillet and the face it runs
into), the slope push decided which was nearer beside that edge; a front
face winning left those samples unmarked, the cap skipped them, and the
edge, drawn at its own depth, showed through the hatch. In perspective, and
while orbiting in orthographic too, edges popped in and out from frame to
frame.
A second marking pass draws the back faces only (front faces culled, depth
test <=, no depth write), pulled toward the eye as far as FACE_BIAS pushes
faces back, before the cap. The depth buffer keeps FACE_BIAS, so edges on
visible faces still draw whole (#1434).
Tests in crates/gpu/tests/depth.rs: a pocket 2 mm behind the cut, and a
filleted hub with a bore and a cross hole seen from twelve directions, both
depths, orthographic and perspective. On main the pocket shows 45 px of
edges through the cap in perspective and the hub 25 px over the views; with
this change 0 and at most 2.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5352
* Sketch: circles and arcs honour construction (#5352)
sketch.arc.center_point, sketch.arc.three_point, sketch.circle.two_point,
and sketch.circle.three_point accepted construction: true but drew normal
geometry. They now mark the new curve as construction, like lines and
rectangles, and their parameter docs list construction?.
* claim: #5562
* Intersection curves: a point that won't project no longer panics (#5562)
A quarter disc at a box's corner revolved 360 deg about the box's edge and
cut (a hemisphere whose pole sits on that edge) panicked in truck at
intersection_curve.rs:133: dividing one of the cut's intersection edges
met a point that projects onto neither surface, and the hinted
projection's fallback called subs, which unwraps that same failed
projection. The guard turned it into a failed cut and a crash report.
Now the leader's point stands in, the division stops there, and the
failure is recorded (truck_geometry take_projection_failure). truck-shapeops'
face division checks it after dividing an edge and gives up, so the
Boolean attempt fails as quickly as the panic made it and the kernel goes
on to its other tries: carrying the approximate points on instead made one
kernel test (equal bores crossing) run for minutes. On this design the cut
still fails with an ordinary error, leaving the box whole; nothing panics.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6000
* Sketch: explicit planes take a frame, world or component (#6000)
sketch.create and sketch.redefine take frame: "world" (the default,
as before) or "component" for an explicit {origin, x_dir, y_dir}
plane. The reply says which frame the plane was read in, with a note
when a world plane goes into a moved or turned component. frame on a
named plane or a face is refused: those are where they are.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With Start set to From Object (or a start offset), the automatic
operation was still decided from the profile: the engine probed for
material just past the profile plane, and the dialog kept the operation
it had latched before the start changed. A profile on a plate's top,
extruded 14 mm down from the plate's bottom face, came up as Cut instead
of Join. The engine now probes from the extrude's start (and, to an
object, reaches from there), and the dialog decides the operation afresh
when the start changes.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6145
* Windows: a stuck icacls no longer freezes SolidCraft when it writes the control token (#6145)
Writing the control token runs icacls to drop the file's inherited
permissions, and waited for it without a limit. On some computers the
new icacls process stays suspended, so connecting an agent froze the app
for good. It now waits at most 5 s; an icacls still running then is ended
and the token written anyway, into the user's own local app data folder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Test: a plane through a construction axis follows it in a design with a component
Two offset planes, the axis where they meet and a plane through that axis. Once the design has a
component, a parameter change fails the last plane with "unknown construction `Ax`".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Construction: the root's constructions are found once the design has a component
Problem: once a design has a component, the root's construction features are built in the model
placed in the world. That state carried each component's bodies, sketches and threads, but not
its construction planes, axes and points, so a construction built on another one failed on every
evaluation with "unknown construction".
Solution: the world state carries each component's constructions where the component is placed
(moved by its occurrence's placement, as the bodies are).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Test: configuration rules given together are checked with all of them in place
Suppressing a box and the cut on it: the box's rule alone fails the row check (the cut has
nothing left to cut), so the two rules can't be added one at a time. Given together they pass,
in one undo step; a bad rule among them adds none.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Configurations: config.rule takes several rules at once and checks the rows once
Problem: config.rule adds one rule and checks every saved row with it. Rules that only hold
together (suppressing a body and the cut on it) could not be added at all: the first one alone
fails the check. A design with many rules also rebuilt every row once per rule.
Solution: config.rule takes `rules: [...]`: each rule is added or replaced as with `rule`, then
the rows are checked once with all of them in place, in one undo step; a rule that fails
validation adds none. The docs describe it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Test: joint origins take a long piece name
A face split by a cut whose edge runs in many segments is named after every face around its
piece, several hundred characters. joint.origin refused such a name, as model.faces gives it,
over 512 characters, both for a face centre and as a round end.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Joints: an origin takes a face name as long as the face's piece name
Problem: joint.origin and joint.create took a face name (a face centre, a round end, an x
direction) only up to 512 characters. A face that a cut split is named after the faces around its
piece (`<face>#[<face>,…]`), which reaches several hundred characters when the cut's edge runs in
many segments (a slot with a zigzag edge, a cut of text), so a name model.faces gives was refused.
Solution: the face names a joint origin takes are capped at 16384 characters (MAX_FACE_NAME),
still a bound on hostile input.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sheet metal: test that an unfolded sheet's flat face is named after its base at every size
An unfold keeps the base in place and lays the flanges out beside it; the flat
face spanning them and the bend strips is the base's face. It was named after
the unfold whenever few of its sample points fell on the base (the bend strips
are many small triangles, the base two big ones), and that changed with the
sizes: a size change renamed every top and bottom face the refold made of it,
and joints and sketches on them lost their faces (40 of 85 joint ends of a
real part after one change). The test unfolds and refolds a base with three
flanges at two flange heights and wants the flat face and every refolded
panel face named after the base.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Sheet metal: the panel an unfold or refold keeps in place names the face it is part of
A face a sheet feature made is named by where its sample points lie on the
input's faces: in place, else taken back through the folds. Only nine points
per face are sampled, triangle centroids picked by index. An unfolded sheet's
flat face spans the base, the flanges laid out beside it and the bend strips
between them; its bend strips are many small triangles, so most samples fell
on them. Those lie on no panel, and one of them made the fold fallback give
up for the whole face. The face was named after the unfold, and the refold
passed that name to every top and bottom face; which way it went changed with
the part's sizes.
For a sheet feature that moved panels, a face the plain vote leaves unnamed
is now looked at closer (up to 256 points): if a part of it lies in place on
the body's own face before (the panel the feature kept), that face names it.
Else the fold fallback votes over the points that can be taken back, when at
least half can, instead of none when one cannot. Other features, and faces
the plain vote names, are named as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Test: a pattern copy's piece names match as the original's do
A copy's faces carry the instance after the name, after any piece qualifier
(`top#[left,slot:a]@F9.1`). A stored piece name of a copy whose bounding faces have changed since
finds no face of the copy at all (None at the first assertion), where the original's identical
name matches by its stem. On a real part: the joint ends on a patterned copy fall back to their
picks after a size change, the original's resolve by name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Names: a pattern copy's piece names match as the original's do
Problem: a pattern's copy names its faces after the original's with the instance appended
after any piece qualifier (`top#[left,slot:a]@F9.1`). The piece parser expects a qualified name
to end with its qualifier, so a copy's name was its own stem: when the faces around a piece
changed, a stored name of a copy matched no face of the copy at all, where the original's
identical name still matched by its stem. Joint ends on a patterned part fell back to their picks
after a size change and flipped to the other end of their holes.
Solution: `history_names::matches` takes the pattern instances off the end of both names first
and compares the rest as before, within the same copy only (equal instances; an original's name
never matches a copy's face, nor one copy's another's).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Test: a switch parameter alone switches the features its rules suppress
Configuration rules that switch features' suppression from a parameter (drive_suppression when
`on` = 0) ran only on config.activate: a plain parameters.change of the switch left the features
as they were, so a script had to activate a row for every change. Red until the fix.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Configurations: a parameter edit runs the suppression rules whose condition reads what changed
Problem: configuration rules ran only when a row was activated or a configuration edited. A
design that switches features on and off by a parameter (a rule that suppresses a feature when
the parameter is 0) did not follow a parameters.change of that parameter.
Solution: parameters.change (one parameter or several) runs, before the rebuild, the enabled
rules that switch a feature's suppression and whose condition reads a parameter whose value
changed (or a feature switched earlier in the same pass), in their order as activation applies
them, and lists them as `switched`. Driven parameters and restrictions still wait for a
configuration edit or activation; a rule whose condition reads nothing that changed leaves a
hand-set suppression alone; a rule that cannot be evaluated is skipped and never blocks the edit.
One undo step with the edit. docs/configurations.md says so.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A live drag frame shares one work budget between its held solves, but
only their cluster decomposition was charged. The numerical solve (LM)
ran outside it, so in a larger sketch where the geometry can't follow
the cursor, one frame ran 64 failing held solves of about 150 ms each:
10 s per frame (68 points, 50 curves, 58 constraints, found by
randomised sketch editing). The frame budget now also counts the LM
work of its held solves, without stopping a solve: once a frame's share
is spent, the drag takes no more steps in that frame, keeps what it
reached and asks for the next frame (retry), as the viewport already
handles. The same drag now takes about 0.3 s per frame. Solves outside
live frames are unchanged.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Placing a design's bodies asked each body's owning component once per
placement, and each answer searched the features for the one that made
the body: placements x bodies x features on every command, twice (the
world state and the frames that map a command's picks into components).
Document::model_body_components answers for all bodies at once, and
world_state_reusing and body_frames_with_source use it. The engine's
frames now work out the per-body table only when a command picks a body
or names one; most commands pick none.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Every evaluation fingerprints every feature by the parameters its
expressions name, and works out every parameter, and most of them (named
feature inputs) are literal values such as "10 mm". In a design built
command by command both were lexed and parsed again on every command.
references() now keeps its answers per thread (bounded), eval_params
keeps its last few results for identical definitions (compared in full),
and a parameter that names no other is evaluated once per expression,
unit and design length unit.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A standard part (a nut, washer, bolt, screw, pin or bearing) drew its sketches without
constraints, so each was left with free degrees: 3 for a hex, 8 to 12 for a revolved outline.
A check of a design that holds parts listed every one of them as not fully defined. The part
builder now fixes each sketch's points and curves before finishing it: the outline is the
part's size, and a resize builds it again the same way.
Claude-Session: https://claude.ai/code/session_01XhgPofrmFDeLvMhar6Wsqw
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A new feature is named after the first free number of its stem
(Cylinder1, Cylinder2 …), and every number tried searched all features'
names: the thousandth cylinder of a design tried a thousand names against
a thousand features. A new input's dN name likewise wrote out every d1,
d2 … it tried. Both now gather the numbers taken after their stem once
(params::numbers_after, which counts only names written as the number
is) and take the first number not among them. The names are the same.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Every evaluation fingerprints every feature by its JSON, so a design
built command by command wrote all of its features out as JSON on every
command. The text is now kept (per thread, by feature id) with the
feature it was written from, and written again only when the feature
differs; the text hashed is the same, so the fingerprints are too.
Imports, whose text depends on their STEP digest, are written each time
as before.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
document.inspect with measure (what inspect_design and the
solidcraft://design resource ask for) measured every body one after
the other, and again on every call: on a 1,280-body design each call
took 25 to 60 s. It now measures them on all cores, as
inspect.measure does, and each body keeps its measurement in a cell
shared by its clones (with the body it was measured on, as the outline
box is kept), so a design inspected again measures only the bodies that
changed. Placements taken as before keep their placed body's cell too.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Every joint whose offset, angle or values are expressions is worked out again
on each solve, twice per solve (re-expressing the joints, then the check that
leaves out one whose expression no longer works out), and each expression that
names a parameter evaluated every parameter of the design anew
(`Document::eval`). The cost grows with the joints with expressions times the
design's parameters: on 40 parts jointed to a plate, each joint lifted by an
expression and each part carrying 13 parameters (522 in all), a joint solve
took 0.55 s on main and takes 0.41 s now, a change of the lifts' parameter
1.06 s and 0.81 s (CLI, medians of 5 runs).
`joints::Params` keeps the design's parameter values once the first
expression names one, and `joints::expressed` takes it, so one pass evaluates
the parameters once. Values and error messages are those of `Document::eval`.
Test: `kept_parameters_evaluate_as_the_design_does` compares values and
errors with `Document::eval`.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Test: a label on a path follows its source far
A rectangle of construction lines placed on three points projected from another sketch, with a
centred text on its far side: moved 270 away by the parameters that place the source, its solve
from the stored sketch failed ("the sketch constraints conflict").
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Sketch: a sketch whose links moved far is walked there and keeps the stored sides and turns
Problem: a change upstream moves what a sketch projects (or the points its parameters place)
while its own values stay. The solve from the stored state then either failed ("the sketch
constraints conflict") or landed on the mirror image of a side or turn and reported it solved,
because a solve after a link move checks no orientation when no value changed: a line tangent to
a moving circle hopped a diameter to the circle's other side. And a sketch moved so was stored
only when its own values changed, so every later evaluation solved it from its first stored state
again.
Solution:
- flips_from_stored: the orientations of the stored sketch (unsigned offsets, distance
triangles, and a line's side of a circle it is tangent to) that a solved sketch turned over,
each read where the stored sketch has its points.
- solve_after_link_move: the new values solved where the links were stored, then the moved
points walked to where they are now (predictor-corrector: each step starts from the last two
solutions carried along their secant; steps halve on a failure or a flip and grow while they
solve).
- Document::solve_sketch: when a linked or fixed point moved and the jump fails or flips, the
walk; a flip the walk cannot avoid either is reported as one.
- Model::edited_sketches: a sketch whose linked points a value edit moved is stored as solved, so
the next evaluation starts where this one ended.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #1796
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
* Sketch Symmetry: centre a line between a face's edges (#1796)
Symmetry of two edges of a face (its projected outline, whose sides run
in opposite directions) about a construction line between them was
refused with "neither entity can move". The pairing of the line ends by
mirror image landed on main in #3029; this keeps the tool side and the
tests for #1796:
- The Symmetry tool and a three-line preselection take the only
construction line or centerline as the symmetry line, whatever the
pick order (picking it first used to mirror it out past the other
line). Two points and a line take the line. Otherwise the last pick,
as before.
- Tests: fixed lines drawn the opposite way, pairs by mirror image (a V,
end to end along the axis), every pick order, a three-line
preselection, and a scenario clicking a face's edges.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4813
* Sketch text: measure a line by the rule that draws it (#4813)
`TextMetrics`, made once from a font and a style, parses the face once and gives the scale for a capital height, the
distance between lines, a line's width and the characters the font has no glyph for (in order, each time, including
those its character map sends to `.notdef`). `text_geometry_styled` lays its lines out through it: one rule for
measured and drawn text, no kerning, a missing character still drawn as `?`. Measuring builds no outlines, so a text
too large to draw (`MAX_LINK_ENTITIES`, `MAX_TEXT_CHARS`) can still be measured. Every outline stays bit for bit the
same: a test pins five styled texts, captured before the change. The size and spacing checks are shared, and still
run before the font is looked for; a text whose place or angle is not finite now says so instead of naming its
height.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5523
* Windows resize smoke test: wait for a settled window before hit testing
Startup placement can move the window between CheckBorder reading its rect
and the hit tests, so correct answers for the new position failed against
stale coordinates. Wait for two reads 200 ms apart to agree before the first
pass, compare the rect after each pass, and retry the pass (up to three tries)
when it moved. A pass with a stable rect fails exactly as before.
Fixes#5523
* claim: #5302
* Extrude: a picked profile takes the whole drawn profile (#5302)
A sketch on a solid's side face holds the face's outline as projected
reference geometry, and that outline cuts the drawn profile into pieces
of the sketch's graph. A pick by index or by point resolved against those
pieces, so extruding a rectangle that straddles the face's edge took
only the piece under the cursor.
The pick is now expanded to the drawn profile it belongs to, so it takes
every region of it. The drawn profiles are the disjoint faces of the
drawn graph, so an explicit pick of a drawn hole still fills only that
hole (#5176).
* Extrude: a pick expands across the reference geometry that cuts the drawn profile (#5302)
The expansion crossed every region of the drawn profile, so a projected curve that bounds a region inside the profile (the rim of a counterbore) was crossed too and the cut took more material than intended; now the expansion stops at that boundary. Fixes the regression in cmd::features::prism_local_tests::a_ring_cut_round_a_counterbored_hole_in_a_chamfered_disc, and keeps the behaviour of #5302 and of the #5176 invariant.
* claim: #2722
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Sketch on the sides of a thin extrude (#2722)
A thin wall's sides lie off the sketch curves (half the thickness either
side, say), so no curve was under them: its side faces had no role, no
exact history name, and Create Sketch refused them as "can't keep an
attachment". The roles of a thin wall's sides are now
`wall:<curve>:left|right` (the side of the curve it runs beside, in the
direction the curve is drawn), and its open ends `wall:<curve>:start|end`.
They follow the thickness and side settings and survive a save.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #5822
* Batch: consecutive parameter edits can rebuild the design once (#5822)
An agent's batch of parameter edits, one parameter per step, evaluated the
design once per edit: an edit early in the timeline rebuilds nearly all of
it, and 21 such edits ran for minutes.
engine.batch and the MCP batch tool take coalesce_parameters (default
false). With it, each run of consecutive plain parameter edits
(parameters.change with only a name and an expression, or with changes) is
set as one parameters.change {changes}: one rebuild for the run. Any other
step ends a run, so later steps see the design up to date. Every step
still gets its own result (coalesced: the run's length); the run is one
undo step. A run whose combined edit is refused (a new parameter, which
changes can't make; a bad expression) runs step by step, so a refused edit
is still named by its own step. The app's off-thread evaluation computes
the run the same way. The grouping is solidcraft_engine::batch_params,
shared by both runners.
21 parameter edits on a plate with 21 extruded bosses (in-process MCP,
ci profile): 40.1 s step by step, 3.5 s coalesced, the same volume.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Batch coalescing: say a run skips the states between its edits; test it against step by step (#5822)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5881
* Snap points: the centre, middle and ends an edge or face offers (#5881)
Phase 1 of #5881. `solidcraft_engine::snap::snap_points` gives the points
a picked item offers as picks, from Measure's exact shapes: a round edge
its centre (an arc also its ends and exact middle, joined across pieces
the kernel split it into), a straight or other edge its ends and middle,
a flat face the centre of its area, a cylinder or cone the centres of its
ends, a sphere or torus its centre, a vertex itself. Nothing uses them yet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* doc: persistent names read as their structure; joint.inspect splits an edge with a blend face first
`persistent_name::{FaceName, EdgeName}` parse a face or edge name once (blends, inner walls,
copies, pieces, numbered pieces, `~k`) and write it back as the same text, bounded against
hostile input. A suffix after a blend's edge naming a later feature belongs to the blend face.
joint.inspect split an edge at its first `|`, so an edge whose first face is a blend
(`F4:blend:F3:+x|F3:+z#[…]|F5:-y`) came back as a cut-off blend name and a string that is no
face. It now uses `EdgeName::parse`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* doc: persistent names as typed values: features, roles, suffixes, a lexer and errors that say where
`FaceName` is a base (made by a feature in a role, a blend and the edge it rounds, an inner
wall, a body's face by index, other text) and its suffixes (copy, piece, numbered piece).
Roles are typed where the vocabulary is known (top, side:<curve>, +x, face:<k>, …) and kept as
written otherwise. The text form is lexed into tokens and read by the grammar in the module's
doc; a name that doesn't read says where and what was expected.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* names: read what made a face, its kind and its feature ids through persistent_name
The places that took a name apart by hand to find the feature that made a face, whether it is
a blend, a copy, a numbered piece or a wall, or to remap feature ids on copy, now ask the typed
name (`made_by`, `split_feature`, `is_blend`, `is_copy`, `is_numbered_piece`, `is_wall`,
`features`, `map`). A name that doesn't read falls back to the old text check.
- An offset of a face that only names a blend inside it (`F2:top#[F5:blend:…]`) is no longer
taken for a fillet's.
- Copying a component remaps feature ids inside a piece's `#[…]` too.
- A shell's inner wall split by a later cut keeps the qualifier on the wall.
- joint.inspect finds the sketch curve of a copied side (`side:l1@F7.2`).
Finding a face by name (strip_piece, parent_piece) is unchanged: it reads a trailing
qualifier as the whole name's and moves once the text form says where it belongs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* engine: regression tests for press-pull on a wall that only names a fillet and joint.inspect on a copied side
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* names: an edge holds one face or more; made_by reads only the leading token
An open boundary's edge has one face and an edge several faces meet at has more, so an edge name
is a list of faces (a blend's own edge still takes two at most, as before). `made_by` reads just
the leading `F<n>:`, which every name a feature made starts with, so asking it of every face
each frame stays cheap. Names are read up to 1 MiB: a face engraved with text has piece names
past 4 KiB.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* persistent_name: split into lexer, parser, write, text and tests
No code changed; callers' paths stay the same (`pub use text::*`).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5962
* Timeline: reorder anyway (allow_errors) and report the features the move breaks (#5962)
timeline.reorder and timeline.reorder_many refuse a move after which a
feature that worked would fail. Sometimes breaking a later feature is the
point: an agent restructuring a design wants the move made, then the list
of what broke so it can repair those features next.
Both take allow_errors (default false: refused as before). With it the
move is made and the reply lists broken: [{feature, error}] (empty when
nothing broke); one undo puts the order back. timeline.reorder_many also
takes active, as timeline.reorder does.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A pattern count such as (L-2*Post)/210 is a length when the 210 is unit-less, and was refused with only "expected a unit-less number, got a length". The message now says to divide by a length, e.g. floor(L / (210 mm)).
Co-authored-by: Nathalie Schönwetter <nathalie@cfw-ag.ch>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
render.settings.set stores exposure_ev and the environment controls, but
render.preflight and render.start refused any setup with them because the
CPU image executor doesn't apply them, so such a setup could never render.
The CPU executor now renders it and the replies list the left-out
controls in ignored_controls.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4230
* Ribbon: an active contextual tab looks selected (#4230)
Contextual tabs (Sketch while editing, Sheet Metal for a sheet body) chose
their solid accent fill before checking whether they were active, so
selected and unselected looked the same.
The selected contextual tab keeps the solid accent pill. An unselected one
now looks like any unselected tab with its accent as an outline, so it still
marks the context without outshining the tab that is selected. Ordinary tabs
are unchanged.
* claim: #4363
* A straight 3D sketch line can be picked as an axis (#4363)
3D sketch wires were always offered as curved, and axis inputs resolved
only 2D sketch lines, so a straight 3D line couldn't be an axis (Plane at
Angle and others). A single-segment wire is now offered as straight and
resolves to its endpoints in model space; splines are still refused.
Combine, Loft, Mirror, Circular pattern, Rectangular pattern and Sweep get the
same two-sentence help as the first six (en, fr, ja, zh). Text only.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Removing a design from the Start page's recent list only changed memory, and on the web sync_recents() rebuilt the list from stored designs on each launch, adding it back. Track removed designs in a persisted 'forgotten' set that sync_recents honours; opening or saving a design, or deleting/renaming its file, clears it.
* Viewport: show the running tool's icon beside the pointer
While an interactive tool or a command's dialog runs, its toolbar icon
rides above and right of the pointer in the viewport, so the next click's
command is visible without looking up at the ribbon. It sits clear of the
system arrow and of the sketch tools' coordinate readout below the
pointer, moves left at the right edge and below the readout at the top.
It shows only where a click reaches the model: not over panels, dialogs,
menus or the navigation bar, and not while the camera moves. The icon is
the command's own (the one the ribbon and the dialog header show), so
there is no new artwork. On by default; "Show tool at cursor" in View
settings (`ui.set {"toolCursor": false}`) turns it off.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Viewport: the tool at the cursor is off by default
As asked in review: the badge starts off, and "Show tool at cursor" in
View settings (or `ui.set {"toolCursor": true}`) turns it on. The test
checks the default and turns it on through the View settings checkbox.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5244
* Sketch Create menu: family rows show their icons
The Line, Rectangle, Circle, Arc, Slot, Polygon, and Spline entries in a
sketch's Create drop-down were drawn as plain text buttons. Draw each as
the shared menu row with its first variant's icon and a submenu arrow,
keeping the same popup handling for its variants.
* claim: #5304
* Sketch: undo and Backspace remove the last pending spline point (#5304)
While a spline tool has points placed, Cmd/Ctrl+Z and Backspace take back
the last point instead of undoing sketch history or deleting the selection.
Undo reaches sketch history again once no spline points remain.
* claim: #5288
* Mirror keeps the preselected feature when the object type changes (#5288)
The pre-selection moves into the objects input when a pattern window
opens, and switching Object Type cleared that input, so switching to
Bodies and back lost the selected feature. The window now keeps the
objects of the type not shown and restores them when it is chosen
again. A plain timeline click on a feature already picked keeps it;
Ctrl+click takes it out.
* claim: #5415
* Floating history: the last done feature is no longer drawn as selected (#5415)
The floating history strip passed the last feature before the marker to
HistoryChip as selected, so it looked selected with nothing selected.
Pass only the real selection, and mark the last done feature with a small
tab-accent dot under its chip instead.
* claim: #5546
* Agent chat: hand WSL paths to agents started through wsl.exe (#5546)
An ACP agent started through wsl.exe was handed the session's working
folder and the solidcraft-app bridge's command as Windows paths, which it
can't use: the Claude adapter refuses the session, and the bridge never
starts. When the agent's program is wsl or wsl.exe, convert both to their
WSL form (C:\Users\... -> /mnt/c/Users/...) before session/new, in the
chat and in the Test button. The bridge still runs on Windows and reaches
the app at 127.0.0.1.
The Preferences > Navigation overall speed slider stopped at 3.00x, but
`space_mouse::frame` clamps the applied speed to 0.1..=5.0, so the top of
the range the engine honours could not be reached from the UI. Raise the
slider's ceiling to 5.0x to match that clamp; the step, format and lower
bound are unchanged.
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
* claim: #5680
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Preferences: draw the Z up orientation figure right-handed (#5680)
Z up showed Y pointing toward the viewer on the left, a mirrored
(left-handed) figure. Y now points away on the right, like the axis
gizmo. The axis tips move into orientation_axes and a test checks that
X to Y and Y to Z turn counter-clockwise in both figures.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #5488
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Link the version number in the title bar to its release notes (#5488)
The version at the right of the title bar opens the release page of this
build (v<version>, which exists for stable, candidate and nightly
releases) in the browser; a development build opens the release history.
Its click area stops at the report button, which the long version
strings of candidates and nightlies overlap.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #5828
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Hem: the handle runs along the hem to its tip (#5828)
The hem's on-canvas handle started on the picked edge and pointed square
to the sheet (#4794 only flipped it), so it neither sat on the hem nor
moved the way the Length grows. Length is measured from the outer face
of the fold to the tip, so the handle now starts on the centre line of
the hem layer at that face and points along the hem toward the tip.
The test builds the hem for both Flip states and checks the handle's
direction, its start on the layer's centre line, and that start plus
Length lands on the built tip. It replaces the arrow-side test of #4836,
whose rule (the arrow points to the side the hem builds on) no longer
applies.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #5850
* Find in Timeline scrolls both timeline layouts to the found feature (#5850)
Find in Timeline selected the feature but neither the bottom History strip
nor the history on the left scrolled to it. The handler now leaves a one-shot
reveal request (TreeState::timeline_reveal); whichever layout draws its items
next scrolls the found item into the middle of its scroll area and drops it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5855
* Ignore middle-click clipboard paste during navigation (#5855)
Firefox on Linux sends a primary-selection paste event when the middle
button orbits or pans. Paste events on frames where the middle button is
pressed, held, or released no longer reach the design; Ctrl/Cmd+V still
pastes.
* claim: #5905
* Drive Joints: a Show ghost checkbox (#5905)
Driving a joint draws the moved parts' old place as a translucent ghost,
which gets in the way when driving joints over and over. Drive Joints has
a Show ghost box under its value rows: on by default (nothing changes),
off draws only the moved parts at their new place. The stopped-short
warning stays. The choice is a preference (drive_joints_ghost), so it is
remembered between uses; it affects Drive Joints only.
The preview's Colors carry the choice (ghost), so toggling it recomputes
the preview (its key hashes the colours) and world_preview leaves the
ghost out.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5868
* Design tabs: the tooltip says where the file is saved (#5868)
Hovering a design tab showed the design's name, the same text as the tab.
The tooltip now says where the file is saved, with a path in the home
folder shortened to ~/ (home::friendly_path, as the start page does), or
"Not saved yet" for a design never saved. When the tab is too narrow for
its name, the full name comes first, then the place.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Look At already has a key (Ctrl+8, Cmd+8 on a Mac), but the button's
tooltip didn't say so, as Fit's does (F6).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
desktop_path_can_start_an_agent_and_its_runtime gave the version probe
3 s and the first reply 5 s. Under a full test run on a loaded macOS
machine the probe came back empty (version None, twice in a row), while
the test alone passes in 0.4 s. Both limits are now 20 s; a working
agent still answers at once, so the test takes no longer.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat: a Mammouth AI preset for OpenAI-compatible endpoints (#5815)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat: the key test names the Mammouth AI URL too (#5815)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With the Thickness box cleared, the dialog sent thickness: "" and the
command refused it ("`thickness` must be a number or an expression"),
shown as a red error with Create Issue. The dialog now makes no command
until a thickness is entered.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Where a body's face lies flush with another body's face, the face point
alone could resolve to the other body, and Shell then refused with
"select at least one face to remove". The dialog now passes the body of
the picked faces, as the command already accepts.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With a sketch profile and a body face picked together, OK ran the
profile's extrude first and then looked the face up by its point. Where
the profile's extrude covered the face, it was gone: "no face of Body1
at face". A new extrude now runs its face picks first; an edit still
rebuilds the profile's extrude first.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Plane at Angle's OK with an empty Plane or Axis input showed "select a
plane first" or "pick an axis" as a red error with a Create Issue
button, and people filed it as a bug (#3292, #3773). A refusal that only
names what to pick is now shown as a plain hint, with no issue button,
in every dialog (Sketch Modify dialogs already hid it).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
sheet.export_dxf always wrote the dashed bend lines (BEND_UP,
BEND_DOWN) and split the flat pattern over a layer for each role. A
cutter that cuts every line it is given cut the bend lines too.
bend_lines: false now leaves them out, and layers: single writes every
entity on layer 0. The defaults keep today's output.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Test: a saved model keeps each sheet body once
A sheet-metal design wrote its sheet bodies again with every later feature's state in the saved
model: each feature that leaves the sheet alone (a box beside it) adds more than the whole sheet
to the records. A design of many features after its sheet metal passed the cache's 32 MiB
manifest limit that way and kept only some of its states (#3220). The test also opens the saved
file and wants the sheets taken from it as evaluated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
* Model cache: each sheet body is written once, not with every state
Problem: a saved model writes every feature's state, and each state wrote its sheet bodies in
full; most features leave the sheets as they were, so a sheet-metal design of many features
repeated them over and over (a 280-feature part: 30.5 MB of manifest, 29.6 MB of it 69 distinct
sheet records written 243 times). Past the manifest's 32 MiB limit the save thinned the cache
(#3220) and opening evaluated the left-out features again, although the distinct data was small.
Solution: the records keep a list of sheet bodies, each written once; a state names its sheets by
their place in it, as it already does for bodies and sketches. A thinned model keeps the whole
list and counts it once against its budget; the reader counts sheets with the bodies and sketches
a state holds; the cache format is now /4 (an older cache is a miss).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PsiqyXWBFGTCX9J51bhysc
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6161
* Preferences: the theme strip shows arrows where more themes are (#6161)
The strip scrolls sideways with a hidden scrollbar, so nothing showed that
more themes were past its end. An end with themes past it now shows a
chevron button over the tiles there; a click scrolls most of a page.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #6304
* view.pick: map screenshot pixels to exact geometry (#6304)
A read-only command that answers, for screenshot pixels (window or model
frame) or world-space rays: body, face index, exact world point and
outward normal, depth, occurrence/component for placed parts, and the
nearest edge/vertex within edge_px. Faces are raycast through the same
per-body pick index the mouse uses; flat faces are solved exactly
against their plane, curved faces take the mesh hit with the normal
interpolated from the mesh's surface normals. Section-aware: faces on
the removed side do not stop a pick and the cut cap answers with
section_cap. Never touches the selection, camera, document or undo.
Exposed as control ui.pick and as view.pick through engine.execute and
engine.batch (and through them MCP execute and batch).
---------
Co-authored-by: Dahak Builds <dahakbuilds@users.noreply.github.com>
* claim: #320
* Sketch: toggle a selected dimension between driving and driven (#320)
A selected sketch dimension now offers 'Toggle Driven Dimension' in the
viewport right-click menu, running the existing engine command
sketch.toggle_driven. The over-constrain -> driven placement half landed
in #366; this exposes the engine's toggle in the UI, the remaining gap
from the retest notes on #320.
- context_menu: sketch_entity_items collects selected SketchConstraint
ids that are dimensions (kind.value() is Some) and, for a single
selected dimension, prepends the toggle item aimed at that constraint.
Non-dimension constraints and plain entities get no item.
- Label reuses the existing catalog entry cmd.sketch_toggle_driven, so
every locale translates it with no catalog change.
- Test: the menu offers the item for a selected dimension, running it
round-trips driving -> driven -> driving, and a horizontal constraint
or a plain curve gets no item.
---------
Co-authored-by: Dahak Builds <dahakbuilds@users.noreply.github.com>
* Test: a long configuration table builds the cells of the rows on screen
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Configurations: the table builds the rows on screen and its cells once per change
A table of hundreds of configurations was slow to open and to use: every frame cloned the
whole config.table answer, rebuilt every cell's text, looked each parameter's unit up once
per cell and built an editor for every cell of every row, on screen or not.
The answer and its cells (with units, looked up once per column) are now kept per design
revision, and rows off screen keep their space without building their cells, as the
parameters table already does. A row with an edit in progress stays live, so leaving its
cell still commits the edit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The taskbar asks a window for its icon once, when the window appears. eframe
sets the icon while the window is still hidden and shows it after the first
frame; when that frame is slow (safe mode's software renderer) or startup kept
the UI thread busy long enough for Windows to mark the window as not
responding (building the viewport pipelines), the request isn't answered in
time, the taskbar falls back to the window class's icon, and winit's class has
none: the button showed Windows' generic icon for as long as the window was
open.
Give the class the exe's embedded icon (resource 1) before the window is shown.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* inspect.parameter_sweep: check a design across a parameter's range
Rebuilds a copy of the design at each value of a parameter and reports what
no longer holds compared with the design as it is: features that fail or
warn, joint origins that are lost or turn over on their own part, sketch
planes that turn over, and sketch solves that fail or fold. The design is
not changed; each value's rebuild time is reported too.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* inspect.parameter_sweep: params doc matches what it takes and returns
`steps` takes at most 199 (200 values), as the check says; failed and
warned entries carry the feature's name; a lost origin says which one; a
value the parameter refuses gives {value, ms, ok: false, error}. The
refusal test covers steps 200.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* inspect.parameter_sweep: Japanese and Chinese labels
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Aac7mP9rthK3WsPpPRvCsF
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* claim: #3988 (3D mouse: follow the device that moves, not the first one found)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse: follow the device that moves, not the first one found (#3988)
On Windows and macOS the reader opened the first 3D-mouse interface it
could and read it until it failed. A 3Dconnexion Universal Receiver with
nothing paired opens fine, has a valid layout and never fails, so it kept
the reader for good. A SpaceMouse on a cable next to it was never read
("Not calibrated (hid 256f:c652)", every axis at zero).
The reader now opens every candidate and follows the one in use. The
first is followed from the start, as before. Another takes over once it
shows use (an axis off zero or a button down) while the followed one has
shown none for STALE_AFTER, so two devices moved together don't trade
the reading. A candidate that fails is dropped; if it was the followed
one the reading resets and the next is read. With one device the reader
behaves as before, down to the 100 ms read timeout. Several take 10 ms
turns, and reports queue in the system meanwhile.
The selection is `follow()`, written against a small `ReportSource`
trait, so it is tested with simulated devices on every platform: an idle
receiver next to a SpaceMouse, two devices in use at once, and devices
failing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse: a press that makes a device the one in use counts as a press
Since #3825 a HID device's first button report after it connects is its
baseline: a button down in it is held, not pressed. A device taking over
from another one was connected at that moment, so the press that took
over was swallowed. With an idle Universal Receiver found first, that is
the first button press on the SpaceMouse.
follow() now keeps the buttons each candidate last reported while another
was read, and the reader hands them to the reading as the baseline when
that candidate takes over. A device that reported no buttons before it
took over keeps the held-not-pressed rule.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The ribbon's command-search field had its own area at `Order::Foreground`, so
it painted above every system window: a Preferences card moved under the top
right corner showed the search field floating on top of it, while the rest of
the ribbon went under the card as it should.
Closed, the field is ribbon chrome and now draws at `Order::Background` with
the rest of the ribbon; open (or animating open or shut), it keeps
`Order::Foreground` so the results sheet still floats over every window.
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
On 0.21.0 the facts test failed on rectangle symmetry, which now works
on two sides. Other statements in the modeling and mechanisms skills
also no longer match: edge picks have a distance limit, concentric
circles extrude as a ring, a hole sketch below a part drills into it,
slot ids count on from earlier curves, shell refuses curved blends and
lofts, measure and inspect_design grow with the body count, and
joint.sweep needs occurrences with against.
Each correction has a test in test_facts.py.
* Fillet/chamfer: find each edge on its own body
A blend picked over edges of several bodies was evaluated on one body,
the one nearest the first pick, and every edge was named on that body
when the feature was made. Picks on the other bodies were named after
that body's nearest edge and rounded it instead (or moved to it by
position once an upstream edit moved them). With two stacked boxes, a
height change put the upper box's picks on the lower box's rounded
corners: the upper box's fillet resolved there and its own edges lost
their rounds.
Now each edge is named on the body it was picked on, and evaluation
finds each edge's body by its stored name (a gone name stays on the
feature's other bodies), groups the edges by body and blends each body.
A name an older build recorded on the wrong body (the pick lies outside
the box its reference recorded) is not used; the pick finds the edge.
A pick an upstream edit left at an end of its named edge goes to the
edge's middle, so the kernel cannot take the neighbouring edge.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Fillet/chamfer: skip the per-body lookup when there is one body; name each body's picks in one pass
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Performance overlay: frame-time graph and histogram, triangles, draw calls (#2548)
Net change of the perf-overlay branch, rebased onto main as one commit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Document frame_rate_details in ui.prefs
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #4971
* Thread: the refusal for a fine-only size says what to type (#4971)
M28 and other sizes without an ISO coarse pitch were refused as an
unknown thread, with examples of other sizes. Thread and threaded Hole
now say that the size has no coarse pitch and suggest two to give, such
as M28x2 or M28x1.5. The coarse table is unchanged.
* Thread: automatic sizing past M56 says what to type (#5055)
Automatic thread sizing only knows the ISO coarse sizes up to M56, so a
75 mm shaft was refused with "no metric thread fits a 75.00 mm
cylinder", although M75x2 and M75x3 work when typed. Past the end of
the coarse range, the refusal now says so and suggests two sizes with
their pitch, scaled to the diameter (a hole's size adds one pitch to its
bore). The coarse table is unchanged.
MidPoint on a point and a line keeps the first pick still and moves the second.
With the point picked first (an end of one line, then the line it should sit in
the middle of), the line turned about its far end to bring its middle to the
point: a line from (0,0) to (20,0) went to (0,0)-(10,10) for a point at (5,5).
A point and a line now put the line first, so the point moves and the line
stays; the line still comes to the point when the point can't move.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5009
* Assistant: a request timeout per endpoint for slow models (#5009)
An OpenAI-compatible endpoint now stores how long one model request may
take (request_timeout_s, 300 s by default, clamped to 30 s ..= 1 h). It
is set in the Add agent dialog and in the endpoint's entry under Agents
(a running conversation uses a change from its next request), through
ui.agent_chat add_agent, and reported in the chat's state. A request
that runs out of time says the limit and where to raise it instead of
"timeout: global". The desktop HTTP cap goes from 900 s to an hour.
The tool-call timeout stays 330 s: it covers SolidCraft's own tool work
and the user's Allow, never the model.
* Assistant settings: line up an endpoint's Request timeout with Permissions
The label column of an agent's entry is wide enough for "Request timeout",
so the label isn't cut and both rows share one column.
* Undo: what is hidden lives in the snapshot, so hiding is a step (#2219)
`Session::execute` records a step only when the document `Arc` changes, and what is
hidden lives in `Session::visibility` — saved with the design, but outside the
document. `browser.visibility` was also declared `.noundo()`, and `Snapshot` had no
field for it, so even a recorded step could not put a hidden body back: undo went to
the command before the hide and left it hidden (#2219).
`Snapshot` and `Checkpoint` now carry the `Visibility`, `go_to` restores it, and
`Session::execute` treats a visibility change as a change of its own. That change is
display-only, so it refreshes the view with a revision bump (`refresh_display`)
instead of evaluating a model a hide does not touch.
The command also takes `folder: "all"` (every body, sketch, construction item, origin
item and joint, plus the browser folders and occurrence hides) so Show All can be one
step, and asking for the state an item already has no longer reorders the hidden
lists: a redundant hide is not a step and leaves no unsaved changes.
* Show/Hide: the menus run browser.visibility, so the history has it (#2219)
The hide and show actions wrote `session.visibility` directly instead of running the
command: the body and face menus (`ui.hide` / `ui.show`), the `V` key
(`context_menu::toggle_visibility`), the browser rows' eyes, the Origin rows and
`ui.originAll` / `ui.originPart`. The engine could record a step and there was
nothing to record — which is why the release still answered "nothing to undo" after
hiding from the Browser.
They all run `browser.visibility` now (the repo rule that the UI acts through
commands), and `ui.showAll` is that one `folder: "all"` call, so Show All is one undo
step rather than three. `docs/occurrence-workflows.md` says what visibility does in
the history.
assets/locales/ko.json had 298 ids that en.json has and it lacks (the
help.solid.* texts, the whole Preferences page, the issue-report dialog,
history names, ribs, mouse gestures and navigation presets). Add Korean for
297 of them; "Of" is left untranslated because its context is unclear.
Wording follows the existing ko.json (필렛, 모따기, 돌출, 쉘, 바디, 피처,
비활성화, 환경 설정, 어시스턴트, 에이전트). The file stays in the canonical
sorted one-message-per-line form; no existing message changed.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #4337
* Measure picks projected and intersection sketch curves (#4337)
Dialog inputs turned away every curve or point of a sketch that is not the
active one, which includes projected and intersection curves, so Measure
took the nearest item it did accept instead. Give Measure explicit
acceptance for these references, keep the source sketch on a click, and
accept them from a pre-selection and a box selection. Other inputs keep
their restrictions.
* claim: #3892
* Browser tree hover stays fast on a nested STEP assembly (#3892)
Hovering a placement row lit its whole subtree by walking it again
(Session::occurrence_bodies) on every frame, and opening a STEP file
makes its top placement active, so every highlight rebuild walked the
whole assembly again for the active placement's bodies. On a nested
assembly each walk resolves every placement below it, which cost
0.5 s or more per walk.
- Session::occurrence_bodies remembers its answers per placement until
the document, the evaluated model or the pending moves change
(occurrence_bodies_shared hands out the shared list).
- The highlight looks hovered bodies and faces up by name instead of
scanning the hovered list once per body.
* claim: #3222
* Sketch drag: project a face sketch's links once per live gesture (#3222)
Every live sketch.drag frame refreshed the sketch's links before solving
the drag. For a sketch on a face that projects all of the face's edges
again: 115-460 ms per frame on the top face of a 600 x 500 mm plate with
120 holes imported from STEP, against about 1 ms for the drag's own solve.
The model the links come from cannot change during the gesture.
The session now remembers the sketch, document and revision a live frame
left. The gesture's next frame reuses the links while nothing else has
run: any other command (undo included) or any other change to the design
makes the next frame project them again. A test counts the projections.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch drag test: drop an unread assignment (#3222)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3883
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Find in Browser finds the body; Find in Timeline opens the feature's folders (#3883)
A right-click on a body in the viewport picks a face, and the face menu's
Find in Browser selected the face's creating feature in the Features folder
instead of the body. It now reveals and selects the body, as the edge and
body menus already did.
Find in Timeline now opens any collapsed history folders around the feature,
and opens the Browser's Features folder for the picked occurrence, scrolled
to the feature (the reveal formerly behind Find in Browser).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Find in Browser opens collapsed rows above the body and finds later placements (#3883)
Open the document, placement, Bodies and group rows when the command runs, not while the
Bodies folder is drawn, so a collapsed ancestor no longer hides the body. Match the scroll
target by the placed body's world name, and fix the group row key ({scope}/g{id}).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Find in Browser picks the body's own placement; Find in Timeline leaves the Browser alone from rows (#3883)
Every component's first placement keeps the definition's body name, so the root placement
matched before the body's own one and Find in Browser on a first copy opened nothing above it:
the placement must belong to the body's component too.
Find in Timeline from a sketch, construction plane or feature row has no body to say which copy
was meant, so it opened the Features folder of whichever placement came first (the other copy),
or one that never lists sketches. From those rows it now only reveals the feature in the
History; from a face, edge or body it still opens that body's copy's Features folder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With an opened STEP assembly's imported root active (it owns no body), a
Join revolve, sweep or loft from a sketch on a nested part's face made a
separate body, and a hole at the points of such a sketch found no body to
cut. A Join with no targets now takes the body its profile sketch is on,
through the same active scope as an extrude, and a hole counts as a cut
for the rule that a cut from a component with no body works on every body.
Refs #3223
* claim: #4236
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Coil: store an optional sketch and profile selection
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Coil: sweep a sketch profile along the helix
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Coil: accept a sketch and profile selection
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Coil dialog: pick a sketch profile for the section
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Coil: cargo fmt
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Coil: refuse a loaded sketch coil that also sets the primitive section
A saved design could add section, section_position or start_angle to a
coil that already has a sketch; eval ignored them instead of refusing the
contradiction as the command does.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Coil: assert the kernel refusal and exercise hostile profile selections
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Coil dialog: keep the sketch-profile check after rebase
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Format
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Code <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* feat: extrude a face or profile by pressing on it and pulling
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* feat: pull-to-extrude keeps the window open and sets the step by sideways motion
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Pull to extrude: coarser step changes, a step ladder below the pointer, the step in the distance box
The step label sat on top of the distance box at the arrow tip. It is now a ladder of the
step in use between its neighbours, below the pointer, and the distance box also names the
step ("Step 1 mm"). One step change takes 44 px of sideways travel instead of 22.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Pull to extrude: a setting, off by default (Preferences > Navigation)
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Rename the pull setting to Direct interaction
It will also cover other direct gestures in the viewport later (grabbing an edge to round it).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Nathalie Schönwetter <nathalie@cfw-ag.ch>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #4853
* Sketch on a turned occurrence: reuse moved bodies across edits (#4853)
A sketch with a face anchor is evaluated in its component's frame.
state_in_frame_at moved every body placed differently from that frame
with a fresh transform on every evaluation, and the face-anchor frame
moved the anchored body again, so each sketch edit on a turned or moved
part transformed and re-meshed it.
The model now keeps the bodies it moved into a feature's frame, keyed on
the source body and the exact matrix (and checked against the source's
mesh, geometry, colour, paint and split lines, as world placement does).
An unchanged body under an unchanged placement is moved and meshed once.
Each evaluation starts a new generation and drops entries that neither
it nor the one before used, so the cache holds only placements in use.
Model states hold a weak reference, so the cache never keeps earlier
states alive.
Fixes#4853
A circular or rectangular pattern of a hole and the chamfers on its edges failed with
"no edge near ..." when a chamfer was picked before the hole: the chamfer was copied
first, onto a body that had no hole there yet.
Co-authored-by: Nathalie Schönwetter <nathalie@cfw-ag.ch>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
projection() reads the model's box for the scene's depth, several times a
frame, and model_bounds() sampled every sketch's curves for it each time,
hidden sketches too. A 533-feature design with 445 sketches spent about 45 ms
of every frame there (stack samples: model_bounds -> Sketch::bounds -> segs ->
spline polylines), so orbiting ran near 16 fps. The sketches' box is now kept
for the world state it was made from, as the bodies' boxes already are.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4099
* Measure: a face or edge picked by its point alone (#4099)
inspect.measure (and the MCP measure tool) refused a face or edge item
without an index ("missing field `index`"). Face and edge items are now
read as body + point with an optional index: the point finds the face
(the same lookup face picking uses) or the nearest edge within the pick
tolerance, before the selection is built. A missing body or a point on
no face or edge is a clear error. Explicit indices still work.
* claim: #4098
* Extrude: document which way a positive taper goes (#4098)
A positive taper slopes the walls outward (the far end is larger along the
extrusion direction), a negative one inward. Say so in the params doc, the
JSON schema and the guide, and pin the sign with a test.
* Offset of free-form curves: trim folds, round chain corners
Offsetting a spline by more than its tightest bend failed with "offset spline
has no interior certificate correspondence": the plain offset folds over itself
there and a single linked spline cannot follow it. A chain of Bézier curves (any
imported SVG outline) was refused outright.
Such an offset is now the parallel boundary: the plain offset of every piece,
joined round each corner by an arc about it, trimmed where it folds or crosses
itself (a sharp corner is left), each kept run fitted with an open spline and
certified to the offset tolerance, and the whole proven free of crossings. Tiny
kinks are joined directly; zero-length Bézier handles (infinite curvature at
the end) are replaced by their chord within a quarter of the tolerance. A
single curve whose plain offset is smooth still gets the linked offset; the
trimmed or chained result is ordinary curves, not linked to the source, so the
command adds no distance parameter for it.
* Offset chains mixing lines or arcs with free-form curves
A chain offset read every curve with Sketch::exact_segs, which only covers free-form
curves, so a chain with straight or circular pieces (a text glyph, an SVG path with
L commands) failed with "offset source changed shape". Read them with exact_pieces,
which gives lines and arcs from their analytic form; the trimming already handles
those pieces.
The solver takes at most 4000 unknowns in one connected group, and a
pattern joins all its copies into one, so a hexagon patterned 17 × 20
(4000+ unknowns) failed as "that would over-constrain the sketch
(conflicts with )", naming no constraint. The solve report now carries
the size of the largest group it could not take, and the failure says
how many unknowns are joined, the limit, and that fewer copies or
several patterns or sketches stay within it.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3786
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Sketch: the first dimension scales only a connected drawing (#3786)
The first length dimension scaled the whole sketch about its origin, so a
second rectangle that nothing links to the first was resized and moved too.
Now the sketch scales only when its curves form one chain, joined by shared
points or constraints (the origin counts as a point); separate shapes keep
the previous behaviour: the dimension sizes the shape it measures.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Dragging one end of a line tangent to a circle or arc let the solver move the
line's other end too, sometimes far ("goes flying"): only the dragged point was
held, and the least-squares step spread the change over the free points. The
drag now first tries holding the line's far end as well, and falls back to the
old solve only when that can't be met. Dragging the end of a tangent line to
(5, 12) moved its other end 0.399 mm on 0.19.0; it now stays put.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Release 0.22.0: version bump, design format 35
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Release 0.22.0: tests no longer expect the folded render and material changes to be pending
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A cut starting on a face of its target runs its tool on behind the start plane with the
tool's slope. A taper that widens the pocket narrows the tool there, and a corner arc
smaller than the run on times tan(taper) closed: "the taper closes the profile", though
the pocket itself only grows. The run on is halved, quartered, then dropped (the tool
starts on the face with its straight foot) until the tool builds.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The fastener tables stopped at M3 and M20. The socket head cap screw (ISO
4762), hex bolt (ISO 4014), hex nut (ISO 4032) and plain washer (ISO 7089)
tables, and the coarse pitches they use, now have M2, M2.5, M14, M24 and M30
too, in size order. Set screws and dowel pins are unchanged.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5202
* History: show the sketch at the marker while it's the current step (#5202)
A sketch a later feature uses is hidden, so moving the History marker
onto it changed nothing in the view. The sketch immediately before the
marker is now drawn even when consumed or hidden; its stored visibility
is untouched, and earlier sketches, hidden components, groups and the
Sketches folder still apply.
* claim: #4717
* Canvas: a plane seen edge-on doesn't take the image (#4717)
Placed with a click at a sketch's origin, the canvas went onto the origin's XZ square, which
is under the cursor there too and was the first plane hit. From the Top view that plane is
edge-on, so the image was out of sight and Canvas seemed to do nothing. A plane hit counts only
when the view shows it (its normal more than ~11.5 degrees off the view's edge-on direction); the
sketch plane, or another plane, takes the image instead.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With Sketch Dimension on and nothing picked yet, a click on a dimension
that is already in the sketch opens its value box, the same in-place
editor a double-click opens without a tool. Enter applies, Esc cancels,
and the tool stays on for the next dimension.
The dimension's text always takes the click; its lines only away from
sketch points and curves, which still start a new dimension (the rule
plain selection uses). Once a new dimension is under way, clicks keep
their meaning. Dimensions off in the selection filter are left alone.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3032
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Create Sketch leads the right-click menu on empty space too (#3032)
A right-click on empty space in the viewport now starts with Create Sketch, in
the same place as a right-click on a plane (#1354): first, then a separator.
Over a plane it sketches on that plane, as before; on empty space it is the
toolbar's Create Sketch, so the next click on a plane or planar face starts the
sketch there. The menu offers it once.
Tests: empty_space_leads_with_create_sketch (the list) and
create_sketch_from_empty_space_sketches_on_the_plane_clicked_next (the harness:
right-click empty space, pick it, click XY).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Slightly fix the extend tool
* Extend: add the line the preview draws, and stop an arc on another curve
sketch.extend reported "nothing to extend to" whenever nothing stood in the
line's way, even though the viewport previewed a line past its end; with
nothing in the way the extension now runs as far as the preview shows, on the
same measurement, so the click adds what was drawn.
An arc extension that ran into another circle or arc refused the target when
its crossing already belonged to another curve's end (a shared crossing point
of two circles): that point is now joined instead, so the extension closes onto
the geometry that is there.
---------
Co-authored-by: SolidCraft Dev <dev@solidcraft.local>
Co-authored-by: n0tfr3shm1lk <n0tfr3shm1lk@ukr.net>
* claim: #4854
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch palette: the Tie button shows "Tie", not its catalog key (#4854)
The button passed the key itself to ui_kit::button with Button::Normal,
which draws its label as given; it now passes the translated label. The
loose-ends test checks the button by its label.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A move changes no shape, but its preview copied the session, evaluated
the move and meshed the bodies again for every change of the values. While
the triad was dragged the part trailed the arrow by one of those
evaluations, and longer for a larger part. The preview now takes the
listed bodies' world meshes once and draws them moved by the move's
transform (about the world origin, as the Move feature turns bodies), the
original shown as removed and hidden as replaced; a copy keeps the
original. Anything it can't take (values that don't evaluate, a body not
shown, over 2 M triangles) goes to the scratch-design preview as before.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5020
* Insert SVG: refuse geometry scaled to infinity, and bound the profile grid (#5020)
A transform such as scale(1e308) turned a circle's radius into infinity.
The SVG reader now refuses non-finite transforms and geometry with a clear
error, and Grid::cells in the profile finder uses checked arithmetic, so an
infinite or huge box is checked against all shapes instead of wrapping the
cell count and allocating until memory runs out.
* Insert SVG: refuse lines with infinite ends; fix the grid test's cell count
A line whose transformed ends are infinite was dropped silently, because the
distance between them is NaN. It is now refused as out of range, like curves.
The new grid test expected one cell for a box spanning its own one-box grid;
such a box touches the 2 x 2 cells at its corners, so it expects 4.
* Ctrl+1…7 turn the view to the standard views (#5057)
Front, back, left, right, top, bottom and isometric, handled with the
other view keys beside Ctrl+8: animated like the view cube, reserved so
no command can take them, and ignored on the flat pattern page.
* Front … Isometric View are commands, Ctrl+1…7 their default keys (#5057)
The standard views become seven commands (view.front, view.back,
view.left, view.right, view.top, view.bottom, view.isometric) whose
default shortcuts are Ctrl+1…7, instead of keys the window handles
itself. They show in Keyboard Shortcuts, can be moved to other keys or
cleared, and the digits can be given to other commands. In the app they
turn the view as the view cube does; headless they set the session
camera, keeping target, distance and projection.
A design file keeps its evaluated model so that it opens without
evaluating the timeline. A feature whose boolean needed a retry records
it, and the kernel's BooleanDiagnostics leaves `rejected` out when it is
empty but has no default for it when read, so reading those records
failed ("saved model: diagnostics"), the saved model was refused, and
the design was evaluated on every open (17.6 s for the reporter's
17-feature design, 1.2 ms with this).
The reading side now takes an absent list as an empty one, without a
change to the kernel crate.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Open or close every Browser folder from the design or a component menu,
including folders that start closed. Display only: not saved, not an undo step.
A face orthogonal to the sketch plane projected every edge onto the same
line: coincident and overlapping pieces, one per notch, that the face
boundary check then refused on the next evaluation ("boundaries cross,
touch or have ambiguous topology") and marked lost.
A link layout whose straight pieces all fall on one line is now one line
between its two farthest points; vertices along it can be projected one
by one. The face-boundary and loop-identity checks skip such a silhouette,
which encloses nothing. Coincident lines are laid out once.
* Orbit about the cursor: over empty space, turn about the part in view (#2616)
With Orbit around → Point under the cursor, an orbit begun over empty space
turned about the middle of everything shown. For a part far from the origin and
anything at the origin (a small body, a sketch), that middle is halfway between,
far off screen, and the part flew out of view: a box at (50000, 50000, 0) with a
10 mm box at the origin went from (872, 535) on screen to (-277084, -279746) after
a 180 px orbit. It now turns about the middle of the bodies the view shows part
of, and only falls back to everything shown when no body is in view.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Restore LF line endings in the files this change edits
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5851
* Add a contribution skill for first-time contributors (#5851)
skills/solidcraft-contributing walks a new contributor from checking their
tools through fork setup, choosing and claiming an issue, a focused change
with checks and screenshots, and review. The skills index lists it and
the install commands point at tangency-labs/solidcraft.
* Skills install: copy the modelling skills only
The contributing skill is read from its link when someone wants to
contribute, so the install commands no longer put it in every project's
global skills folder.
#4349 fixed the start page treating a sample's shipped picture, which comes back under the
sample's key without a design, as its build failing. That race dropped the request with "the
sample could not be built" and made the s15_start_page_sample_save_reopen scenario fail
depending on timing. This adds the test that sends the picture while the sample is wanted and
checks that the sample still opens.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3034
* Pattern of features: show the direction arrow on the patterned feature (#3034)
A rectangular pattern whose Object Type is Features had no drag arrow
and no value box: the canvas anchor knew bodies, faces, edges and
sketch geometry but not a whole feature, so it returned nothing.
A picked feature now anchors at the middle of the geometry it made
(its bodies, or its faces on a body later features changed). The
pattern's preview stands in for the body the feature lies on, so the
anchor reads the feature's geometry including replaced bodies, as a
body pick reads its body.
* claim: #3713
* Exploded views export as an animation (#3713)
motion.export_animation takes an exploded view (view: id|name) as well as a
study or a joint: the parts come apart at a steady pace from the assembled
placement to where the view shows them, and back together with back: true
(exploded at the middle frame). 41 frames by default, 81 with back. The home
view frames the parts both assembled and exploded. As with explode.show, the
design keeps its placements, its history and its saved state.
explode.show and the export share the moves; moves a design file gives with
no finite translation are left out.
* Mechanisms skill: an exploded view's default frame count
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #2873
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Press Pull: route curved faces to Offset Face, with an arrow (#2873)
Q sent every non-edge selection to Extrude, which takes only profiles and
planar faces, so a hole's wall or a boss's side was refused. The engine's
solid.press_pull already offsets curved faces; the window now does too:
- Q on a curved face opens Offset Face; its Faces input takes curved faces.
- Q with nothing selected opens Extrude as before, and a first pick of a
curved face or an edge reopens Press Pull on it (Offset Face or Fillet).
- Offset Face's arrow follows a curved face's outward normal where it was
picked, so it can be dragged like an extrude's.
Fixes#2873
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4582
* Keyboard Shortcuts: a bound Shift key is named "Shift", not "Shift+ShiftLeft" (#4582)
The shortcut's name was the modifiers held plus the key itself, so binding the
Shift key alone showed "Shift+ShiftLeft": the modifier named twice, and the
left and right keys as two different shortcuts.
A modifier key is now the shortcut itself, and its two sides are the same one
(the app's modifiers collapse them), so the page shows what the user pressed.
A shortcut bound before the change ("Shift+ShiftLeft") keeps working: the
stored name is migrated when the preferences or a shortcuts file are read.
* Keymap: the bound-modifier test passes the new repeat flag to press
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #4664
* Control: name the token file and spot a stale one when a call is unauthorized (#4664)
When the app refuses a token that the MCP client read from a token file,
the error now names that file, when it was written and how long ago,
without printing the token. The app also writes `<port>.pid` beside the
token file (its process id and, on Linux, its start time) and removes
both at exit; when that process has exited, the error says the token
file is stale. The process check runs only after a refusal, and the
token file itself stays one line, so existing clients keep working.
Document force-quit leftovers and the MSIX AppData redirection.
* Bench assistant: read the token out of the new ClientToken
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
The Origin folder's eye, a finished sketch's Show Profile and Show Dimensions,
and body locks move from UiState into the design's saved visibility
(origin_hidden, hidden_profiles, shown_dimensions, locked_bodies), so each tab
keeps its own and they survive closing the design. The Sketches folder uses the
saved browser folder (c0/sketches in hidden_folders); ui.set/ui.inspect keep
showOrigin, showSketches, hiddenProfiles, shownDims and lockedBodies.
Preferences > Design > "Hide origin by default": a design that never toggled
its Origin eye follows the preference. New fields are a pending format change
(format/pending/browser-view-state.rs).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3862
* Coil: its parameters show and can be edited (#3862)
The Coil dialog now shows every setting solid.coil takes: the section
(circular or square), where it sits on the diameter (inside, center or
outside), the start angle and a Clockwise option. Edit Feature on a coil
reopens the dialog filled in from the feature, and OK redefines it;
before, it fell back to a timeline.edit line in the command search.
Settings left at their default are not sent, so OK on an unchanged edit
keeps the coil as it was.
* claim: #2224
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Agent chat: keep Send in view for a long message, open "View attached" (#2224)
The message box sat in a fixed 92-point reserve, so a long message grew it
past the panel and took Send, Stop and Attach view with it. The box is now a
panel of its own at the bottom: it grows upwards up to a cap, then scrolls,
and the conversation takes what is left.
The picture sent with a message is kept on it (once the agent starts on the
message), and the "View attached" caption opens it in a window.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Agent chat: a message's pictures as thumbnails with a hover preview and a viewer (#2224)
Instead of a window on the far side of the screen: the pictures of a message
show as small thumbnails under it, hovering one shows it bigger for as long as
the pointer stays, and a click opens a viewer over the app with the message's
pictures one at a time (previous and next, arrow keys, Esc). A picture carries
a note shown with it, so attachments of other kinds can use the same viewer.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #2286
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Agent chat: Attach selection sends what is selected in the model with the message (#2286)
A checkbox beside Attach view. The selection is read when the message is
sent (also for a queued one) and described: faces and edges with the names,
indices and points the tools take plus their size and place, features by name
and kind, bodies by name; at most 64, the rest counted. The agent gets it as
text after the message; the conversation shows the message as typed with the
things as grouped chips (hover for what the agent was told).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* WIP: the chat's context picker dialog (#2286)
* Agent chat: Pick context, a picker for elements, features and bodies (#2286)
A button beside Attach selection opens the picker: a dialog with Elements,
Features and Bodies inputs, so clicks in the model, the timeline and the
browser pick into the active one (clicking again takes one out), with the
usual highlight. OK puts the picks as chips with an x above the message box;
they go with the next message like an attached selection (text for the agent,
chips in the conversation) and the tray is empty again afterwards. Cancel
keeps what was picked before. The control channel opens it with pick.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Agent chat: Attach selection counts and describes what was clicked, not the pieces of a rim (#2286)
A click on a circular rim or a split surface selects all its pieces for the
commands. The checkbox counted them one by one and the agent got one entry
per piece (reported on macOS). They are now grouped back into the click they
came from: one chip, one entry with the whole size, and the indices of every
piece in its detail. Edge details are looked up by position as well as
index.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Agent chat: a message's picture is kept on exactly that message (#2224)
The turn now carries its message's place in the conversation (also through a
queue and a send again after a sign-in), so its picture is kept there, once,
and never on an older message that also wanted one (review of #2225).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Agent chat: use the UI kit's checkbox and button and the font tokens (guard tests)
Main's UI-kit guard tests reject raw egui checkboxes, buttons and font sizes in the
chat panel; Attach selection, Pick context, the chip remove button and the chip fonts
now come from ui_kit and Tokens.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* wip(render): reconcile completed material and image pipeline source with 0.20 (#5493)
Source-only integration reference for staged review, not a merge-ready aggregate.
Donor: 0xlillith/solidcraft@aea20b0115.
Upstream: 858e5995d6.
Retain original CPU/material contracts and explicit GPU donor provenance in
docs/render-materials-lab.md (#2067, #2191, #2780, #3306). Preserve released
format migrations and current navigation/startup/shell behavior. New saved
records use pending descriptors. The physical-card descriptor follows the
maintainer's material-engineering-cards naming request, with a legacy-lab
format-29 regression. Replace raw new panel hosts with shared containers.
Repository checks, complete workspace/all-target Clippy, and the independent
physical/resource authoring checks pass for this source. The fresh full local
CI invocation has reached test compilation and is not yet complete. Do not
transfer donor native output or prior test totals to this integration.
Generated review screenshots, video, package ZIP and machine logs remain at
the immutable personal evidence pin. Uncommitted geometric displacement is
backed up separately, not silently included. No kernel/vendor change, direct
main write, release or independent contributor-branch rewrite.
Co-authored-by: Christian Svensson <blue@cmd.nu>
* test(cli): pin the Compact camera-sample admission diagnostic (#5493)
The bounded request still refuses 1024x1024x64 samples before host initialization and writes no output. Update only the stale expected diagnostic to the existing Compact budget wording; production limits and code are unchanged. Exact-tree CLI regression suite: 14 passed, 0 failed. The full repository gate has been invoked separately and is not yet complete.
* Normalize UI container violation paths across platforms
Report repository-relative paths with forward slashes so diagnostics and the existing nested-package regression assertions agree on Windows and Unix. Keep the container detection rules unchanged.
Validation: official ui-containers gate and all seven focused xtask tests passed; rustfmt and diff checks passed. Only xtask/dependencies were compiled; no application build or heavy tests.
(cherry picked from commit 44b58d8e10)
* fix(render): preserve camera contracts and correct acceptance regressions (#5493)
Preserve Design view.apply semantics with a separate strict Render camera command. Retain existing translations when English-only aliases collide; add material-axis entries and camera-specific pitch labels. Compare complete validated engineering cards and retain job handles before cancellation.
Focused and native acceptance is pinned to tree f8a423b21f. Full CI has been invoked separately and is not claimed green. Landing follows the render integration policy in #5493.
* fix(render): resolve showcase licence and upstream resource hygiene (#5493)
Record SychPL's explicit MIT OR Apache-2.0 grant, retaining source/model bytes and the Ubuntu lettering notice. Remove 80 absent attribution rows, personal fork records, dead delivery attributes and three fork/resource-only workflows. Existing xtask assets now performs bounded offline hash checks and rejects orphan render attribution; no parallel draft CI or package installation. Original authoring reproduction remains explicit. Scoped xtask tests, assets, Clippy, formatting and prepared SELENE validation pass; no application or renderer-wide result is inferred.
* fix(render): align saved bounds and progressive sample resolution (#5493)
Address review r4364211669/r4364211677/r4364211681/r4364211691/r4364211695. Persisted light/frame domains respect the existing 1e9 document bound; related LED/rectangle/direction cases use the same bound. Workstation primary resolve shares the 4096 admission cap and film counters use checked conversions. Exact candidate: 485 CPU render tests, 21 render document tests and strict scoped Clippy pass, including positive boundary round trips and 2048/4096 end-to-end checkpoints. No released format/migration edits, relaxed test limits or full-stack acceptance claim.
* wip(render): isolate optional resources and harden sample handling (#5493)
Review work in progress: opt-in encoded render library with reproducible runtime catalog exports; native-only robust dependency through the workspace; checked bloom slices; per-sample point-light diagnostics; independent area-light spatial/film keys. Resource-module binaries in both configurations, generation, Cargo production feature graphs and attribution checks pass. Full engine/render regression and combined acceptance are pending: do not merge this donor as a constituent or treat scoped resource proof as runtime acceptance.
* fix(gpu): preserve donor allocations through the scope-safety repair (#5722)
Port the native-validated #5723 scope-cleanup correction into the experimental donor, retaining its adaptive, volume, aging, photographic and bloom allocations. Same reverse-drain helper and regression cases; the donor combination still needs its own compile/test gate and is not covered retroactively by render-base results.
* claim: #1369
* Reject unknown pending concepts after format folding (#1369)
* test(render): close bounded sampling and bloom review regressions (#5493)
Validate the optional-resource/lighting/dependency donor at its exact tree. Required Vulkan GPU 127/0/3; CPU renderer 487/0; engine Render 72/0; document render 21/0; lean library 7/0; strict scoped Clippy, assets/layers/lists/containers and lean-engine WASM pass. Preserve original failed controller/typing attempts. This is the draft donor/disposition source, not a replacement for constituent render PRs or final main acceptance.
* wip(render): preserve stable aging target validation candidate
Exact staged source from prior review work on donor 3b892eab5e. This is a recovery/review checkpoint, NOT accepted delivery. An actual command-path regression currently fails because the selected executor refuses ordered surface aging. No assertion, capability or admission limit is weakened. Native/UI/full combination acceptance remains pending. Saved/runtime split adds one pending descriptor and retains original file scope; no main/render branch is changed.
* test(render): preserve numeric setup intent at the admitted coordinate bound
Follow up jyviko review on #5494 without weakening the corrected saved-coordinate contract. The positive clicked fixture uses 1e9 instead of inadmissible 2e9; explicit out-of-range cases still refuse, and the full saved setup round-trips through JSON and the native container. Exact-source Render workspace 95/0 and strict all-target UI Clippy pass. This does not incorporate the separately failing stable-aging candidate or claim full current-render acceptance.
* test(render): use the real GPU executor for saved aging
Move successful native cold-reopen Check/Start coverage from the CPU-backed protocol fixture into the production GPU executor tests. Verify capture-index changes preserve the stable target and that enabling aging changes both PNG and raw EXR output. Retain explicit, allocation-free refusal coverage for executors without aging support. Production capabilities and admission checks are unchanged.
* fix(render): persist stable identities for targeted aging (#5493)
Separate saved RenderBodyTarget identities from captured mesh ordinals. Reject legacy ambiguous targets, resolve Check/Start after visibility changes and preserve native reopening, copies and undo. Unsupported executors still refuse; real AMD Vulkan regression completes saved aging after reordered capture.
Source tree 6b84c8ad passes document 23, engine Render 75, GPU effects 5 and Render UI 95 with no failures; scoped strict Clippy, WASM and repository gates pass. Fresh CLI/native reopen and clean/aged GPU exports pass with identical physical totals and inspected actual images. Whole current-render integration is separate.
* fix(render): reconcile combined CPU/GPU contracts on current upstream
Unify captured resources/status and CPU-vs-GPU request schemas after the real two-parent integration; remove obsolete duplicate Vulkan selector and retain combined platform-filter regression. Deduplicate only identical or superseded test definitions, retaining the stronger card metadata and reserved-executor assertions. No assertions, admission limits or feature refusals are weakened.
All targets of render dependencies, GPU, engine, UI, CLI and native app pass cargo check with raytrace-cpu and bundled-render-library at tree c70d00d1. Native executable tests and full current-source acceptance remain pending. This is a private integration rehearsal for subsequent constituent PRs, not an aggregate official render/main merge.
* test(render): retain one complete host lifetime regression set
Deduplicate byte-identical DropProbe/task-lifetime tests and retain the adapted environment/texture ownership regressions with every assertion and complete memory estimate. Keep the registered-geometry admission regression once; only struct field order differed. Replace a one-element diagnostic loop with its equivalent block, preserving all wrap/width assertions.
Test-only integration repair. Formatting and diff checks pass; refreshed executable validation is required. No production behavior, limits, assertions or ignores changed.
* fix(doc): recognize the proven format-29 card descriptor alias
Normalize only the historical format-29 physical-material-cards spelling, and only when material-engineering-cards is registered. Admission and migration de-duplication use the same identity. Unknown concepts, wrong versions and incapable readers still refuse before mutation. Six controlled-history Rust regressions and actual CLI migration/no-clobber acceptance pass. No version, released migration or CI changes. Refs #5793, #1369, #5493.
* test(render): preserve grounding refusal before explicit instance moves
Update the two Render binding fixtures for upstream first-occurrence grounding. Verify grounded movement refuses without changing document/history, then explicitly unground through the public command before the intended move. Retain every target, resource, copy and persistence assertion. Refs #5493.
---------
Co-authored-by: Christian Svensson <blue@cmd.nu>
Co-authored-by: Victor Santos Oliveira <56417388+Vittuu@users.noreply.github.com>
Co-authored-by: cfaulkingham <colin.faulkingham@gmail.com>
* Normalize UI container violation paths across platforms (#5599)
Report repository-relative paths with forward slashes so diagnostics and the existing nested-package regression assertions agree on Windows and Unix. Keep the container detection rules unchanged.
Validation: official ui-containers gate and all seven focused xtask tests passed; rustfmt and diff checks passed. Only xtask/dependencies were compiled; no application build or heavy tests.
(cherry picked from commit 44b58d8e10)
Scoped render-branch prerequisite from #4507, coordinated under #5493. No unrelated translation/kernel changes are included. Fresh Windows fmt, seven ui_containers regressions and the production container gate pass on render base 3a6153d18f. Full CI is reserved for final render-to-main acceptance under the maintainer policy.
Co-authored-by: Victor Santos Oliveira <56417388+Vittuu@users.noreply.github.com>
* Preserve translated source-label aliases (#5619)
Render integration prerequisite for #5493. Fixes#5618. Retain stable-ID fallback and translated legacy captions; include controlled regression coverage.
* Persist validated engineering cards independently of mass and optics (#1060)
Render constituent under #5493. Fixes#1048. One pending material-engineering-cards descriptor, immutable released migrations, independent assigned density and source conditions. Nine persistence and ten physical-card regressions plus the fresh quick gate pass. Additional native follow-up is reported separately; no full-stack acceptance is implied.
* Render: progressive GPU path-tracing executor (#2067)
Land the unchanged contributor GPU foundation into render under the maintainer policy in #5493. Fresh quick gate 39993339605 passed. Original author and co-author attribution retained; no workflow file changes. Final render-to-main CI and jyviko sign-off remain separate.
Co-Authored-By: Christian Svensson <blue@cmd.nu>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Render: asynchronous GPU setup and completed-image panel (#2780)
Integrate the original asynchronous setup, progress and completed-image UI after #2067. Refreshed source passes quick gate 40009901424. Original attribution retained. Final render-to-main full CI and jyviko sign-off remain separate.
Co-Authored-By: Christian Svensson <blue@cmd.nu>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Render: studio lighting, clear coat and glass on GPU (#2191)
Integrate the original studio renderer after the GPU executor and async UI foundations, under #5493. Fresh render-target quick run 40011408339 passes. Preserve attribution, bounded resource admission and documented glass/denoising limits. Historical native benchmarks retain their source pins. Full renderer-to-main CI and jyviko sign-off remain separate.
Co-Authored-By: Christian Svensson <blue@cmd.nu>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Codex <noreply@openai.com>
* GPU path tracing: use compatible Vulkan adapters (#3306)
Land the final Vulkan-only path-executor policy after #2191 under the #5493 constituent route. Fixes#3217. Fresh quick run 40012565783 passed. Preserve adapter/software policy and CPU fallback; no DXC dependency or workflow changes. Full native/hardware acceptance and final jyviko sign-off remain separate.
Co-Authored-By: Christian Svensson <blue@cmd.nu>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test(material): initialize optional cards in upstream density regression
Fix the E0063 found by quick run 40015318093 while combining main #4785 with render #1060. The new upstream fixture explicitly has no engineering card. Keep every density/name/default-material assertion and production implementation unchanged.
* feat(render): expose independent anisotropic GGX core (#1694) (#1699)
Extract the original numerical source and analytic tests from published #1698 at 463a3e0d into its existing #1699 constituent. Keep current GPU/shared jobs, format and kernel untouched. Add scoped numerical-contract documentation. Full render library 158 passed, strict scoped Clippy and wasm pass, as do assets/layers/lists/format; source tree pinned in the acceptance record. No runtime consumer or final combined renderer acceptance is claimed.
* Sync current main into render, preserving release and viewport changes (#5705)
* Regenerate THIRD-PARTY-LICENSES.txt and docs/features.md [skip ci]
* Viewport: whole-body highlights reuse the model's own triangles (#4554) (#5109)
Hovering or selecting whole bodies -- a browser row, a History group, a body
under the pointer in body-pick mode -- built a CPU copy of every body's
triangles into the highlight scene. On a 40-body 4.3 MB STEP assembly that was
147 bodies, 70 MB of triangle copies and a 519 ms frame.
The model already has the same geometry on the GPU. The highlight now names it
and the renderer draws it again in the highlight colour:
- bodies drawn from a display plan: the model's evaluated plan chunks, with a
hover/selection tint uniform (the chunks stay resident in the model scene);
- bodies drawn as placements of shared triangles: the same placements of the
model's shared triangles, coloured per instance;
- bodies the model builds on the CPU: their vertex range in the model scene's
block.
See-through and faded bodies keep the CPU feedback, which colours them; faces
and edges highlighted on their own are unchanged. The highlight looks the same
(verified side by side).
On the assembly the highlight build drops from 519 ms to 9 ms and no longer
copies any triangles; selecting the whole assembly takes 9-20 ms.
Co-authored-by: Claude Code <noreply@anthropic.com>
* Offer to stand up SolidWorks files instead of turning them silently; add Turn Up Axis to the component menu (#5655)
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* Sketch on a picked origin plane uses the plane shown after standing an import up (#5659)
* Sketch on a picked origin plane uses the plane shown, also in a moved component
After a STEP file opens its component stays the active one, and standing it up
turns that component a quarter about X. Starting a sketch on XY picked in the
viewport then sent `{"plane": "XY"}`, which names the active component's own
XY: upright after the stand-up. The sketch landed on the world XZ, Look At
turned to face Y and the extrude went along Y.
The viewport and browser only show the design's origin planes, so a picked
origin plane now goes to sketch.create written out as that world plane while
the active component is moved; sketch.create maps it into the component's
frame once. The flip toward the viewer uses the same plane.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Stand-up axes tests: stand the file up through the offer when one is waiting
#5655 offers to stand up SolidWorks files instead of turning them silently.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* gpu tests: give the two test frames the new hover and selection tints (#5698)
#5109 added Frame::hover_tint and selection_tint; two test frames that landed alongside it didn't set them, so solidcraft-gpu's tests stopped compiling on main.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Release 0.21.0-rc.2 (#5691)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Jackson Oursland <jacob.oursland@gmail.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* GPU: preserve error scopes on cancelled preparation (#5722) (#5723)
* claim: #5722
* fix(gpu): unwind preparation error scopes on cancellation (#5722)
Move fallible CPU HDR packing before GPU scopes and drain the remaining scope stack in reverse order before propagating denoiser/resource allocation failures. Native regressions cover cancellation and allocation-error cleanup plus subsequent scope reuse. Exact candidate Vulkan-required GPU suite: 101 passed, 0 failed, 3 existing ignored; strict GPU all-target Clippy and repository assets/layers/lists/container checks pass. No driver exception, shader transport or test limit is suppressed.
* library: transport PBR and engineering cards without lossy copies (#1086)
* wip(library): reconcile exact rich-catalog replacement semantics (#1062)
Source-only scoped extraction from original 0f801962 / #1086 on the refreshed #1060 card prerequisite. Preserve current asynchronous picker guards and source-ID handles. Scalar v1 and rich-v2-without-PBR clear existing PBR; supplied rich PBR is retained. No new saved concept or duplicate card descriptor. Formatting and source review completed; compile/native acceptance is still pending.
* test(library): preserve explicit downgrade and fallible UI regressions (#1062)
Request scalar v1 explicitly in the card downgrade refusal regression; rich v2 remains the default lossless writer. Propagate fallible harness actions with ? in the shared clicked-catalog helper, preserving all assertions. No production code, limits or ignore changes. Fresh grouped IO/engine/UI tests, scoped strict Clippy and WASM pass on this exact tree.
* Render: defer path-host startup until first use (#5742) (#5743)
* claim: #5742
* fix(render): defer GPU path initialization until first use (#5742)
Retain the selected adapter factory during Design/safe-mode startup; instantiate once on first Render entry, including late viewport installation. Preserve the deferred/preparing/ready/failed state across document changes and retain selected fallback adapter policy. Exact tree passes 32 render-workspace and 1966 full UI tests (0 failures; 7 existing ignored), strict native all-target Clippy, wasm check and repository gates. Three wasm warnings remain in untouched upstream files. Fresh native desktop acceptance is separate and in progress; no whole-renderer/main sign-off implied.
* GPU: adapt dispatches and retire lost devices (#5754) (#5755)
* claim: #5754
* fix(gpu): adapt dispatch size and retire lost path devices (#5754)
Start with a 64-item quantum and adapt bounded work toward 20 ms from observed completion, including denoiser tiles. Never exceed 16384 work items; hardware-specific shader cost means this is not a universal TDR guarantee. Publish device loss/uncaptured failure/poll failure through live executor availability without reinitializing automatically. Source-pinned required-Vulkan GPU suite: 105 passed, zero failed, three existing ignores, including actual device.destroy retirement and session availability. Strict GPU/engine all-target Clippy passes. Broader engine and repository/WASM continuation remains separately recorded. No geometry baseline, CI rule or published format history change.
* fix(render): wrap complete backend diagnostics (#5714) (#5715)
Preserve full backend/job/decoded-image diagnostics in the existing shared panel without ellipsis. The painted-galley regression covers Floating/Compact and light/dark, complete text, wrapping and fixed panel width. Exact-tree Render UI 33/0, strict UI Clippy, WASM and repository gates pass. Real safe-mode window captures, camera restore, CPU export and normal exit are recorded; no backend behavior or unrelated UI policy changes.
* Sync pinned main into render (#5869)
* Viewport: show reference geometry through bodies (Preferences, off by default) (#2697)
A new Graphics preference, "Show reference geometry through bodies", draws
inactive sketch contours, origin and construction planes and their axes over
shaded bodies. It is off by default, so the viewport looks as before: reference
geometry is depth-tested and a section leaves it whole (#1229); Create Sketch
and dialogs picking a plane still draw the planes through the model.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* ui-egui tests: offscreen GPU tests start their devices one at a time (#3989) (#3990)
* claim: #3989
* ui-egui tests: offscreen GPU tests start their devices one at a time
Every offscreen test started a wgpu instance of its own, and starting one
makes the Vulkan loader scan its drivers. With several test threads two
tests doing that at once crashed the test process with SIGSEGV inside the
loader (loader_icd_scan). offscreen_device now holds a process-wide lock
while it starts the instance, adapter and device; each test keeps its own
device and the rendering still runs side by side.
* DXF: write and read a Bézier at its own degree (#5269) (#5281)
* claim: #5269
* DXF: write and read a Bézier at its own degree (#5269)
The DXF writer and reader held a spline's degree to 7, the value MAX_BEZIER_DEGREE had before
#1642 raised it to 9. A G4 blend (one Bézier of degree 9) was written as a degree-7 SPLINE over
its first 8 control points, so a flat pattern's cut-out stopped 4.444 mm short in the file.
- seg_shape takes its limit from MAX_BEZIER_DEGREE: a Bézier is written whole, at its own degree.
- read_dxf reads SPLINE degrees 1 to MAX_BEZIER_DEGREE, the degrees a sketch spline takes.
- Tests: every Bézier degree 4 to 9, a degree-9 round trip, the reader's degrees, hostile SPLINE
degrees and too few control points, the G4 cut-out exported and read back with its areas, and a
degree-9 SPLINE with 2 to 10 points inserted into a sketch.
Fixes#5269
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Renaming a plane keeps its projections, canvases and hidden state (#4353) (#4354)
* claim: #4353
* Renaming a plane keeps its projections, canvases and hidden state (#4353)
timeline.rename only renamed the feature in other features' references. A sketch's
projection of a construction plane, a canvas placed on it and the hidden list keep it
by name too, so they lost it: the sketch warned that the plane no longer exists, the
canvas left the view and a hidden plane showed again. Document::rename_feature_keys
renames it in sketch links and canvases, as rename_body_keys does for bodies; the
selection follows in the command, and the hidden list in Session::refresh_from, so a
hidden plane stays hidden through undo and redo too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Browser: add Rename and Delete to feature menus (#3839, #4343) (#4342)
* claim: #3839
* fix(ui): add Rename and Delete to browser feature menus
* Browser: a second click on a Document Settings entry closes its window (#4150) (#4152)
* claim: #4150
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Browser: a second click on a Document Settings entry closes its window (#4150)
Each entry under Document Settings only opened its window. A second click on
Units closed the menu on the press (the row is outside it) and opened it again
on the release; Precision left Preferences open; Parameters and Design
Properties came back as a fresh copy. Now a click on an entry whose window is
open closes it the way the window's own x does:
- Units: context_menu::toggle_for with the row as the opener, the fix#1522
made for the workspace button's menu. A right-click still opens it at the
pointer.
- Precision: prefs::close(app, true), the Cancel/x path that puts back what
changed since the window opened (now pub(crate)).
- Parameters and Design Properties: the window is dropped, as its x does. A
Parameters value typed but not applied yet is dropped with it, as today's
second click already drops it.
Fixes#4150
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Preserve expression-positioned sketch fillet corners (#4097) (#4106)
* claim: #4097
* Sketch fillet keeps a corner placed by an expression (#4097)
Filleting a corner whose position comes from parameter expressions deleted
the point and its expressions, so the sketch gained two degrees of freedom
and later parameter changes distorted the outline. keep_corner now also
keeps fixed or expression-positioned points as virtual sharp corners.
* Fix feature editor picks after upstream sketch resizing (#4083) (#4084)
* claim: #4083
* Feature editors refill named edges and faces where they are now (#4083)
After an upstream sketch change, Fillet, Chamfer, Shell and Draft evaluated
correctly, but their editors refilled the originally clicked points and
could then fail with "no edge at ...". Resolve the stored names through the
document's naming helpers first, keep the resolved index, and map the pick
into the displayed body's frame; the point lookup stays as the fallback.
* ui-egui tests: clear the texture deltas of the frames they run (#4039)
Nine tests dropped an egui frame's output with its texture delta still
unapplied (the expression completion editor's frames, the slot preview's
three, the browser's feature rows). epaint asserts against that in debug
builds, so they failed under the dev profile and passed only where debug
assertions are off. They clear the delta as the other tests do; one more
hover test that dropped its output the same way does too.
* Frames: profile library, members along frame sketch lines, and the Frame Member dialog (#4003) (#4004)
* claim: #4003
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: the profile library (#4003)
The first slice of frames and weldments: the cross-sections structural members
will be made of, each SolidCraft's own parametric definition built in code
(no vendor or standard size tables are shipped).
- solidcraft_doc::frame: bar, round bar, rectangular and round tube, angle,
channel, I-beam and a generic T-slot extrusion as exact line/arc regions,
with defaults worked out from the profile's size, refusals that name the
dimension that does not fit, anchors and placement, and exact section
properties (area, centroid, Ixx/Iyy/Ixy by Green's theorem).
- frame.profile.list, frame.profile.properties (with mass per metre for a
density) and frame.profile.insert (draws a profile into a sketch, locked by
default, at a point, anchor and angle).
- Tests against hand-computed values (rectangles, circles, composite
rectangles by the parallel-axis theorem, fillet spandrels), hostile input,
extrusion volume, undo; catalog entries in every locale; docs/frames.md;
ROADMAP M15.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: drop the built-in T-slot profile (#4003)
Aluminium extrusions and other vendor sections come from the user's own
sketch (drawn or inserted from the vendor's DXF), so the library keeps the
plain steel shapes only.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: members along frame sketch lines (#4003)
frame.member lays one body per straight line of a frame sketch (2D lines or
3D sketch lines), its cross-section square to the line from end to end:
- profile: a built-in profile, or profile_sketch, a sketch whose material
(nested loops alternating, so two circles make a tube) is the section in
its own x and y around its origin; profiles picks override it.
- anchor (origin, centre, centroid or a bounding-box corner or edge middle),
angle (an angle expression, so a parameter), mirror.
- Orientation: the profile's y as near world up as the line allows, x
square to both; a vertical member's x is world X (frame::member_axes).
- Members follow sketch and parameter edits, undo as one step, save and
reopen (format: pending/frame-members.rs), and are named after the profile
(Rect Tube 40x40x3, (2)…), numbering on past names already taken.
Tests against hand-computed volumes and bounding boxes (rectangle, 3D
tower lines, parameter edit, sketch tube profile and picks, anchor/angle/
mirror), refusals, undo/redo and reopen; walk fixtures and goldens; docs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: fixes from the adversarial review (#4003)
- Default member lines are what was drawn: projected and other linked curves
and 3D wires (a face sketch's edges) and lines of no length are left out.
- A profile sketch's material starts from its drawn profiles, as
solid.extrude's "all" does, so a section sketched on a face works; the
even-odd test uses outlines and boxes worked out once, capped at 256.
- Member names are pinned to their lines (`names`, by line id; `body_name`
is the base): deleting or adding a line, or deleting another feature, no
longer renames other members, and frame.member stores every name it gives.
body.rename renames a member on its line (also for later member features);
a copy whose body_names were set still names its members by position.
- Built-in profile dimensions are expressions (ProfileSpec): they follow
parameters, bare numbers take the design's units, and they are feature
inputs. frame.profile.properties/insert read them the same way.
- Default root/toe radii shrink to the room a thick section leaves; radii
given are checked by name; a radius must be 0 or at least 0.01 mm; error
messages report the wall or flange value given, not twice it.
- A line within 1e-6 rad of vertical orients as vertical, so rounding never
turns a post a quarter turn.
- A 2D frame sketch in a differently placed component is mapped into the
member's frame, and member features recompute when placements change.
- Caps before lookups for a design's `lines`; a line listed twice is
refused; a 3D wire that runs back on itself is not a line.
Regression tests for each, from the review's scenarios.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: the Frame Member dialog on the ribbon (#4003)
- Frame Member (MODEL > Create) opens a dialog: the frame sketch's lines
picked in the view (or All Lines), a built-in profile and its dimensions or
a section picked in another sketch, anchor, angle, mirror and name, with a
live preview. Edit Feature reopens it filled in.
- Lines picked in the view keep their sketch (a SKETCH_LINES input picks
ExternalSketchCurve): line ids are each sketch's own, so two 3D sketches
both have a `w2`, and the members went on the newer sketch's lines.
- timeline.redefine takes frame.member. Members keep their names through an
edit unless it gives them a new body_name; frame.member takes `names`.
- A member whose section is a whole profile sketch keeps it when edited
(the dialog shows the sketch instead of picking every region, which would
fill a tube's hole).
- An original frame_member icon, open for review, on the ribbon, timeline
and browser.
- The angle profile is listed as L-Angle, and web and flange as Web/Flange
thickness, so they translate apart from the existing Angle, Web and Flange;
the dialog's labels are in en, ja and zh. en-GB no longer repeats the frame
command labels it spells the same.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Rectangular and circular pattern: copies that miss the body are reported (#3042) (#3919)
* claim: #3042
* Warn when cut pattern copies remove no material
A rectangular or circular pattern whose copies miss the body succeeded silently and those copies did nothing. The pattern now reports how many of its cut copies removed no material, and keeps the copies that land.
* Delete in a rolled-back history warns about the features after the marker (#3471) (#3698)
* claim: #3471
* Delete in a rolled-back history warns about the features after the marker (#3471)
With the timeline rolled back, the checks a delete runs first (the
confirmation's "these features will fail", timeline.dependents and
selection.delete's would_fail) compared models evaluated only up to the
marker. Features after it, broken by the delete, were not listed, so their
error marks turned up only once the timeline moved to the end.
newly_broken_in_history evaluates both sides over the whole history when
either has a marker (on copies; the model and document are unchanged) and is
used by the delete's report and by timeline.dependents.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* macOS: Fusion navigation preset does not support Shift + two-finger orbit (#3446)
* Macos: keep touchpad/key modifiers true to fusion
* fix: remove obsolete orbit snap reset from trackpad navigation
Upstream removed ViewportState::orbit_snap. The new trackpad orbit path retained one reference, causing E0609 in the PR merge build.
Validation: all seven trackpad tests pass; cargo check -p solidcraft-ui-egui --all-targets passes on the exact failing CI merge 66a2791b with this fix applied; rustfmt and diff checks pass.
Co-Authored-By: GitHub Copilot <noreply@github.com>
* fix: retain Fusion trackpad help in refactored preferences
Preserve upstream's new preferences pages and apply the translated Fusion trackpad tooltip in prefs/navigation.rs.
Validation: cargo build -p solidcraft passed. Tests and full CI were not run, as requested.
Co-Authored-By: GitHub Copilot <noreply@github.com>
* style: format Fusion trackpad preferences tooltip
Apply rustfmt's required wrapping for the relocated tooltip. No behavior changes.
Validation: cargo fmt --all -- --check and git diff --check passed.
Co-Authored-By: GitHub Copilot <noreply@github.com>
---------
Co-authored-by: GitHub Copilot <noreply@github.com>
* 2-point circle takes a typed diameter (#3278) (#3335)
* claim: #3278
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 2-point circle takes a typed diameter (#3278)
After its first point the 2-point circle tool showed no value box, so the
circle could only be sized with the cursor. It now shows a Diameter box,
like the centre circle: the typed value puts the second point that far
along the cursor's direction, and Enter draws the circle.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Reorder design tabs by dragging them (#3187) (#3188)
* claim: #3187
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Reorder design tabs by dragging them (#3187)
A tab dragged along the strip follows the pointer while the others make
room, and lands where it is released; Escape, or releasing down the
window, leaves the order as it was. With the custom title bar a drag that
starts mostly up or down (or on the only tab) still moves the window.
The control channel's ui.documents gains a move action (from, to).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Renaming a body keeps the evaluated timeline, off the window thread (#2467) (#2468)
* Renaming a body that later features change evaluates nothing (#2467)
Model::rename_body now keeps the results when later features change the
body (a pattern, a chamfer, a join). It still evaluates when a body of
that name comes and goes or isn't the same body throughout, when a later
body's name is worked out from it (X (2)), or when a message quotes it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Rename box: rename off the window thread, and locked bodies follow a rename (#2467)
The rename box commits body.rename through run_heavy, so a rename that
still evaluates later features (one names the body in its targets)
computes beside the window instead of freezing it. Locked bodies now
follow a body.rename from anywhere (the rename box, the control channel,
scripts), not only the rename box.
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Toolbar: a constraint dragged out of the small box becomes a full button (#5254) (#5456)
* claim: #5254
* Toolbar: a constraint dragged out of the small box becomes a full button (#5254)
The Constrain panel's small box of constraint icons was a fixed list, so
dragging one onto the full-size buttons saved a new order but it was still
drawn in the box. The toolbar settings now remember, per tab, which box
commands were moved out; those are drawn as full buttons, the box closes
up, and dropping one back on a box icon returns it. Reset Toolbar restores
the default box, and settings saved before this load unchanged.
* Sketch offset keeps its source lines free to turn (#4464) (#4466)
* claim: #4464
* Sketch offset keeps its source lines free to turn (#4464)
An offset adds a Parallel between each source line and its offset line.
Since offsets became associative, the offset's points are held during a
solve, so each Parallel pinned its source line to the offset line's
current direction: a closed chain lost its turning freedom (8 degrees of
freedom became 6), and Horizontal or an angle edit on a source line was
refused as conflicting.
The solver and the constructive reading now leave out a Parallel that
ties an offset line to the source it follows while the association lives;
the association already keeps them parallel. After Break the stored
Parallel drives again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Flat Pattern view draws perforations (#4440) (#4441)
* claim: #4440
* Flat Pattern view: draw perforation holes (#4440)
The flat view drew the outline, cut-outs, pockets and bend lines but not
perforation fields, though the flat-pattern data and the DXF export list
their holes. The view now reads the hole centres from the sheet through
SheetBody::perforation_holes, caches them per model state and body, and
draws them as circles in the cut-out style. Holes outside the viewport or
smaller than one pixel are skipped, and the cache is capped at one million
holes.
* "Don't Save" looks like a button in the Save changes prompt (#4380) (#4400)
* claim: #4380
* "Don't Save" looks like a button in the Save changes prompt (#4380)
The prompt's third action used the kit's low-emphasis text style, so it
didn't read as clickable and people thought Cancel was the only way out.
Add an opt-in secondary_left style to the confirm footer (a framed
secondary button from the theme tokens) and use it for Don't Save.
* Save prompt test: clear each frame's texture delta once it is applied
egui 0.36 panics when a TexturesDelta with unapplied deltas is dropped;
the test kept only the last frame's output and dropped the earlier ones.
* 3D mouse test: clear the frame's texture changes, so it passes in debug builds (#4382)
each_frame_hands_the_view_its_motion_through_the_drivers_mapping dropped
the output of a frame it ran (`let _ = ctx.run_ui(...)`). epaint panics
in debug builds when a TexturesDelta is dropped with deltas left ("Dropped
TexturesDelta with 1 unapplied deltas"), so the test failed under a plain
`cargo test` on every platform while the ci profile passed. It now clears
them, as the other UI tests do since #60.
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Spanish (es-ES): catch up with en.json (#4292)
Co-authored-by: carlymx <24634598+carlymx@users.noreply.github.com>
* Esc ends the navigation bar's Pan, Orbit and Zoom modes (#4226) (#4233)
* claim: #4226
* Esc ends the navigation bar's Pan, Orbit and Zoom modes (#4226)
The nav bar toggled a mode on, but nothing cleared it on Esc, so a left drag
kept panning, orbiting or zooming. Esc now ends an active nav mode first;
the next Esc reaches the tool or dialog as before.
* Keep long Select Other labels inside the popup (#4209) (#4225)
* claim: #4209
* fix(ui): bound Select Other labels and expose full names (#4209)
* Deleting an inserted file removes its components, placements and joints (#4100) (#4113)
* claim: #4100
* Deleting an inserted file removes its components, placements and joints (#4100)
A STEP or mesh file inserted into a design makes a component tree. Deleting
its feature left that tree behind: the components, their occurrences, the
rigid joints that hold them and the bodies' component assignments. A later
body that took an imported body's name was then placed in the stale component.
Document::remove_component_tree removes a component subtree and everything
that refers to it: the features made inside it, its occurrences, the joints
on them with their motion links, joint origins, contact set members, motion
study keys and exploded moves, body component and offset entries, standard
part and link records, and configuration columns. Deleting an inserted
file's feature and component.delete both use it; component.delete now also
removes joints left pointing at a deleted occurrence. The body transfer
refusal moves with it, unchanged.
* Only an inserted file's feature takes its component tree when deleted
Mark the feature a STEP, IGES or mesh insert makes (`inserted`), and remove
the component tree on delete only for those and for a linked design's
feature. Deleting a copied component's independent base deletes just that
feature, as before. Older designs get the mark as they open (pending format
change), except copied components' independent bases.
* Rigid joint saved with the wrong occurrence when two occurrences share a component (#3857) (#3939)
* claim: #3857
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Joint dialog: a body of a further instance picks that occurrence, not the component's first (#3857)
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* Cancel Sketch: leave a sketch without keeping its changes (#3838) (#3938)
* claim: #3838
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
* Cancel Sketch: leave a sketch without keeping its changes (#3838)
sketch.cancel (Cancel Sketch, ribbon Exit group and the sketch's
right-click menu) puts the design back as it was when the open sketch
was created or opened for editing: a new sketch is removed with its
dimension parameters, an edited one returns to how it was. It is one
undo step, so Undo goes back into the sketch.
The session remembers the design and the undo history at the point a
sketch is entered. Cancel uses that only while the history still runs
through it (the same steps before, every step since taken in this
sketch); otherwise it refuses and points to Undo, so a stale entry can
never throw away later work.
The icon is the approved close glyph in the sketch accent (alias
sketch_cancel); ja and zh carry the new label.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* 3MF import: name a Bambu Studio project's bodies after its objects (#3858) (#3859)
* claim: #3858 (3MF import: names from a Bambu Studio project)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3MF import: name a Bambu Studio project's bodies after its objects (#3858)
Bambu Studio leaves the name off the root model's objects and keeps it in
Metadata/model_settings.config (<config><object id="N"><metadata key="name">),
so every object of a project came in as Mesh1, Mesh2, ... A build item whose
root object has no name of its own now takes the object's name from those
settings. A name in the model still comes first, and part names, other
settings and objects of other model parts are left alone.
The settings are read only when a body needs a name, within what the model
parts left of the 1 GB budget. Settings that can't be read (malformed, nested
too deeply, too large) never fail the import: the bodies keep the generic
names, with a warning saying why.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Viewport: right-click keeps a multi-selection of faces and edges (#3733) (#3788)
* claim: #3733
* Viewport: right-click keeps a multi-selection of faces and edges (#3733)
---------
Co-authored-by: AhmetV <344941595+ahmetv10@users.noreply.github.com>
* Parameter import and image pickers offer their own file types (#3216) (#3314)
* claim: #3216
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Parameter import and canvas image pickers offer their file types (#3216)
Both opened the picker with no file type of their own, so it listed only
designs, CAD files and meshes (and All files). macOS lets you pick only
the types a filter names: CSV parameters, and PNG or JPEG images for a
canvas, were greyed out. They now ask for csv/json and png/jpg/jpeg,
which the picker lists first (#1801).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Return to the originating tab after finishing a sketch (#3250) (#3264)
* claim: #3250
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Return to the originating toolbar tab after finishing a sketch (#3250)
A sketch started from SHEET METAL or PLASTIC always ended on the SOLID
tab. Remember the tab that was active when the sketch began
(UiState::sketch_return_tab, not saved) and go back to it when the
sketch ends. Falls back to SOLID when that is not one of the normal
toolbar tabs. A sketch started from SOLID behaves as before.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Test: no one-element loop in the return-tab test (clippy single_element_loop)
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* STEP import: let a cancelled import release the import slot (#3171) (#3189)
* claim: #3171
* Release cancelled file import slots (#3171)
* Dragging a part along its joint no longer highlights what's under the pointer (#5137) (#5140)
* claim: #5137
* Dragging a part along its joint no longer highlights what's under the pointer (#5137)
The viewport's hover pick was skipped during a box selection and camera
moves, but not while a jointed part was dragged, so faces and edges under
the pointer lit up on every frame of the drag. The pick now waits until
the joint drag ends.
* Browser: a body picked in the viewport is opened to and scrolled to (#3005) (#3081)
* claim: #3005
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Browser: a body picked in the viewport is shown in the Browser (#3005)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Merge main; viewport::select takes a toggle flag now
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Line tool snaps a diameter through the circle centre (#4326) (#4327)
A line drawn across a circle only snapped its end onto the curve, so a
diameter could miss the centre at any angle. Near the centre the line now
locks onto the radial line: on the rim it stays on the circle and the centre
is kept on the line, and on the centre point the end lands there.
* Flatpak: find the host's fonts for CJK text and the font list (#5124) (#5127)
* claim: #5124
* Flatpak: find the host's fonts for CJK text and the font list (#5124)
Inside the Flatpak sandbox the host's fonts are mounted at /run/host/fonts
and /run/host/user-fonts, which no font search looked at. Japanese and
Chinese text fell back to boxes, and the font list missed the host's
families.
- Search both folders for the CJK and Chinese fallbacks and for the
installed-font list.
- The CJK fallback reads SOLIDCRAFT_FONT_DIRS (a path list, as the font
list already does), with CRAFT_FONTS_DIR kept as a fallback.
* Crash reports: keep component.activate in the journal (#5079) (#5121)
* claim: #5079
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Crash journal keeps component.activate (#5079)
Activating a component changes only the session's active component and
occurrence path, not the design, so the journal dropped the command and a
crash report's replay built later parts under the previously active
component. The journal now also keeps a command that changes either.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Control channel and MCP on by default, with a setting to turn it off (#5561)
The app opens its localhost control channel on port 27878 (or the next free
one) unless Preferences > Assistant turns it off; --control and
SOLIDCRAFT_CONTROL_PORT still win. A current.port discovery file lets
solidcraft-cli mcp find the running app with no arguments.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* Tests: moving an occurrence recomputes only the cuts and joins it touches (#4965) (#4982)
* claim: #4965
* Joins and cuts follow only the placements they depend on (#4965)
Tests that turning a joint between parts cuts no hole again, and that a
tool cutting another part cuts again when that part moves relative to it,
while an unrelated part moving keeps the result.
Booleans and Combine now hash only the placements of the bodies they take
as inputs, relative to the frame that evaluates them (#3397), so these
pin that behaviour for joins and cuts with named targets across parts.
* fix(ui): preview shortcut chords and wait for complete modifier input (#1810)
Keep physical modifier events out of the ordinary-key slot, preview held keys immediately, and confirm after release and a one-second settling interval. Preserve legacy control-channel modifier behavior while allowing explicit physical Control/Command input. Add regression coverage, a UI scenario, localized guidance and documentation.
Fixes#1809.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* Move: scope box selection to the active component (#4459) (#4527)
* claim: #4459
* fix: scope Move box selection to active component (#4459)
* Delete removes a selected construction axis (#4692) (#4695)
* claim: #4692
* Delete removes a selected construction axis (#4692)
Delete on a construction axis skipped it with "the origin axes can't be
deleted". It now deletes the construction axis feature, like construction
planes; the origin axes X, Y and Z are still refused.
* Agent chat: end the agent's process group where `kill` is busybox' (#4587) (#4588)
* claim: #4587
* Agent chat: end the agent's process group with the system call (#4587)
The group was signalled with a `kill` program: `kill -s TERM -- -<pgid>`.
That works with procps' `kill` and with BSD's, but busybox' `kill` takes no
`--`: it prints its usage, exits 1 and signals nothing. The agent's helpers
(the CLI an adapter wraps, the MCP servers it starts) then stay alive after
the agent is ended.
The group is signalled with the system call instead, so no `kill` program and
no shell are needed and the behaviour is the same everywhere. `nix` (MIT) is
already in the tree through zip, and only on Unix: the browser build does not
take it.
The two tests that cover this now pass where they failed before:
`agent_chat::process::tests::dropping_the_process_ends_its_group` and
`agent_chat::process::tests::stopping_the_agent_leaves_no_helpers_running`.
* Extrude: picking a face, vertex or parallel plane off the profile switches the extent to To Object (#2855)
* Extrude: picking a face, vertex or parallel plane off the profile turns the extent to To Object
With profiles picked, a face off their plane (or curved), a vertex off it, or a
construction plane parallel to it is taken as the To Object target instead of
another profile: the extent switches without the Extent Type choice, and the
distance shows how far the target is. Coplanar faces still join the profiles.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Extrude: automatic To Object is a Preferences setting, off by default
Preferences > Design > "Extrude to a picked target" (`extrude_auto_to_object`, default
false) turns the automatic To Object on. Off, picks in the Extrude dialog only add
profiles and the distance is not synced to a target, as before this change.
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Regenerate THIRD-PARTY-LICENSES.txt and docs/features.md [skip ci]
* FreeCAD/Rhino import: decode caps with room for detailed parts (#1783) (#3209)
* claim: #1783
* FreeCAD/Rhino import: decode caps with room for detailed parts (#1783)
cadmpeg decoded FreeCAD and Rhino files under its service profile, whose
million collection items and 2.56e8 work units a detailed part (a sprocket,
a threaded rod, a part with a few hundred faces) runs past: the import was
refused with "resource limit on CollectionItems: BudgetExceeded". Allow
8 million items and 2e9 work units, half of cadmpeg's desktop profile; the
input-size cap stays SolidCraft's STEP limit and every other cap stays at
the service profile's.
decoding_is_capped checks the two raised caps (and that they stay below the
desktop profile's) and that every other cap is still the service one.
Fixes#1783
* A spherical shell revolves into a solid with a void, and cuts and joins with hollow tools work (#451) (#4182)
* Assembly: ground the first part; drag free joint chains (#2804) (#2805)
* assembly(feat): ground first parts and drag free joint chains
Ground the first top-level part when it is created and let face dragging reposition free occurrences or any member of a floating joint chain in one undo step. Preserve exact repeated-placement frames, joint values, modifier selection and Escape cancellation. Keep the accumulated fixture, viewport layout and documentation integration fixes from the existing PR.
Signed-off-by: Karl Von Richter<karl@embeint.com>
* assembly(fix): update movable fixtures for first-part grounding
Accept the documented grounded override in component creation, explicitly free movement fixtures, and keep linked clipboard pastes movable even in an empty document. Cover first and reused linked pastes, default insertion grounding, and single-step undo in the existing link tests.
Signed-off-by: Karl Von Richter<karl@embeint.com>
* assembly(test): keep the motion-cost crank free
Unground the first component in the document-level joint-face measurement fixture so its motion samples can rotate after first-part grounding. Preserve the rotation and shared-measurement assertions that caught the Linux CI failure.
Signed-off-by: Karl Von Richter<karl@embeint.com>
* assembly(test): free the frame placement fixture
Create the frame-member placement fixture ungrounded before moving it above its root sketch. Keep its world-space member assertions while respecting first-part grounding in the frame tests added on main.
Signed-off-by: Karl Von Richter<karl@embeint.com>
---------
Signed-off-by: Karl Von Richter<karl@embeint.com>
* UI: integrate theme radii, borders, palettes and viewport refresh (#5045) (#5046)
* claim: #4922
* UI: let themes own corner radii
* UI: use four-point panel and window radii in SolidCraft 95
* UI: align the 95 radius assertion with four-point panels
* claim: #4953
* UI: let themes own structural and control borders
* UI: single Blueprint outlines and dotted secondary rules
* UI: separate container outlines and add theme-owned 95 bevels
* claim: #5007
* UI: apply theme palettes to icons, workspace tabs and Gruvbox settings
* claim: #5045
* Viewport: invalidate cached geometry colours when the theme palette changes
* UI: preserve legacy theme logo colours with the borderless mark
* Section Analysis: OK with no plane picked turns the section off (#2410) (#2411)
While a section is on, Section Analysis opened with no plane picked now
says "No plane picked: OK turns the section off." and OK runs
inspect.section {"clear": true}. With no section on, OK with no plane
still asks for a plane.
Fixes#2410.
* Regenerate THIRD-PARTY-LICENSES.txt and docs/features.md [skip ci]
* Close Change Parameters with Escape (#3372) (#3399)
* claim: #3372
* Fix Escape cancellation in Change Parameters
Consume Escape before parameter editors can commit pending text on focus loss, while preserving autocomplete and shortcut capture priority. Add 11 UI regressions covering draft cancellation, preview cleanup, document and history preservation, and reopening saved values.
* Preserve construction references when axes and points are renamed (#4419) (#4449)
* claim: #4419
* fix: preserve renamed construction references (#4419)
* Browser: route body picks into open command dialogs (#2560) (#4371)
* claim: #2560
* fix(ui): route Browser body picks into active dialogs
* Fix midpoint picks in sketch dimensions (#4872)
Retain MID snaps across both picks and materialize reusable midpoint
constraints for point-to-line and point-to-point dimensions. Let midpoint
pairs choose horizontal or vertical spans with pointer placement before
entering a value.
Cover projected model edges, overlapping geometry, helper reuse,
undo/cancel/reopen, and existing angular and line-circle workflows.
* Sketch slots: provide typed dimension inputs for every variant (#2423) (#4318)
* claim: #2423
* fix(sketch): add typed dimensions to every slot tool (#2423)
* Version 0.21.0 on main (released from rc.2's code) (#5774)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* docs(mcp): Windows setup for Claude Desktop (#3794)
* docs(mcp): Windows setup for Claude Desktop (paths, two servers, Store config file)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* docs(mcp): say the Store build can keep the config elsewhere, not that the other path fails
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Regenerate THIRD-PARTY-LICENSES.txt and docs/features.md [skip ci]
* MCP: wait out long requests, honest batch outcomes, batch tool aliases (#2901)
Rebased onto main as one commit (was 120d26f7, df488f9a, db9d1073 plus
merges of main).
- A request the app reports as still running (#1618) is waited for on the
same connection through `control.pending`; past the wait the error says it
was not cancelled and will still apply (or only write its file), later calls
are not sent behind it, and the new read-only `pending` tool returns its
result once it lands.
- A failed batch says what really happened: `rolled_back` only when the
design is back where the batch started (as #3288 / #3242 on main: the app's
`rolled_back` is believed; for an older app, every completed step undone;
only a rollback asked for is reported as one), plus `applied`, `not_run` and
`failed.still_running`.
- A batch step may name a tool (set_parameter, list_commands,
describe_command, execute, measure, inspect_design, undo, redo, export, save,
open, new_design, crash_reports); screenshot, body_topology, batch and
pending are refused before anything is sent.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse: driver choice, guided calibration, button mapping and a picture of the device (#1463) (#3825)
* Add a small test boat to steer around with the 3D mouse
A little hull in its own camera, drawn by the UI itself and not the CAD
viewport, so a 3D mouse can be tried out without opening a design. It
brings the model (generated by assets/models/make_test_boat.py) and the
tests for how the camera moves it.
Nothing uses it yet; Preferences picks it up with the rest of the 3D mouse
work.
* UI kit: keep a choice row from flipping with the scrollbar
A choice row picks segments or a dropdown by whether its labels fit. A page
scrollbar that comes and goes narrows the row by its own width, so the row
could flip back and forth between the two on a page that sits right at the
edge. It is now judged on the whole row and through a small helper,
fits_steadily, that only changes its answer once the miss is wider than the
scrollbar.
Also tests for that, for typed numbers in a slider's value box, and a
test hook to put up the issue-report input barrier by hand.
* 3D mouse: device model, guided calibration, buttons and a picture
The 3D mouse becomes a device we know something about instead of six raw
numbers.
The reader picks between spacenavd and the direct hidraw driver, falls back
from one to the other, and says so when it can't open the device. Guided
calibration asks whether the device is a spring cap or a trackball, adds a
touch step, and uses a settle filter so a ball let go of stops reading as
motion. Buttons get jobs (keys, held modifiers, commands), saved per
device, with defaults for the SpaceMouse Pro.
Preferences gets a picture of the device that moves as it does, next to the
test boat, and the same picture can sit in the 3D view at Left Viewport,
Right Viewport or Left Window, above the floating History card like the
legend and readouts.
These ride together because the crate's calibration types, the Preferences
page, the Linux reader and the button hook all use each other.
* 3D mouse: offer to calibrate an uncalibrated device
When a 3D mouse shows up that has not been calibrated, a card in the 3D
view offers to calibrate it, to swap drivers when the other one would work
better, to close, or to never ask again. It is made once per device, goes
away after 30 seconds in view, and gives way to a status or computing
toast instead of stacking on it.
The card is the shared toast, which now takes buttons: they group by
purpose, wrap inside the card at narrow widths, and do nothing while an
issue report holds CAD input. The picture of the device in the 3D view
keeps clear of the card.
* 3D mouse: translations and attribution
The 3D mouse strings (ui.3d_mouse_*, the calibration steps and the offer
card) in every shipped language, the Japanese catalog note, and the credit
for the test boat model in ATTRIBUTION.md.
This is last because the code before it only looks keys up: it builds and
runs without these, falling back to English until they land.
* 3D mouse: draw its preferences with the settings kit like the rest of Navigation
The 3D mouse section was a folded header with its own layout. It now
uses the same kit as the sections above it: a settings_section card,
settings_form rows with the shared label column, dividers between the
device, axes, picture and buttons groups, Text::Hint for the helper
lines, and settings_control_width for the dropdowns.
The section is open like every other one (the header still folds it),
so the device badge on the folded header is gone; the calibration row
already names the device. Every setting, string and key is unchanged.
Tests that unfolded the section no longer click it, and the hint lines
are looked up as hints instead of notes.
* 3D mouse: show each axis's live tilt in Preferences, like main's axis rows
Each view axis row is main's axis row again: a meter that follows the tilt of the motion driving it, with the dead zone marked, next to Invert, Speed and the "driven by" menu. The meter reads the calibrated travel when the device has a calibration and the raw reading over full scale when it does not, and asks for a frame whenever the reading changes. The helpers are back as they are on main, and the calibration state is a badge by the enable toggle.
* 3D mouse: line its preferences up with the rest of Navigation
- The two checkboxes sit side by side in the same two-column grid as Wheel and trackpad.
- Row labels stop repeating "3D mouse" and use sentence case: Calibration, Driver, Restore defaults, Reset view, Left viewport, Right viewport, Left window.
- Hints line up under the control column instead of the left edge.
- The six view axes are named Pan X, Pan Y, Zoom, Orbit X, Orbit Y and Roll so they fit the name column.
- Buttons are a divider and rows in a form like the other groups, without the bold sub-heading.
- The driver, dead zone, axes, picture and buttons sit behind a collapsed Advanced section, since most people never touch them.
* 3D mouse: tidy the Advanced test for clippy
* Timeline: suppressing a feature keeps later features' body names (#4302) (#4311)
* claim: #4302
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Timeline: suppressing a feature keeps later features' body names (#4302)
Bodies numbered by the counter (`BodyN`) take the next free number on every
recompute, so suppressing a feature that made one moved every later counted
body down one (`Body2` became `Body1`) and broke what named it, such as a
Move of `Body2`. timeline.suppress now stores the current names of later
features' counted bodies in their `body_names` first, as body.rename already
does for a renamed counted body (#1109). Engine only: no evaluation or
design-format change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Timeline: a body made while a feature is suppressed keeps its name (#4302)
Suppressing a feature freed its counted name (`Body1`), so a body made
meanwhile took it and became `Body3` once the feature was back. Suppress
now also stores the suppressed feature's own counted names, and the
evaluator adds the names stored on earlier suppressed features to the
state's `reserved_body_names` (the names a rename freed, #4957), which the
`BodyN` counter passes by. They are taken from the design before each
feature is computed and are part of the state digest, so a cached result
is only reused when they match.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* cadmpeg 0.6.3: FreeCAD and Rhino import in the browser too (#4482)
cadmpeg 0.6.3 is pure Rust (ruzstd instead of the zstd C library, zip 9), so
solidcraft-io depends on it on every target and the wasm build reads .FCStd
and .3dm files like the desktop app. The browser's file pickers offer them.
0.6.3 replaces DecodeOptions::container_only with a decode depth; ask for
Geometry, what its STEP writer reads. zip 8 leaves the tree, and with it the
three cargo-deny advisory exceptions kept for it.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Pattern dialogs: Quantity fields get up/down step arrows (#4260) (#4264)
A new kit control, `ui_kit::count_field` / `count_row`, is an expression field for a
whole-number count with stacked step arrows at its right edge. An arrow steps the
evaluated count by one within the field's limits: a plain number is replaced, an
expression keeps its parameters and gains or adjusts a trailing `+ n`. The text
changes as typing changes it, so the live preview follows.
Used by the Quantity rows of the sketch Rectangular (1-500, both directions together
at most 500) and Circular (2-500) patterns, and the solid Rectangular, Circular and
Path patterns (1-1000).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Regenerate THIRD-PARTY-LICENSES.txt and docs/features.md [skip ci]
* Join: say so when the joined piece doesn't touch the body (#4434) (#4436)
* claim: #4434
* Join: say so when the joined piece doesn't touch the body (#4434)
A Join (extrude, revolve, sweep, loft, pipe, coil, primitives, thicken,
boundary fill, surface network) whose material meets none of its targets
still joins it to the first target as a separate piece. The result stays
as it was; the feature now warns: "the joined piece doesn't touch `Body1`,
so `Body1` is now in 2 separate pieces (New Body makes it a body of its
own)". History marks it and the command reply carries it.
The check compares pieces (outer shells, not voids) of the bodies the
feature changed or absorbed before and after it, so it holds for the
batched multi-profile path too, stays quiet for a join that bridges
pieces or closes a cavity, and leaves Combine (which joins the bodies it
is given) alone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Hole: snap to face and round-edge centres, drag placed holes, section in the dialog (#2171) (#2912)
Placing a Single hole put it exactly where the click landed, with nothing to line it up by and no
way to move it but taking it out and picking again; the dialog was a list of size fields.
- Preferences › Design › Create Hole has two settings, both off by default, so a new installation
places holes as before:
- Snap to face and edge centres (`hole_snaps`): the planar face under the cursor offers its
middle and the centres of its round edges (a boss, a rounded corner, a hole already there) as
marks; a pick near one, or on or beside such an edge, takes that centre and the mark is named.
Sketch points on the face still come first; Alt or Ctrl held places it where clicked.
- Move holes by clicking or dragging (`hole_moves`): each placed hole has a handle that drags it
(snapping when snaps are on) and takes it out when clicked, and a pick elsewhere on its face
moves it there. Edit Feature shows the handle too, so a hole can be moved later (#1047).
- The Hole dialog draws a section of the hole as its options make it, with each size in a field
beside the dimension it sets; a hovered or focused field lights its dimension. The layout is a
new kit element, `ui_kit::figure_rows` (a caller-painted figure beside captioned fields).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Hole: place a Single hole by distances from reference edges (#37) (#2931)
A Single hole can take up to two straight edges of its face as references,
each with the distance of the hole's centre from that edge's line, measured
in the face's plane. `solid.hole` takes them as `references: [{edge, distance}]`;
the edges are found again by their persistent names, else their points, so
the hole follows them when the part changes, and each distance is a
parameter. A bad reference (no edge there, a curved edge, one square to the
face, parallel edges, a negative distance, a centre off the face) is refused.
In the Hole dialog, Reference 1 and 2 follow the face, each with its Distance
row; the canvas shows each distance as a dimension from its edge to the hole,
with a box just past the edge to type it in. Picking an edge measures its
distance (and its box takes the keyboard), typing one moves the hole, and
moving the hole measures them again. Edit Feature picks the edges again with
their distances.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
---------
Signed-off-by: Karl Von Richter<karl@embeint.com>
Co-authored-by: Frank <58932224+plexi32@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: raykholo <raykholo@users.noreply.github.com>
Co-authored-by: CamWight <91156642+CamWight@users.noreply.github.com>
Co-authored-by: SS-EN <84559909+SS-EN@users.noreply.github.com>
Co-authored-by: Colin Faulkingham <colin.faulkingham@gmail.com>
Co-authored-by: Karl Von Richter <karl@von-richter.net>
Co-authored-by: HiroKenEz <59478073+HiroKenEz@users.noreply.github.com>
Co-authored-by: Peppe Bergqvist <p@bergqvi.st>
Co-authored-by: GitHub Copilot <noreply@github.com>
Co-authored-by: phanoob <lesur.Stephane@wanadoo.fr>
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Kai Evans <nemshilov@gmail.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Tyler Chiriboga <tylerchiriboga.apps@gmail.com>
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: CaRLymx <carlymx@gmail.com>
Co-authored-by: carlymx <24634598+carlymx@users.noreply.github.com>
Co-authored-by: Loohi <78334654+Loohi@users.noreply.github.com>
Co-authored-by: leValdemar19 <valdemarnord@hotmail.com>
Co-authored-by: AhmetV <mihendiz10@gmail.com>
Co-authored-by: AhmetV <344941595+ahmetv10@users.noreply.github.com>
Co-authored-by: rlartot <94693724+rlartot@users.noreply.github.com>
Co-authored-by: Christian Svensson <blue@cmd.nu>
Co-authored-by: Thodoris Tsagkogeorgas <thodoristsagko@gmail.com>
Co-authored-by: ahmetalpcelik <80778880+ahmetalpcelik@users.noreply.github.com>
Co-authored-by: ealitt <ealitt@gmail.com>
Co-authored-by: Homero Thompson <vdirienzo@gmail.com>
Co-authored-by: Diogo Soares Rodrigues <dsrodrigues20@gmail.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Artem Sushchev <Suschev.Artem+github@gmail.com>
Co-authored-by: Nathan Dane <nathandane.dev@gmail.com>
Co-authored-by: Kai Evans <hi@kaievans.co>
Co-authored-by: Victor Santos Oliveira <56417388+Vittuu@users.noreply.github.com>
Co-authored-by: cronos <crono2250@users.noreply.github.com>
Co-authored-by: rbousigues88 <rbousigues@gmail.com>
Co-authored-by: The Nomad <nomad@positron3d.com>
Co-authored-by: Jackson Oursland <jacob.oursland@gmail.com>
Co-authored-by: Jacques <59315319+Jacquesvdberg92@users.noreply.github.com>
* feat(render): reconcile definition-local visual mapping (#1214) (#1223)
Retain the original mapping implementation, all eight analytic/hostile-input tests and mapping contract byte-for-byte from #1223. Reconcile only module registration and the current roadmap on render. Actual current render library 166/0; strict all-target render Clippy, wasm32 and repository gates pass. No persistence, sampler, new capability, kernel/vendor change or external code is added. Historical mixed-base full-CI failures are not relabelled.
* feat(render): add bounded linear film accumulation (#1696) (#1697)
Extract the independent FilmAccumulator from owned donor 3b892eab5e. Keep the existing job cancellation and caller-owned admission/work contracts. PNG resolution, primary camera coverage and progressive consumer wiring remain separate. Add analytic statistics, atomic refusal, allocation/range boundaries and real 4096/1048576-count tests. No kernel, document schema, workflow or capability change.
* Sync main into render and preserve GPU CLI help (#5976)
* CLI: --help on each subcommand prints its usage and exits 0 (#4538) (#4545)
* claim: #4538
* CLI: --help on each subcommand prints its usage and exits 0 (#4538)
Every command now answers `--help` and `-h` with its own usage on stdout and
exit 0, the way the top level has since #1828. Before, each subcommand
answered differently: `fuzz --help` started a fuzz run with the default seeds
and ran until it was stopped, `recipe --help` took the flag for a file name,
and `snapshot`, `render`, `oracle`, `run`, `eval`, `script` and `crashes`
answered with an error on stderr and exit 1.
The help is read before the command runs, so `--help` never starts work: no
fuzz run, no MCP server, no file read.
`fuzz` also refuses an argument it doesn't take, as `snapshot` already did: a
flag it doesn't know and a stray word are errors, not arguments to ignore
while a hundred seeds run.
The usage lives in one place now: each command's lines are a constant, and the
top-level usage is composed from them, so the two cannot drift apart. The
top-level usage is unchanged, byte for byte.
# Conflicts:
# apps/solidcraft-cli/src/main.rs
* Renaming a parameter follows it into sheet-metal and plastic rules and canvases (#4420) (#4423)
* claim: #4420
* Renaming a parameter follows it into sheet-metal and plastic rules and canvases (#4420)
Document::rename_param rewrote other parameters, feature inputs, dimensions,
configurations and joints, but not the expressions of sheet metal and plastic rules or
the plane a canvas lies on, so after a rename the sheet metal and plastic features
using them failed with "unknown parameter" and the canvas left the view. It renames
them there now, and param_users lists them (a rule's field, a canvas's input), so a
parameter they use can't be deleted.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Parameters: rules' features in drives and the graph; shared rule field lists (#4420)
A sheet metal or plastic rule that uses a parameter makes the sheet metal (or plastic)
features rebuild when it changes, so parameters.users lists them in drives and
parameters.graph links them. SheetRule::fields/fields_mut and PlasticRule::fields_mut
replace the hand-written copies of the rules' fields in params.rs, eval.rs and
sheet.manage_rules. Suggested by @CamWight.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fmt (#4420)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Mirror and pattern: a pattern or mirror feature as the source (#395) (#3998)
* claim: #395
* Mirror and pattern: a pattern or mirror feature as the source (#395)
A pattern or mirror of features refused another pattern or mirror as a
source ("Pat cannot be patterned or mirrored yet"). Such a source now
stands for what it copied: its own sources are replayed at each of its
transforms moved by each of ours, to any depth up to eight. Copies of
a feature picked directly keep the instance numbers they had, so
existing designs keep their face names; copies made through a nested
pattern or mirror are numbered after them, so one feature reached two
ways never names two copies alike. A pattern or mirror of bodies as
the source is still refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Expression completion: keep longer names visible when the typed word is a unit (#3663) (#3664)
* claim: #3663
* Expression completion: keep longer names listed when the typed word is a unit (#3663)
Typing `rad` into an expression box showed nothing because `rad` is the radians unit and
a word that is itself a known token closed the popup. The popup now stays open on such a
word. A complete name (`width` with `width_extra` defined) opens it unarmed: the longer
names are listed with no row selected, Enter submits what was typed and Tab moves on; Up
or Down picks a row, after which Enter or Tab complete it as before. A unit means
something only after a number, so a unit typed anywhere else (`rad` on the way to
`radius`) is a prefix like any other: the popup is armed and Enter completes `radius`
instead of submitting one radian.
A complete token is extended only by names that start with it (`in` lists `inner`, not
`rounding`), a typed unit offers no other units (`in` does not list `inch`), and a unit
right after a number (`2 in`, the form every dialog starts its sizes in) opens no popup at
all, so nothing sits over a dialog's OK button.
Fixes#3663
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Homero Thompson <vdirienzo@gmail.com>
Co-authored-by: SS-EN <84559909+SS-EN@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Martin Pålsson <martin.palsson@outlook.com>
Co-authored-by: Daniel <daniel.owsianski@gmail.com>
* Render: isolate matched GGX visible-normal sampling (#1654)
Independent numerical constituent of #5493. Part of #1654; the CPU reflection consumer remains #1698, so this does not close its image acceptance. Original sampler, matched PDF and analytic/hostile-input tests are preserved from the reviewed donor. 189 render tests, strict Clippy, wasm32 and repository gates pass; fresh quick run 40086624771 passes. No CI configuration, saved-data or kernel change.
* GPU: match index-matched Fresnel in path and viewport (#6036)
Fixes#6035. Preserve actual GPU/f64 material-oracle checks in both translations. Scoped Vulkan, render, Clippy, WASM and fresh quick checks pass. No CI configuration or gate changes.
* feat(render): expose bounded original appearance recipes (#1722)
* claim: resolve the main/render synchronization conflict for #6050
* Settle appearance panel scrolling before targeting Search (#6079)
* claim: #6076
* test(render): settle appearance scroll before resolving Search (#6076)
Expanded starter swatches moved under the scenario click while the existing scroll animation was active. Advance 20 deterministic harness frames before resolving the Search handle. No wall-clock sleep, timeout change, assertion deletion or production change. Original native reproduction fails at the empty-state assertion; all 124 scenarios now pass with one pre-existing ignored adhoc entry.
---------
Signed-off-by: Karl Von Richter<karl@embeint.com>
Co-authored-by: Victor Santos Oliveira <56417388+Vittuu@users.noreply.github.com>
Co-authored-by: Christian Svensson <blue@cmd.nu>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Jackson Oursland <jacob.oursland@gmail.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Frank <58932224+plexi32@users.noreply.github.com>
Co-authored-by: raykholo <raykholo@users.noreply.github.com>
Co-authored-by: CamWight <91156642+CamWight@users.noreply.github.com>
Co-authored-by: SS-EN <84559909+SS-EN@users.noreply.github.com>
Co-authored-by: Colin Faulkingham <colin.faulkingham@gmail.com>
Co-authored-by: Karl Von Richter <karl@von-richter.net>
Co-authored-by: HiroKenEz <59478073+HiroKenEz@users.noreply.github.com>
Co-authored-by: Peppe Bergqvist <p@bergqvi.st>
Co-authored-by: GitHub Copilot <noreply@github.com>
Co-authored-by: phanoob <lesur.Stephane@wanadoo.fr>
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Kai Evans <nemshilov@gmail.com>
Co-authored-by: Tyler Chiriboga <tylerchiriboga.apps@gmail.com>
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: CaRLymx <carlymx@gmail.com>
Co-authored-by: carlymx <24634598+carlymx@users.noreply.github.com>
Co-authored-by: Loohi <78334654+Loohi@users.noreply.github.com>
Co-authored-by: leValdemar19 <valdemarnord@hotmail.com>
Co-authored-by: AhmetV <mihendiz10@gmail.com>
Co-authored-by: AhmetV <344941595+ahmetv10@users.noreply.github.com>
Co-authored-by: rlartot <94693724+rlartot@users.noreply.github.com>
Co-authored-by: Thodoris Tsagkogeorgas <thodoristsagko@gmail.com>
Co-authored-by: ahmetalpcelik <80778880+ahmetalpcelik@users.noreply.github.com>
Co-authored-by: ealitt <ealitt@gmail.com>
Co-authored-by: Homero Thompson <vdirienzo@gmail.com>
Co-authored-by: Diogo Soares Rodrigues <dsrodrigues20@gmail.com>
Co-authored-by: Artem Sushchev <Suschev.Artem+github@gmail.com>
Co-authored-by: Nathan Dane <nathandane.dev@gmail.com>
Co-authored-by: Kai Evans <hi@kaievans.co>
Co-authored-by: cronos <crono2250@users.noreply.github.com>
Co-authored-by: rbousigues88 <rbousigues@gmail.com>
Co-authored-by: The Nomad <nomad@positron3d.com>
Co-authored-by: Jacques <59315319+Jacquesvdberg92@users.noreply.github.com>
Co-authored-by: Martin Pålsson <martin.palsson@outlook.com>
Co-authored-by: Daniel <daniel.owsianski@gmail.com>
* claim: #3083
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* WIP: choose the window system on Linux in Preferences (#3083)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Preferences: a window system set over the control channel is saved for the next start again
The merge kept main's check in ui.prefs, which only looked at the anti-aliasing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Preferences: the window system section drops the restart badge its note already states
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2459
* Sketch tools: pressing a tool's shortcut again cycles its creation modes (#2459)
R starts 2-Point Rectangle; R again, before the first corner, switches to
3-Point, Center, Rounded, then back to 2-Point (likewise for Circle,
Line, Arc, Slot, Polygon and Spline keys). A held key's auto-repeat
doesn't cycle, the first press always starts the bound mode, and once a
point is placed the key keeps the tool and its points.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch tools: the mode cycling is a Sketch preference, off by default (#2459)
A tool's shortcut pressed again keeps restarting the tool, as before;
Preferences > Sketch > "Pressing a tool's shortcut again switches its
mode" turns the cycling on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* ci: repair local nextest scheduling and Windows fixtures
* test: refresh MSVC robustness references and isolate GNU policy
Integrate the Windows reference records from #3361 and the target-environment
selection and recording policy from #4159. Keep shared records and comparison
tolerances intact, and retain seed 36 as an existing failure tracked by #3583.
Co-authored-by: Ray <raykholo@users.noreply.github.com>
Co-authored-by: chris <christopher.nourse@gmail.com>
* test: make CI subprocess and revolve fixtures deterministic
---------
Co-authored-by: Ray <raykholo@users.noreply.github.com>
Co-authored-by: chris <christopher.nourse@gmail.com>
* claim: #4985
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Equal with three or more curves selected makes them all equal (#4985)
With more than two curves selected, Equal fell back to the pick tool and
applied nothing. sketch.constraint.equal also takes curves: [2..200],
each made equal to the first in one step (strict, so a conflict changes
nothing), and the toolbar sends a selection of three or more curves as
that list.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5074
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Trim cuts a line at its tangent point on a circle (#5074)
A line the solver made tangent to a circle touches it within rounding
noise, and the exact intersection missed the touch whenever the noise
fell outside, so trim found no cut there and deleted the whole line.
Trim takes its cut points from intersections_within at its own merge
distance, as profile detection does (#588).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3330
* Linked components: joints keep their face after Update Links (#3330)
A linked design comes in as STEP text, which keeps no face names, so the
faces of its bodies were named by their place in the file (`face:<k>`).
An edit of the part that adds a face earlier in the file (a second hole
of the same size) gave that place to another face, and a joint on the
linked part moved there silently.
When a link is made or updated, each imported face is matched to the
part face it was written from (the same geometry), and the Import
feature keeps that face's name in the part (`Feature::link_faces`, a new
pending format change). The linked face is named
`F<import>🔗<digest of that name>`, pieces keeping their qualifiers,
so references follow the part's own history through Update Links as they
do inside the part, and a face the part no longer has is reported.
Linked faces still answer to `face:<k>`, so older references keep
resolving; Update Links warns about joints that use them and reports
joints that no longer solve.
* Linked faces: a solid's twin keeps its own names; renamed faces count as an update
Two solids of a part alike in one place each take the names of the solid
the file names them after. Update Links reports a link whose part named
its faces differently since as updated (a joint may follow that name to
another face), not as unchanged; only a link getting its first names
stays unchanged.
* Linked faces: Extract to File carries the design's face names too
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #3663
* Expression completion: keep longer names listed when the typed word is a unit (#3663)
Typing `rad` into an expression box showed nothing because `rad` is the radians unit and
a word that is itself a known token closed the popup. The popup now stays open on such a
word. A complete name (`width` with `width_extra` defined) opens it unarmed: the longer
names are listed with no row selected, Enter submits what was typed and Tab moves on; Up
or Down picks a row, after which Enter or Tab complete it as before. A unit means
something only after a number, so a unit typed anywhere else (`rad` on the way to
`radius`) is a prefix like any other: the popup is armed and Enter completes `radius`
instead of submitting one radian.
A complete token is extended only by names that start with it (`in` lists `inner`, not
`rounding`), a typed unit offers no other units (`in` does not list `inch`), and a unit
right after a number (`2 in`, the form every dialog starts its sizes in) opens no popup at
all, so nothing sits over a dialog's OK button.
Fixes#3663
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* claim: #395
* Mirror and pattern: a pattern or mirror feature as the source (#395)
A pattern or mirror of features refused another pattern or mirror as a
source ("Pat cannot be patterned or mirrored yet"). Such a source now
stands for what it copied: its own sources are replayed at each of its
transforms moved by each of ours, to any depth up to eight. Copies of
a feature picked directly keep the instance numbers they had, so
existing designs keep their face names; copies made through a nested
pattern or mirror are numbered after them, so one feature reached two
ways never names two copies alike. A pattern or mirror of bodies as
the source is still refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4420
* Renaming a parameter follows it into sheet-metal and plastic rules and canvases (#4420)
Document::rename_param rewrote other parameters, feature inputs, dimensions,
configurations and joints, but not the expressions of sheet metal and plastic rules or
the plane a canvas lies on, so after a rename the sheet metal and plastic features
using them failed with "unknown parameter" and the canvas left the view. It renames
them there now, and param_users lists them (a rule's field, a canvas's input), so a
parameter they use can't be deleted.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Parameters: rules' features in drives and the graph; shared rule field lists (#4420)
A sheet metal or plastic rule that uses a parameter makes the sheet metal (or plastic)
features rebuild when it changes, so parameters.users lists them in drives and
parameters.graph links them. SheetRule::fields/fields_mut and PlasticRule::fields_mut
replace the hand-written copies of the rules' fields in params.rs, eval.rs and
sheet.manage_rules. Suggested by @CamWight.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fmt (#4420)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4538
* CLI: --help on each subcommand prints its usage and exits 0 (#4538)
Every command now answers `--help` and `-h` with its own usage on stdout and
exit 0, the way the top level has since #1828. Before, each subcommand
answered differently: `fuzz --help` started a fuzz run with the default seeds
and ran until it was stopped, `recipe --help` took the flag for a file name,
and `snapshot`, `render`, `oracle`, `run`, `eval`, `script` and `crashes`
answered with an error on stderr and exit 1.
The help is read before the command runs, so `--help` never starts work: no
fuzz run, no MCP server, no file read.
`fuzz` also refuses an argument it doesn't take, as `snapshot` already did: a
flag it doesn't know and a stray word are errors, not arguments to ignore
while a hundred seeds run.
The usage lives in one place now: each command's lines are a constant, and the
top-level usage is composed from them, so the two cannot drift apart. The
top-level usage is unchanged, byte for byte.
# Conflicts:
# apps/solidcraft-cli/src/main.rs
A Single hole can take up to two straight edges of its face as references,
each with the distance of the hole's centre from that edge's line, measured
in the face's plane. `solid.hole` takes them as `references: [{edge, distance}]`;
the edges are found again by their persistent names, else their points, so
the hole follows them when the part changes, and each distance is a
parameter. A bad reference (no edge there, a curved edge, one square to the
face, parallel edges, a negative distance, a centre off the face) is refused.
In the Hole dialog, Reference 1 and 2 follow the face, each with its Distance
row; the canvas shows each distance as a dimension from its edge to the hole,
with a box just past the edge to type it in. Picking an edge measures its
distance (and its box takes the keyboard), typing one moves the hole, and
moving the hole measures them again. Edit Feature picks the edges again with
their distances.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Placing a Single hole put it exactly where the click landed, with nothing to line it up by and no
way to move it but taking it out and picking again; the dialog was a list of size fields.
- Preferences › Design › Create Hole has two settings, both off by default, so a new installation
places holes as before:
- Snap to face and edge centres (`hole_snaps`): the planar face under the cursor offers its
middle and the centres of its round edges (a boss, a rounded corner, a hole already there) as
marks; a pick near one, or on or beside such an edge, takes that centre and the mark is named.
Sketch points on the face still come first; Alt or Ctrl held places it where clicked.
- Move holes by clicking or dragging (`hole_moves`): each placed hole has a handle that drags it
(snapping when snaps are on) and takes it out when clicked, and a pick elsewhere on its face
moves it there. Edit Feature shows the handle too, so a hole can be moved later (#1047).
- The Hole dialog draws a section of the hole as its options make it, with each size in a field
beside the dimension it sets; a hovered or focused field lights its dimension. The layout is a
new kit element, `ui_kit::figure_rows` (a caller-painted figure beside captioned fields).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A new kit control, `ui_kit::count_field` / `count_row`, is an expression field for a
whole-number count with stacked step arrows at its right edge. An arrow steps the
evaluated count by one within the field's limits: a plain number is replaced, an
expression keeps its parameters and gains or adjusts a trailing `+ n`. The text
changes as typing changes it, so the live preview follows.
Used by the Quantity rows of the sketch Rectangular (1-500, both directions together
at most 500) and Circular (2-500) patterns, and the solid Rectangular, Circular and
Path patterns (1-1000).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4434
* Join: say so when the joined piece doesn't touch the body (#4434)
A Join (extrude, revolve, sweep, loft, pipe, coil, primitives, thicken,
boundary fill, surface network) whose material meets none of its targets
still joins it to the first target as a separate piece. The result stays
as it was; the feature now warns: "the joined piece doesn't touch `Body1`,
so `Body1` is now in 2 separate pieces (New Body makes it a body of its
own)". History marks it and the command reply carries it.
The check compares pieces (outer shells, not voids) of the bodies the
feature changed or absorbed before and after it, so it holds for the
batched multi-profile path too, stays quiet for a join that bridges
pieces or closes a cavity, and leaves Combine (which joins the bodies it
is given) alone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
cadmpeg 0.6.3 is pure Rust (ruzstd instead of the zstd C library, zip 9), so
solidcraft-io depends on it on every target and the wasm build reads .FCStd
and .3dm files like the desktop app. The browser's file pickers offer them.
0.6.3 replaces DecodeOptions::container_only with a decode depth; ask for
Geometry, what its STEP writer reads. zip 8 leaves the tree, and with it the
three cargo-deny advisory exceptions kept for it.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* claim: #4302
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Timeline: suppressing a feature keeps later features' body names (#4302)
Bodies numbered by the counter (`BodyN`) take the next free number on every
recompute, so suppressing a feature that made one moved every later counted
body down one (`Body2` became `Body1`) and broke what named it, such as a
Move of `Body2`. timeline.suppress now stores the current names of later
features' counted bodies in their `body_names` first, as body.rename already
does for a renamed counted body (#1109). Engine only: no evaluation or
design-format change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Timeline: a body made while a feature is suppressed keeps its name (#4302)
Suppressing a feature freed its counted name (`Body1`), so a body made
meanwhile took it and became `Body3` once the feature was back. Suppress
now also stores the suppressed feature's own counted names, and the
evaluator adds the names stored on earlier suppressed features to the
state's `reserved_body_names` (the names a rename freed, #4957), which the
`BodyN` counter passes by. They are taken from the design before each
feature is computed and are part of the state digest, so a cached result
is only reused when they match.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Add a small test boat to steer around with the 3D mouse
A little hull in its own camera, drawn by the UI itself and not the CAD
viewport, so a 3D mouse can be tried out without opening a design. It
brings the model (generated by assets/models/make_test_boat.py) and the
tests for how the camera moves it.
Nothing uses it yet; Preferences picks it up with the rest of the 3D mouse
work.
* UI kit: keep a choice row from flipping with the scrollbar
A choice row picks segments or a dropdown by whether its labels fit. A page
scrollbar that comes and goes narrows the row by its own width, so the row
could flip back and forth between the two on a page that sits right at the
edge. It is now judged on the whole row and through a small helper,
fits_steadily, that only changes its answer once the miss is wider than the
scrollbar.
Also tests for that, for typed numbers in a slider's value box, and a
test hook to put up the issue-report input barrier by hand.
* 3D mouse: device model, guided calibration, buttons and a picture
The 3D mouse becomes a device we know something about instead of six raw
numbers.
The reader picks between spacenavd and the direct hidraw driver, falls back
from one to the other, and says so when it can't open the device. Guided
calibration asks whether the device is a spring cap or a trackball, adds a
touch step, and uses a settle filter so a ball let go of stops reading as
motion. Buttons get jobs (keys, held modifiers, commands), saved per
device, with defaults for the SpaceMouse Pro.
Preferences gets a picture of the device that moves as it does, next to the
test boat, and the same picture can sit in the 3D view at Left Viewport,
Right Viewport or Left Window, above the floating History card like the
legend and readouts.
These ride together because the crate's calibration types, the Preferences
page, the Linux reader and the button hook all use each other.
* 3D mouse: offer to calibrate an uncalibrated device
When a 3D mouse shows up that has not been calibrated, a card in the 3D
view offers to calibrate it, to swap drivers when the other one would work
better, to close, or to never ask again. It is made once per device, goes
away after 30 seconds in view, and gives way to a status or computing
toast instead of stacking on it.
The card is the shared toast, which now takes buttons: they group by
purpose, wrap inside the card at narrow widths, and do nothing while an
issue report holds CAD input. The picture of the device in the 3D view
keeps clear of the card.
* 3D mouse: translations and attribution
The 3D mouse strings (ui.3d_mouse_*, the calibration steps and the offer
card) in every shipped language, the Japanese catalog note, and the credit
for the test boat model in ATTRIBUTION.md.
This is last because the code before it only looks keys up: it builds and
runs without these, falling back to English until they land.
* 3D mouse: draw its preferences with the settings kit like the rest of Navigation
The 3D mouse section was a folded header with its own layout. It now
uses the same kit as the sections above it: a settings_section card,
settings_form rows with the shared label column, dividers between the
device, axes, picture and buttons groups, Text::Hint for the helper
lines, and settings_control_width for the dropdowns.
The section is open like every other one (the header still folds it),
so the device badge on the folded header is gone; the calibration row
already names the device. Every setting, string and key is unchanged.
Tests that unfolded the section no longer click it, and the hint lines
are looked up as hints instead of notes.
* 3D mouse: show each axis's live tilt in Preferences, like main's axis rows
Each view axis row is main's axis row again: a meter that follows the tilt of the motion driving it, with the dead zone marked, next to Invert, Speed and the "driven by" menu. The meter reads the calibrated travel when the device has a calibration and the raw reading over full scale when it does not, and asks for a frame whenever the reading changes. The helpers are back as they are on main, and the calibration state is a badge by the enable toggle.
* 3D mouse: line its preferences up with the rest of Navigation
- The two checkboxes sit side by side in the same two-column grid as Wheel and trackpad.
- Row labels stop repeating "3D mouse" and use sentence case: Calibration, Driver, Restore defaults, Reset view, Left viewport, Right viewport, Left window.
- Hints line up under the control column instead of the left edge.
- The six view axes are named Pan X, Pan Y, Zoom, Orbit X, Orbit Y and Roll so they fit the name column.
- Buttons are a divider and rows in a form like the other groups, without the bold sub-heading.
- The driver, dead zone, axes, picture and buttons sit behind a collapsed Advanced section, since most people never touch them.
* 3D mouse: tidy the Advanced test for clippy
Rebased onto main as one commit (was 120d26f7, df488f9a, db9d1073 plus
merges of main).
- A request the app reports as still running (#1618) is waited for on the
same connection through `control.pending`; past the wait the error says it
was not cancelled and will still apply (or only write its file), later calls
are not sent behind it, and the new read-only `pending` tool returns its
result once it lands.
- A failed batch says what really happened: `rolled_back` only when the
design is back where the batch started (as #3288 / #3242 on main: the app's
`rolled_back` is believed; for an older app, every completed step undone;
only a rollback asked for is reported as one), plus `applied`, `not_run` and
`failed.still_running`.
- A batch step may name a tool (set_parameter, list_commands,
describe_command, execute, measure, inspect_design, undo, redo, export, save,
open, new_design, crash_reports); screenshot, body_topology, batch and
pending are refused before anything is sent.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* docs(mcp): Windows setup for Claude Desktop (paths, two servers, Store config file)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* docs(mcp): say the Store build can keep the config elsewhere, not that the other path fails
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Retain MID snaps across both picks and materialize reusable midpoint
constraints for point-to-line and point-to-point dimensions. Let midpoint
pairs choose horizontal or vertical spans with pointer placement before
entering a value.
Cover projected model edges, overlapping geometry, helper reuse,
undo/cancel/reopen, and existing angular and line-circle workflows.
While a section is on, Section Analysis opened with no plane picked now
says "No plane picked: OK turns the section off." and OK runs
inspect.section {"clear": true}. With no section on, OK with no plane
still asks for a plane.
Fixes#2410.
* claim: #4922
* UI: let themes own corner radii
* UI: use four-point panel and window radii in SolidCraft 95
* UI: align the 95 radius assertion with four-point panels
* claim: #4953
* UI: let themes own structural and control borders
* UI: single Blueprint outlines and dotted secondary rules
* UI: separate container outlines and add theme-owned 95 bevels
* claim: #5007
* UI: apply theme palettes to icons, workspace tabs and Gruvbox settings
* claim: #5045
* Viewport: invalidate cached geometry colours when the theme palette changes
* UI: preserve legacy theme logo colours with the borderless mark
* assembly(feat): ground first parts and drag free joint chains
Ground the first top-level part when it is created and let face dragging reposition free occurrences or any member of a floating joint chain in one undo step. Preserve exact repeated-placement frames, joint values, modifier selection and Escape cancellation. Keep the accumulated fixture, viewport layout and documentation integration fixes from the existing PR.
Signed-off-by: Karl Von Richter<karl@embeint.com>
* assembly(fix): update movable fixtures for first-part grounding
Accept the documented grounded override in component creation, explicitly free movement fixtures, and keep linked clipboard pastes movable even in an empty document. Cover first and reused linked pastes, default insertion grounding, and single-step undo in the existing link tests.
Signed-off-by: Karl Von Richter<karl@embeint.com>
* assembly(test): keep the motion-cost crank free
Unground the first component in the document-level joint-face measurement fixture so its motion samples can rotate after first-part grounding. Preserve the rotation and shared-measurement assertions that caught the Linux CI failure.
Signed-off-by: Karl Von Richter<karl@embeint.com>
* assembly(test): free the frame placement fixture
Create the frame-member placement fixture ungrounded before moving it above its root sketch. Keep its world-space member assertions while respecting first-part grounding in the frame tests added on main.
Signed-off-by: Karl Von Richter<karl@embeint.com>
---------
Signed-off-by: Karl Von Richter<karl@embeint.com>
* Extrude: picking a face, vertex or parallel plane off the profile turns the extent to To Object
With profiles picked, a face off their plane (or curved), a vertex off it, or a
construction plane parallel to it is taken as the To Object target instead of
another profile: the extent switches without the Extent Type choice, and the
distance shows how far the target is. Coplanar faces still join the profiles.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Extrude: automatic To Object is a Preferences setting, off by default
Preferences > Design > "Extrude to a picked target" (`extrude_auto_to_object`, default
false) turns the automatic To Object on. Off, picks in the Extrude dialog only add
profiles and the distance is not synced to a target, as before this change.
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #4587
* Agent chat: end the agent's process group with the system call (#4587)
The group was signalled with a `kill` program: `kill -s TERM -- -<pgid>`.
That works with procps' `kill` and with BSD's, but busybox' `kill` takes no
`--`: it prints its usage, exits 1 and signals nothing. The agent's helpers
(the CLI an adapter wraps, the MCP servers it starts) then stay alive after
the agent is ended.
The group is signalled with the system call instead, so no `kill` program and
no shell are needed and the behaviour is the same everywhere. `nix` (MIT) is
already in the tree through zip, and only on Unix: the browser build does not
take it.
The two tests that cover this now pass where they failed before:
`agent_chat::process::tests::dropping_the_process_ends_its_group` and
`agent_chat::process::tests::stopping_the_agent_leaves_no_helpers_running`.
* claim: #4692
* Delete removes a selected construction axis (#4692)
Delete on a construction axis skipped it with "the origin axes can't be
deleted". It now deletes the construction axis feature, like construction
planes; the origin axes X, Y and Z are still refused.
Keep physical modifier events out of the ordinary-key slot, preview held keys immediately, and confirm after release and a one-second settling interval. Preserve legacy control-channel modifier behavior while allowing explicit physical Control/Command input. Add regression coverage, a UI scenario, localized guidance and documentation.
Fixes#1809.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #4965
* Joins and cuts follow only the placements they depend on (#4965)
Tests that turning a joint between parts cuts no hole again, and that a
tool cutting another part cuts again when that part moves relative to it,
while an unrelated part moving keeps the result.
Booleans and Combine now hash only the placements of the bodies they take
as inputs, relative to the frame that evaluates them (#3397), so these
pin that behaviour for joins and cuts with named targets across parts.
The app opens its localhost control channel on port 27878 (or the next free
one) unless Preferences > Assistant turns it off; --control and
SOLIDCRAFT_CONTROL_PORT still win. A current.port discovery file lets
solidcraft-cli mcp find the running app with no arguments.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #5079
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Crash journal keeps component.activate (#5079)
Activating a component changes only the session's active component and
occurrence path, not the design, so the journal dropped the command and a
crash report's replay built later parts under the previously active
component. The journal now also keeps a command that changes either.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5124
* Flatpak: find the host's fonts for CJK text and the font list (#5124)
Inside the Flatpak sandbox the host's fonts are mounted at /run/host/fonts
and /run/host/user-fonts, which no font search looked at. Japanese and
Chinese text fell back to boxes, and the font list missed the host's
families.
- Search both folders for the CJK and Chinese fallbacks and for the
installed-font list.
- The CJK fallback reads SOLIDCRAFT_FONT_DIRS (a path list, as the font
list already does), with CRAFT_FONTS_DIR kept as a fallback.
A line drawn across a circle only snapped its end onto the curve, so a
diameter could miss the centre at any angle. Near the centre the line now
locks onto the radial line: on the rim it stays on the circle and the centre
is kept on the line, and on the centre point the end lands there.
* claim: #5137
* Dragging a part along its joint no longer highlights what's under the pointer (#5137)
The viewport's hover pick was skipped during a box selection and camera
moves, but not while a jointed part was dragged, so faces and edges under
the pointer lit up on every frame of the drag. The pick now waits until
the joint drag ends.
* claim: #3250
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Return to the originating toolbar tab after finishing a sketch (#3250)
A sketch started from SHEET METAL or PLASTIC always ended on the SOLID
tab. Remember the tab that was active when the sketch began
(UiState::sketch_return_tab, not saved) and go back to it when the
sketch ends. Falls back to SOLID when that is not one of the normal
toolbar tabs. A sketch started from SOLID behaves as before.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Test: no one-element loop in the return-tab test (clippy single_element_loop)
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #3216
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Parameter import and canvas image pickers offer their file types (#3216)
Both opened the picker with no file type of their own, so it listed only
designs, CAD files and meshes (and All files). macOS lets you pick only
the types a filter names: CSV parameters, and PNG or JPEG images for a
canvas, were greyed out. They now ask for csv/json and png/jpg/jpeg,
which the picker lists first (#1801).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3372
* Fix Escape cancellation in Change Parameters
Consume Escape before parameter editors can commit pending text on focus loss, while preserving autocomplete and shortcut capture priority. Add 11 UI regressions covering draft cancellation, preview cleanup, document and history preservation, and reopening saved values.
* claim: #3858 (3MF import: names from a Bambu Studio project)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3MF import: name a Bambu Studio project's bodies after its objects (#3858)
Bambu Studio leaves the name off the root model's objects and keeps it in
Metadata/model_settings.config (<config><object id="N"><metadata key="name">),
so every object of a project came in as Mesh1, Mesh2, ... A build item whose
root object has no name of its own now takes the object's name from those
settings. A name in the model still comes first, and part names, other
settings and objects of other model parts are left alone.
The settings are read only when a body needs a name, within what the model
parts left of the 1 GB budget. Settings that can't be read (malformed, nested
too deeply, too large) never fail the import: the bodies keep the generic
names, with a warning saying why.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3838
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
* Cancel Sketch: leave a sketch without keeping its changes (#3838)
sketch.cancel (Cancel Sketch, ribbon Exit group and the sketch's
right-click menu) puts the design back as it was when the open sketch
was created or opened for editing: a new sketch is removed with its
dimension parameters, an edited one returns to how it was. It is one
undo step, so Undo goes back into the sketch.
The session remembers the design and the undo history at the point a
sketch is entered. Cancel uses that only while the history still runs
through it (the same steps before, every step since taken in this
sketch); otherwise it refuses and points to Undo, so a stale entry can
never throw away later work.
The icon is the approved close glyph in the sketch accent (alias
sketch_cancel); ja and zh carry the new label.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3857
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Joint dialog: a body of a further instance picks that occurrence, not the component's first (#3857)
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #4100
* Deleting an inserted file removes its components, placements and joints (#4100)
A STEP or mesh file inserted into a design makes a component tree. Deleting
its feature left that tree behind: the components, their occurrences, the
rigid joints that hold them and the bodies' component assignments. A later
body that took an imported body's name was then placed in the stale component.
Document::remove_component_tree removes a component subtree and everything
that refers to it: the features made inside it, its occurrences, the joints
on them with their motion links, joint origins, contact set members, motion
study keys and exploded moves, body component and offset entries, standard
part and link records, and configuration columns. Deleting an inserted
file's feature and component.delete both use it; component.delete now also
removes joints left pointing at a deleted occurrence. The body transfer
refusal moves with it, unchanged.
* Only an inserted file's feature takes its component tree when deleted
Mark the feature a STEP, IGES or mesh insert makes (`inserted`), and remove
the component tree on delete only for those and for a linked design's
feature. Deleting a copied component's independent base deletes just that
feature, as before. Older designs get the mark as they open (pending format
change), except copied components' independent bases.
* claim: #4226
* Esc ends the navigation bar's Pan, Orbit and Zoom modes (#4226)
The nav bar toggled a mode on, but nothing cleared it on Esc, so a left drag
kept panning, orbiting or zooming. Esc now ends an active nav mode first;
the next Esc reaches the tool or dialog as before.
each_frame_hands_the_view_its_motion_through_the_drivers_mapping dropped
the output of a frame it ran (`let _ = ctx.run_ui(...)`). epaint panics
in debug builds when a TexturesDelta is dropped with deltas left ("Dropped
TexturesDelta with 1 unapplied deltas"), so the test failed under a plain
`cargo test` on every platform while the ci profile passed. It now clears
them, as the other UI tests do since #60.
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4380
* "Don't Save" looks like a button in the Save changes prompt (#4380)
The prompt's third action used the kit's low-emphasis text style, so it
didn't read as clickable and people thought Cancel was the only way out.
Add an opt-in secondary_left style to the confirm footer (a framed
secondary button from the theme tokens) and use it for Don't Save.
* Save prompt test: clear each frame's texture delta once it is applied
egui 0.36 panics when a TexturesDelta with unapplied deltas is dropped;
the test kept only the last frame's output and dropped the earlier ones.
* claim: #4440
* Flat Pattern view: draw perforation holes (#4440)
The flat view drew the outline, cut-outs, pockets and bend lines but not
perforation fields, though the flat-pattern data and the DXF export list
their holes. The view now reads the hole centres from the sheet through
SheetBody::perforation_holes, caches them per model state and body, and
draws them as circles in the cut-out style. Holes outside the viewport or
smaller than one pixel are skipped, and the cache is capped at one million
holes.
* claim: #4464
* Sketch offset keeps its source lines free to turn (#4464)
An offset adds a Parallel between each source line and its offset line.
Since offsets became associative, the offset's points are held during a
solve, so each Parallel pinned its source line to the offset line's
current direction: a closed chain lost its turning freedom (8 degrees of
freedom became 6), and Horizontal or an angle edit on a source line was
refused as conflicting.
The solver and the constructive reading now leave out a Parallel that
ties an offset line to the source it follows while the association lives;
the association already keeps them parallel. After Break the stored
Parallel drives again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5254
* Toolbar: a constraint dragged out of the small box becomes a full button (#5254)
The Constrain panel's small box of constraint icons was a fixed list, so
dragging one onto the full-size buttons saved a new order but it was still
drawn in the box. The toolbar settings now remember, per tab, which box
commands were moved out; those are drawn as full buttons, the box closes
up, and dropping one back on a box icon returns it. Reset Toolbar restores
the default box, and settings saved before this load unchanged.
* Renaming a body that later features change evaluates nothing (#2467)
Model::rename_body now keeps the results when later features change the
body (a pattern, a chamfer, a join). It still evaluates when a body of
that name comes and goes or isn't the same body throughout, when a later
body's name is worked out from it (X (2)), or when a message quotes it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Rename box: rename off the window thread, and locked bodies follow a rename (#2467)
The rename box commits body.rename through run_heavy, so a rename that
still evaluates later features (one names the body in its targets)
computes beside the window instead of freezing it. Locked bodies now
follow a body.rename from anywhere (the rename box, the control channel,
scripts), not only the rename box.
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3187
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Reorder design tabs by dragging them (#3187)
A tab dragged along the strip follows the pointer while the others make
room, and lands where it is released; Escape, or releasing down the
window, leaves the order as it was. With the custom title bar a drag that
starts mostly up or down (or on the only tab) still moves the window.
The control channel's ui.documents gains a move action (from, to).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #1783
* FreeCAD/Rhino import: decode caps with room for detailed parts (#1783)
cadmpeg decoded FreeCAD and Rhino files under its service profile, whose
million collection items and 2.56e8 work units a detailed part (a sprocket,
a threaded rod, a part with a few hundred faces) runs past: the import was
refused with "resource limit on CollectionItems: BudgetExceeded". Allow
8 million items and 2e9 work units, half of cadmpeg's desktop profile; the
input-size cap stays SolidCraft's STEP limit and every other cap stays at
the service profile's.
decoding_is_capped checks the two raised caps (and that they stay below the
desktop profile's) and that every other cap is still the service one.
Fixes#1783
* claim: #3278
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 2-point circle takes a typed diameter (#3278)
After its first point the 2-point circle tool showed no value box, so the
circle could only be sized with the cursor. It now shows a Diameter box,
like the centre circle: the typed value puts the second point that far
along the cursor's direction, and Enter draws the circle.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Macos: keep touchpad/key modifiers true to fusion
* fix: remove obsolete orbit snap reset from trackpad navigation
Upstream removed ViewportState::orbit_snap. The new trackpad orbit path retained one reference, causing E0609 in the PR merge build.
Validation: all seven trackpad tests pass; cargo check -p solidcraft-ui-egui --all-targets passes on the exact failing CI merge 66a2791b with this fix applied; rustfmt and diff checks pass.
Co-Authored-By: GitHub Copilot <noreply@github.com>
* fix: retain Fusion trackpad help in refactored preferences
Preserve upstream's new preferences pages and apply the translated Fusion trackpad tooltip in prefs/navigation.rs.
Validation: cargo build -p solidcraft passed. Tests and full CI were not run, as requested.
Co-Authored-By: GitHub Copilot <noreply@github.com>
* style: format Fusion trackpad preferences tooltip
Apply rustfmt's required wrapping for the relocated tooltip. No behavior changes.
Validation: cargo fmt --all -- --check and git diff --check passed.
Co-Authored-By: GitHub Copilot <noreply@github.com>
---------
Co-authored-by: GitHub Copilot <noreply@github.com>
* claim: #3471
* Delete in a rolled-back history warns about the features after the marker (#3471)
With the timeline rolled back, the checks a delete runs first (the
confirmation's "these features will fail", timeline.dependents and
selection.delete's would_fail) compared models evaluated only up to the
marker. Features after it, broken by the delete, were not listed, so their
error marks turned up only once the timeline moved to the end.
newly_broken_in_history evaluates both sides over the whole history when
either has a marker (on copies; the model and document are unchanged) and is
used by the delete's report and by timeline.dependents.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3042
* Warn when cut pattern copies remove no material
A rectangular or circular pattern whose copies miss the body succeeded silently and those copies did nothing. The pattern now reports how many of its cut copies removed no material, and keeps the copies that land.
* claim: #4003
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: the profile library (#4003)
The first slice of frames and weldments: the cross-sections structural members
will be made of, each SolidCraft's own parametric definition built in code
(no vendor or standard size tables are shipped).
- solidcraft_doc::frame: bar, round bar, rectangular and round tube, angle,
channel, I-beam and a generic T-slot extrusion as exact line/arc regions,
with defaults worked out from the profile's size, refusals that name the
dimension that does not fit, anchors and placement, and exact section
properties (area, centroid, Ixx/Iyy/Ixy by Green's theorem).
- frame.profile.list, frame.profile.properties (with mass per metre for a
density) and frame.profile.insert (draws a profile into a sketch, locked by
default, at a point, anchor and angle).
- Tests against hand-computed values (rectangles, circles, composite
rectangles by the parallel-axis theorem, fillet spandrels), hostile input,
extrusion volume, undo; catalog entries in every locale; docs/frames.md;
ROADMAP M15.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: drop the built-in T-slot profile (#4003)
Aluminium extrusions and other vendor sections come from the user's own
sketch (drawn or inserted from the vendor's DXF), so the library keeps the
plain steel shapes only.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: members along frame sketch lines (#4003)
frame.member lays one body per straight line of a frame sketch (2D lines or
3D sketch lines), its cross-section square to the line from end to end:
- profile: a built-in profile, or profile_sketch, a sketch whose material
(nested loops alternating, so two circles make a tube) is the section in
its own x and y around its origin; profiles picks override it.
- anchor (origin, centre, centroid or a bounding-box corner or edge middle),
angle (an angle expression, so a parameter), mirror.
- Orientation: the profile's y as near world up as the line allows, x
square to both; a vertical member's x is world X (frame::member_axes).
- Members follow sketch and parameter edits, undo as one step, save and
reopen (format: pending/frame-members.rs), and are named after the profile
(Rect Tube 40x40x3, (2)…), numbering on past names already taken.
Tests against hand-computed volumes and bounding boxes (rectangle, 3D
tower lines, parameter edit, sketch tube profile and picks, anchor/angle/
mirror), refusals, undo/redo and reopen; walk fixtures and goldens; docs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: fixes from the adversarial review (#4003)
- Default member lines are what was drawn: projected and other linked curves
and 3D wires (a face sketch's edges) and lines of no length are left out.
- A profile sketch's material starts from its drawn profiles, as
solid.extrude's "all" does, so a section sketched on a face works; the
even-odd test uses outlines and boxes worked out once, capped at 256.
- Member names are pinned to their lines (`names`, by line id; `body_name`
is the base): deleting or adding a line, or deleting another feature, no
longer renames other members, and frame.member stores every name it gives.
body.rename renames a member on its line (also for later member features);
a copy whose body_names were set still names its members by position.
- Built-in profile dimensions are expressions (ProfileSpec): they follow
parameters, bare numbers take the design's units, and they are feature
inputs. frame.profile.properties/insert read them the same way.
- Default root/toe radii shrink to the room a thick section leaves; radii
given are checked by name; a radius must be 0 or at least 0.01 mm; error
messages report the wall or flange value given, not twice it.
- A line within 1e-6 rad of vertical orients as vertical, so rounding never
turns a post a quarter turn.
- A 2D frame sketch in a differently placed component is mapped into the
member's frame, and member features recompute when placements change.
- Caps before lookups for a design's `lines`; a line listed twice is
refused; a 3D wire that runs back on itself is not a line.
Regression tests for each, from the review's scenarios.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Frames: the Frame Member dialog on the ribbon (#4003)
- Frame Member (MODEL > Create) opens a dialog: the frame sketch's lines
picked in the view (or All Lines), a built-in profile and its dimensions or
a section picked in another sketch, anchor, angle, mirror and name, with a
live preview. Edit Feature reopens it filled in.
- Lines picked in the view keep their sketch (a SKETCH_LINES input picks
ExternalSketchCurve): line ids are each sketch's own, so two 3D sketches
both have a `w2`, and the members went on the newer sketch's lines.
- timeline.redefine takes frame.member. Members keep their names through an
edit unless it gives them a new body_name; frame.member takes `names`.
- A member whose section is a whole profile sketch keeps it when edited
(the dialog shows the sketch instead of picking every region, which would
fill a tube's hole).
- An original frame_member icon, open for review, on the ribbon, timeline
and browser.
- The angle profile is listed as L-Angle, and web and flange as Web/Flange
thickness, so they translate apart from the existing Angle, Web and Flange;
the dialog's labels are in en, ja and zh. en-GB no longer repeats the frame
command labels it spells the same.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Nine tests dropped an egui frame's output with its texture delta still
unapplied (the expression completion editor's frames, the slot preview's
three, the browser's feature rows). epaint asserts against that in debug
builds, so they failed under the dev profile and passed only where debug
assertions are off. They clear the delta as the other tests do; one more
hover test that dropped its output the same way does too.
* claim: #4083
* Feature editors refill named edges and faces where they are now (#4083)
After an upstream sketch change, Fillet, Chamfer, Shell and Draft evaluated
correctly, but their editors refilled the originally clicked points and
could then fail with "no edge at ...". Resolve the stored names through the
document's naming helpers first, keep the resolved index, and map the pick
into the displayed body's frame; the point lookup stays as the fallback.
* claim: #4097
* Sketch fillet keeps a corner placed by an expression (#4097)
Filleting a corner whose position comes from parameter expressions deleted
the point and its expressions, so the sketch gained two degrees of freedom
and later parameter changes distorted the outline. keep_corner now also
keeps fixed or expression-positioned points as virtual sharp corners.
* claim: #4150
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Browser: a second click on a Document Settings entry closes its window (#4150)
Each entry under Document Settings only opened its window. A second click on
Units closed the menu on the press (the row is outside it) and opened it again
on the release; Precision left Preferences open; Parameters and Design
Properties came back as a fresh copy. Now a click on an entry whose window is
open closes it the way the window's own x does:
- Units: context_menu::toggle_for with the row as the opener, the fix#1522
made for the workspace button's menu. A right-click still opens it at the
pointer.
- Precision: prefs::close(app, true), the Cancel/x path that puts back what
changed since the window opened (now pub(crate)).
- Parameters and Design Properties: the window is dropped, as its x does. A
Parameters value typed but not applied yet is dropped with it, as today's
second click already drops it.
Fixes#4150
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4353
* Renaming a plane keeps its projections, canvases and hidden state (#4353)
timeline.rename only renamed the feature in other features' references. A sketch's
projection of a construction plane, a canvas placed on it and the hidden list keep it
by name too, so they lost it: the sketch warned that the plane no longer exists, the
canvas left the view and a hidden plane showed again. Document::rename_feature_keys
renames it in sketch links and canvases, as rename_body_keys does for bodies; the
selection follows in the command, and the hidden list in Session::refresh_from, so a
hidden plane stays hidden through undo and redo too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5269
* DXF: write and read a Bézier at its own degree (#5269)
The DXF writer and reader held a spline's degree to 7, the value MAX_BEZIER_DEGREE had before
#1642 raised it to 9. A G4 blend (one Bézier of degree 9) was written as a degree-7 SPLINE over
its first 8 control points, so a flat pattern's cut-out stopped 4.444 mm short in the file.
- seg_shape takes its limit from MAX_BEZIER_DEGREE: a Bézier is written whole, at its own degree.
- read_dxf reads SPLINE degrees 1 to MAX_BEZIER_DEGREE, the degrees a sketch spline takes.
- Tests: every Bézier degree 4 to 9, a degree-9 round trip, the reader's degrees, hostile SPLINE
degrees and too few control points, the G4 cut-out exported and read back with its areas, and a
degree-9 SPLINE with 2 to 10 points inserted into a sketch.
Fixes#5269
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3989
* ui-egui tests: offscreen GPU tests start their devices one at a time
Every offscreen test started a wgpu instance of its own, and starting one
makes the Vulkan loader scan its drivers. With several test threads two
tests doing that at once crashed the test process with SIGSEGV inside the
loader (loader_icd_scan). offscreen_device now holds a process-wide lock
while it starts the instance, adapter and device; each test keeps its own
device and the rendering still runs side by side.
A new Graphics preference, "Show reference geometry through bodies", draws
inactive sketch contours, origin and construction planes and their axes over
shaded bodies. It is off by default, so the viewport looks as before: reference
geometry is depth-tested and a section leaves it whole (#1229); Create Sketch
and dialogs picking a plane still draw the planes through the model.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#5109 added Frame::hover_tint and selection_tint; two test frames that landed alongside it didn't set them, so solidcraft-gpu's tests stopped compiling on main.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch on a picked origin plane uses the plane shown, also in a moved component
After a STEP file opens its component stays the active one, and standing it up
turns that component a quarter about X. Starting a sketch on XY picked in the
viewport then sent `{"plane": "XY"}`, which names the active component's own
XY: upright after the stand-up. The sketch landed on the world XZ, Look At
turned to face Y and the extrude went along Y.
The viewport and browser only show the design's origin planes, so a picked
origin plane now goes to sketch.create written out as that world plane while
the active component is moved; sketch.create maps it into the component's
frame once. The flip toward the viewer uses the same plane.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Stand-up axes tests: stand the file up through the offer when one is waiting
#5655 offers to stand up SolidWorks files instead of turning them silently.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Hovering or selecting whole bodies -- a browser row, a History group, a body
under the pointer in body-pick mode -- built a CPU copy of every body's
triangles into the highlight scene. On a 40-body 4.3 MB STEP assembly that was
147 bodies, 70 MB of triangle copies and a 519 ms frame.
The model already has the same geometry on the GPU. The highlight now names it
and the renderer draws it again in the highlight colour:
- bodies drawn from a display plan: the model's evaluated plan chunks, with a
hover/selection tint uniform (the chunks stay resident in the model scene);
- bodies drawn as placements of shared triangles: the same placements of the
model's shared triangles, coloured per instance;
- bodies the model builds on the CPU: their vertex range in the model scene's
block.
See-through and faded bodies keep the CPU feedback, which colours them; faces
and edges highlighted on their own are unchanged. The highlight looks the same
(verified side by side).
On the assembly the highlight build drops from 519 ms to 9 ms and no longer
copies any triangles; selecting the whole assembly takes 9-20 ms.
Co-authored-by: Claude Code <noreply@anthropic.com>
* claim: #5295
* An empty intersect is refused instead of deleting the body (#5295)
An intersect whose boolean came out empty removed the target body and
reported success. The shared evaluator now refuses it when nothing of a
named target would remain, when nothing of any target would remain, or
when the tools meet a target and the result is still empty. An unnamed
target the tools miss altogether is still removed, as an intersect with
every body does (#5160). Sheet metal flat machining follows the same rule.
* Sheet metal: an intersect with already removed material is refused (#5295)
* claim: #133
* mesh: Save As Mesh declares the panel that shows it (#133)
The ribbon redesign (#2176) removed the MESH tab, but three things still
pointed at it:
- `file.save_mesh` declared `.at("MESH", "MODIFY")`. The command inventory
prints the placement a spec declares, so `docs/features.md` advertised a
`MESH > MODIFY` location no ribbon has. Its real home is already
INSERT / FILES (`SECTIONS` lists it), so the declaration now agrees with
the ribbon; `destination()` resolved it there before and after, so nothing
moves on screen.
- The user guide sent people to "the Mesh tab under Modify" for Save as mesh.
- A test now refuses any command that declares a tab the ribbon removed, and
pins where Save As Mesh lives.
`PLACEMENTS` keeps its MESH entries: those migrate toolbars users saved under
the old panels, which is a different question from what a spec declares.
---------
Co-authored-by: Geraldrunion <geraldrunion@users.noreply.github.com>
* claim: #4563
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Section Analysis: a Hatch option, off for a plain cut in the body's colour (#4563)
The Section Analysis dialog gets a Hatch check box beside Flip, and
inspect.section a hatch?: bool (default true), so scripts, the CLI and the
control channel reach it too. Hatched, a cut is its body's colour tinted
toward the cap colour with 45 degree stripes, as before; off, it is the
body's own solid colour. The viewport (cap alpha in the uniform), the CPU
rasterizer and ui.render follow it, and the section keeps it when hidden
and shown again or edited.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4589
* Expression suggestions stay closed while another control is dragged (#4589)
A drag elsewhere leaves a text box its focus (egui gives it up on a click,
not a drag), so dragging the Drive Joints slider rewrote the focused Angle
box under its old selection. A letter before the caret ("5.0 d|eg") made a
new word, a fresh undismissed popup, and the list opened over the slider.
Dismissed suggestions now stay closed until the box is typed in, a drag of
another widget dismisses them, and a hidden list forgets its row rects.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4609
* Mesh Convert works without activating the component first (#4609)
Inserting a mesh puts it in a new component while the parent stays
active, so Convert Mesh always refused the first time. Convert now
resolves the mesh body's owning component and adds the conversion
there when the active component is that component or an ancestor,
leaving the active component unchanged. From another branch it still
refuses, and the error names the component to activate.
* Control: a whole-design measure reads a copy off the window thread
`inspect.measure` and `document.inspect` (which measures every body by
default) ran on the window thread. On a large assembly that took about a
minute, the window answered nothing meanwhile, and Windows closed it as
not responding (#4604). They change nothing, so they now run on a copy of
the session on a worker and are answered from there, like exports, but
whatever the "agent commands off the window thread" preference says: an
answer from the design as it was when asked has no order to keep. The
window keeps answering `ui.*` and drawing meanwhile.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Control: clearance and interference checks read a copy off the window thread too
inspect.clearance and inspect.interference only read the design, and on a
large assembly measure every pair of bodies: on a 153-body design a clearance
check computed for over five minutes on the window thread, the window showing
"not responding". They join inspect.measure and document.inspect in running on
a copy, answered from there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4030
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch Symmetry: a dimensioned shape slides onto the axis (#4030)
With a rectangle's width dimensioned, Symmetry of its two sides about an
axis was refused: the second side can't move on its own, the axis never
moves, and the ordinary solve that slides the whole rectangle only ran
when every first-picked entity could move. One movable first entity is
now enough.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4328
* Fix sheet conversion picks on tangent bend edges (#4328)
Resolve cylindrical bends through the nearest edge's incident faces when a
pick on a tangent boundary resolves to its planar panel. This keeps valid
remote bends connected, including picks on the other skin of a shell.
Keep conversion geometry and picks in the body's local frame, and format
conversion diagnostics through its world placement. Cover rotated bend
picks and disconnected-pick coordinates with synthetic regressions.
A join failed with the kernel's internal error "tolerance must be no less
than 1e-6" at vendor/truck-geometry/src/decorators/intersection_curve.rs:172:
IntersectionCurve::parameter_division asserted truck's minimum tolerance,
and the guard turned the panic into a failed operation. A tolerance below
truck's TOLERANCE, or one that is not a number, now divides at TOLERANCE:
the curve's points are only found to within it anyway.
The test divides the intersection edges of a hole cut through a B-spline
face at 5e-7, 1e-12, 0, -1 and NaN: each gives the division at 1e-6. Without
the fix it panics with the reported message at the reported line.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4593
* Pattern and mirror: a rib as the source (#4593)
A pattern or mirror of features refused a rib as a source ("Rib1
cannot be patterned or mirrored yet"). A rib's walls run until they
meet the body, so each copy is built again where it is: the rib's
sketch is moved by the copy's transform and the walls are run against
the bodies there, as a hole is drilled again at each copy. The copies
join the body, as the rib does. The rib's own tool building moves into
rib_tools, which the rib feature calls unmoved.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch Point: a point put on a midpoint or a curve snap is held there (#4691)
The Point tool sent only a snapped point's coordinates and sketch.point
took only [x, y], so a point clicked on a line's midpoint (or on a curve,
or a model vertex) was free to move off it. sketch.point now takes the
snap forms the other draw commands take ("mid:<line>", "on:<curve>:x,y",
{vertex}, {on_edge}); the snap makes the point with its constraint. The
tool passes those snaps as themselves; a snap to an existing point still
goes as coordinates, where `connect` joins them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch Point: refuse a snap it can't read instead of answering as if it made a point
sketch_project::snap_points leaves a snap it can't parse ("on:c1:abc",
"on:c1", "on:c1:1,2,3") as it was given, and sketch.point then read
that string back as the point it had made: no error, no point, and an id
a script would go on to dimension. It now takes the snap only when it
names a point of the sketch, and otherwise refuses, listing the snap
forms (found by the test agent on #4745).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The app's background loader (file_load::control) kept only `up` and
`names` of a control-channel open or import, so file.insert_step's
`matrix` and `name` were dropped and the part landed at the origin, and
file.insert_mesh lost all its options. It keeps every parameter but the
path, as given; each command checks them as it does when run directly.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
timeline.rollback / timeline.roll_to read a negative position as 0, so
an agent passing -1 for "one step back" rolled the whole timeline back
and every body disappeared, with no error. A negative position is now
refused. The two commands shared one function that refused in
roll_to's name; each now refuses in its own.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Nine messages: the pattern tools, their headings and two sample
descriptions. "Réseau" can be mistaken for a network.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #4452
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Combine takes bodies clicked in the Browser (#4452)
A body row in the Browser only replaced the selection, so with Combine
open its target and tools stayed empty and overlapping bodies could only
be picked in the view. When the open dialog's active input takes bodies,
a body row now goes there, as plane and axis rows already do.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Combine Browser picks: test both layouts and control-channel clicks (#4452)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The Keyboard Shortcuts page refused a key another command used, even a
sketch tool and a model command that never run in the same mode, so
Shift+S could not open Create Sketch in the model and place a point in a
sketch. Edited by hand, the press ran Create Sketch inside the sketch.
A sketch tool (SKETCH tab) and a model command (the other tabs) may now
share a key. A press runs the enabled command of the mode the view is in
first: the tool in a sketch, the model command outside. Two commands of
one mode still clash, and commands on no tab (undo, views, files) clash
with everything.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4783
* Physical material: one rule for a body's material and its density (#4783)
inspect.measure, component.properties, occurrence.properties and the
Properties window took a body's material name from the raw assignment,
but its density from Document::density, which falls back to the default
material when the assigned one is unknown or has an unusable density. So
a damaged or hand-edited design read one material and weighed another: a
design material "Broken" at -3 g/cm³ read "Broken" at 1.29 g/cm³.
One rule, in one place:
- DEFAULT_MATERIAL (new, beside DEFAULT_DENSITY in the material library)
is the default material's name. The library's first row uses it.
- Document::body_material (new, beside Document::density) is the material
a body's mass is computed with, as the library or the design spells it:
the assigned one when it is known and has a usable density, else
DEFAULT_MATERIAL. Document::density is that material's density, with
the same value as before for every input.
inspect.measure and the physical properties name body_material.
inspect.measure now names every body's material, measured or not, since
naming it takes no measuring; the density stays with the mass, on the
bodies measured. So the Properties window shows the engine's name without
a fallback of its own. The places that keep the assignment as stored
(configuration cells, material.assign, material.edit, material.delete
and the dialogs) only use the constant, so a configuration still keeps a
damaged assignment as it is.
The visible change: a body whose material is unknown, or has an unusable
density, reads "Default", the material it weighs, and a miscased name
reads as the library or the design spells it. A body inspect.measure
can't measure now has its material too. Masses and densities don't
change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The Move dialog's floating value box always showed and edited the X
distance, whichever arrow was dragged, and Tab left the box for another
widget. Now:
- dragging an arrow shows that axis's distance, turning a ring (Free Move)
that axis's angle; a plane square or the ball leaves the box as it is;
- Tab in the box goes to the next value (X, Y, Z, then in Free Move the
X, Y and Z angles) and Shift+Tab back, the keyboard staying in the box.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2987
* Flange on a plain solid says how to start sheet metal (#2987)
Flange, Flat Pattern, and Export DXF on a design with no sheet metal body
now say how to make one: Flange > Base, or Convert to Sheet Metal on a
plate's large flat face. An edge flange checks for a sheet metal body
before it adds a feature. The message is in the catalogs, and engine
messages that match a catalog entry are shown in the interface language.
* Dialogs show a refusal without the engine's parameter prefix (#2987)
The Flange dialog showed "invalid parameters for `sheet.flange`: " in front
of the new guidance. The dialog already names its command, so the red
refusal text leaves that prefix out and follows the interface language
when the message is in the catalogs. Create Issue still gets the full text.
* claim: #4116
* Web: a trackpad swipe pans instead of zooming (#4116)
Browsers deliver trackpad swipes as wheel events without scroll phases, so
the shared navigation treated every vertical swipe as a mouse-wheel zoom.
The web raw-input hook now brackets point-mode swipes (small or horizontal
deltas) with gesture start/end phases, ending after 100 ms idle or before
line/page wheel or pinch input. Wheel notches and pinch still zoom, and
camera input still goes through navigation::handle.
* claim: #221
* Units of new designs apply to the blank design already open (#221)
Changing Preferences > Units of new designs only took effect at the next
File > New, so the empty design open at launch stayed in millimetres. When
the preference changes (the dialog, Cancel, Restore Defaults, or
preferences.set) and the open design is blank and unsaved, it takes the new
units at once, without an undo step or unsaved changes. A design with
content or a file keeps its units.
* claim: #4794
* Hem: the direction arrow points the way the hem builds (#4794)
The Hem dialog shared the edge flange's arrow, which points out of the
sheet's larger neighbouring face. A hem folds back the other way, so the
arrow pointed away from the material it adds. Flip the hem's arrow so it
matches the geometry; edge flanges keep theirs.
A regression test compares the arrow with the side the finished hem adds
material on, for both Flip states, top and bottom edge picks, and both
dialog layouts.
* claim: #3220
* Model cache: keep fewer states of a design too big for its file instead of none (#3220)
A model cache over the container's budget keeps the final state, then the
slowest features' states while the bodies they add fit (a body several
states hold counts once), and is left out only when not even the final
state fits. Opening evaluates just the features whose state the file left
out, each from the state before it, so the model is whole afterwards.
* Model cache: keep each encoded body apart, copy only what a cache stores
Encoding a model too big for its file no longer copies every distinct body
into one buffer before thinning: the payload is copied from the encodings
once its size fits, so saving holds no more than before.
* Model cache tests: canonical runs in the timeline fixture; rustfmt
* Model cache: count the records a thinned cache keeps; always try the final state
Thinning counted only the bodies, so a model of many copies of one part
(records, not bodies, filling the room) shrank its budget to nothing and
saved no cache although its final state fitted. The choice now counts the
records of what it keeps, always keeps the final state, and says what it
counted, so each retry keeps less and the final state alone is always tried.
* docs: a thinned model cache counts its records and always tries the final state
* Model cache test: the slowest feature's state holds bodies 16 to 23
* Model cache: thinned opens with fingerprints that read the state before a feature
Since #3397 a feature's fingerprint reads where the bodies before it lie.
After a state the file left out, adopt takes the saved fingerprint and the
evaluation that works the state out checks it. A result evaluated ahead
is never taken as one the file left out. The thinned round trips include
the placed components' design, whose cut reads those placements.
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #4880
* Rename a body: keep the names of siblings a later feature removed (#4880)
Renaming a body sets the stored names of all bodies its feature made, so
the siblings are not renumbered. Those names were read from the final
model, where a body that a later feature removed no longer exists: its
stored name was lost, it came back as `Ext (3)`, and the later Remove
failed with "no body `Body3`". Read the siblings from the feature's own
result instead, as the later features already are.
* Test: renaming one half of a split leaves the other
Renaming one half keeps the other half's name, also once a later feature
joins that other half away, after both an incremental and a full
evaluation.
* claim: #4846
* MCP tools: unknown command ids suggest the closest ids (#4846)
An unknown command id in `execute` or `batch` now lists up to three closest
ids, and an id with parameters packed into it says to pass them in `params`.
* claim: #4804
* Keyboard Shortcuts page: draw only the visible rows (#4804)
Off-screen rows keep their height but skip their cells, the list is
filtered once before measuring and drawing, and each row's shortcut
comes from the command spec already in hand instead of a registry
lookup.
* claim: #4899
* Sketch: offer Break Offset Association for offsets of lines (#4899)
Trim and Extend refuse an associated offset output and point to Break
Offset Association, but the right-click menu offered it only for
free-form offsets. Offer it whenever the curve is an offset output
(analytic, topology or free-form), using the sketch's shared lookup.
* claim: #4851
* solidcraft-cli bench-assistant: the standard Assistant model test
Runs the eleven v1 tasks of #4851 against a desktop app that is already
running with a control port. Each task starts a new design and a new chat
with All permissions, runs only the prescribed setup, sends the prompt
once and waits for the turn. The result is checked on the design
(document.inspect with measurements): body count, volume within 1%
(0.3% for the fillet), bounding-box size in any orientation, the lowest
corner, the body name, live feature kinds, and changed or unchanged
content; the volume question is checked on the reply. Failures get the
issue's categories from the conversation's items. Prints a table, and
writes the issue's comment template (--markdown) and the raw results
with each conversation (--json).
Dropping a feature just behind the history marker now rolls it back, and
dropping it just before the marker makes it active again, in both the
bottom strip and the left panel. timeline.reorder and timeline.move_many
take an optional `active` param that sets the marker in the same command,
so the move is one undo step and one recompute. A position on the wrong
side of the marker for the requested `active` is refused.
Drop feedback:
- bottom strip: the drop line sits centred between two features (it
used to touch the previous one) and shows at the first position
(it used to be drawn outside the strip there)
- left panel: a drop slot below the marker when it is at the end
- no drop line when the marker is dragged back to where it is
- a dragged feature is outlined in the drop line's green
- grab cursor on the marker, grabbing cursor while dragging the marker
or a feature
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4932
* icon_audit: compare the requested-icons manifest with CRLF checkouts ignored (#4932)
On Windows, Git for Windows checks out requested_icons.txt with CRLF line
endings (core.autocrlf; the repo has no .gitattributes), while the test
builds the manifest with \n, so the comparison fails although the lists
are the same. Read the file with \r\n normalised to \n, as #1931 did for
the icon tools. Regenerating still writes \n.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4966
* Agent chat: serve the bundled skills' reference through an MCP tool (#4966)
The bundled skills start with "read reference.md", but SolidCraft's own
chat turns the agent's file tools off, so its agent could not follow them.
The MCP server gains a read-only `skill_reference` tool that returns a
skill's documents from copies built into the binary. It takes a skill
name and a file name from a fixed list, never a path, and refuses an
unknown skill or file with the available ones. The skills point to it,
and docs/mcp.md and skills/README.md describe it.
#5187 gave the scene its own offscreen target with the app's sample count,
which skipped the OpenGL clamp to 1x from 6dd202d. The scene's count now
goes through effective_msaa_samples as well.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* test(mcp): reproduce whitespace bypass of request-size limit
Add a source-level regression that rejects a JSON-RPC request larger than
MAX_LINE even when most of it is whitespace, plus bounded-stdio framing
coverage and invalid UTF-8 recovery.
* fix(mcp): bound stdio line buffering and validate raw request size
Consume oversize JSON-RPC lines in bounded memory while keeping the
following request available. Reject whitespace-padded lines before
trimming, preventing a request-size cap bypass.
* docs(mcp): keep server framing comment idiomatic
* claim: #4585
* Sketch drag: a linkage held by redundant parallelograms still moves (#4585)
Cursor motion looks for the sketch's freedom in the null space of the
constraint Jacobian. Redundant constraints (a parallelogram held both
parallel and equal on each pair of sides) are dependent only at an exact
solution; after a correction to the solver's tolerance their rows differ
by about 1e-9, which the 1e-9 pivot tolerance counted as rank, so the
null space came out empty and a one-DoF linkage stayed locked. Use the
rank analysis' 1e-7, so the drag sees the freedom the sketch reports.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #4991
* Browser: Shift+click above the anchor selects the whole range (#4991)
The browser rebuilds its row order while drawing, and a Shift click looked
the anchor up in that half-built list, so an anchor further down was not
there yet and only the clicked row was selected. Resolve the range against
the previous frame's complete row order instead.
* claim: #5005
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Timeline: delete the selected features, Shift+click ranges, delete after the history marker (#5005)
Delete in a feature's menu deletes the selected features when the
right-clicked one is among them ("Delete 3 Selected Features").
Shift+click selects the range from the last feature clicked (or the last
selected one), in the order taken before drawing. The menu of a feature
after the history marker deletes every feature after the marker, with
their number in the label. All of it goes through delete_features: one
timeline.delete and one undo step. The confirmation no longer lists
features that are themselves being deleted.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5015
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Assistant: add Anthropic-compatible API agents (#5015)
The Add agent wizard now offers Anthropic-compatible endpoints beside OpenAI-compatible ones. The API backend speaks the Anthropic Messages protocol with the same MCP bridge, tool gate, limits, undo and keychain rules: x-api-key and anthropic-version, a top-level system note, input_schema tools, content-block answers and one user tool_result message per round. Test sends a short message because compatible endpoints (Qwen's) need not list models.
Co-Authored-By: Claude Code <noreply@anthropic.com>
---------
Co-authored-by: Claude Code <noreply@anthropic.com>
* claim: #5021
* Opening a design: range-check integer-valued coordinates too (#5021)
The file-level number check skips every integer-valued number so that ids
and counters may be large, which let a coordinate such as
9223372036854775808 through. The typed design is now checked as well:
float fields (coordinates, sizes) are serialized as floats and must be
within the same +-1e9 limit, while integer fields stay integers and are
not limited.
* claim: #5022
* Plastic features: refuse sizes under 0.01 mm (#5022)
A snap fit with a near-zero thickness (1e-9 mm) got past the size checks,
which only asked for a positive number, and panicked in the kernel's
intersection code. The boss, lip, groove and rest had the same gap.
Every length of a plastic feature that becomes geometry (snap fit length,
thickness, width, catch_depth and catch_length; boss diameter, height,
hole_diameter, hole_depth and rib sizes; lip and groove width and height;
rest width, length, height and thickness) must now be at least 0.01 mm.
The check sits where the feature is evaluated, so a parameter changed later
is refused the same way, and the message names the parameter.
* claim: #5023
* Extrude and loft: refuse repeated profile indexes (#5023)
`profiles: [0, 0, ...]` with 10 000 copies built 10 000 identical bodies in
about 35 s, and a loft section listing one profile 10 000 times took 25 s
before it was refused.
A list of profile indexes now refuses a repeated index while it is parsed
("profile 0 is listed twice"), as offset association ids already were.
Before the profiles are resolved, a selection naming more profiles than its
sketch has, an index the sketch doesn't have, or two points in the same
profile is refused too. A design file that repeats an index still opens,
using the profile once.
Catch up with en.json: the new Preferences pages, assistant settings,
3D mouse axis assignment, bug report form, history names and the tool
help texts. pl now covers every message of en.json.
Coincident with two circles (or arcs) picked was refused: "two curves must both
be straight lines; otherwise choose a point and a curve". It now lays the second
pick on the first: Concentric and Equal together, in one step, with the first
staying put. A pair already concentric or equal gets only the other half; one
already lying on the other is refused as nothing to add.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* A renamed body keeps its number (#4957)
The design's body counter starts over at every evaluation, so a body
renamed after a join (the join changed it, so the rename cannot be
followed without evaluating the design again) gave its number back: the
next body took the name the rename freed, and a script that names a body
found another one.
The design now saves the names a rename freed, and the counter skips
them.
* A renamed body keeps its number (#4957)
The design's body counter starts over at every evaluation, so a body
renamed after a join (the join changed it, so the rename cannot be
followed without evaluating the design again) gave its number back: the
next body took the name the rename freed, and a script that names a body
found another one.
The design now saves the names a rename freed, and the counter skips
them.
* Format
A favourite is listed twice, under Favourites and under User. Double-clicking its name opened a
name editor in both rows, under one widget id; the two took each other's focus, one lost it at
once, and the rename closed with the name unchanged. The rename now opens only in the table it
was started in, under an id that includes that table.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5087
* view.apply: keep the camera's roll, and refuse an up axis (#5087)
An explicit camera was read as a named view, which has no roll or up, so
roll was silently reset to 0 and up was ignored. Read roll as a finite
number in radians, clamped to ±π, and refuse up with a message saying the
up axis comes from the design settings.
A docked dialog is placed from the previous frame's drag offset, so a fast
drag can leave the pointer outside the dialog, over the model. The viewport
then treated the pointer as hovering the model and highlighted faces and
edges under it.
drag_dock now marks the frame in egui temp data while its title band is
dragged, and navigation::handle treats the pointer as not hovering the view
while that mark is current and a button is held. This covers every dialog
docked through drag_dock.
NVIDIA's EGL answers wgpu's request for a GL 3.3 context with exactly 3.3
(GLSL 3.30) yet lists GL_ARB_compute_shader. wgpu-hal then reports compute
shaders, wgpu-core builds its indirect-call validation pipeline, naga cannot
emit compute GLSL below 4.30, and request_device fails as Device(Lost).
The test draws nothing indirect, so it opts out of that validation.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5049
* GPU tessellation: an empty plan no longer fails buffer validation (#5049)
Empty tessellation inputs were padded to 16 bytes, smaller than the
shader's 32-byte vertex and 48-byte surface records, so wgpu refused the
dispatch. Skip the dispatch when there is nothing to evaluate, and pad
each empty buffer to one whole element of what the shader reads.
The component's sketches, features, bodies, subcomponents, the joints among
them and the parameters they use move to a new .solidcraft design, keeping
their ids and names, and the component becomes linked to it as Insert Design
makes it. The new design must build the same bodies and the assembly must
look the same afterwards, else nothing changes and no file is written. Joint
faces on the component are found again on the linked bodies, and components
that build the same bodies can become placements of the link.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A STEP file inserted into a new design sits in its own component under the
empty root, and a sketch on its face must join that part going out and cut it
going in rather than make a separate body beside it. With a distance, main
already resolves this; Through All with no distance gave `auto` no reach to
probe, so it fell back to New Body and cut nothing. Through All now reaches
past every body in its direction, and `auto` reads it as a cut (#1234).
Co-authored-by: Steven <steven2021067613@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The 14 new built-in themes still listed four logo colours, so none of them read after #5465 and main's theme test failed.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5461
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* App icon: draw the borderless mark everywhere (#5461)
app_icon.rs drew the older mark with white lines between the faces and a
white outline around it, so the Windows, Linux, web and README icons did
not match the macOS icon, the social card or the title bar. It now draws
the borderless mark from the faces and gradients of the macOS artwork
(03-solid-blue-cutout.svg): five faces with transparent gaps between them.
The filler fills even-odd spans so the concave faces work.
The Linux hicolor PNGs and both Windows .ico files are regenerated with
`cargo xtask app-icons`. The README logo, the web app's icons and every
package pick them up from there.
The logo colour token's fourth entry (the white edge) is gone: nothing
draws it any more.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The agent cursor only acted out engine commands: pointer input sent over the control channel
reached the app as synthetic events without moving it, so an agent's clicks, right-clicks and
drags were invisible to a person watching. While the cursor is shown, each synthetic pointer event
now moves it to where the input lands, a press leaves a ripple, and a drag is followed step by
step. A request being acted out keeps the pointer; with the cursor hidden nothing changes.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #5172
* Extrude without profiles keeps nested loops as holes (#5172)
With no `profiles`, solid.extrude took every region of the sketch, so
circles drawn inside a plate outline were extruded as well and filled the
plate's holes. The default now applies even-odd nesting to the drawn
profiles: a loop inside an outline is its hole, a loop inside that hole is
material again, and separate outlines all extrude. Sketches without nesting
keep the `all` selection, and explicit `profiles` (including "all") are
unchanged. Saved designs store their selection, so they recompute as before.
* claim: #5174
* component.create refuses parameters it does not take (#5174)
A call with `bodies` (or a misspelled key) used to succeed and make an empty
component. Unknown keys are now refused before anything changes, the error
lists the accepted parameters, and `bodies` points to component.move_bodies
and component.from_bodies.
* claim: #5175
* Joints: a joint origin's name given as a face says to pass it as the origin (#5175)
When the name in `face` or `circle` matches no face but is a joint origin,
the error now says so instead of "no face is named".
* Viewport: show the picture it drew last when nothing changed (#4820)
A frame redrew the whole scene whenever egui asked for a frame, even when
the camera, scene, highlight and overlays were identical: pointer moves over
empty space and UI animations over a still view paid for the full scene.
The renderer now hashes everything the picture depends on (scene keys,
camera, size, section, images, analysis, accessibility map, occlusion) and
blits the last offscreen picture when the hash matches. The first still
frame filters a supersampled picture down once and later frames copy it as
it is. A frame with the same content costs a blit instead of a scene draw.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Viewport: draw the scene offscreen, into a 1x window pass (#4820)
The scene keeps the app's depth, stencil and MSAA, but egui's window pass now
has one sample and no depth. Whenever the two targets differ, the scene
(product previews too) is drawn offscreen and blitted, so the UI pass no
longer clears and resolves a multisampled window.
Over 8 s of pointer moves on the 450E rotor head, the app's GPU intervals fell
from 1325 ms to 1006 ms and egui_render from 892 ms to 509 ms; idle stays at
0%. A GPU test checks the split path's picture matches the direct one.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* 3D mouse: a repeated report at rest asks for no frame (#4820)
A device may repeat its latest report while the cap rests. Publishing it
again asked egui for a frame although nothing changed. Only a report that
changes the reading now asks for one; while the cap moves, the view asks for
the next frame itself.
Co-Authored-By: Claude Code <noreply@anthropic.com>
---------
Co-authored-by: Claude Code <noreply@anthropic.com>
The File view's recent row already shows its path on hover; Home added
a second hover text with the raw path, so both appeared.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Add the format registry: one table of formats, extensions and options in crates/io.
The hosts' Open-dialog filters and the web accept list derive from it, and
registry tests pin it to the path predicates and the command docs.
* Fix format-registry tests that compared FormatSpec without Eq.
assert_eq on Option<&FormatSpec> does not compile; the none cases
only need is_none, matching the other rows that compare ids.
* claim: #5230
* Viewport: draw the ground shadow on the direct path (#5230)
The ground shadow went into the model's translucent triangles, so every
frame with a body counted as see-through and took the order-independent
transparency path. Give it its own batch, drawn alpha-blended after the
opaque geometry as before, so only truly see-through bodies use that path.
* claim: #5248
* feat(motion): regenerate kept traces when source inputs change
* feat(motion): add a parametric Chebyshev trace demo
* perf(joints): reuse face measurements across motion samples
Face-centred joint origins called Body::faces on every solve, bypassing the
reference-mesh reuse used for rigid placements. Share face measurements with
unchanged model geometry at the existing tolerance, while guarding replaced
bodies and applying offsets after the cached measurements.
Cover repeated named/nearest-face solves, shape edits, retained snapshots and
direct replacement with deterministic work counts. The 145-point Chebyshev
dimension and range updates retain identical sampled and interpolated curves.
* perf(naming): skip preparation for cached face names
Both face lookup paths spawned fresh workers over every body even when their name cells were ready. Prepare only cold cells, with direct handling of zero or one cold body. Preserve the existing parallel naming path and all face selection semantics.
Deterministic regressions cover repeated picked and unique-name lookups, misses, partial warmup, renames, snapshots and moved copies. Release Chebyshev medians fall from 859 to 50 ms for a dimension edit and 871 to 34 ms for a sweep edit; all 145 samples and interpolated points remain identical. Full cargo xtask ci passes with RUST_TEST_THREADS=4 and the existing local Windows validation overlays.
* claim: #5287
* web-offline-test: wait only for the files the browser can ask for at once
The fail-fast step waited for every other file of build v2 to reach the
server. Chrome sends at most six requests to one server at a time, and
the page's failure cancels the download, with it the files still waiting
for a connection. A real build keeps nine files, so only six were ever
asked for and the Pages job failed with "only 6 of 8 slow files were
asked for" whenever the installing worker didn't happen to ask again.
Wait for as many as can be in flight (less the page's connection), and
give the stand-in app more files than that, so the default run covers it.
* claim: #5227
* Web: leave Ctrl+Tab and Ctrl+Shift+Tab to the browser (#5227)
eframe cancels the default action of every Tab keydown on its canvas
(egui moves focus with Tab), so browsers that let a page cancel their
shortcuts, such as Vivaldi, could not switch tabs while SolidCraft had
the keyboard. A capture listener on the window now stops Ctrl+Tab and
Ctrl+Shift+Tab before they reach the canvas, without cancelling them,
so the browser acts on them as usual. Plain Tab and Shift+Tab still
move between the app's fields. Desktop builds are unchanged.
web-api-test checks it in headless Chrome: Ctrl+Tab and Ctrl+Shift+Tab
keydowns are not cancelled, Tab still is.
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #4765
* Agent chat: explain the Flatpak sandbox when no agent is found (#4765)
In the Flatpak the sandbox has no home-folder or network access, so agents
installed through nvm or ~/.local/bin are never discovered and the chat only
said that no agent was found. With FLATPAK_ID set and nothing discovered, the
picker and Preferences > AI now say why and offer the exact
`flatpak override --user ai.storyteller.solidcraft ...` command with a Copy
button. The message is in every catalog, and docs/agent-chat.md gains a
Flatpak section listing what each permission is for. The manifest is
unchanged.
* claim: #5150
* Stand it up: don't offer it for SolidCraft's own STEP files (#5150)
STEP replies from file.insert_step and doc.open now carry the file's
originating system, and the window skips the stand-up offer when it is
exactly SolidCraft, as SolidCraft's STEP export writes it.
* Stand up Y-up STEP and IGES files from SolidWorks automatically
The STEP header (originating system, preprocessor version, FILE_DESCRIPTION) and the IGES
global section's native system ID say which program wrote a file. A SolidWorks header means
Y up: the import's component is stood up as it opens, as one undo step, and the message offers
Undo. SolidCraft's own files are left alone; an unknown exporter keeps the offer.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Ray <raykholo@users.noreply.github.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
The viewport's frame-rate readout asked for a new frame every frame, so with
it on an untouched window redrew as fast as the display allowed and kept the
GPU busy: 77% GPU at 60 fps on an M1 with a 5K display showing the sample,
against 0% with the readout off. It now asks for no frames of its own and
shows how often the view really draws; one frame a little over a second after
the frames stop shows the view idle at 0 FPS. The preference's note no longer
says the view redraws continuously.
The status toast also asked for a frame every half second for the six
seconds it shows; it now asks for one frame, when it is due to go.
* claim: #4172
* GPU: a lost device stops drawing instead of crashing (#4172)
After the driver resets the GPU (Windows TDR), the first queue write to meet the
lost device panicked inside wgpu-core (PendingWrites::activate unwrapping
DeviceError::Lost). Nothing watched the device for its loss.
- egui-wgpu: RenderState::device_lost (DeviceLost) is set by the device's lost
callback. The painter draws nothing on a lost device, and before a frame's
uploads it submits nothing (DeviceLost::probe), so a loss the driver reports
is found by that submission instead of the first write.
- Pictures (thumbnails of saves, ui.render) watch the same flag: on a lost
device the capturer refuses with GpuError::DeviceLost without touching it,
and the picture comes from the CPU rasterizer. Headless capturers watch
their own devices.
- The app, seeing the loss, closes keeping unsaved changes for recovery (as a
forced quit does) and starts again; the new start says what happened. A
second loss within ten minutes starts on the software renderer. The loss is
noted in the crash journal.
`Body::size` walked a mesh body's vertices on every call, and it is asked
for every frame. A mesh body's triangles never change after `mesh_body` or
`mesh_transformed` makes it, so both now measure it then (`TriMesh::span`)
and `size` returns it like a B-rep body's. `ModelBody::mesh` returns a mesh
it already has before working out the display tolerance.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* claim: #4107
* Keep bottom viewport overlays above the floating History card (#4107)
In the floating layout the History card covered the analysis legend and
the frame-rate readout (#3896). One bottom inset, computed from the
History card (zero when History is hidden or docked), now positions every
bottom-anchored viewport overlay: the surface-analysis and Continuity
legends, the frame-rate readout, the command-progress chip and the
navigation bar fallback.
On main, 120 label literals (95 distinct) can't be translated in any
locale: the English catalog doesn't reach them. Catalogue them in en,
ja, zh and de, translate indented row labels once for display and
measuring, look up the thread profile by its id, and give each
navigation preset's help, shown by the mouse mapping picker, a catalog
message.
A guard tokenizes the UI sources and checks every label literal of a
kit call, of context_menu::button and of tr_label/tr/SelInput against
the English catalog, resolving kit calls by their imports.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3567
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019Yz6dHZiXcgTgrxjVNC6hx
* Edit Offset Plane: keep a face base, refuse a plane on itself (#3567)
Reopening an Offset Plane built on a face looked the face up at the
plane's origin, which can lie off the face, so the Plane input came up
empty. Look it up at the point picked on the face. A redefine that
builds a feature on itself is refused with a plain message instead of
failing with "plane offsets nested too deeply".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019Yz6dHZiXcgTgrxjVNC6hx
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2874
* Sketch fillet and chamfer refuse an offset's corners (#2874)
An analytic offset's output ends are recomputed from its source curves on
every solve and held fixed there. Filleting an output corner moved those
ends to the tangent points, the next refresh pulled them back to the sharp
corner, and the new arcs closed into full circles with the sketch left
unsolved.
Fillet and chamfer now refuse a corner on an offset output before anything
changes (a batch fillet applies all of its corners or none) and say how to
get the rounded result: round the source corners before offsetting, or
break the offset association first, after which the same corners round
normally.
* claim: #2885
* Circular Pattern and Revolve: a construction axis can be the axis (#2885)
The dialogs send a picked axis by its name, but solid.pattern.circular
only knew X, Y and Z, and solid.revolve / surface.revolve read any other
name as a sketch line id. Picking a construction axis therefore failed.
- solid.pattern.circular takes a construction axis name, saves it as
`axis_feature` and turns about the axis as evaluated, so the copies
follow it when it is edited.
- AxisRef::Construction { name } for revolve and surface revolve,
resolved at evaluation and checked to lie in the sketch plane.
- construct.plane.at_angle accepts a construction axis name (its fixed
axis line is taken from the axis as it is now).
- Reopening these features shows the construction axis as the pick.
- Design format: pending/construction-axis-refs.rs.
* Construction axis references follow the axis in the feature's component frame (#4506)
- Revolve, Surface Revolve and Circular Pattern resolve a construction axis
where it is evaluated, mapped into the feature's component frame.
- A line of the revolved sketch with the same id as a construction axis
still wins, so existing scripts keep their meaning.
- Tests: the revolve keeps the reference through save and reopen and
follows an edited axis (60π to 80π mm³); perpendicular, offset and
profile-crossing axes are refused without changing the design.
* Joint drive moves parts without evaluating them
A joint drive (also the Drive dialog and motion steps) changes only occurrence transforms, yet it
re-evaluated nearly the whole timeline: every feature that might read another component's
geometry (a sketch with any link, even to its own axis) hashed every occurrence's transform into
its fingerprint, so any motion invalidated it and everything after it, and no new pose was ever
in the cache.
- Placement digest: a feature now hashes only where the geometry it actually reads lies relative
to its own frame (link sources, anchors, profiles and paths from other components, mirror
placements; To Object and face/edge-picked planes read every component, relatively). Moving
parts that keep those relative placements leaves every result in place: nothing is evaluated
or re-meshed. In-context features whose referenced part really moved still rebuild.
- Evaluated ahead, applied once: `Ahead::evaluate` returns what it evaluated and
`Session::prime` lets the window-thread pass take it (history commands and dialog OK), so an
off-thread command no longer costs twice on designs past the cache's budget.
- Contact meshes are kept per body geometry across drives.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Joint drive: booleans read only their input bodies' placements
A join, cut or intersect with named targets, and a Combine, read every
placement, so a joined part transferred to a component re-ran its boolean
at each new joint pose. Each now hashes only the placements of the bodies
it actually takes as inputs, resolved in its timeline input state (before
any later BodyTransfer), relative to the frame that evaluates it; Combine
reads its tools relative to its target. Unresolved and legacy component or
path reads keep the conservative dependency. 24 new joint_pose_ tests
cover new angles, transfers, targets, save/reopen, undo and real relative
motion (900 -> 800 mm3).
Supplement offered on #3397 by takuto-NA.
Co-Authored-By: takuto-NA <183465875+takuto-NA@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Creating a component recomputes nothing
A new, empty component changes nothing a feature reads, yet it changed
the placement fingerprint of every targeted boolean, Combine and To
Object feature, so `component.create` evaluated the timeline again:
- the first component of a design: `placement_digest` returned 0 while
the design had no placements and a nonzero value once it had one;
- To / From Object and legacy face- or edge-picked planes hashed every
component, not just those holding the bodies, sketches and
construction geometry the feature can reach.
Co-Authored-By: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Test: creating a component after a legacy face sketch recomputes nothing
A face extruded after its construction certificate is gone (a patterned
cut, a pad joined flush with it, then one more cut) is sketched on a
legacy face plane, which reads every placed body. A new, empty component
must not change what it reads either (reported by Tyler Chiriboga).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: takuto-NA <183465875+takuto-NA@users.noreply.github.com>
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
* Viewport: levels of detail for large meshes (#4724)
A large instanced mesh (20 k triangles or more) gets coarser copies, made
once on a worker (`mesh_lod`): a quarter, a sixteenth and a sixty-fourth of
its triangles, by quadric edge collapse (`geom::simplify`, after Garland and
Heckbert 1997; no fold-overs, no sheets joined, open boundaries kept), each
with its measured deviation from the full mesh. Each frame, as placements
are culled, each is drawn from the coarsest copy whose deviation stays
under 0.75 px where it stands (`placement_scale`, `level_for`). Until the
copies are made the mesh is drawn in full; the scene is built again when
they are. Display only: picking, measuring and exports keep the reference
mesh. Outside the kernel.
25 placements of a 2 M-triangle STL, 3600 x 2190 px, 4x MSAA, M2 Max
(`instanced_scene_gpu_benchmark`): 20.2 -> 1.2 ms a frame with all in
view, 21.9 -> 2.9 ms close up. The copies take 10 s to make, on a worker.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: see-through and faded parts drawn as instances (#4888)
See-through (an appearance's opacity) and faded (context around an active
component) bodies in one colour are drawn as instances like opaque ones,
their alpha in each placement's colour: one copy of the triangles, levels
of detail and culling, no per-placement copies built on the window thread.
Their placements are ordered back to front each frame (not their
triangles one by one) for devices that sort see-through surfaces.
The GPU benchmark gains `SOLIDCRAFT_BENCH_OPACITY` and the CPU share of
each frame, and its pictures each placement's own triangles too.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: order-independent transparency for see-through surfaces (#4734)
The model's see-through surfaces (glass, faded context; instanced or not)
were sorted back to front triangle by triangle on the window thread
whenever the view turned. They are now drawn with weighted blended order-
independent transparency (McGuire and Bavoil 2013): when there are any,
the frame is drawn into the renderer's own textures (as supersampling
does), the see-through surfaces gathered into a weighted colour and a
revealage target against the frame's depth, laid over the frame in one
full-screen pass, then the overlays drawn over them. Nothing is sorted.
A device that cannot blend into float targets keeps the sorted way.
25 placements of a 2 M-triangle STL at opacity 0.5, M2 Max, 3600 x 2190 px:
about 10 s a frame sorted, 21.6 ms (17.2 ms instanced, 2.8 ms with levels
of detail).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* claim: #2614 (3D mouse: find devices on Windows without asking every HID device for its strings)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse: find devices on Windows without asking every HID device for its strings (#2614)
On Windows the 3D mouse reader rescanned every two seconds while no 3D
mouse was plugged in, with hidapi's refresh_devices(). hidapi's Windows
enumeration opens every HID interface and reads its serial number,
manufacturer and product strings: on USB, three control requests to
every keyboard, mouse and touchscreen, forever, on a machine without a
3D mouse. That fits the report of an idle machine whose USB keyboard
drops off and returns and whose cursor jumps.
Windows now asks Plug and Play which present HID interfaces are
multi-axis controllers (hardware ID HID_DEVICE_UP:0001_U:0008), which
sends nothing to any device, and opens only those with open_path().
hidapi's own discovery is turned off before the reader makes its one
context, so making it touches no device either. macOS keeps hidapi's
enumeration (IOKit's registry); Linux is unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* AGENTS.md: list the Plug and Play scan among the files allowed unsafe code (#2614)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Finish Sketch, double Esc (including a blank sketch discarded), and Cancel no longer
restore the pre-sketch view. Entering a sketch still turns the view to face the plane.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #2027
* Redesign the home page as the File view (#2027)
* Match the File view with the accepted component design (#2027)
* Restore the Start subtitle and simplify the recovery icon (#2027)
* Integrate File view with main's shared UI containers
Every edit clones the document for undo, and `MeshData`'s positions and
triangles (plain `Vec`s) were copied with it: ~165 MB a command for a 2 M-
triangle STL. They are now `Arc<[f64]>` and `Arc<[u32]>`, shared by clones;
the few places that rewrite them build new arrays. Saved designs are
unchanged (an `Arc<[T]>` serializes as the array).
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The pick buffer drew the model's surfaces and edge ids for every still
view to narrow which edges a pick tests. Measured against the BVHs alone
(release, M2 Max, 1400 x 900 view, 600 cursor positions), a pick costs
2-51 us through the BVHs and the tile saved 0-4 us of it, while each pass
cost 0.7-12.5 ms of GPU time per still view, plus gathering its scene on
the window thread, read-backs a frame or two late and repaints to wait for
them (and, before, a copy of every triangle, which made it refuse models
past 6 M triangles). Picks are unchanged: the BVHs give the same answers.
`pick_timing_on_curved_models` gains eight layers of bodies seen from above
(the pick buffer's best case) and `SOLIDCRAFT_BENCH_MESH` placements.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The highlight's key hashed the hovered item with its point under the
pointer, so every move across a face built and uploaded its highlight
again: for an imported mesh (one face) all its triangles, every frame.
The key now leaves out a face's and a plane's point, which the highlight
does not draw; an edge's middle stays in (it picks the part lit).
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
`SharedTris::new` used the standard SipHash hasher and collected each
triangle's corners into a Vec: 451 ms for a 2 M-triangle STL on the window
thread. A quick hasher for the coordinates' bits and a fixed array make it
278 ms, and the background mesher that meshes an opened or inserted
file's bodies now welds their large meshes too (`warm_shared_tris`), so the
first scene that shows them takes the shared vertices ready-made.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
`toolbar_custom::panel_lists` is asked for every frame and matched every
registered command against every section by string each time (~337 us a
call for MODEL, several calls a frame). The registry and the ribbon layout
never change while the app runs, so each tab's lists are kept: 0.2 us a call.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* claim: #4616
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: draw a part placed many times from one copy of its triangles (#4616)
A part placed by many occurrences was drawn, framed and picked from a moved
copy of its mesh per placement: on the CPU (the moved mesh and its scene
bytes) and on the GPU (its vertices uploaded once per placement).
- The model scene draws bodies that share a mesh as GPU instances: the
shared triangles once, a rigid move and colour per placement
(`InstancedMesh`, `vs_tri_inst`), in the solid, section and AO passes.
`GpuScene::merged` expands them, so scene tests see world triangles.
- `ModelBody::mesh_in_place`, `plan_mesh_in_place_at`, `mesh_bounds` and
`display_identity` give the shared mesh, its placement, the exact box and
an identity without making the moved copy.
- Display refinement's body ids, Fit/Home bounds, the viewport's body boxes,
the background mesher and picking use them (picking moves the ray into
the part's frame and shares one pick index).
25 placements of a 2 M-triangle STL on an M2 Max: orbit 45 -> 37 ms a
frame, footprint 41 -> 16 GB, 24 copies 39 -> 18 s.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: draw large meshes from shared, indexed vertices (#4637)
An instanced mesh's vertices are shared between its triangles and drawn
indexed (`SharedTris`). A flat-faceted mesh (an STL's facets) shares them by
position alone and lights each fragment with its triangle's own plane
(`vs_tri_inst_flat`, `fs_*_flat`), taken from the fragment's position
relative to the eye so it stays exact at any zoom; its vertex outputs are
kept few, since a tiled GPU stores every vertex's outputs. A body alone is
drawn this way from 50 k triangles. A mesh too large for one buffer falls
back to unshared vertices.
`instanced_scene_gpu_benchmark` (manual, ignored): 25 placements of a 2 M-
triangle STL at 3600 x 2190 px, 4x MSAA, M2 Max: 21.3 ms a frame for the
placements' own triangles, 16.1 ms shared and indexed.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Viewport: leave out placements of an instanced mesh that are out of view (#4646)
Each frame `Batches::cull` keeps the placements whose box may be seen
(`placed_box_in_view`: all eight corners past one side of the view in clip
space, so it holds behind the eye too) and writes them again only when
that set changes. The manual benchmark gains `SOLIDCRAFT_BENCH_CLOSE=1`:
close up on the middle of 25 placements of a 2 M-triangle STL (13 in
view), 3600 x 2190 px, 4x MSAA, M2 Max: 28.2 ms a frame for the
placements' own triangles, 17.2 ms instanced and culled.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The slow steps waited on the clock (6 s sleeps, a 1 s timer, a 30 s cap), so on a busy
runner a launch could outlast them and time out at 240 s. The server now holds a response
until the test releases it, and the test waits on events:
- overlapping downloads: build v5's own files are held; files both builds share (index.html,
names no build changes) are served, since the browser queues a request behind an in-flight
one for the same URL. The held files are released once v6 has started.
- fail-fast step: the other files wait until the test has seen them asked for.
- update in another tab: the other tab updates once the launch is waiting for build.json, and
the browser's own sw.js check waits for the same moment.
- hung requests no longer give up after 20 s.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Registers them after the Gruvbox set in the Theme list, adds attribution rows,
and keeps the heading-band ordering check passing by nudging heading/subheading
(and, for High Contrast, dialog) backgrounds.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* Vertical history strip: left of the browser, compact rows, hollow-circle marker, collapsed rail hover
* Vertical history strip: seam handle to expand and collapse names, drag to resize with snap
* Vertical history strip: no empty gap below the transport
* Show the history seam arrow only on hover (#2028)
* claim: #2021
* Redesign the Assistant chat panel with shared UI components
* Show Assistant in the shared floating panel shell
* Extend floating Assistant to viewport height and shift navigation
* Dock Assistant before history and match docking preview bounds
* Align Assistant top margin with floating shell cards
* Extract a shared panel drag grip
* Rebuild responsive history sizing and three-way docking
* Share browser and History collapse hover styling
* Show the History title as soon as its text fits
* Animate stacked browser and History expansion and collapse
* Make panel folding responsive and independent of frame delays
* Let folding panel contents shrink continuously to header height
* Use four explicit History transport actions in every layout
* Resize the shared Browser and History column from either card
* claim: shared compact-mode containers (#2029)
* Centralize layout-aware UI containers and enforce their use
* Keep docked Assistant at the outer edge beside tool sheets
* Use shared surface policy for Assistant tool space
* Let tool sheets drag over shell panels while avoiding Assistant
* Update Parameters drag bounds assertion for shell overlap
* Keep dragged tool sheets below ribbon and workspace selector
* Enforce universal UI containers in mandatory repository checks
* Remove legacy report windows and require shared window composition
* Remove retired UI styling and require shared controls
* claim: compact layout preference (#2029)
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Chamfer: a cutter that would reach into a nearby wall is not used
A straight chamfer's cutter reaches several times the chamfer's size out
from its edge, and further on a larger part. On a pin standing in a socket,
that far side lands inside the socket's walls, and the cut took bites out
of them. A cutter whose far side is inside the part is now refused, as one
whose ends are, and the chamfer is built locally instead.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Robustness goldens: seed 9 re-recorded for the chamfer cutter check
A chamfer in seed 9 was refused because its result changed more of the
part than a chamfer can. With its cutter refused, it now builds locally.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Navigation: opt-in free orbit and shared Normal to Selection
Preserve the complete reviewed feature, shared camera setting propagation,
rigid pivot behavior and focus/drag regressions from PR #1912.
Consolidate the prior main-integration commits before rebasing, so their
merge-resolution fixes remain in the linear feature history.
* docs: move free orbit help beside viewport controls
* Merge main into fix/solidworks-free-orbit
Camera animation is gone on main, and with it the orbit snap spring: the Shift
snap keeps its unsnapped orbit in orbit_free, which the free-orbit checks now read.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #3708
* Export Current View as DXF includes shown sketches (#3708)
The view export refused a design with only a sketch and left shown
sketches out when bodies were present. It now draws the sketches the
viewport shows (profile curves only: no construction geometry, points,
dimensions or constraints) on a SKETCH layer, one entity per curve.
Finished sketches are hidden behind bodies through the body-edge
occlusion; the active sketch is drawn on top, as in the viewport.
Sketch points use the same projection, budgets and cancellation.
The export is refused only when there is neither a visible body nor a
shown sketch curve; the dialog and the command share that check
(cmd::view_dxf_exportable).
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* Sketch Dimension: `name = value` defines a user parameter (#2916)
In a dimension's value (the D tool's box, and `sketch.dimension`'s `value`
from scripts, the CLI, MCP and the control channel), `name = expression`
adds the user parameter `name` with that expression and dimensions with
`name`: typing `width = 40` on a line gives a 40 mm user parameter `width`
and a dimension whose expression is `width`. A bare number is read in the
design's unit, as a bare dimension value is; an angle's parameter is in
degrees. The expression may be anything the parameter editor takes.
It is one command: one undo removes the parameter and the dimension
together, and a dimension the sketch refuses leaves no parameter behind.
A taken name is refused with a hint (type the name to use it, or change
it in Parameters), a reserved name with the parameter editor's message,
and a left side that is not a name (`2 = 3`) still fails to parse as
before. A driven dimension defines nothing. The reply carries `defined`,
and the status line names the parameter.
The splitter lives in the doc crate (`expr::definition`) so other value
inputs can take the same form later; editing an existing dimension, the
inline shape boxes and feature dialogs are follow-ups.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Sketch Dimension: say why a name is refused, in the command's own words (#2916)
The dimension editor showed a refused definition raw, in the box and the
status line: "invalid parameters for `sketch.dimension`: value: `mm` is a
unit, constant or function name". It now reads "Sketch Dimension: `mm` is
a unit name, so it cannot be a parameter name": the editor passes the
command's errors through `friendly_error` like every other command, and
drops the `value:` field tag, since the box is the value. The taken-name
hint reads the same way: "Sketch Dimension: there is already a parameter
`w`: type `w` to use it, or change it in Parameters".
A reserved name says what kind of word it is and what that means
(`param_name_problem`, shared with the Parameters dialog): a unit name, a
function name, or a constant, "so it cannot be a parameter name";
`expr::reserved_kind` tells them apart.
`friendly_error` translates the command label with the interface language
(the engine's sentence stays English), so the prefix reads "Wymiar szkicu:"
in Polish.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* Sketch Dimension test: keep the third line off the ribbon (#2916)
On current `main` the ribbon floats over the top of the canvas, and the
test's third line at sketch y = 50 projected under it: the click opened
Parameters instead of picking the line. The line now goes below the
first one.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #3085
* Extrude preview: a cut from an empty root is not refused (#3085)
A cut or intersect sketched where the active component has no body of
its own (the root of an assembly, or of a design whose part was
imported into a component) takes part in every body (#1234): the
command makes the cut. The Extrude preview checked the targets itself,
looked only in the active component, and showed "there is no body to
cut or intersect" beside a tool going through the part. Pressing OK
still cut, so the dialog said one thing and the model another.
The rule now lives in one place, Session::cut_takes_every_body, which
both the command and the preview ask, so they cannot disagree again.
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #3313
* fix(ui): the Rename box keeps the new name when it loses focus (#3313)
Only Enter committed a rename: Tab (and anything else that moved focus away)
dropped the typed name. Leaving the box now keeps it, as a click elsewhere
already did; Escape still cancels.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #2581
* fix(view): dialogs picking a plane show the origin planes (#2581)
Create Sketch showed the origin planes while it picked its plane; now any dialog
whose active input takes origin planes does (Mirror's plane, Offset Plane,
Midplane, the construct tools), so the plane can be picked in the view.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Vertical history strip: left of the browser, compact rows, hollow-circle marker, collapsed rail hover
* Vertical history strip: seam handle to expand and collapse names, drag to resize with snap
* Vertical history strip: no empty gap below the transport
* Show the history seam arrow only on hover (#2028)
* Extract a shared panel drag grip
* Rebuild responsive history sizing and three-way docking
* Share browser and History collapse hover styling
* Show the History title as soon as its text fits
* Animate stacked browser and History expansion and collapse
* Make panel folding responsive and independent of frame delays
* Let folding panel contents shrink continuously to header height
* Use four explicit History transport actions in every layout
* Resize the shared Browser and History column from either card
An offset's output curves have their ends recomputed from the source
curves on every solve. Trimming a piece off one reported success and left
the curve whole, because the refresh pulled the new end straight back.
Extending one was refused as "connected to other geometry", which does not
say why. Both now refuse with the reason and the way out: edit the source
curves, or break the offset association first. Trimming away a whole
offset curve, which removes it from its association, still works.
The fuzzer chamfers randomly picked edges and expected every chamfer to
shrink the body, so a chamfer of a concave edge, which fills the corner,
was reported as a wrong volume: 53 of the 6,000 designs of a run (seed
7100465: a box, a shell, a chamfer of an inside edge). Fillets already
expect any volume change for the same reason; chamfers now do too.
The regression test checks both directions on a shelled 20 mm box (an
inside-edge chamfer of 1 mm adds 8 mm³, an outside one takes 10 mm³
away) and that seed 7100465 is no longer a volume finding.
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3099
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Centre circle: a second click snapped to a point keeps the circle on it (#3099)
The tool sent the centre's snap reference but only a radius for the
second click, so a circle drawn from a line's midpoint through its end
was left free to change size. sketch.circle.center takes through: a point
reference, kept on the circle (PointOnCurve), and the tool passes it when
its second click snapped to a point.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3257
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Clicking a parameter suggestion inserts it (#3257)
A press on a suggestion took the focus from the expression box before
the list could see the click: the list stopped drawing, the click never
landed, and the box gave up what was typed. The rows are kept from the
frame before; a press on one picks it and gives the box its focus back
(again on the release, which the box takes as a click away). The rows
are no longer focusable. The completion tests' frames now drop their
texture updates on purpose, so they run where egui checks that.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2919
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Parameters: a new parameter's expression starts blank, and Tab selects a box of the add row (#2919)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Parameters tests: the add row's blank expression in the manage dialog tests (#2919)
The two inventory tests typed a name and pressed Enter, counting on the
10 mm the expression used to start with; they now type it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2440
* feat: choose ISO metric tapped-hole sizes and derive the drill diameter
* fix: label the tap drill and hide the unused diameter editor
* test: cover tapped hole catalogue and drill sizing
* fix: omit coarse metric pitch from generated designations
* Keep custom tapped-hole nominal diameter and pitch editable
Preserve Custom as dialog state, derive the existing metric designation from numeric inputs, and refuse invalid drafts without applying the last valid size. Retain manual text editing, physical tap-drill geometry and saved-design compatibility.
* fix(i18n): register custom tapped hole labels in source catalogs
* Adapt tapped-hole Custom fields to the current form kit
* fix: preserve tapped designation focus with dynamic custom rows
* solid.mirror: copies join their source's component; follow the saved placement
Three follow-ups to mirroring bodies of placed components:
- A copy joins the component of the body it mirrors (or, when that
component is placed more than once and a copy there would show at
every placement, the innermost component around it placed once), so
checks that group bodies by component see it beside its source.
- The saved placement path, not the body's name, says which component
a mirrored body is in: inserting the same file again reuses the body
names, and the copies of the first insert vanished.
- The mirror is evaluated again only when its own placements or its
component's frame change, not on every occurrence move: each later
command paid for one more mirror evaluation.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* solid.mirror: a saved placement must still lead to the body's component
A stale or edited placement path that leads to another component's
occurrence mirrored the body silently where that other component is
placed. The mirror now reports that the placement no longer exists.
The body's component comes from the evaluated body where a transfer
recorded it, else from the component its name is recorded in or its
feature's. When a later insert of the same file has taken over the name
(the recorded component lies outside the components the body's own
feature made), the placement has to lie inside those components instead,
so mirrors of the first insert keep working.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* solid.mirror: the copies' component change is a pending format change
v26 shipped in 0.15.0, so its mirror-placements note stays as released;
the change of where a Mirror with placements puts its copies is noted in
a new pending file instead.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* 3D mouse calibration: find each axis from a guided recording (#1463)
A recording of the cap at rest and moved along each axis both ways gives which raw axis is which,
which way it runs, where it rests, how much it wanders at rest and how far it goes each way.
Hands aren't exact: on our SpaceMouse Pro, sliding the cap right moved Ry further than X. So each
raw axis is picked by the one-to-one assignment that best explains all six motions together, each
raw axis scored against its own range so a wider one doesn't win through cross-talk. Each
direction is scaled on its own, so a device that goes further one way still reaches full scale
both ways; a direction under a twelfth of the furthest axis counts as not having responded.
A small state machine walks the user through it and moves on by itself once the cap has been held
out for 0.4 s in one go and let go. "Held out" is the least the solver accepts, twice an axis's own
rest wander, so whatever the solver would accept can be recorded and a noisy axis doesn't hold up a
quiet one.
A calibration also carries a dead zone the user sets, a share of full deflection taken off the
bottom of the travel, with full deflection still reading full scale.
Serde support is behind an optional feature, so the crate still needs nothing by default.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse calibration in Preferences, saved per device (#1463)
Preferences > Navigation gets a 3D mouse calibration row: whether the device being read is
calibrated, Calibrate... and Forget calibration. The window asks for each motion in turn and
saves what it found under the device (hid, web, spacenavd or 3dxware, with vendor and product ids), so
switching devices or turning spacenavd on doesn't apply the wrong mapping. A calibration is saved
for the device it started with, and ends if another one is read instead.
navigate_frame puts the reading through the device's calibration before the driver mapping, so
the per-axis Invert and speed and everything after them are unchanged. While the window is open
the reading goes to it and the view stays put; it closes with Preferences. Restore Defaults on
the Navigation page keeps calibrations, since they describe the device.
The web app hands over the gamepad's readings as they are, so calibration sees where the cap
rests; navigate_frame now leaves a reading inside the dead zone alone after calibration, so a
cap resting just off zero still doesn't keep redrawing. Whether the cap is moving is asked of the
calibrated reading there, in the web page's idle waker and in both wheel filters. Those run before
the frame, so they keep the last frame's calibration with its device and use it only for that
device. On the desktop, an axis with nothing fresh (timed out, or not reported since the device
connected) reads as the device's calibrated rest, so a device resting off zero doesn't drift.
A saved calibration that can't be read or used (a hand edit) is dropped on its own when the
preferences load, with a log line.
The window shows the device's raw readings live and has a dead zone slider, a share of full
travel on top of the measured noise; it's saved with the calibration and adjustable in Preferences
afterwards.
New strings are in all 14 catalogs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse: choose which motion drives each axis (#1463)
After calibrating, which motion of the cap should drive which view axis is partly a matter of
taste: rolling and twisting swapped, say, or Y and Z. Each axis row in Preferences > Navigation
now has a "driven by" choice next to Invert and speed. Picking a motion another axis uses swaps
the two, so each motion always drives exactly one axis, and Reset axes puts them back.
The choice applies after the device's calibration and before the driver mapping, so Invert and
speed stay with the view axis rather than moving with the motion. It's a preference, not part
of a device's calibration, and stays in force when recalibrating; the calibration window says
so while one is set. A saved assignment that doesn't drive each axis once loads as the default.
New strings are in all 14 catalogs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse: axis choice review follow-ups (#1463)
- An axis's speed, like its Invert, stays with the view axis when a motion is reassigned; tested.
- Calibrating records the device's own reading, before the axis choice, so a saved swap isn't
baked into a new calibration and then applied twice; tested through a whole calibration.
- A settings file without the axis choice, or with one of the wrong length, loads with the
default and keeps every other setting; tested.
- Each axis row says "driven by" in sight before its choice, which otherwise only explained
itself on hover (no help on a touchscreen or the web).
- "Reset axes" is now "Reset axis assignment", so it doesn't read as forgetting the
calibration just below it; the "driven by" tip reads more clearly in English.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse: name axis rows by what they do, and motions without a device in mind (#1463)
The axis rows and the "driven by" choices showed the same names, "Push left / right" and so
on, which read as one thing mapped to itself, and they described a cap that springs back to
centre, which a trackball or a knob that stays put isn't.
Each row is now named for what it does to the view: Pan left / right, Zoom, Pan up / down,
Orbit up / down, Orbit left / right or Roll, or Off. The name comes from the driver mapping
navigation used last frame, so it stays right where the mapping swaps axes (object mode swaps
the last two). Before navigation has run there's no mapping to read, and a row is named for
its own motion instead of a guessed effect. The choices are the device's calibrated motions:
Translation X / Y / Z and Rotation X / Y / Z, under a header row, and the calibration window
names a motion that barely moved the same way. This renames #1814's existing rows, which had
no translations; every new name is in all 14 catalogs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse: rows never fall back to a motion name (#1463)
Without a driver mapping, a row was named for its own calibrated motion,
which is also what its dropdown lists. It now takes the default mapping's
view axis, so the two sides never share a name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse: grey out Invert and speed on rows the driver turns off (#1463)
A row the driver profile turns off says so on hover, and its Invert and
speed are greyed out, since they do nothing there. Its motion choice stays
live so a motion can still be swapped out. The column header is shorter.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse: unmapped rows say Axis 1-6, and the off-row tip explains the swap (#1463)
Without a driver mapping, a row is now just "Axis n" rather than the
default mapping's view axis, which would be a guess there. The tip on a row
the driver turns off says that a motion swapped into it stops moving the
view, and a test swaps one in and back out.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3D mouse calibration: a device that stops short of rest after a release (#2778)
Through spacenavd, a relative-axis device like the Space Navigator never
reports its return to zero, so the reading stays at its last small report
after every release. The wizard never saw the cap let go, and once
calibrated, the stuck value crept the view.
A reading that stays exactly the same, close to rest, now counts as let go,
and where the cap settled after each step goes into the calibration: such a
device rests at the middle of where it settled, with a dead zone out to the
furthest of them. A device that comes back to rest calibrates as before.
* 3D mouse calibration: a steady hold through spacenavd stays a hold (#2778)
* 3D mouse calibration: stopping short of rest only counts from spacenavd and 3DxWare (#2778)
Recentring on where the cap settled kept the rest stage's wander radius but
not its offset, so a rest extreme could fall outside the dead zone; the
radius now grows by how far the centre moved. A reading that stops changing
close to rest only ends a step from a source that sends nothing at rest, so
HID and web devices calibrate exactly as before.
* 3D mouse calibration: settling off-centre counts on any device (#2778)
A drifting stick settles off its rest after a release just as spacenavd's
stuck reading does, but it keeps reporting, noisily. A step now ends once
the reading stays steady, each raw axis within twice its rest wander (so
exactly still from a silent source), close to rest, on any device. The
dead zone takes that wander on top of the furthest settle point.
* 3D mouse calibration: steadiness by the run's spread, settled on the step's own axis (#2778)
An axis that read flat at rest may dither once off-centre, so steady
readings may spread at least 2 units and a 50th of the furthest any step
went. Steadiness is the spread of the whole run, not distance from its
first reading. Close to rest is judged on the axis that set the step's
peak, since a device's axes may have different ranges, and the settle point
is the middle of the steady run.
* 3D mouse calibration: readings close to rest never count as held (#2778)
* 3D mouse calibration: the dead zone reaches every reading of each release (#2778)
Saving one reading of each release and padding it by the rest wander could
miss a reading seen in the same steady run. Each release now keeps the whole
run, and the dead zone is centred on and reaches everything read at rest
and after the releases.
* 3D mouse calibration: settle for half a second before a step counts as let go (#2778)
A hand held still close to rest is less likely to end a step early; a
device that comes back to rest never waits for it.
* 3D mouse calibration: show the dead zones as a game controller tester would (#2778)
Once calibrated, the window, and Preferences under the Dead zone slider,
show three pads: slide, press and twist, and tilt. Each has a circle for
the full travel, the dead zone shaded inside it with each direction its
own size and the user's dead zone on top, and a dot where the cap is now,
lit once the view moves.
* 3D mouse calibration: settle through cross-talk, per-axis steadiness floor (#2778)
The axis that went furthest in a step needn't be the one asked for:
cross-talk can move another raw axis further. A step now counts as let go
once the furthest axis is at least halfway back and some axis that moved
about as far is within a quarter of its own peak; a hand still holding the
push keeps the furthest axis out. The steadiness floor follows each axis's
own reach, and each stage's steadiness starts with its own readings.
* 3D mouse calibration: steady readings may dither ±2 on any axis (#2778)
* space-mouse: make the settle check's near band match the steadiness spread, test the 4-unit floor
near() measured distance from the settled run's middle against the whole tolerance,
so it accepted twice the band the steadiness check does. Use half. Add a short-range
test where only the .max(4.0) floor lets a +-2 dither settle, and one where dither
the spread check rejects is also not skipped as untouched.
* space-mouse: settle on the middle of the steady run so a lopsided dither isnt movement
near() allows half the steadiness spread around the settled point, but the point was the run mean. A dither that is mostly one value with an occasional far one (10, 10, 10, 14) is steady, yet the far reading sat 3 from the mean and started a hold in the next step. Use the middle of the run range instead.
Adds a lopsided-dither test and a test pinning the documented cross-talk limit ([100,300] then [5,90] stays open).
* Judge an untouched cap by the settled run's spread, not a rounded midpoint
Rounding the midpoint to i16 could leave one end of an odd-width run more
than half the tolerance away, so the cap read as touched while sitting at
the end of the run it settled on. Keep the run itself and ask whether one
more reading keeps its spread within tolerance.
* space-mouse: pin the wider skip band of a narrow settled run and a single-reading settle
* space-mouse: keep a step held while a lesser axis stays out as the larger one relaxes
Settling needed one axis furthest and returned, and one axis about as far and settled, which a single relaxed cross-talk axis satisfied alone. Every axis that moved at least half as far as the furthest now has to be back by RETURNED_SHARE of its own peak too.
* 3D mouse: the calibration tests run in a focused window, as the focus check now requires
* 3D mouse: the axis tests run in a focused window, as the focus check now requires
* 3D mouse: the calibrated-rest test runs in a focused window, as the focus check now requires
* Settle a spacenavd reading that is nearly at rest (#3523)
spacenavd sends a record only when the values change and nothing when the cap springs back, so the last small leftover of a release was held forever and the view kept drifting.
A held reading now reads as rest once every axis is within 150 raw units of the device rest and no record has arrived for 0.5 s. A firmer push stays as long as it lasts, and so does a small value next to a large one. 150 is the smallest round number above the worst leftover in the issue (129 on one axis).
* Choose which 3D mouse driver to read on Linux
When both spacenavd and the device itself (hidraw) are there, Preferences now has a "3D mouse driver" choice: Automatic, spacenavd or Direct. Automatic is what it always was, spacenavd when it runs and the device otherwise, so nobody's setup changes. If the chosen one isn't available the other is read and the log says so once. Each driver keeps its own calibration, as before.
To switch without a restart both reads now wake up when idle and look at the choice: spacenavd with a 250 ms read timeout, hidraw opened non-blocking and polled every 5 ms when it has nothing. A record that arrives is read as before, and one split across a timeout is kept. The reader also publishes which of the two exist, which is what shows the combo box; with only one it says which one is in use.
* Build the 3D mouse row header from existing labels
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Sketch tools come back in the variant used last, and it is kept in preferences
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Scenario: a new design keeps the remembered sketch variant
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
A sample picked before its picture arrives still opens; the language test waits for the file check.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* Revert "Start page: sample pictures and the existence check of recent files run off the window thread (#4295)"
This reverts commit d9059d8da5.
* Test: an older design is marked format 28 now that 0.18.3 folded face pieces into 29
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Thread: turn a shaft thicker than a modelled thread down to its major diameter
A modelled external thread on a shaft thicker than the thread no longer refuses: the shaft
is turned down to the major diameter over the threaded length (run past a free end of the
face), then the thread is cut as before. The turned face is named F<n>:turned and the step
it leaves F<n>:shoulder:start/end. A shaft the part goes on around is refused rather than
cut into, shafts under the minor diameter and cosmetic threads on thicker shafts are still
refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Thread preview test: an 8 mm shaft is turned down for M6; M10 still refused
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Face names: pieces of a split face keep their numbers across evaluations (#3902)
Pieces sharing a name were numbered in the kernel's order, or by the index of the operand face
they came from. A boolean can list its result's faces in another order on the next evaluation,
and an operand made by one (a closed sweep tool) carries that order into the indices, so the
#1/#2 of a split face could swap between two evaluations of the same design. Pieces that
nothing else tells apart are now numbered along the spread of their centres (as edge pieces
already were), and pieces from different faces of one operand by kind, then by place.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Face names: older designs keep the piece numbers they were saved with
Numbering pieces by place renumbers split faces in many existing designs, and a stored name
matched exactly would then land on the other piece. Features of a design saved before this
change are marked `kernel_piece_order` by a pending format step and number their pieces as
before; features added afterwards number by place.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* STEP import: mesh band faces that wrap around a closed surface correctly
Fixes two related bugs that caused SolidCraft to fail to mesh a face wrapping
all the way around a closed, non-periodic surface (a revolved ridge band, or a
tube-shaped B-spline surface): a missing face, a folded/self-overlapping
triangulation, or a hole in the rendered body.
1. Ring edges that run slightly off their surface. STEP files write a surface
and the curves bounding it to the file's own declared uncertainty, so a
ring edge can legitimately sit a few micrometres off the surface it
bounds. The code finding where a ring edge crosses the planned split line
compared against the edge's raw (off-surface) point using a hardcoded
tolerance far stricter than the file's own precision, rejecting a valid
cut. Now matches against the point on the surface directly below the ring
crossing, respecting the file's declared tolerance.
2. Closed B-spline surfaces, their slits and their slivers. truck only treats
the rotation of a revolved surface as periodic; a face on a B-spline
surface whose control net closes on itself (a tube) wasn't recognized as
wrapping around at all. This produced a folded face, a hole (a boundary
loop landing a full period away from where it should cut), and one
near-zero-width sliver face costing the entire body its tessellation
plan. Adds periodicity detection for closed B-spline surfaces, period-
aware hole-loop alignment, dropping of zero-width slit boundaries, and
isolating a too-thin-to-triangulate face so it no longer takes the rest
of the body down with it. The GPU shader's periodic wrap mirrors the CPU
math so the two don't drift apart.
Regression tests: band_faces_split_when_ring_edges_run_off_their_surface,
band_face_with_a_zero_width_slit_is_still_meshed,
a_sliver_face_too_thin_to_triangulate_does_not_cost_the_body_its_plan.
Ablation-tested: disabling either mechanism in (2) reproduces its target
test's original failure message while leaving the other test unaffected.
* STEP import: mesh the band's split faces, not just split them
jyviko asked whether a band wall missing from the interactive app's render
was a shading issue or a genuinely missing face. It was missing: the ring-
edge split fix made the band's topology correct, but plan_face still
produced zero triangles for the two new sub-faces.
Two causes, both in plan_face (crates/kernel/src/tess/mod.rs):
1. The tolerance for how far an imported boundary may lie off its surface
was tied to the requested mesh's chord tolerance instead of to the
body's own size. A finer display tolerance could then reject a boundary
a coarser one had just accepted, so the face would drop out exactly when
zooming in on it.
2. Near a sharp corner of this flat-panelled band, several boundary points
that are legitimately a few micrometres apart (within the STEP file's
own declared precision) all projected onto the same surface point. The
boundary loop then repeated a point in (u, v), which a triangulator
can't follow, so the face's plan failed outright. One face without a
plan costs the whole body its plan, and the fallback mesher can't
handle a closed B-spline surface either, leaving both halves of the
split band with no triangles at all: present in the topology, invisible
in every renderer, and absent from the body's volume.
Fix: scale the off-surface tolerance to the body's own extent (new
Rules::extent), and collapse a run of boundary nodes that land on the same
surface point to one (drop_collapsed) before triangulating.
Regression test additions to band_faces_split_when_ring_edges_run_off_their_surface:
the body's volume now matches the analytic pentagon prism (was a third of
it - only the top cap - before this fix); display_plan leaves no face
without triangles from coarse to fine tolerance; and the actual tolerance
and facet angle the viewport draws with gives the split band its full
lateral area on both halves.
Confirmed on review_band_big.step, the fixture behind the review screenshots:
volume's relative error against the analytic pentagon prism drops from
1.4e-4 to 4.1e-6, and the band wall now occludes the floor grid behind it
in a snapshot the same way the top and bottom caps always did.
Full kernel suite: 739 tests, 0 failed.
* STEP import: hold an imported boundary to the file's own tolerance, not the part's size
jyviko: the off-surface acceptance added in the last commit
(rules.extent * OFF_SURFACE) scaled with the part, so a 2 m part
silently accepted a boundary 2 mm off its surface while a 20 mm copy
of the same part accepted only 0.02 mm. A STEP file's own declared
uncertainty does not grow with the part, so the bound shouldn't either.
Body gains boundary_tol: Option<f64>, Some(file's declared tolerance)
for a body read from STEP (crate::step_in::Units::tol, already used
by the ring-split fix in split.rs), None for anything this kernel
built. It's set before the import's own unmeshed-face check runs, and
carried over by the two face-split operations that re-cut an imported
surface (splitface.rs, outline.rs), since the file's uncertainty still
describes those boundaries afterward.
plan_face's accept now reads: the file's own tolerance when the body
declares one, or the old chord-tolerance/extent bound otherwise. The
old bound's tol * 10.0 term had to go too, not just the extent term:
at a 2 m part's typical display tolerance that term alone is ~20 mm,
nearly ten times the extent bound jyviko objected to, so keeping it
would have left the exact 2 m/2 mm example still silently accepted.
Two new tests in step_in/tests.rs:
- band_faces_split_when_a_scaled_copy_runs_off_its_surface: the same
band fixture at a realistic 2 m scale with the same absolute slop
and declared uncertainty as the small one, meshes the same way at
every chord tolerance from coarse to fine.
- off_surface_budget_follows_the_file_not_the_part_size: the actual
regression. The same off-surface error is accepted or refused the
same way whether the part is 20 mm or 2 m; a 2 mm error against a
1 µm declared uncertainty is refused at both sizes now (accepted at
2 m under the old bound, refused at 20 mm - that asymmetry was the
bug). Checked at both a coarse and a fine display tolerance.
Known gap, not fixed here: a STEP-imported body saved into a
.solidcraft design and reopened loses boundary_tol (native.rs's
decoder has no field to read it from) and falls back to the old
extent-based bound on reload. Today's design format has no concept of
this value at all; persisting it is a real format change, out of
scope for this fix.
Full kernel suite: 776 tests, 0 failed. clippy -D warnings clean.
* Display plans for STEP threads and pointed ends (#3847)
Three faces cost a threaded or pointed STEP part its whole display plan, so it
was drawn from the reference mesh: cracked, faceted, and never refined up close.
- A loop through a pole (a cone's apex on the boundary) seemed to wind around
the surface: the jump at the apex, unwrapped to the nearest copy, was
arbitrary. The loop is turned to end where it leaves the pole, so the step
that closes it is an ordinary one (`seam::close_through_pole`).
- Projection onto a spline with hundreds of spans (a helical flank) restarted
from a 16 x 16 grid and settled on another turn; it now starts from two
points per span, capped.
- Sag refinement no longer inserts a point exactly on a boundary side, which
split the side and failed the face.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* tess: a refused face's edges are divided finer and it is planned again (#2814)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* tess: sliver faces plan empty; a face no method plans is filled flat from its nodes (#2814)
A face whose boundary runs out and back along itself (two edges between the
same vertices, a rounding apart, as STEP files write slivers) encloses nothing
and is planned with no triangles. A face still refused after its edges are
divided finer (a cusp, a sliver whose sides sag across each other, a small cap
at a pole) is filled flat from its shared boundary nodes, so the rest of the
body keeps its display plan instead of falling back to the reference mesh.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* display mesh tests: a rim the chord tolerance leaves a line is divided again (#2814)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* portable B-rep: up to 8192 control points along a spline direction (#2814)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* solidcraft-cli: --debug logs what the engine reports to stderr
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* STEP import: a spline written over a tiny parameter range is read over one (#2814)
Some writers parameterize splines over a few millionths (a humanoid robot's
edges span 5.4e-6). The kernel tells parameters apart only to 1e-6, so over
such a range it evaluated the curve's end wrongly, and trimming an edge to its
vertices dropped a control point: the bodies could not be stored exactly and
the whole model cache was left out of the design. The knots are now stretched
to span one; the curve is the same and edges are trimmed by their vertices, so
nothing read from the file refers to the old parameters.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* model cache: a body keeps its file's declared accuracy (#2814)
A body restored from the cache dropped the accuracy the file it was read from
declares, so the planner held its boundaries to the kernel's own bound and a
reopened design drew differently from the import. It is stored with the body's
runtime values, and only when there is one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* tess: a boundary within a hundredth of its body lies on its surface, at every quality (#2814)
#1721 held a body read from a file to the accuracy the file declares. A writer
declares what it aims for, not a bound its edges keep: a 251-body robot
assembly declaring 0.01 µm has boundaries 10 to 30 µm off their surfaces, and
95 of its bodies lost their display plans. The bound is a hundredth of the
body again, what Normal quality always allowed, and now the same at every
quality, so zooming in still never drops a face (#1721's point). A file
declaring more raises it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* STEP import: keep faces whose edge vertices sit slightly off their curves
Some exporters write an edge's vertex beside the end of its curve, further off
than the accuracy the file declares (a surface export declares
0.01 mm and puts a vertex 0.19 mm off two spline ends). The reader refused
such edges and left the face, and with single-face bodies the body, out.
A vertex may now sit up to the file's tolerance or 1/1000 of the model's size
(the diagonal around its VERTEX_POINTs), whichever is larger, never more than
the existing 1 mm ceiling. Spline and open-conic edges have the end control
point moved onto the vertex, so every such edge meets its vertices exactly.
The file's declared accuracy still governs closed-curve, seam and circle
checks.
* tess, io: review clean-ups (#2814)
- A sliver face is one whose loops are all slit: drop_slits finds it in the
retry after a failed plan, so the pre-check on every face (folded) is gone.
- The flat fill uses cdt::triangulate (a lenient fill that leaves out sides
crossing one already kept) and its flood-fill inside test, and the plane
from geom::Plane.
- Refinement only builds its indexes when a face was refused, takes the
faces to plan again from edge_faces, and re-samples edges on the pool;
Rules::finer for the finer rules.
- One PackageWriter::fits for the model cache, the preview and the thumbnail.
- The band split's ring tolerance lives in split.rs; comments say what the
off-surface bound is now.
- The CLI logger prints Level::as_str().
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
---------
Co-authored-by: David Lackovic <lackovicdavid1234@gmail.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Ray <raykholo@users.noreply.github.com>
While a new geometry kernel is evaluated, a workflow labels new issues, pull
requests and discussions about kernel work kernel-paused, so a maintainer finds
them and answers in person. A pull request counts when it changes
crates/kernel/ or vendor/truck*, anything else when its title names a kernel
operation. Crash and data-loss reports and anyone who can push are left alone.
CONTRIBUTING.md says so.
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Two OK-off-the-window-thread tests checked that OK was computing right
after Enter without holding the evaluation. The preview's evaluation is in
the cache, so the worker was usually in before OK's zero wait was over and
OK applied at once: cancel_while_computing_applies_nothing failed in most
runs (5 of 5 in a debug build, 4 of 5 with the ci profile). Both now hold
the evaluation with the test gate, as their neighbour does, and let it go
once the computing state is checked.
Consecutive pieces of a trim loop can miss each other by micrometres to a tenth of a
millimetre, past the file's resolution, so the loop did not close and the face was
dropped. Faces are now read first and sewn after: each small gap's ends, and the other
piece ends at them, move to one vertex. Gaps over 0.1 mm or not small next to their
pieces stay open; the sewing tolerance stays the file's. The stated accuracy covers how
far a vertex moved, so the STEP reader keeps those edges.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
An agent's design changes over the control channel (engine.execute, engine.batch) are evaluated
ahead on a copy of the session on an `agent` worker, then run on the window thread reusing every
feature computed there (doc::KeptResults, whatever the bounded evaluation cache dropped); the
worker also makes the display meshes of the bodies they end with. Exports run on a copy. While a
change computes, or the agent cursor acts one out, ui.*, inspection and screenshots answer; later
design changes wait their turn. A chip names the command with Stop. A person's edit meanwhile goes
first; a replaced design refuses the request. Preference: Preferences > AI, agent_off_thread, off
by default.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
transform_matrix took any finite 4x4 matrix. A perspective entry in the
last row, or a singular linear part, was built into faces that mesh to
nonsense: a revolved ring placed with such a matrix measured hundreds of
millions of mm3, and a boolean on it ran for over half an hour before
failing with "no result". Matrices come from design files and command
parameters, so they are refused with KernelError::Invalid instead.
* claim: #3394
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* STEP export: a merged cylinder or cone face's loop starts on the side its angle grows from (#3394)
A whole-turn cylinder or cone face is written as one loop: ring, seam, ring, seam reversed. Readers lay the ring that follows the seam's first use out from the surface's angle 0, so that ring must turn the way the angle grows (anticlockwise about the written axis). Three of the four tapered legs in #3394 came out the other way and other readers dropped their sides. When the second ring turns against the angle, the loop now starts with it.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* claim: #2966
* Expressions: a leading decimal comma such as ,5 is accepted (#2966)
In a comma-decimal locale `,5` means 0,5, but the lexer only started a
number on a digit or a point. A comma now starts a number outside
parentheses when a digit follows and an operand is expected (at the
start or after + - * / ^). Commas after an operand keep their meaning,
so function arguments and the existing error cases are unchanged.
* claim: #2882
* Holes from a sketch on a body's face drill that body (#2882)
A multi-point hole from a sketch attached to a body's face chose its
side and moved its points onto whichever body lay along the direction,
then cut every body, so points on a front body's face could drill the
body behind it. `targets` was accepted and ignored.
- solid.hole: holes at the points of a sketch on a body's face store
that body as their participant (the lookup a face-sketch Join uses);
`targets` sets the bodies to drill, `["*"]` every body; explicit
participants still override.
- Evaluation: for sketch-point holes with participants, the drilling
side, the surface each point moves onto and pattern copies are found
on those bodies only.
Sketches on planes, position and Single holes, and saved holes without
participants recompute as before; no design-format change.
* claim: #2998
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* io: the 3MF and DXF budget tests do not check wall-clock time
nested_fan_out_of_empty_objects_is_bounded spends about 20 million
visits reaching the work budget: 1.5 s alone, up to 40 s and more when
every test binary runs at once, past its 30 s limit. The time checks
run only after the reader returns, so a missing budget hangs these tests
with or without them; the error, the empty geometry and the three
vertices per placement are what show the budgets work. CONTRIBUTING.md:
a test must not depend on wall-clock timing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: kezmdor <15539254+kezmdor@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Measure: a body whose meshes leave a gap keeps its volume wherever it is
A part ring cut by solid.combine (two prisms through its ends, a notch
through its chamfer) measured 992.03 mm3 at the origin, 999.17 moved
100 mm and 1063.48 moved 1000 mm; it encloses 993.92. The reference
mesher leaves a crack where the notch meets the chamfer's cone: the
volume beside it is left out, and the divergence-theorem sums, taken
about the world origin, close the crack by a cone to the origin, so the
error grows with the body's distance from it.
A closed surface's area vectors sum to zero. When the picked meshes'
sum could move the volume by more than 1e-5 within the body's size, and
the display plan (which shares every edge's nodes between its faces) is
closed, the body is measured on the plan, about its box centre. Every
other body measures exactly as before.
Tests: the cut ring at the origin and moved by up to 10000 mm, against a
grid quadrature of the cut; a revolved ring of cones and cylinders moved
100/1000/10000 mm along each axis (1e-6); a cylinder and a frustum far
from the origin against hand values. Robustness seed 64 moves by 1.4e-6
and is recorded.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Measure: a gap the display plan can't close is measured about the box's centre
When the meshes leave a gap and the display plan isn't closed either, the
sums used to fall back to the origin-based volume, which drifts with the
body's distance from the origin (7 % at 1000 mm for the part ring). The
per-face sums are now taken about the fine mesh's box centre, which closes
the gap by a cone to a point inside the body; over a closed mesh nothing
changes. The gap decision moves into gap_closed, and a unit test with a
cube left open by hand covers closed vs open meshes, the closed plan, and
the fallback with an open plan or none.
Seeds 14, 20, 40, 55 and 61 end with such a body and are recorded with
ROBUSTNESS_GOLDENS_BLESS (volumes move by -0.51 %, +0.16 %, +1.9e-4,
+9.0e-5 and -1.5e-5; steps, outcomes and counts unchanged).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Record robustness seed 86 after measuring bodies whose meshes leave a gap
Seed 86 was added on main by #2387 after this branch's goldens were recorded. Its last
body leaves a gap in its meshes, so the measure change here moves its volume by -0.27 %
and its area by +0.13 %; steps, outcome and counts are unchanged. Seed 86 passes
unchanged on main 57d21013. Every other seed matches on the merged branch.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* measure: take the display plan only where it agrees with the sums to within what the gap could move
A closed display plan meshed less finely than the sums differs from them by its own chord error,
which can be larger than anything the gap explains: an imported part with small blends measured
0.03 % low on its plan. The plan is now taken only where the two differ by at most what the
gap's area could move the volume within the body's size; otherwise the sums about the box's
centre stand.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* measure: sum a mesh left open about its box's centre only, without the display plan
The closed display plan meshes some faces of an imported B-spline part
more coarsely than the sums do, so a body and its boolean pieces no
longer add up: clearance's union cross-check then refused a 17.3 mm3
overlap between two such parts (the pieces implied 20.5). Summing about
the box's centre alone keeps the gap closed by a cone to a point inside
the body and stays consistent with the pieces. Seeds 64 and 86 are
re-recorded with the reason.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* measure: placement tests in the area's test files
The ring and frustum tests join measure.rs's tests, and the cut ring
joins measure_centre_tests.rs (AGENTS.md: a regression test goes into
the existing test file for its area). The goldens note for seeds 14, 20,
40, 55 and 61 describes the centre-only sums.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* measure: keep the origin for a body left open on purpose
A surface, or a cup whose lid is still a separate sheet, has no volume of
its own. Measured about its box's centre, two such bodies stopped
comparing once the centre moved (a crowned lid rises), which broke the
patch crown test. Only a mesh with a small crack is measured about the
centre now.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Aac7mP9rthK3WsPpPRvCsF
* Robustness goldens: keep main's notes, one note for this change's seeds
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Aac7mP9rthK3WsPpPRvCsF
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* claim: #3654
* Control channel: control.running lists what the app is running, across connections (#3654)
control.pending answers only for the connection that asks, so a second client,
the MCP bridge after a reconnect or a script saw {"pending": null} while the app
was busy. control.running answers for the whole app, the same on every
connection and at once (the connection's own thread answers it, also while it
is busy with a timed-out request): every request the app has been sent and not
answered yet, oldest first, with its connection, id, method, command id (or a
batch's step count), seconds since it arrived and whether it still waits its
turn. A timed-out request stays listed until the app answers it, also after its
connection closed. Parameters are never shown.
The window marks a request started when it takes it (ControlRequest.started),
which tells queued from running. The MCP server gets a read-only `running` tool.
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
#3609 added Applying::seen; dialogs_sketch_tests' apply_keeps_the_panel_open_after_asynchronous_evaluation
still built the struct without it, so ui-egui's tests stopped compiling on main. The test stands
for a job that is already being shown, so seen is true (the behaviour it had before).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2929
Co-Authored-By: Claude Code <noreply@anthropic.com>
* OK off the window thread: a job that has just started waits a frame (#2929)
The off-thread evaluation of a feature dialog was taken up in the same frame
it started, so a job whose worker had already finished was applied at once
and the dialog never said it is computing. Under load the window thread is
held up between starting the job and the check, and the feature appears
where the computing state belongs - which is what made
`apply_off_thread_tests::cancel_while_computing_applies_nothing` fail with
`assertion failed: h.app.applying.is_some()`.
A job that has just started now waits one frame: its result is taken up from
the next one, so the dialog says it is computing before the feature appears.
The app waits APPLY_AT_ONCE (150 ms) for the evaluation before it lets the
window go on, so a job only exists when the evaluation took longer than
that: the change shows the computing state for a frame in the 150-151 ms
window, and nothing else moves.
Co-Authored-By: Claude Code <noreply@anthropic.com>
---------
Co-authored-by: jeremymiribung-blip <jeremymiribung-blip@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
A sketch attached to a face is evaluated in the face's frame. When that frame is the body's
own, the body was still rebuilt as a new ModelBody with empty mesh cells, so projecting the
face's edges display-meshed a throwaway copy and the next cut or join meshed the same body a
second time. On a body whose cut edges are slow to mesh this added about a second to each
feature after a sketch on a face. In place, the anchored body is now the model's own clone and
shares its mesh.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3223
* Extrudes from a nested STEP part's face join and cut it with the imported root active (#3223, #2872)
Opening a STEP assembly leaves its root component active, and the root
owns no body: its parts are components inside it. A profile sketched on
such a part's face was therefore treated as outside the active scope.
Join made a separate body instead of joining the part, and the automatic
operation saw no material and offered New Body in either direction; an
automatic extrude that came out as a cut found nothing to cut.
A component with no body of its own now works on the bodies of the
components inside it whose geometry lies in its own frame (as an import
places them): Join targets the body whose face the sketch is on, the
automatic operation looks at those bodies, and an extrude whose own
operation is Cut or Intersect takes part in every body as an explicit
cut already did.
* claim: #2089 (3MF production extension: review follow-ups)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3MF import: production extension review follow-ups (#2089)
Follow-ups from the review of #2127:
- Only the root model's build is read; other parts' build sections are
ignored as the spec asks (a slicer's child build pointing at a thumbnail
no longer matters).
- A model part that can't be read fails only the placements that need it;
the package-wide caps (part names, 1 GB of model XML) still refuse the
package. The root XML is dropped once parsed and each part is read
against the remaining budget.
- Loops are reported apart from depth, keyed by (part, object).
- Messages name parts as written; objects are kept in id order so parts are
read in the same order every run; %XX escapes in part names are decoded;
the model unit is a local of the parser.
- Tests: a component without p:path stays in its part (T1), the same id in
two parts is no loop (T2), a slicer-shaped project (T3), the placement
budget across parts (T4), a loop through three parts and a chain past the
depth cap (T5), broken parts and child builds (T6), escaped names; the
hostile-parts test asserts its messages; the mutation test also mutates
the parts' XML and zips it again.
- Guide: what slicer-only data is not read.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Split the schema contracts into one module per command file
`schema/commands.rs` becomes `schema/features.rs` (solid.box, solid.extrude,
FeatureResult) and `schema/inspect.rs` (engine.describe), each with its own
tests file holding its `schemas_and_the_key_check_agree` key lists and its
`strict_keeps_todays_errors` cases, so conversions of different command files
touch different files. The shared loops become helpers in `schema/tests.rs`;
the assertions are unchanged. `engine.describe` gains a key-list test, and
`schema_defaults_are_valid` now covers every command with a contract.
No behaviour change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Add shared schema field types for the next conversions
For the feature commands to be converted next, each deciding acceptance with
the reader the handlers already use:
- `FaceRef` and `FaceRefs` (`face::face_ref`, `features::face_points`)
- `EdgeRef` and `EdgeRefs` (`features::edge_points`)
- `SketchRef` (`features::sketch_id`)
- `FeatureCommon`: `name`, `body_name`, `body_names` and `participants`,
which `add_feature` reads for every feature, flattened into a contract
with `deserialize_with = "feature_common"`, which keeps the key in the
message and the issue's path.
The shape checks of `face_ref`, `edge_points` and `sketch_id` (and
`feature_sketch`) move into pure helpers (`face_ref_form`, `edge_ref_form`,
`sketch_key`) that the handlers and the types share; the handlers behave as
before. Tests check that each type and its schema accept what the reader
accepts (apart from a face name's byte cap and an index written `1.0`),
that the contract's messages match the handlers' errors, and that the
handlers still look picks up one at a time.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3510
* Split Body: each piece can be renamed on its own (#3510)
A split named its new pieces after the split body every time (`Top (1)`),
so renaming a piece did nothing, and renaming the split body renamed the
other piece too (and broke the features that used it). The pieces now
take the names body.rename stores, and a rename keeps the names later
features derive from the renamed body.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3383
* Combine: tools from another component meet the target where they are shown (#3383)
Bodies are evaluated in their own component's frame and an occurrence's
placement is applied only for display. Combine passed tool bodies to the
boolean untouched, so a tool in a component placed differently from the
target's was read as if its coordinates were in the target's frame: a
join that overlapped on screen gave two separate shells, and cuts and
intersects missed the same way. Any placement difference did it, a turn
or a plain move, in either direction.
Combine now moves each tool by the inverse of the target's placement
times the tool's own, the same placement used to show the body, keeping
face lineage. A tool placed exactly like the target is left untouched,
so combines within one component give the same result as before.
Moving an occurrence now re-evaluates a combine that depends on it.
* Joins, cuts and intersects draw their tools where they are shown (#3383)
A primitive, extrude, revolve, loft, sweep, pipe or coil that joins,
cuts or intersects bodies draws its tool in its own component's frame,
but each body is kept in its own component's. A target in a component
placed differently (an occurrence moved or turned, or a body moved into
it) met the tool as if the tool's coordinates were in the target's
frame: a join that overlapped on screen gave a separate lump 100 mm
away, and cuts missed the same way.
While such a feature evaluates, the tool now goes into each body's frame
(the inverse of the body's placement times the feature's component's)
wherever the join, cut or intersect meets it, and a body joined into one
placed differently is moved into its frame. Bodies placed as the feature
is are left untouched, so results within one component are unchanged.
Features that name the bodies they change re-evaluate when an occurrence
moves.
This also settles the moved-fastener cut of #439 (Combine, earlier commit).
* Through all and an extrude's contact tests see targets where they are shown (#3383)
An extrude or revolve measured its through-all length, and tested where
its tools meet the targets (a profile on a target's face, coplanar
growth, holes over empty space, taper run-on), on the targets as stored
in their own component's frame. With the tool now moved into a target
placed differently, a through-all cut drawn from the root under a cube
in a component moved 100 mm up was cut only as far as the cube is tall
in its own frame, and missed it.
These features now see the bodies they change that are placed
differently moved into their own component's frame, the same moves the
tools are taken back through, so the length reaches the target where it
is shown. Bodies placed as the feature is are read as before.
* claim: #3353
* Sheet metal: mirrored and copied sheet bodies stay sheet metal
Mirroring a sheet metal body into a new body gave a plain solid: the
opposite-hand bracket had no flat pattern, DXF or bend table
("`Body1 (2)` is not a sheet metal body"). Copies from a body pattern or
Move with copy lost their sheet model the same way.
A copy now takes the sheet model with it (SheetBody::copied). A rigid
copy moves the flat frame. A mirror moves the frame with its x axis
turned round, so the thickness side stays, and mirrors everything in
flat coordinates (x -> -x): base, flange bend lines (run from their far
end, with their end extensions, slopes and tears swapped), panel shapes,
cut-outs, pockets, bores, letters and base sketch lines. Arcs stay exact
and loops keep their sense. A perforation field of a mirrored copy runs
its grid from the right, so its holes are the mirror image too. The flat
length, bends and K-factor are unchanged, and the copy takes further
sheet metal features of its own.
Mirror with combine is unchanged: the joined body has no single sheet
model yet. Mirroring or patterning flange features is still refused.
Sheet models are derived from the timeline, so no design format change.
* Format: note the saved from_right of a mirrored sheet's perforation field
The new `from_right` flag is saved in the design, so it gets a pending
format note. No migration: a field without it runs from the left as
before.
* claim: #3221
* Save designs with large STEP imports: compress the STEP part (#3221)
A STEP between 256 and 512 MiB imported fine but could not be saved: a
design stored the STEP text uncompressed, and a design part is capped at
256 MiB. STEP sources are now always Deflated (level 2), may be as large
as an import accepts inside the file, and an import that could not be
saved even compressed is refused with a clear message. Reading stops at a
part's declared size, so a decompression bomb is refused early.
* claim: #3749
* Edit Joint: deleting the joint being edited closes the dialog and deletes it (#3749)
A joint delete was dropped without a word while any dialog was open, and
Edit Joint is a dialog. Deleting only the joint Edit Joint edits now
cancels the dialog first (as its Cancel does) and goes through the usual
selection.delete dry run; Delete with nothing else selected in Edit Joint
targets that joint. Every other refusal (another dialog, a tool, a
computing history, a loading file) now says why in the status bar.
Edit Section Analysis (its menu item, or a double-click on its browser row)
opened the dialog at distance 0 with no plane, so moving the cut on by a
millimetre meant picking the plane and typing the whole distance again.
OK now keeps the dialog's plane pick, distance and flip with the cut they
made, and Edit starts from them while the section shown (or hidden) is still
that cut. The toolbar still starts a new section afresh, and a section made
some other way (a command, an agent) opens the dialog as before.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3798
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* fix(ai-chat): replay once with reasoning_effort none when tools are refused (#3798)
Newer OpenAI models answer HTTP 400 to function tools on /v1/chat/completions
unless reasoning_effort is "none". On that exact refusal the request is sent
again once with it, and later requests of the chat carry it directly.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #3483
* Settings > AI: Add agent opens from the Home tab too (#3483, #3513)
The Add agent dialog was drawn by the chat panel, which Home doesn't
show, so Add agent did nothing while Home was open. It is now drawn
from the frame itself, before Home and the workspace part ways.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Type-first (pick, then type), the dimension's text follows the pointer, and where it goes
picks the kind: across from two points or a slanted line the distance between them, above or
below the horizontal span, beside them the vertical one. The draft kept the value of the kind it
started as, so moving from diagonal to horizontal previewed (and Enter applied) the aligned
length as the horizontal span: the solver moved the picked points (the first dimension of a
sketch even scaled it), so the dimension's extension lines no longer met them.
While nothing has been typed, moving the text now re-derives the draft for the new place, so
its value is what that kind measures and stays selected for typing. A typed value is kept.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Ribbon: Finish Sketch follows the last sketch panel
The Sketch tab's exit group sat alone at the far right of the ribbon,
about 600 px from the sketch tools on a 1600 px window. It now comes right
after the last panel, as before the redesign. Its room is still kept at
every width, and when panels go into the » menu it moves to the right end
after it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* ui_layout test: Finish follows the last sketch panel, right end only with the » menu
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The menus no longer ask for offset_one_side, offset_symmetric,
offset_two_sides, pattern_center or pattern_flip, so the manifest check
failed on main and on every UI pull request.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch: a quick double Esc finishes the sketch
One Esc still puts the tool down, then clears the selection, and with
nothing left to cancel keeps the sketch open (#1804). A second press within
0.4 s finishes the sketch, keeping it, the same as Finish Sketch, also when
the first press cancelled a tool.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch: a double Esc takes away a blank sketch; only Esc twice in a row counts
A quick double Esc on a sketch the user added nothing to removes it in one
undo step, the view and tab going back as on Finish, so no empty sketch is
left behind. A sketch with anything in it is still finished and kept.
The double press counts only for two Esc presses in a row with no tool or
dialog left after the first: any other key in between resets it, and while
a tool is still out (its first Esc only ended a chain) the press puts it
down as before. Tests that press Esc on its own now wait past the window.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Sketches drive surface extrudes, revolves, sweeps and lofts, so the Surface
tab gets the same Sketch panel as Model and Sheet Metal. A sketch started
there keeps the tab until it opens.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2501
* English (UK) as a language: a plain catalogue of the 25 messages UK English spells differently (#2501)
assets/locales/en-GB.json holds only the English messages whose spelling differs in UK English (Centre, Customise, Organisation, Licence, colours, towards, ...), in UK spelling; every other message falls back to English. One registration in i18n.rs (English first, then by id), a test that shows UK spelling and English for the rest, a row in ATTRIBUTION.md, and the docs/localization.md note. The translation-coverage checks skip a regional English catalogue, which has no gaps to report.
Fixes#2501
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(skills): refresh Boolean checking facts that fail on current release (#1835)
* fix(skills): restore warning coverage in test_run and use present tense in Boolean facts
* fix(skills): the slot keeps its side at zero since the solver takes the shortest step
The upstream solver now takes the shortest step that fits (#2060), so the slot
that other constraints also held slides to the origin when the distance reaches
0 and keeps the side it was drawn on when the value comes back. It no longer
comes back mirrored, so the trap and its test state the current behaviour.
---------
Co-authored-by: jeremymiribung-blip <jeremymiribung-blip@users.noreply.github.com>
* Extrude: a join from a whole face rounded by fillets sweeps the face itself (#2040)
A face's sketch fits an edge that is no line or arc (the quarter-ellipses
where fillets meet a slanted face) as splines a hair off the body's edges,
so a join from the face touched the body along a face it did not match and
failed with "boolean Union: no result" after minutes. When the profile is
a whole planar face of the one target, the join now sweeps the body's own
face, which shares its edges and is sewn on along it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Extrude: a cut into a whole face keeps the walls that cut and moves the others off (#2040)
A cut from a whole planar face sweeps walls from each of its edges. Where
the face beside an edge turns away from the sweep (n_N . -n_P >= 0) the wall
only touches the body along that edge, or lies in the face beside it, and
the boolean failed on it with "no result". The cut now keeps the walls that
go into the body, exactly (the body's own edges), runs the outline round
the others well outside the body, and starts a little out from the face.
A cut whose walls all only touch takes a slab beyond its depth plane.
Not covered: a cutting wall tangent to the face beside a touching one (a
round's wall meeting an end face, as in the issue's part); the boolean
still fails there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Extrude: leave a whole face outlined by lines and arcs to the profile's own tool
Only a face whose outline the projection had to fit needs the body's own face as the tool. A square drawn over a box's top already meets the body exactly, and the swept face lost the names a sketch on the new top attaches to (stacked_face_tests, #489).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Run the whole cut tool from a face under the kernel guard (#2040)
Review: `builder::vertex`, `builder::line`, `Face::try_new` and
`builder::translated` ran outside `crate::guard`; only the sweep was
guarded. `cut_tool` now runs all of it under one guard.
Adds an end-to-end test of a cut into a face the body stands out past
below it (the block's top strip between its rounds): it takes only
what lies over the face.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test: the slab cut of whole_face_tool, on a spline prism (#2040)
A prism on a closed spline standing on a wider plate, cut 3 into its whole
top face: the slab takes the face's area times the depth and leaves the
plate alone. A test-only counter asserts the slab was the tool, so the test
cannot pass through the profile's own tool. Shortens the name of the
rounded block's top-face cut test.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Kernel: the face's outline as a Vec now that polyline returns a closed chain (#2040)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Build on #1426's import components and nested STEP assembly instances.
Replace placeholder solid labels with meaningful part or instance names.
Translate IGES groups and subfigure instances through the existing STEP
component path, preserving nesting and placements and sewing surfaces
within each group. Bound assembly walks and reuse work buffers.
Related to:
- tangency-labs/solidcraft#1386
- tangency-labs/solidcraft#1426
Co-authored-by: Codex <noreply@openai.com>
* claim: #2570
* Loft: a circle lofted to a non-star-shaped profile is split once round (#2570)
A circle section was split at the lead section's vertex directions, kept in
the lead's order and forced to increase by whole turns. A non-star-shaped
lead (a C, a comb) turns back on itself, so the circle wound round many
times and closed with a negative arc: self-crossing walls and a huge mesh
(which froze the web build while meshing it).
The circle is now split exactly once round: at the lead's directions when
they go round once in order (clockwise on a plane facing the other way),
otherwise at the lead's perimeter fractions, evenly as the fallback. Every
arc is finite and positive and they total one turn; the unbounded loop is
gone. Section preparation is split out as loft_section_loops so the
resampled sections can be checked.
* claim: #2847
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* loft: keep sections turned past 90 degrees facing the same way (#2847)
The loft oriented each section by the sign of its normal against the first section's, so on planes turning about an axis every section beyond 90 degrees was reversed and the side faces folded. Follow the facing neighbour to neighbour instead.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* Ribbon: New Sketch on the Sheet Metal tab again; tests pin sketch enter/exit
Before the ribbon redesign the Sheet Metal tab started with New Sketch. It
gets its own Sketch panel again, like the Model tab's, and a sketch started
there keeps the tab until it opens.
Workspace tests pin the ways into and out of a sketch: New Sketch then a
plane, a double-click in the timeline or the browser, and the browser's Edit
Sketch all open the SKETCH tab looking at the sketch. Finish Sketch from the
palette or the ribbon lands on MODEL with the earlier view back. Escape and
Enter keep the sketch open (#1804), and Undo after Finish reopens it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* docs: ribbon layout lists Sheet Metal's Sketch panel
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Workspace test: press the keys well apart
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* CI: build and run the tests in one job, sccache on a Blacksmith sticky disk
The tests job compiles and runs the tests itself on the 32 vCPU runner, so a pull request or merge
group uploads and downloads no 3 GB archive (the download took 45 to 385 s per shard, and once
14 minutes) and starts beside the repository checks instead of after them. The archive is built
only for the nightly and manual jobs that need it. sccache moves from GitHub's 10 GB cache, where
about 1100 per-ref entries were evicting each other, to a sticky disk that only main commits.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* TEMP: let this branch save caches so the warm run can be measured
* Revert "TEMP: let this branch save caches so the warm run can be measured"
This reverts commit 1bbbfcf5ff.
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #3378
* Pattern and mirror copies of a join that merged its targets join the merged body (#3378)
Since #2755 a pattern or mirror copy of a primitive reuses the seed's targets. When the
seed was a join that merged two targets, the second one no longer exists, so the copies
failed with `unknown body`. The copies now keep the seed targets that are still there;
when none is, the names stay and the error says which body is missing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Rerun CI: the previous run was cancelled before any check ran
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3242
* MCP batch: rollback restores the design the batch began in (#3242)
A failed batch with rollback: true reported rolled_back: true even when the
design never went back. Two defects, both fixed:
- file.new (and any document-replacing command) replaced the session with it,
taking its checkpoints down too, so a batch that failed after file.new had
nothing to restore. Session::replace_design now carries the checkpoints over
to the fresh session: the checkpoint belongs to the batch, not to the
session that happened to be holding the document.
- rolled_back echoed the requested flag. It now reports whether the design is
actually back: the app answers engine.batch with a new restored field, and
the MCP tool reports restored && rollback. An app that predates the field
falls back to undone/completed.
* Batch: answer with main's rolled_back field after the merge (#3242)
Upstream #2900 put the real restore check on engine.batch as
`rolled_back` while this branch named it `restored`; the merge kept
main's control test and this branch's field, and CI caught it.
Standardize on main's wire name: the app answers `rolled_back`, the
MCP tool reads it (fallback for older apps unchanged), docs and tests
follow. The checkpoint carry-over across file.new is unchanged.
---------
Co-authored-by: DahakBuilds <dahakbuilds@users.noreply.github.com>
* claim: #3141
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 2-tangent circle: start from the touching circle, not the click (#3141)
The tool passed its second click as the place to start solving, and a
click lies on a curve: the circle began centred on that curve, where its
tangency has no side yet. The solve then moved the user's curves to fit
(a rectangle came out skewed), or failed outright when they were fixed.
The command now starts from the centre, nearest the requested point, of
a circle of that radius touching both curves (where the curves moved by
the radius meet), and the tool asks for the circle between its two clicks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* CI: give the slowest tests a shard each
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: spread the slow tests on pull requests too, and keep both junit files
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: pull requests skip the goldens and the action matrix sample unless the kernel changes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Docs: which pull requests run the goldens
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: run the tests in one job on a big runner, slowest first
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: allow the 32 vCPU runner label in actionlint, read junit.xml in the golden hint
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The origin point was drawn in the view and listed in the browser, but
nothing could pick it: the viewport never offered it as a hit and the
browser's O row had no selection. A new ORIGIN accept bit lets an input
take it as a vertex of no body; Move's point inputs (From, To, Point,
Pivot) accept it, from the view or the browser's Origin folder.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A curve written over a knot span of 1e-7 or less was refused on the STEP side
("knot vector consists single value"), and both faces it bounds were lost. Its
knots are moved and scaled onto (0, 1), which leaves its shape as it is.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
cone_point and cone_point_vertex join the sweep now that #3871 reads a cone
bounded only by its base circle up to its point.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Fix recovery for unsaved background tabs (#2884)
* Recovery: scheduled autosaves of every tab write on the worker thread (#2884)
After #3139, Autosaver::save waits for the write and Autosaver::tick runs
it on a worker. The scheduled path now goes through tick (which skips a
tab whose previous write is still running), so several dirty tabs no
longer write on the window thread each interval (#3044). save stays for
Save, close and shutdown. Each tab's autosaver has no clock of its own:
the shared cadence decides when it is due. The tests wait for the
writes a tick started.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Recovery: closing a tab waits for its running autosave before removing the entry (#2884)
With the autosave on a worker thread, discard could remove the files and
the lock while a write was still running; the write then recreated the
entry without a lock, and the next launch offered a design closed with
Don't Save. discard now ends the running write first, as Drop does.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A BREP_WITH_VOIDS void's faces face into the cavity, so it encloses volume of
the opposite sign to the outer shell. Its ORIENTED_CLOSED_SHELL flag alone
doesn't tell: SOLIDWORKS writes .F. over a shell already facing into the
cavity, SolidCraft .T., and flipping by the flag added the cavity's volume. A
void enclosing volume of the outer shell's sign is now turned inside out.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* STEP import: regression sweep over the step-conformance parts of known shape
Fifteen generated parts from step-conformance (CC0): plain, NURBS, seamed,
placed, inch and far-away cylinders, domes, a grooved shaft, a plate with 100
holes, a 3.6 m disc, a 0.5 mm pin, a B-spline half-torus and a washer. Each
reads back with its exact volume, its box and its face count. The half-torus
on a TOROIDAL_SURFACE and the cones to a point are left out until they read
right.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* STEP conformance sweep: the pin's volume as FRAC_PI_8 (clippy)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A cone face written with no vertex at its point has the base circle as its only
loop (or that circle and a VERTEX_LOOP at the point, which has no edges). Its
surface was spanned over the circle alone, a sliver, so a solid cone came in as
a flat disc. Such a face is now spanned to the apex, and its circle is joined to
the apex by a seam line out and back, as a cone with a seam is written.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Boolean: a small hole in a large body's face stays a hole when the face is divided
A boolean divides faces at a tolerance that follows the larger operand (a few
millimetres on a part a few metres long). A small hole's rim, two half circles
when read from STEP, then became two chords: its polygon in the face's
parameters had no area, and the sign that tells a hole from a face piece came
out at random. The hole became a face of its own, inside out, and the face it
was cut in stayed whole. A box cut round such a hole gave a piece without it,
and the shared part of that piece and a pin in the hole came out negative
("malformed") or the boolean failed.
divide_face now divides a loop whose polygon is flat again, finer (down to a
hundredth of the loop's own size), before reading its sign. Only the polygon
that places the loop changes; loops that have an area are untouched.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Kernel: the long-plate hole test in tests.rs; a flat loop's size from its box
Review: the test moves into the kernel's test file (no file of its own),
and finer_if_flat takes the loop's size from the diagonal of the box round
its sample points instead of the largest distance between any two.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Aac7mP9rthK3WsPpPRvCsF
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* claim: #3358
* IGES import: trim curves with reversed pieces no longer drop their faces
A curve on a surface (142) gives each trim loop twice: B in the surface's
parameter space and C in model space, running the same way. Some exporters
write C as a composite curve (102) whose pieces are in order but some of
them end to start. The reader passed those pieces on as they were, so the
loop did not close and the STEP reader left the face out ("face boundary is
not a closed loop").
The pieces of C are now turned to run head to tail, starting at S(B(start)),
the point where the parameter-space curve begins on the surface. That keeps
the loop's sense from the file, so holes still run against the outer loop.
A piece whose ends are one vertex at the sewing tolerance keeps its
direction.
On a 9,276-face trimmed-surface car model, 7,887 faces now import instead
of 3,882.
A surface of revolution (120) wrote its arc generatrix as a whole STEP circle, which
turns into a whole sphere twice over when the arc is centred on the axis; the trim could
not pick the cap out of it and the face meshed as the full ball. An arc in a plane through
the axis now comes in as a sphere or torus, any other arc as an exact rational B-spline of
just the arc.
Hovering an edge lights its circular rim, which starts from the edge's
tangent chain. That looked at every edge of the mesh for each edge in the
chain (and searched the chain itself each time), so a long helical edge in
many pieces, a thread's crest, took seconds: 7.6 s inside the frame on the
first hover of the stem's thread in a real watch-movement STEP file. The
edges are now found by the grid cells of their ends, two join tolerances
across, so each step looks only at the edges that end near it: the same chain
in the same order, in 29 ms for that thread.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2757
* fix(params): autocomplete offers a parameter for any part of its name (#2757)
Typing `side` now offers `cab_side_thickness`: parameters match the word
anywhere in their name, after those that start with it, both in the expression
boxes and in parameters.complete. Functions, constants and units still match
from the start.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
The first hover over a large body made two things in the frame: the body's
pick index (its triangles and edges in hierarchies; about 0.3 s for a long
modelled thread, a quarter of a million triangles) and, with the GPU viewport,
the plan mesh a highlighted face is drawn from (about 0.7 s for the same
thread). The window froze for the sum.
A large mesh's pick index is now made on a worker, and picks scan the mesh
until it is there (a few milliseconds); a body's plan mesh is made on a worker
too (`ModelBody::plan_mesh_soon`), and until it is there the hovered face is
lit on the mesh the viewport already has. Small meshes, and the browser build,
make both at once as before.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* ui: spacing, radius and type tokens, per-tab accents, agent and on-accent tokens
* ui: per-tab accent behaviour and a larger application bar with a Start page button
* icons: silhouette master and flat assistant and agent link icons (open for review)
* icons: apply review notes (sparkle assistant, refined plug, silhouette halo)
* icons: wire the silhouette master to inspect.silhouette; record the accepted icons
* ui: application bar to the design: separators, Save, house, assistant icon, readable hovers
* ui: ribbon structure: Model to View tabs, contextual Sketch and Sheet Metal, migrated pins
* ui: ribbon painting to the compact design
* ui: smaller icon button highlight and assistant mark on the bubble corner
* ui: short ribbon labels that wrap at spaces
* ui: ribbon buttons fit their labels and Finish Sketch shows its full name
* ui: constraint gallery, workspace menu, section list headings and richer tooltips
* ui: lighter selected tab label, ribbon surface and dimmer bar separators match the dark design
* ui: document tabs as rounded chips with a dimmer inactive label
* ui: document tabs fit their names and show close only when active or hovered
* icons: close and plus masters for the document tabs (open for review)
* ui: centre the new-tab plus and overflow button on the bar
* ui: a quiet rule under the ribbon tabs on non-contextual tabs
* ui: the application bar mark separates its faces with the bar colour, not white
* ui: open ribbon section label uses the selected tint and a chevron
* ui: dim the tab close and plus icons until hovered
* ui: every ribbon dropdown uses the chevron, the workspace card included
* ui: the active document tab uses the design's surface colour
* ui: the bug button's right padding mirrors its icon side
* ui: square constraint cells and a rule before the grid
* ui: constraint cells back to their earlier size, keeping the rule
* ui: smaller constraint cells and icons, centred in the section
* ui: constraint cells fit the height of the large buttons' icon and label
* ui: sketch Create titles share the other labels' line
* ui: format the application bar and ribbon changes
* ui: active tab surface token and the colour and hover tests that follow the redesign
* tests: the merged parameter and narrow toolbar tests follow the compact ribbon
* i18n: the Malay Gap label has an English source and its Japanese and Chinese translations
* apply_off_thread tests: the evaluation lingers in tests, so still computing never depends on machine speed
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* perf: renaming a feature evaluates nothing (#3044)
A feature's name labels it and builds nothing, but a timeline rename evaluated
the whole design again (about 70 s on the Prusa Core One assembly, and again
for undo). When the only change is one feature's name, the model renames its
result, sketch and construction geometry in place and takes the renamed
design's fingerprints. A rename other features follow (a plane a sketch is on)
or of a failed feature still evaluates the design.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Guard cheap commands on designs with thousands of bodies, plus a scaling bench (#3044) (#3121)
Cheap commands stay cheap on designs with thousands of bodies, with a scaling bench.
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Save As and Open use the browser's file dialogs where the File System Access API exists; Save writes back to the same file, its handle kept in IndexedDB across reloads. File > Download design added. Other browsers keep storing designs in OPFS.
A closed loop whose joints meet a few nanometres apart along their parallel
axes (faces modelled that far apart) passes the conflict check, but the loop
solver wanted 1e-7 mm before it moved anything. No value of a planar loop
takes such a gap out, so every drive of it stopped at once (stopped_by_loop,
fraction 0) and joint.reach called the loop open.
The solver now closes such a loop as far as it comes and, when what is left
is within the conflict check, keeps it that closed on the way. joint.drive,
joint.reach and the nearby pose of the dof count share it.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Tessellation: trimmed faces of display meshes stay on their surfaces
Display meshes triangulate a trimmed face in parameters scaled by the
surface's speed along each, so a strip stretched along one parameter is
not fanned across, and every boundary segment stays a constraint: one that
crosses another is split where they meet. Boolean meshes keep truck's way.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* Tessellation: one boundary rule for every mesh; drop the #1648 retry
A boundary segment that crosses one already added is split where they
meet in Boolean meshes too, which keeps a pinched face (#1648) whole
without the snap-and-retry. Faces whose segments all go in mesh as
before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* Robustness goldens: seeds 1, 23, 55, 64 and 86 re-recorded on current main for #1653
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
* Thread preview test: wait for the stale fit error to clear before the next preview
After switching back from M6x1.3 to M8x1.3, wait_feature_preview stopped at the M6 error still on
screen when the new preview started a few frames later (seen on Linux CI with #1653's meshing).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
* Robustness goldens: the slow seeds 14, 20 and 40 re-recorded for #1653 from the merge queue
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
---------
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
On an imported part with 2.5 million triangles, every frame of a Section
Analysis drag looked up the picked face's plane several times (each lookup
tessellates the whole body, about 0.1 s) and cut every triangle again for
the outline (38 ms).
- planar_face keeps its last answers for the model state they were found in.
- The outline cuts through a SectionIndex: the mesh's triangles sorted into
slabs along the cut's normal, kept per mesh while the normal stays, so a
cut looks at one slab (0.06 ms on that part) and finds the same segments.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3249
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* inspect.interference/clearance: list mesh bodies as unchecked
Mesh bodies (STL, 3MF and OBJ imports) were filtered out before any pair
was formed, so two overlapping STL imports reported count 0 and an empty
unchecked list, which reads as "no collision". Each mesh body is now
listed in `unchecked` as {a: name, b: null, reason: "mesh body: not
checked"} by both commands. inspect.clearance also stops refusing a mesh
body named in `bodies` or `against` as "no solid body"; it is listed
instead. The parameter docs and the assemblies guide say so.
Fixes#3249
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* docs: separate the mesh-body note from the paragraph above
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #3393
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0158is7fBbAVh538YmVF3MpZ
* Sketch: a radius dimension past an arc's end extends the arc to its arrow (#3393)
A radius or diameter dimension on an arc whose leader is dragged past the arc's
ends had its arrow pointing at empty space. The radial frame now keeps the arc's
span, and an arrow outside it gets the arc continued from the nearer end round
to the arrow tip and a little past it, as dimension drawings do. Full circles,
ellipses and arrows on the arc are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0158is7fBbAVh538YmVF3MpZ
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
While editing a dimension's value, a single click on another driving
dimension now inserts its parameter name (a double-click no longer needed,
and its second click does not insert it twice). A lone value such as `10`
or `10 mm` is replaced by the name; an expression in progress keeps what
was typed and gets the name at the cursor (replacing selected text), so
`10 + ` becomes `10 + d10`. The cursor ends after the name and the field
keeps the keyboard.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2716
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Editing a sketch past the history marker rolls the marker to it (#2716)
Edit Sketch rolled the timeline back to just after the sketch only when
the marker was later. With the marker before the sketch it stayed there:
the sketch wasn't built, so it opened with no palette and couldn't be
edited. The marker now moves to just after the sketch either way, and
Finish Sketch puts it back as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3323
* Contact: a check finds bodies crossing face to face, not only a corner inside (#3323)
The collision probe sampled each tessellation's points (vertices, edge
midpoints, triangle centres) against the other body, so plates crossing
face to face - where the crossing strip is smaller than a face triangle -
put no sample point inside and went unreported while inspect.interference
found the overlap. Where an edge of one body passes through the other's
surface, the middle of the chord it cuts lies inside; its distance from
that surface is the depth. Touching and grazing are unchanged. The sweep's
collision work model pays for the added edge pass.
---------
Co-authored-by: DahakBuilds <dahakbuilds@users.noreply.github.com>
* claim: #2724
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0158is7fBbAVh538YmVF3MpZ
* Sketch: an angle between two lines from one point measures the angle between them (#2724)
An angle dimension on two lines that share an end (two radii of a circle, the
two sides of a corner) took the acute reading of the lines' directions, so
radii 120° apart measured 60° and 150° apart 30°. With no text position given,
the angle is now measured from the shared end towards the lines' other ends,
the reading anyone would mean; the supplement stays the alternative reading.
Lines that share no end, and dimensions placed with a text position, are
unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0158is7fBbAVh538YmVF3MpZ
* ci: re-run the checks; the run on the last push was cancelled
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0158is7fBbAVh538YmVF3MpZ
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A body rename changes names only, but committing it went through the general
"design changed" path: the whole timeline was evaluated again. On an imported
assembly that means reading the STEP file again and meshing every body (about
100 s on the Prusa Core One, and the same again for undo).
When the only change from the previous design is one body rename (the
features' stored body names and the body-keyed maps the rename updates), the
model now renames the body in its results instead: every body keeps its
geometry and meshes, and the fingerprints are set to the renamed design's so
the next evaluation reuses everything. Undo and redo of the rename take the
same path. A rename the model can't follow that way (the body is changed or
named by a later feature, sheet metal, threads) still evaluates the design as
before.
`bench-open --ops script.json` times commands and their undo on an opened
design, with the view work each leaves.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The test could lose a race: with a zero wait the evaluation sometimes finished and applied
before the check that the dialog was still computing. The evaluation is now held at a test
gate until the dialog has been seen saying so.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* WIP: adopt the saved model on open (#3053)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* WIP: tests
* WIP: bench-open opens as Open does
* docs: opening from the saved model
* Tests: an edit after opening evaluates from the feature it changes
* clippy
* Read SOLIDCRAFT_NO_MODEL_CACHE through the env helper
* Open: no picture from the model when the model itself opens; cancel test evaluates
* Saved model: bodies passed through found without writing their record; states list runs of bodies
* Adopted results go into the evaluation cache: an undone edit, or the marker rolled forward, takes them again
* Model cache size guard: the saved model's writer
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3248
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* sketch.create: reject a custom plane origin that is not three numbers
A custom plane {origin, x_dir, y_dir} whose origin was given as an
expression ("w/2") or a number with a unit ("20 mm") silently became
[0, 0, 0], while a bad x_dir or y_dir was an error. A missing origin is
still the world origin; one that is present but is not three numbers now
returns a clear error. Expressions are not evaluated, so the saved
format is unchanged.
Fixes#3248
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* joint.inspect: what each joint is attached to, for agents
A read-only `joint.inspect {joint?}` gives each joint's definition: for each
side the occurrence, its component and path, the snap as saved, and what
each persistent face/edge name refers to (the feature that made it, and the
sketch and curve a swept side came from), whether the origin resolves and
why not; plus kind, offset, angle, flip, values with limits, expressions,
the motion links it is in, and its solve errors and conflicts. Its id ends
in `.inspect`, so agent gates treat it as reading only.
MCP `inspect_design` includes the definitions when the design has joints
(`joints: false` leaves them out). joint.list and joint.inspect share the
motion-link JSON.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* i18n: catalog entries for Inspect Joint (en, ja, zh)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
`sketch.trim.path {points, dry_run?}` walks a polyline and, at each place it
crosses a curve trim may change, removes that piece exactly as `sketch.trim`
would there, in the order crossed and in one undo step. A piece already gone
is not trimmed again; projected geometry, infinite lines, offset outputs and
pieces trim refuses are passed over (reported in `skipped`). It returns each
trim's result and each removed piece as a polyline; `dry_run` reports them
without changing the design.
With Trim on, a press and drag in the viewport draws that path: the pieces it
would remove light up (a dry run of the command) and letting go runs it. A
click still trims the one piece under it.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The merge queue reads the required ci check from the newest run of a commit. Adding any label but
approved-for-ci starts a run whose jobs all skip, and its ci job was skipped too (as
'ci (ignored label)'), so a pull request that had passed could not enter the queue (#1850, #3182,
#3252). On such a run the ci job now waits for the commit's other runs to finish and passes only
when the newest one that ran its checks passed.
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3234
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Sheet metal: refuse an edge flange on a flange's bend line (#3234)
A pick on the edge where a bend meets the plate (or the flange panel)
found a panel edge that is already a bend line and made a second flange
on it, leaving a gap and a sheet of two shells. add_flanges_each now
refuses such an edge before anything moves.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
The test turned a temp path into a link with `file://` + the path, which on
Windows gives `file://C:\Users\...`, not a file URL, so the test failed there.
It now builds `file:///C:/Users/.../my%20part.stl` on Windows (what Explorer
and browsers put on the clipboard) and `file:///tmp/...` on Linux and macOS,
as before. `files_in_text` already reads Windows file URLs correctly; the
fixed test now checks that on Windows.
Co-authored-by: Ray <raykholo@users.noreply.github.com>
Every engine.execute was acted out stop by stop (0.45 s glide + 0.25 s click
per toolbar button and per on-screen point), so a 720-point sketch.line took
minutes while the app answered no other control request. Now the pointer
visits the toolbar button and at most three picks (a long point list is
summarised by its first, middle and last point on screen), and a request is
acted out within 1.5 s whatever its size (the legs get shorter).
engine.batch is no longer refused while the cursor is shown: the pointer acts
out a short summary (the toolbar buttons of up to four of its distinct
commands, within the same budget), then the whole batch runs in one request.
A failed engine.batch now says whether the design really went back
(rolled_back).
(cherry picked from commit 55535e01af45cc9ee7555320863ff5ddcd7c151f)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
BIZ UDPGothic's Cyrillic glyphs are full-width (1.0 em), so promoting the
CJK font to the front of Proportional made Ukrainian text look wide with
large gaps between letters, broke word wrapping, and truncated UI labels.
When CJK text loaded on the web, install_cjk_font hard-coded "ja" and
stayed promoted after switching to Ukrainian or English.
Pass the live UI language to install_cjk_font_for so the CJK face only
leads a CJK UI. Keep fetched CJK font bytes on the web shell and re-apply
fonts when the language changes, and add a regression test that a
non-CJK UI keeps the CJK face as the last fallback.
* claim: #3244
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Fail an unverifiable intersection instead of keeping what it should trim (#3244)
When every route of the general boolean fails, it falls back on the sampled answer
(one body whole, or nothing) with a "not verified ... that part is ignored" warning,
even when the bodies' facets cross well past their faceting error. For an
intersection that keeps exactly the material it was asked to trim: teeth standing
0.27 mm out of a trimming cylinder came back untrimmed, and `solidcraft-cli run`
exited 0 with a part 16.918 mm in radius.
An intersection whose sampled answer is contradicted by facets crossing by more than
twice the depth the warning reports (6 x the check mesh tolerance, three times the
faceting error of two meshes) now fails, and the model is kept. Cuts and unions keep
the fallback: shelling a body in two pieces relies on it (a cavity tool grazing the
other piece), with the same depth and ignored volume as this case.
Fixes#3244
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Leave an empty unverified intersection as it was (#3244)
Only a doubted intersection that keeps one body whole is refused. An empty
one ("the bodies are apart") stays a result reported unverified: the clearance
and interference checks rely on it, and main's new
a_pin_through_a_d_bore_of_its_size_still_overlaps_the_flat (#2714) expects it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3270
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* A body dropped on a component's Bodies folder moves into that component (#3270)
The Bodies (and Sketches) folder row only took items of its own
component, to take them out of groups. A body dragged there from another
component was refused, though the folder looks like the place to drop
it, so moving a body back to a parent seemed impossible. Items of the
same folder from another component now move into the folder's
component, as a drop on the component's own row does.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2167
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Delete Face asks for its faces instead of failing (#2167)
Clicking Delete Face in the toolbar ran face.delete with no parameters and
reported "`faces` must be a list of points or face names", ignoring any
preselected face. It now opens a dialog with a Faces input, like Shell,
Draft and Offset Face, takes the preselected faces, and reopens from the
timeline.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* Canvas and Decal: the file picker opens on PNG and JPEG images
canvas.insert and canvas.decal opened the general Open dialog with no file
type, so it started on "Designs, CAD files and meshes" and PNG/JPEG files
were greyed out. Offer the image types first, as Insert SVG/DXF do (#1801).
* Canvas and Decal: rustfmt the file-picker test
- The repository is tangency-labs/solidcraft (origin and the claim check pointed at the old one).
- PRs land through the merge queue (`gh pr merge N --squash --auto`) after a maintainer adds
approved-for-ci; merging your own PR straight away is gone.
- A PR is rebased onto main, never merged with it.
- New "Before you start": an authenticated gh, the full open PR and issue lists, and a look for the
same feature or the same files before opening a second PR for one change.
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Mouse gestures: a ring of tools on the right button, per studio and customizable
Preferences > Navigation gets a Mouse gestures switch and a Customize window (number of
tools and their place, per studio). Gestures need the SolidWorks mouse mode and turn the
right-drag orbit off. The ring is 280/350 pt, one zone per tool, alternating tints (three
when the count is odd). The radial context menu is unchanged.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Dwcq9eNUMZ1jNeKvWd1bcZ
* Mouse gestures: the ring opens on the right-button press; the right-click menus are off while gestures are on
The ring opens at once on the press, even while a tool or dialog is active. While gestures are
active the viewport's right-click context menu, the right-click that finishes a tool and the
radial context menu no longer fire. With gestures off nothing changes.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Dwcq9eNUMZ1jNeKvWd1bcZ
* Mouse gestures: the ring opens after 4 px with the right button held; a plain right-click keeps its menu
The ring no longer opens on the press. It opens where the button went down once
the pointer has moved 4 points with the button held, even while a tool is in
progress. A right-click that does not move opens the usual context menu (or
finishes the tool) again; the release that ends a gesture opens no menu. The
radial menu stays off while gestures are active.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Dwcq9eNUMZ1jNeKvWd1bcZ
* Re-trigger CI (the run was cancelled when approved-for-ci was added)
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Dwcq9eNUMZ1jNeKvWd1bcZ
---------
Co-authored-by: Claude <noreply@anthropic.com>
* Open: show the design as its bodies are meshed, off the UI thread (#3053)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Open: hold screenshots until the design is meshed; profile lines
* Tests: time-bounded opens
* Open: a Loading parts n/N note while bodies are meshed
* Open: frame the design from outlines while it goes into the window
* Outline bounds: vertices and edges only; test
* Frame watch: the window thread's frame times by open phase (SOLIDCRAFT_PROFILE_FRAMES)
* Open: the picture grows in mesh order and stays once complete; refreshes pace themselves; named frame steps; watchdog test
* Open: outlines measured on the file worker and kept per body
* Open: draw the Home picture under the opened file's own key once meshed
* Format the open wait
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3324
* Sheet metal: refuse a flat pattern that no longer matches the body
A feature that is not sheet metal can change a sheet body's material
without its sheet model knowing: a mirror combined into it, an extrude
joined onto it, a fillet or chamfer on a corner. The flat pattern, the
DXF and the bend table then wrote the sheet as it was before, a part
other than the one in the design, with no warning.
The evaluator now notes, per sheet, the first such feature: one that
changes the body's shape while its sheet model stays the same. That
feature warns, and sheet.flat_pattern, sheet.export_dxf (also with
per_body) and sheet.export_bend_table refuse with "not supported yet",
naming it. A sheet metal feature, or machining the sheet keeps as
cut-outs, rebuilds the body from the model: it clears the note and
warns that the change is not kept. Moving the body, splitting its faces
and cosmetic threads leave the flat pattern as it is.
* Sheet metal: a modelled thread is refused in the flat pattern too
Only a cosmetic thread leaves a sheet body's material as it is. A
modelled thread cuts the thread profile into the hole, which the flat
pattern does not have, and a later sheet metal feature dropped it
without a word. It now marks the sheet like a fillet does: it warns,
the flat pattern, DXF and bend table refuse naming it, and the feature
that rebuilds the body says the thread is not kept.
A loft through two sketches on the same plane built a flat body of zero
volume and reported success, so the mistake surfaced later, if at all
(a measure of 0, or a boolean that did nothing).
`loft_wires` now checks each pair of consecutive sections: parallel
planes with the second section's centroid in the first's plane are
refused with "loft sections i and i+1 lie in the same plane (no volume
between them)". The tolerance scales with the section's distance from
its plane's origin.
Test: two circles on XY; the loft fails with that message and leaves no
body.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
`PlaneRef::walk_exprs` stopped at `Flipped`, so a sketch created with
`flip` on an offset plane (`{plane: "XZ", offset: "d", flip: "x"}`) did
not list the offset as an input. Changing `d` rebuilt nothing that read
the sketch, and the model went stale.
`Flipped` now continues to its base, as `Offset` and `AtAngle` do, in
both `walk_exprs` and `walk_exprs_mut`.
Test: a loft between XZ and a flipped XZ plane offset by `d` has the
volume for d = 30, and for d = 50 after `parameters.change`.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: Convert to Sheet Metal keeps a plate where it is
* Convert to Sheet Metal keeps a plate where it is
Converted without bend edges, a plate moved by one thickness: a 100 x 60 x 2
plate on XY came out at z -2..0 picked on its top face, 2..4 picked on its
bottom. The sheet's frame (its base panel's bottom face, thickness along its
normal) was put on the plate's far face facing away from the picked face, so
the material grew out into empty space.
The frame now lies on the picked face with z into the material, as Convert
with bend edges does. The sheet is the old one moved back by the thickness:
same orientation, flat pattern, flange directions and pocket layers.
A test converts the plate from either face and checks its bounding box; the
engraved-plate test now checks the plate's position, not only its thickness.
* claim: #3376
* Fonts: find a Chinese fallback font on Linux and Windows (#3376)
The Chinese fallback was searched for at a few fixed paths, and in the
font folders only on macOS. On Linux and Windows a Chinese font anywhere
else (Fedora's NotoSansCJK-VF.ttc, WenQuanYi under truetype/wqy/, Source
Han Sans, a per-user font) was never found, so a Japanese CJK font drew
the Chinese interface alone and its simplified characters (设, 语, 实)
showed as boxes.
Search the system and per-user font folders on every desktop system,
three folders deep, and take a font only when it draws simplified
Chinese, so a Japanese-only file with a matching name is skipped. The
"sourcehanss" keyword matched no real file; match SourceHanSans* and
SourceHanSerif*, and Noto Serif CJK/SC.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2672
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Chamfer: one pick follows a straight edge's curved tangent chain at any size (#2672)
chamfer_tool built a single-edge cutter for a straight edge even when the
edge runs on tangentially into an arc (a slot side into its end). The
cutter's overshoot (max(2*distance, 1% of the part size)) is probed for
material beyond the edge ends; on a small part the probe hit the arc's
material and the cutter was refused, so the blend followed the whole
chain. Past a bounding-box diagonal of about 200 mm the overshoot cleared
the arc, the cutter was accepted, and only the straight edge was chamfered,
silently.
A straight edge whose smooth chain contains a curved edge no longer gets a
cutter; it takes the chain-following blend, as at small sizes and as a
fillet does.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
The two reliefs that would overlap are still resolved one after the other: the
second one is left out, with a note. The corners come out in flange-index
order, so which of the two keeps its relief depends on the order the flanges
were made in.
Two corners that share an edge never come this far: each relief reaches
exactly r along that edge, and a relief is refused unless 2r fits in it. Two
corners that do not share one can: the square relief at a corner of 120°
reaches 2r from it along the bisector, so at the two 120° corners opposite
each other in a rhombus the two reach each other.
Two reliefs that would overlap are now left out together, each with a note
naming the other, so that which of them is cut does not depend on the order
the flanges were made in.
Fixes#3417.
Co-authored-by: jeremymiribung-blip <jeremymiribung-blip@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
* claim: #3267
* fix(view): a finished sketch shows its standalone points (#3267)
Points were drawn only for the sketch being edited, so a point placed for a
hole vanished once the sketch was finished although it could still be picked.
The standalone points of every finished sketch shown are now drawn.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Moving an occurrence places every body under it again, and each placed copy
searched every revolved face for where it sits to turn it away from its seam,
although its source had been turned already. On the Prusa Core One assembly
that search was most of the 12.7 s a move of the top occurrence took. A body
now records that its revolved faces are turned, and world placement
(`place_matrix`) skips the search for such a body. Modelling moves (pattern
copies, hole tools) keep `transform_matrix`, which still turns them in the
copy's own frame: where a face wraps all the way round the turn it finds
depends on those coordinates, and later booleans depend on it.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Later instances of a part showed a mesh shared through the first instance,
but the first instance meshed its own, so every repeated part was meshed
twice. The first instance of a part placed more than once now shows the
shared mesh too; a part placed once keeps its own (no second copy).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Autosave serialised and wrote the whole design on the window's thread: about
120 ms on the Prusa Core One assembly (its 61 MB STEP text inside), a hitch
once per interval after every change, and more once saves keep the evaluated
model. A tick now takes a snapshot of the design (its Arc and what is hidden)
and writes it on a worker; a later tick takes up the outcome, and no new write
starts while one runs. Crash replays rebase onto the autosave only when the
design is still the one written. Save and close still wait for the write, and
dropping the autosaver lets a running write end.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3198
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Look At frames the sketch's geometry, not its plane's origin (#3198)
Look At on a sketch (its menu, Ctrl+8, the navigation bar, entering a
sketch) centred the view on the sketch plane's origin and kept the zoom,
so a sketch drawn away from it - an imported part placed far from the
origin - ended up out of view. Both paths now share onto_sketch: centred on
the sketch's bounds, zoomed out when they would not fit (never in); an
empty sketch keeps the plane's origin.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Register the Look At tests
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Look At tests: Vec3::len
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The retries in boolean_whole_inner mesh every candidate to compare its
volume. A nudged union at the coarse tolerance could come out with a face
on a cone whose trim loop ran metres along the cone's continuation (edges
up to 12.7 m outside a 155 mm part); meshing it took seconds to minutes per
face and once never finished, though its volume (+-1e10 mm3) failed the
check anyway.
check() first samples the candidate's edges (16 per edge) and discards a
candidate with a point further outside the operands' mesh box than 5% of
their size (or a non-finite point): no edge of a union, cut or
intersection can lie there. The reason goes into the new
BooleanDiagnostics::rejected (omitted from JSON when empty). Results of
sound candidates are unchanged.
On the reporting design the PitchTail join drops from 18.7 s to 4.3 s with
the same result; the regression tests synthesize a cone face whose loop
runs off its surface.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Spike: typed parameter contracts and JSON Schemas for solid.box and solid.extrude
The schema RFC's spike (plan/schema-spike.md). solid.box and solid.extrude get
typed contracts with generated JSON Schemas (2020-12), checked in front of the
unchanged handlers by a per-session level: off (the default, which changes
nothing), warn (the call runs and gains param_warnings) or strict (the call is
refused and the design left alone). Also engine.describe and the MCP tool
describe_command, `solidcraft-cli mcp --param-check off|warn|strict`, a
FeatureResult type for feature replies, and an env-var audit recorder
(SOLIDCRAFT_PARAM_AUDIT) used to check the contracts against every test.
Local spike branch; not for upstream as is (see the results report).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Parameter contracts: put the audit recorder behind a feature; drop two ignored test keys
The audit recorder is now on only in a build with the engine's `param-audit` feature
and `SOLIDCRAFT_PARAM_AUDIT` set. Its code is still compiled, linted and tested without
the feature. `scripts/param_audit.py` summarises a test run's audit logs, and the schema
module's docs give the steps for converting a further command.
Two tests passed keys their command ignores, which the audit found: `position` on
`solid.box` (`hole_face_tests.rs`) and `symmetric` on `solid.extrude` (`face_tests.rs`).
Both are removed, so the tests build what they always did. Comments no longer call
the contracts a spike.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
truck placed a face that no intersection divides (every face of a shell
that meets nothing of the other) by counting the faces a ray crosses,
taking each face's surface as stored. A revolved body's reversed faces
counted the wrong way, and an inward shell (a void, a cut's inverted
tool) was read as a solid: a cut beside a sealed void came out empty,
refused or, with a tool whose bounding box covers the target, taken with
the body gone. Intersections of such bodies came out empty as well.
In an AND the crossings now count as each face is oriented, and an
inward shell holds what lies outside it. An OR keeps truck's count.
The corrected cut let a folded offset through: an arc edge next to a
face it meets tangentially had its ends moved past its middle, and the
arc went the long way round (robustness seed 73's shell). The offset
refuses an arc whose middle crossed its chord, as it does a straight
edge that turned round.
Goldens: seeds 0, 45 and 77 had a cut refused as empty that now takes
what it should; their volumes are checked by hand in the PR.
A face pinched to a point, where its boundary touches itself or another of its loops within
TOLERANCE (a sphere touching a box's edge leaves one on the box), had a constraint refused by the
CDT, and the fallback cut across the face: it came out with no triangles or without those at the
pinch. When a constraint is refused, the vendored truck-meshalgo now puts every boundary point
within TOLERANCE of a segment into it (at most 8192) and triangulates the face again. Faces whose
segments all go in mesh as before.
The regression tests live in crates/kernel/src/thin_face_tests.rs, and the #1648 property test in
boolean_props.rs is no longer ignored. Robustness seeds 20 and 55 are recorded with their Linux
values.
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #2678
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* WIP: timeline.dependents shares the evaluation cache (#2678)
The check before a delete evaluated the edited design on a fresh model:
it recomputed the whole timeline and left nothing for the delete. It now
evaluates on a copy of the session's model. Adds Ahead::replies for
running such checks off the window thread (the UI part is still to come).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Delete: check what a delete breaks off the window thread (#2678)
Both checks a delete makes first (timeline.dependents for the timeline's
Delete, the selection.delete dry run for Delete on the selection) evaluate
the history after the deleted feature, on the window thread. A part with a
slow feature froze the window for as long as everything after it took
(5.9 s in the report). They now compute off the window thread like undo and
the history marker: within the app's 150 ms wait they carry on at once,
otherwise the window says it computes and the delete (or the confirmation)
follows when the report is in. A design changed meanwhile is checked again;
a second Delete meanwhile is left out.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* ci: run the checks again (the run for 3b94e8e0 was cancelled when the PR left draft)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* kernel: a blend runs along chains of several faces and lines closed rims up (#764)
A blend's chain may run along the boundaries of several faces (a tube's rim
along the halves of its wall): each edge continues from its own best-facing
face, read station by station. Before, a chain had to be the edges of one face,
so every rim of a revolved or cylindrical surface was refused.
Two closed chains blend from matching seams: the second chain's start (searched
by arclength, then narrowed) and direction are the ones that put each station
closest to its partner, so rims whose seams sit apart no longer twist. A
picked `seam` point on each chain fixes the start instead. A closed chain
against an open one is refused. Closed chains leave towards each other's
centroid, not a point on the rim.
Stations are read on the edges' own curves, not on the chords of the length
table (they sat up to the chord's sagitta inside a curved edge), and the
reported gap is measured to the curves.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* kernel: closed blend chains run round the same way; Hermite strip and seam tests (#764)
The second of two closed chains runs round the same way as the first (by
their area vectors), so a picked seam turns the blend by just the angle
between the seams instead of folding it over. The nearer pairing decides only
when the loops stand almost at right angles.
Tests for the issue's gates: the blend between two offset box rims is the
analytic cubic Hermite strip (control points from the chord) and keeps the
top faces' normal along both seams; picked seams a quarter apart turn the
blend a quarter round.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* surface.blend: `seam` picks where each chain's stations start (#764)
A point on each chain, saved with the feature (one pending format change),
moved with the design like `align`, and refused when it is not two points.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* surface.blend: a test between two tube rims, their seams picked and saved (#764)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* Blend dialog: edges at each end, continuity and bulge per end, stitch (#764)
SURFACE › CREATE › Blend Surface opens a dialog: the edges the blend leaves
and the ones it arrives at, each end's continuity (G0–G3) and bulge, and
Stitch to both bodies. Edit Feature brings it back and keeps what the dialog
does not show (align and seam points). Labels in every locale catalog.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* ROADMAP: M8 Blend runs along several faces, pairs closed rims and has a dialog (#764)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* kernel: a blend station on the vertex between two support faces is read on the neighbour (#764)
A station where a tube wall's two halves meet sits at a corner of one half's
parameters; with a leave direction that turns round the wall (tubes off each
other's axis) the shape could not be read there and the blend was refused. The
neighbouring chain edge's face holds the same point and reads it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* blend: review clean-ups (#764)
- ChainLine keeps one list of samples instead of four parallel lists.
- frames_on and boundary_gap take the chain's ChainLine instead of rebuilding it.
- A station is read on the neighbouring face only when it sits on the vertex the two edges
share, each face once, and each face keeps its own walk.
- The seam rule (a closed chain starts at it, an open one runs from the nearer end) lives in
ChainLine::start_at_seam; pair_with reuses the start of its both-ways search.
- support_faces projects with curveblend::project; loop_area sits with plane_through.
- surface.blend parses `align` and `seam` through one point_pair; the Blend and Ruled dialogs
share body_and_edges.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
---------
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A cut whose tool ends exactly on a face of the target (a hole cut from a pocket's floor up to the
plate's top face) could come out of the general boolean with the target's face whole over the hole
and the tool's end face under it: a skin of no thickness that keeps the hole closed. No volume
shows it and the sampled membership checks see the hole either way, so the result was kept. The
result check now refuses a body with a planar face lying back to back on another of its own planar
faces with nothing behind part of it (`coplanar::has_skin`), and the retries take the cut.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #3254
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Preview: nothing is drawn once the current inputs fail (#3254)
The last good preview stayed up when the current inputs made the feature
fail, so the view contradicted the dialog (Extrude at 0 mm still showed
the 1 mm extrusion beside its error). Once the result for the current
inputs is an error, the preview scene is hidden and the bodies it stood
in for come back; while the next preview computes, the last good one
still stays up as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Extrude preview: a zero distance is refused at once (#3254)
The straight extrusion feedback gave up on a zero distance, so the
preview built a whole scratch design only for the kernel to refuse it,
and the old preview hung on until then. It now returns the extrusion's
own error straight away.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Iterate instead of indexing in nine loops
Replace index loops that look up neighbours by position with enumerate,
zip(cycle().skip(1)) for wrap-around pairs and zip(skip(1)) for
consecutive pairs. No behaviour change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* geom: ring_edges for the edges of a closed ring
Every walk over a polygon's edges, the last back to the first, spelled
x.iter().zip(x.iter().cycle().skip(1)); eighteen of them now call
geom::ring_edges. Three redundant take(len) calls go with them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* geom: Ring, a closed ring that knows it is closed
A Ring holds items in order around a closed polygon, loop or fan; edges()
walks each with the next and the last with the first, corners() each
with both neighbours. Ring<Vec2> has signed_area and even-odd contains.
Loop2::polyline returns a Ring, as do doc's face_loops and
boundary_loops, kernel's coplanar wire samples, CapLoop's segments,
tess pieces' rings, the gizmo's plane quads and the navigation cube's
facets. Their walks take edges() or corners() instead of wrapping an
index; sketch's text-hole test and doc's poly_area become contains and
signed_area. Loop2::joints pairs each segment with the next.
Replaces the ring_edges free function.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* geom: Chain, Ring and Path; bounds for a chain of points
Ring becomes Chain<T, Closed> beside Path = Chain<T, Open>: edges()
closes a ring and stops at a path's end, Path::close makes the ring,
and bounds() (the empty box for no points) serves both. Eight
hand-rolled min/max folds over a loop's polyline call bounds(); doc's
flat_polyline returns a Ring.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* geom: area and inside on Ring for any planar point
Ring's signed_area and contains take any Planar point (Vec2 or
[f64; 2]). The cam profile in joints_cam is a Ring: its own even-odd
test goes, and its edge walks, gap and area take edges(). face_params'
loop area, the text path's winding, tess pieces' ring area and the
copies in three tests call signed_area or contains; the formulas are
the same term for term.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* doc: sheet's closed outlines go through Ring
dedup takes and returns a Ring (dropping a last point back on the first
is what makes it a ring's), near_outline walks a Ring's edges, and the
copy of it beside fold checks calls it. point_in goes: its callers ask
the Ring. Its formula anchored each crossing on the other end of the
edge, so the two can differ only for a point on an edge to the last
digit; one caller sets such points aside first, the others test a
region's interior point, a cut's midpoint or another outline's corners.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* doc: cam tests build their outlines with Ring::from
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* geom: area_vector on Ring<Vec3>; three Newell sums use it
Ring<Vec3>::area_vector is half the sum of a × b round the ring. heal's
loop_area, lip's outer-loop turning and the revolve's loop orientation
summed it by hand with a wrapped index; they take it from the Ring
(the half is exact, and two of them only read its sign).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* geom: Aabb3 from points; boxes built and asked through it
Aabb3 collects from points (skipping non-finite ones, as add does) and
gains inflated, contains and meets. Fifteen hand-rolled min/max folds
over points or boxes in kernel, doc and engine become collect or union,
and their "within tol of the box" tests become inflated(tol).contains.
blend's refused-edge boxes and naming's face index hold an Aabb3.
A NaN point is now outside every box, where the old negated
comparisons let it in; an infinite point no longer stretches a box.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* geom: Aabb2, the plane's box; 2D folds and box tuples use it
Aabb2 mirrors Aabb3 (collect, add, union, inflated, contains, meets)
and adds relative, a point's place in the box. sketch's link and
link_loops each had that formula; link_loops' boundaries, prism's
segment and loop boxes, profiles' shape boxes and grid, and ends'
segment box hold an Aabb2 instead of a (low, high) pair. Folds in the
cam, sketch bounds, sheet's perforation grid and the sketch clipboard
collect one.
Boxes are built from the same terms (a round's box keeps c - r - tol).
As with Aabb3, a NaN point is outside every box and an infinite point
no longer stretches one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* kernel: prism's segment boxes are plain Aabb2s
The SegBox alias named a (low, high) pair; Aabb2 says it itself.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* geom: Chain::bounds is an Aabb2; the boxes #2382 left as pairs follow
Chain::bounds returns an Aabb2. Its callers keep it: sheet's hole and
cut boxes (loop_box), letter boxes, island outlines and line-near-cut
test use meets, contains and inflated; profile's outline overlap and
max_clearance, profiles' box tests and the cam test read its corners.
flat_size and the flat pattern's JSON size are the union of the loops'
boxes (0 × 0 for no loop, where the pair fold gave -∞).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* geom: Aabb3::intersection; hand-grown and hand-met boxes use the box
Aabb3::intersection is the box both hold. ops' four overlap boxes of a
boolean's operands take it, and seven boxes grown by hand (ops' body
boxes, the distance set's bounds, coplanar's face box, eval_more's
padded mesh box, compare's group and region boxes) call inflated with
the same amount on every side.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
A line and a circle or arc can now be dimensioned to the circle's edge
(ConstraintKind::LineCircleDistance, saved as line_circle_distance): the gap
while the line misses the circle (0 is tangency), or how far the edge reaches
past the line when it cuts through (stored as `inside`).
Rebased onto main as one commit (the branch's own changes; its merges of main
dropped). Conflicts resolved: the test mod lines in cmd/sketch.rs (both kept,
sorted), and README.md, whose feature list main moved to docs/capabilities.md;
the line-to-circle dimension is listed there now.
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Co-authored-by: jyviko <jyviko@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Preferences: a Plugins page lists installed plugin folders
Each folder under the per-user plugins directory that has a plugin.json
appears in Preferences. Turning one off is saved with the other settings.
The page reads the manifest and does not run any code from the folder.
* Retrigger checks after the cancelled run
The required ci check failed because the workflow was cancelled before
any job started. This commit does not change the plugins page.
* claim: #3087
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Viewport: the ground grid is centred where the view's axis meets the ground (#3087)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Grid: its centre stays within reach of the point under the target
At a grazing slant in perspective, where the view's axis meets the
ground runs far off, and the ground in front of the camera had no grid.
The centre now moves no further than the grid's half-size from the
point under the target (review of #3124).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Two empty numerical cut attempts stopped the boolean retry sequence before
the exact geometric strategies ran. When the result guard refuses that empty
candidate, the exact strategies (sliced planar, pushed walls, local prism,
sphere lens, seam halves) now try the cut first; if none recovers it, the
original empty candidate and diagnostics are kept, before any displaced
numerical retry.
Rebased onto main as one commit (the branch's own changes, its merges of main
dropped). The only conflict was the new engine regression test next to
main's #1985 STEP export tests in boolean_regression_tests.rs; both are kept.
* kernel: loft guide rails (#201)
A loft follows one or two guide rails between its sections: stations are added between each two
sections, each the plain loft there moved so that the point a rail guides lands where the rail
crosses the station's plane — turned about the loft's direction and scaled (or stretched across
to the rail) with one rail, carried, turned and sized between two. The sections and caps stay
exact; the smooth loft runs through sections and stations. Solid and sheet lofts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* doc, engine: solid.loft and surface.loft take guide rails (#201)
`rails` (one or two {sketch, curves}) and `rail_fit` (scale | stretch), stored on Loft and
SurfaceLoft (one pending format note), parsed by the sweep's rail parser (now shared) and built
through the kernel's railed loft. Rails on a loft to a point, or on a surface loft with edge
sections or end continuity, are refused as not supported yet. The rails' sketches are walked as
the feature's references and their components placed like a sweep's.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* ui: Rail, Second Rail and Rail Fit in the Loft and Surface Loft dialogs (#201)
The rail inputs and fit row the sweep dialogs have, shared through dialogs_sweep; Edit Feature
brings the rails back.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* docs(roadmap): loft guide rails (#201)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* loft rails: review clean-ups (#201)
- Sweep and loft rails share one turn-and-size map (`sweep_path::rail_frame`).
- A station's middle and the loft's direction come from one curve through the sections'
middles (exact derivative) instead of two more plain-loft evaluations per station; at most
33 wires, so railed faces stay light.
- Nearest points on a section use `param_on` (kept only where the search comes nearer); the
size uses `Aabb3`; one B-spline builder; anchors keep their nudge.
- The railed and plain lofts share one path in `loft_with_rails` and `loft_sheet_with_rails`.
- The engine leaves the not-yet-supported refusals to the design's evaluation; both lofts start
their rails from the first section's first point.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* loft rails: refuse rails on sections with holes, with a test (#201)
Main's lofts now take sections with holes (#2447); rails shape only the
outlines, so a railed loft of rings is refused until the rails carry the
holes too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
---------
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Adjacent \X2\...\X0\ runs, as step_out writes one per character, now
decode: the backslash opening the next run was pushed as a literal, so a
name exported as 圆角 came back as 圆\X2\89D2\X0\ (also every linked
component insert, which round-trips through STEP).
STEP and IGES bytes that are not UTF-8 are decoded in their detected
legacy encoding (chardetng + encoding_rs) instead of from_utf8_lossy, so
GBK or Shift_JIS names from vendor files no longer turn into U+FFFD.
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #2448
* Sketch: a click that moves on the way places its point (#2448)
egui reports a press that moves past its click distance, or is held past its
click time, as a drag, not a click. The viewport only placed points on clicks,
so a hand that moved while clicking placed nothing with the Slot, Line, arc,
three-point, Ellipse, Spline and the other tools, and on the first point of a
Rectangle, Circle or Polygon the few pixels became the press-drag-release
gesture of #1630 and drew a sliver.
- Any tool's press that turns into a drag (no navigation, no Select Other
hold) clicks where the button is let go.
- A Rectangle, Circle or Polygon drag let go within 24 screen points of its
press is such a click too: the first point goes where it was let go and the
shape goes on by clicks. A longer drag still draws the shape (#1630).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Model cache written by default: exact bodies, stored once, scaled caps (#3053)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* bench-open: say why bodies were not kept
* Exact planes and revolutions; budgets for large bodies (#3053)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Native B-rep: rebuild shells as the kernel made them (closed and seam edges) (#3053)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* docs: the model cache and the exact native B-rep (#3053)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* bench-open: time a save without the cache too
* Model cache entries are Deflate-compressed (#3053)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Model cache: one that does not fit the design file is left out, never a failed save (#3053)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sheet metal: a corner relief no longer depends on the order the corners are cut in
The corner reliefs are planned against the base as it is before any of them is
cut, and the two reliefs at the ends of one edge are compared with each
other. Before, every relief was cut into the base as it went, so the corner
after it saw the relief of the corner before it: the square relief reaches
further from its corner than the round one does (4 / cos(α / 2) against 4), so
on a short edge it was left out where the round one was cut, and which of the
two corners kept its relief depended on the order the flanges were made in.
The reliefs themselves are unchanged: a round about the corner through the
angle the base fills there, or the square with that side (a kite where the
corner is not square, its cuts square to the two edges). What is new is that
they are planned first and cut together, so that a corner's relief depends on
the base and on the other reliefs, and not on the order they are cut in. Two
reliefs that would overlap are still refused, now with a note that says so.
Covered: the four relief shapes at corners of 60° and 120°, with the flanges
of equal and of unequal height, through the kernel and through the engine
(where the flat pattern and the DXF keep one clean loop).
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Sheet metal: cover the corner reliefs at a corner that is not square
The engine and the UI had no test at all for a base whose corners are 60° and
120°: the flat pattern, the DXF, the unfold and the rule manager's own square
relief were all untested there. They work; this says so.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Sheet metal: say what happens when a corner relief is left out
The guide now says that the reliefs at the corners of one edge are cut
together, and that two of them that would overlap leave the second one out.
A relief that cannot be cut is reported, not dropped.
Co-Authored-By: Claude Code <noreply@anthropic.com>
* Sheet metal: write the corner relief's expected corners out
The engine test hard-coded the corners of the trimmed base; it works them out
instead, and the numbers the test and the sketch share are written with the
precision they need.
Co-Authored-By: Claude Code <noreply@anthropic.com>
---------
Co-authored-by: jeremymiribung-blip <jeremymiribung-blip@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
Converting an assembly went one occurrence at a time: a part's solid was read
from the file and oriented on the calling thread, and only runs of four or
more solids in one representation used the workers. Each step now takes the
next solids across occurrences and parts (up to 64, stopping where a part not
yet oriented comes back), and the workers read and orient the new parts side
by side, each taking the next one when it is done. Later instances still only
place the part their first instance oriented.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
World placement already shares a part's reference mesh between its
occurrences (#2924), but the display plan the GPU viewport draws was made
again for every occurrence from its placed B-rep: planning every face of
every instance of the part. A placement now carries the definition's plan
cells too, and a placed body's plan (and its CPU mesh) is the definition's,
moved by the placement's matrix (`TessPlan::transformed`): the part is
planned once however often it is placed.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
`a_long_message_keeps_the_buttons_in_the_window` failed now and then on CI
(Send 914.5 -> 888.5, also on #2893): looking for agents runs on its own
thread, and what it finds (and the status it reports) can land between the
two measurements and change what is around the panel. The test now waits for
the search to finish and measures the buttons from the panel's bottom, which
is what the box growing upwards keeps fixed.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Save ran on the window's thread: about 130 ms on the Prusa Core One design,
and more than a second once saves keep the evaluated model. The toolbar's
Save and Save As now run the command on a scratch copy of the session on a
worker; the design waits for it as it waits for a file picker (commands are
refused meanwhile), and the outcome is taken up as running the command here
would have left it: the name and links the save gave the design, its file,
its saved state (`Session::adopt_save`), the echo, the error status, the
recent files and the home thumbnail. A tab closing or the app quitting after
Save waits for the write, as does the app's exit. Scripts, the control
channel and the browser build still save in place.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* inspect.clearance: pairs whose bodies did not change are not measured again
Each clearance check measured every close pair again, though between two
checks usually only a few bodies change or move. Results are now kept per pair,
keyed by both bodies' geometry and placement and the check's settings; a pair
whose two bodies are unchanged is taken from that cache. With 40 overlapping
parts, a check after moving one of them took 0.8 s instead of 23 s (741 of 780
pairs reused). Pairs that could not be worked out are tried again each time.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* inspect.clearance: boxes and placed keys exactly as before the reuse
The pair boxes are taken over the mesh's triangles with finite corners, as a
distance shape's are, so pruning is unchanged. A placed body's identity now
includes its model body's own frame, so two placements are never mistaken for
one another.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* inspect.clearance: a large body is meshed once for all its pairs
Each pair that touches cuts the larger body down to the pair's region with a
kernel boolean, and that boolean meshed the whole body twice: once for its
checks and once inside the intersection. A sheet with 600 faces paid about
2.5 s per pair for meshes that were the same every time.
A body can now keep its meshes for reuse (Body::keeping_meshes); its booleans
take the meshes from there, and the intersection takes its shell meshes
from there too (truck-shapeops: and_with_history_meshed). Clearance uses such
copies for one check. Results are unchanged to the bit. On a 3 x 2 m sheet
with bends and 300 holes, the clip took 0.23 s per pair instead of 2.6 s, and
24 rivets took 5.6 s instead of 16.5 s.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* inspect.clearance: pins on one cylinder with their hole are answered or reported in seconds
A pin seated line to line in its hole, on its axis, and with a blend running into its wall
(a fillet under a head) sent the boolean through minutes of retries that ended unchecked; a
pin a hair off its hole's axis was passed as clear. Clearance now asks first whether the pair
holds such a pin (from the faces' surfaces alone). On its axis the boolean gets a few tries,
which answer plain pins at once; a hair off, or when those tries fail, the shared volume is
worked out apart by coaxial_overlap, which now also takes the zero-offset, equal-radius case,
moves the hole's wall when a blend keeps the pin's from moving, and checks its band more
tightly. What neither answers is reported as unchecked, naming the pin.
The kernel gains with_attempt_budget: the booleans run inside it get a number of tries of
the kernel's own intersection between them, the same on every machine.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* inspect.clearance: the pin tests sit beside the other measured pairs, not at the file's end
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* inspect.clearance: the bodies left whole meet their booleans with their meshes kept
A body that meets several others at one size (a bracket with balls pressed into it) was
meshed again for each pair's intersect: only the clipping used the copies that keep their
meshes. The intersect, the pin work and the union check now take those copies too.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Near-coaxial overlap: ten tries per intersection, enough for a flush head pulled back
#1984's headed pin made before its plate needs more than eight.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Aac7mP9rthK3WsPpPRvCsF
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Fix large modeled threads and thin crest reconstruction
Helix stations follow the pitch (radial chord error under 1% of it, 48 to
720 stations), and the faceted rebuild retries at the BSP's precision when
the usual weld tolerance collapses a thin crest facet.
Rebased onto main as one commit (the branch's own commits touched UI files
that later moved to #2263 and #2264). Adapted to main's UnitVec3 thread
axis: `*k` in thread_groove and `*cylinder.axis` in the new flange test.
* Fix icon review revisions for same-timestamp edits
* claim: #2430
* feat: copy saved designs between documents as linked components
* fix: keep occurrence movement controls with their document tab
* test: cover linked design clipboard across tabs
* fix: place linked design pastes at the viewport pointer
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
#1949 joined the straight runs of a profile into one side face for straight
extrudes; tapered extrudes (extrude_tapered, extrude_tapered_between) still
sloped each piece as its own face, so a drafted mirrored profile kept the
edge. They now join the runs the same way (NumberedLoop::merged_lines), the
sloped side generated by every piece under it.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #2301
* fix: resolve face context actions from persistent face provenance
* test: cover feature targeting from picked faces
* fix: reveal picked face creators in the Browser
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Boolean: a pin in a bore of its own size is clear of it, whatever its facets
A pin whose wall is in other faces than the bore's (turned from a profile: three
faces; the bore two) is faceted at other angles. Where one body's faces end on
the shared cylinder (the pin's wall through the end face of the tube, the pin's
end inside the bore) the facets cross by the facetings' error, so the sampled
shortcut doubted its own "apart" answer, every retry failed on the coincident
walls, and the result came back unverified after seconds: inspect.clearance
left such pairs unchecked ("could not be intersected there").
The shortcut now leaves out crossings that lie on a cylinder both bodies have
faces on (axes and radii within the kernel's same-surface tolerance, where both
faces run). A pin a hair off its hole's axis, or one pushed into a narrower
bore, is still not taken for apart.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Boolean: skip only the shared cylinder's own crossings, not a flat along it
The shortcut left out every crossing near a cylinder both bodies have faces
on, from any face: a pin through a D-bore whose flat is 0.022 mm inside it
came back as contact. Mesh::crosses_except now hands the predicate the two
triangles' faces, and only crossings between the faces on the shared
cylinder, or between one of them and an end face across its axis that keeps
to its body's side of the cylinder (within the facetings' error), are left
out. The D-bore's end face reaches into the cylinder to the flat, so the
pin's wall through it counts again; a flat, key or rib along the axis always
counts.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Boolean: a pin in a bore of its size is clear at every placement
The sampled shortcut's check that no facet has material under it on the
wrong side of the other body still caught the facetings of a shared
cylinder. A facet of a pin's end inside the bore (or of the bore part's
end face round the pin) with all three corners on the rim has its middle
just inside the cylinder. Where the other body's chords reach further in
at that angle, the point under it lies inside the other mesh, though the
bodies only touch. Which facets do so depends on the face splits and the
placement, so the same pair read as clear at one placement and as "not
verified" at another.
Such points are no evidence now when they lie on the shared cylinder,
under a facet of one of its end faces: the end faces and band the
crossing check already leaves out. A pin 0.02 mm larger is no shared
cylinder, and its overlap is still found.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
An M8 modelled thread in a 6 mm hole, narrower than M8's minor diameter, was
refused as "the face is too far from the thread's diameter". It now says what
is wrong and what fits, as a hole that is too wide already does (#995): "this
6 mm hole is too narrow for a modelled M8 thread (its teeth would fill the
hole); drill 6.75 mm for M8", with the thread that fits this hole when there
is one. A shaft too thick for its thread gets "make the shaft 8 mm for M8".
Fixes#2688
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #2283
* inspect.clearance: a point straight past an edge's end no longer passes as on the face (#2283)
A witness point on a face's surface is kept only if it lies on the face
itself; near the face's boundary the nearest edge decides which side it is
on. A point straight past the end of an edge (on the line through it) was
meant to be left to the other edge meeting there, but "straight" was an
absolute 1e-9 of the body size. With a slightly tilted axis (1.5e-6 rad, as
a STEP round trip leaves it) a point 0.2 mm past the end of a knuckle's seam
edge is 4.5e-6 mm off that line, so its side came down to rounding and the
point on the knuckle's extended cylinder, level with the frame band's rim,
was taken as contact (0.000) at some angles.
Straight along is now relative to the distance from the edge (1e-3 of it),
so the rim edge meeting there decides. Regression test: a hinge knuckle
turned 30-80 degrees between two bands reads 0.2 mm at every angle, with
both witness points on the bodies (41 degrees read 0.0 before).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Joints: a point origin keeps the x it is given
`joint.origin` and `joint.create` take `x` for a point origin (a
direction, laid square to z) and the point snap saves it. Two origins
given at one point with the same z and x now meet as they are: a part
turned a quarter turn about z no longer turns back (and moves) when
rigid-joined there. A point origin without x takes its x from z alone,
as before. An x along z, or of no length, is refused.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Joint relations test: a point origin has an x too
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Aac7mP9rthK3WsPpPRvCsF
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Browser model: find the placement a body is in
Model::body_path maps a placed body (by its world name) to the occurrence
path of the placement holding it, as the browser's occurrence rows nest.
Every placement is walked once per model, from the placements the
browser already caches, so asking it every frame stays cheap. The next
commits use it to mark the rows leading to a body.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Browser: rows can carry a mark for a lit item and its branch
A row's Mark draws a bar down its left edge: a solid one on a lit item's
own row (which also takes the hover colour unless it is selected), and a
thinner, fainter one on each row leading to it, so the bars trace the
item's branch through the tree. Nothing sets a mark yet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Browser: light the rows of what is selected or hovered in the 3D view
The tree now follows the 3D view. Whatever is selected (in the tree or
in the view) or under the pointer in the view lights its row: the
selection colour when selected, the hover colour when hovered, with a
solid bar; each row leading to it gets a faint bar, also when the
folders are closed.
- A body lights for itself and for a selected or hovered face, edge or
vertex of it.
- A sketch lights for its curves, points and profiles, including those
of the sketch being edited.
- Construction planes, and the origin planes and axes, light their rows;
the Origin folder carries the faint bar.
- The rows leading to an item are the design, the placements above it
(by the placement path of a body; for a sketch or plane, every
placement of its component), its folder and its group.
A row's own selected flag, which drag and the context menu go by, is
unchanged: the selection colour for things picked in the view is drawn
separately.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2920
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Parameters window: the add row lines up with the table's columns (#2920)
The add row was a row of its own under the User Parameters table, so
its fields did not sit under the headings, least of all before the
first user parameter, when the headings had no rows to size their
columns. It is now the table's last row. Its text boxes take the rows'
widths, since a grid cell otherwise gives them only the column's width
so far.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: kezmdor <15539254+kezmdor@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* sketch.edit_text: several texts in one edit, one rebuild
Re-texting the labels of a design took one sketch.edit_text per label, and
each one rebuilt every feature after the sketch (seconds per label in a large
assembly). sketch.edit_text {edits: [{link, text, ...}, ...]} changes them
together in one command: one rebuild and one undo step, all or nothing.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* sketch.edit_text: a live frame takes the edits list too
The live path (live: true) still read one text's link and refused a list.
Both paths now go through one helper.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Add body_name and targets to the solid.box and solid.cylinder rows, the
bodies form of solid.mirror and the names its bodies get, unit-suffixed
parameter values, and the set_parameter argument shape.
Translate 21 messages while keeping Italian partial and preserving existing translations and catalog guards.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Blacksmith's Windows runners start the app but never present a frame (no interactive desktop),
so the startup smoke failed on main after #3343. The shader job stays on Blacksmith.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* CI: merge-queue groups run on Blacksmith, test shards on 16 vCPU
The merge queue builds up to three groups at once, but every group's
checks, Clippy, doc and reference tests and all four shards ran on the one
self-hosted machine, so the groups waited for each other. Merge groups now
run on Blacksmith like pull requests (test archive built once, shards on
16 vCPU). Pushes to main keep the self-hosted runners and their warm build
directories, and a push the queue already tested still runs only the
repository checks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
* CI: every job on Blacksmith: pushes to main and the Windows jobs too
Pushes to main ran on the one self-hosted machine and the Windows jobs on
GitHub's runners. Both move to Blacksmith (Windows Server 2025, 8 vCPU),
so nothing in ci.yml depends on a single box any more. With no
self-hosted job left, the local routing goes: no warm build directories,
every run builds the test archive once and the shards run it, and main
saves the build caches every other run restores.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
---------
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2621
* Sketch: dragging part of a selection moves the whole selection (#2621)
A drag that started on a selected sketch line or point moved only that
entity; the rest of the selection was left to the solver, so geometry that
kept only its length (the reporter's 5 mm tail) trailed behind like a string
and chamfers stretched.
- sketch.drag takes `with`: the rest of a selection dragged by `entity`.
Every free point of all of it slides by the cursor's move and the held
solves keep them all (circles and ellipses keep their radii), so the
selection moves as one (`plan_group`). `with` is checked and capped like
the other parameters.
- The viewport sends the rest of the selection when the drag starts on a
selected point or curve. A curve's own points, picked with it by a window,
are not a group: a circle grabbed on its edge still takes its radius.
- Snapping looks at all the moving points.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A sketch and the feature built from it may share a name, and the sample and
oracle designs do. `timeline.add` refused such a recorded name, so a design's
own exported script failed to replay. `Document::insert_feature` now keeps the
name, as `add_feature` does; a blank name still gets a fresh one.
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Deleting a pocket or a boss from an imported part failed with "face N cannot
close: cannot attach a plane to a wire that is not on one plane". The face
around the gap only loses an inner loop, but it was rebuilt as a new plane
through its outer boundary, and an imported boundary is often a few 1e-5 mm off
the plane: within the planar test, too far for a plane fitted to it. Such a face
now keeps its surface and only drops the loop.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Replace Face rebuilt its body with ModelBody::new and dropped the body's
lineage, so no face of the result was named exactly: a sketch on any face of
it, even one the replacement didn't touch, failed with "this face can't keep
an attachment". Each face it moves is now a lineage step, as in Offset Face
(which does the same geometry), and the result keeps the lineage.
The test of a body without history used Replace Face as its example of an
operation that reports none; it now lifts the block with Align instead, the
same geometry.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #1985
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01L15TLCjpXsQdp8ovC4ZacH
* STEP export no longer hangs on a box with a screw hole (#1985)
The export reads its own STEP text back, and meshing the read-back screw
hole never finished. On a seam cylinder the boundary walk projected points
to angles up to 5e7 rad apart and moved each by two turns at most, so the
11-point loop spanned millions of turns and the polyline closing it across
the seam had 4.3 million points. Each projected parameter is now brought to
within half a period of the point before it, whatever the gap.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01L15TLCjpXsQdp8ovC4ZacH
* Second #1985 regression: the reporter's flanged enclosure
The issue's reporter delta-debugged a second design whose STEP export hung
on main: a flanged enclosure with two standoffs and no screw hole. It hangs
without the get_mindiff change and reads back with 29 faces and the same
volume with it. Both tests share the read-back checks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01L15TLCjpXsQdp8ovC4ZacH
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #3052
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch Trim: the hover shows the piece a click takes away (#3052)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch Trim: the piece is shown in the error colour, thicker
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Trim hover: preview keeps its doc line; the piece is worked out once a frame
The new function had gone between preview's doc line and preview. The
cursor's Trim piece is now kept by design revision, sketch, curve and
cursor, so the viewport's hover and the preview no longer both intersect
the curve with every other curve each frame.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* parameters: import evaluates once; parameters.change takes several names
parameters.import checked every row through change_param, which cloned the
design and evaluated every parameter per row: 2000 rows took 9.5 s, 2823 in
an assembly minutes. Rows are now checked and applied in one pass with the
values carried along, then evaluated once (2000 rows: 0.016 s), still all or
nothing; a row may name a parameter defined by a later row.
parameters.change {changes: {name: expression, ...}} sets several existing
parameters together, rebuilt once and undone in one step.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* parameters.change: the params doc names the `changed` result of `changes`
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
The self-hosted-untrusted pool is being retired for CI, so fork PRs whose
number ends in 7, 8 or 9 no longer route there. Build tests, the test shards
and the reference tests reduce to our own runners for main and the merge
queue, Blacksmith for everything else; the build's 90-minute gVisor timeout
goes. Blacksmith VMs are ephemeral and fork runs get no secrets.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Blacksmith runners report runner.environment == 'self-hosted', and a list of
our machines' name prefixes breaks when a box is added or renamed. Select
checks already decides where each job runs; its local and untrusted outputs
(and the trusted-run condition in the repository checks) now drive the warm
build directory, cache and lavapipe steps. Same results as before on every
event.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Re-applies #2237 (reverted in #2277 while the Blacksmith account awaited verification) on today's
ci.yml: every ubuntu-24.04 job or fallback runs on blacksmith-<N>vcpu-ubuntu-2404, sized as #2237
did (quick-compile, new since, gets 8), and steps that asked runner.environment now tell our own
machines apart by runner name, since Blacksmith runners report 'self-hosted' too. The trusted
self-hosted routing and the fork routing are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
From the review: the guidance said a pull request's run tests it merged with the latest main,
which is only main as of that run; a break between two pull requests that land close together
shows up in main's own run. It also forbade rebasing. Now: don't push only to catch up; when you
push a change anyway or GitHub reports a conflict, rebase onto main (--force-with-lease on your
own branch) or merge it in once.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
From the review: the web app embeds ATTRIBUTION.md, so the Docker image's Markdown exclusion now
keeps it; and THIRD-PARTY-LICENSES.txt also reads the vendored crates' licence files and the
fonts' OFL texts, so changes there regenerate it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
From the review: git patch-id --stable ignores whitespace, so a conflict resolution that only
respaced the change (inside a string, or Python indentation) counted as no change and was
carried; --verbatim compares every byte. And a passing run is now taken only from this pull
request (by number, or for a fork, where GitHub lists none on the run, by head repository and
branch), not from another pull request with the same head commit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
From the review of this PR:
- What counts as code lives only in scripts/ci/affected.py now: nothing selected and not full
means no code changed, on pull requests and pushes to main alike. The untested bash copy in
Select checks disagreed with it. Root files and folders a crate embeds now select that crate
(the licence texts and NOTICE: the engine; skills/: the MCP server's tests), where before a
change to them ran no test at all. Other workflows and the bots' scripts select nothing.
- A push to main skips the native jobs only when nothing a build reads changed since the last
push to main whose run passed. Comparing with the push just before was unsafe: a newer push
cancels a waiting run, so that commit may never have been tested.
- Fail fast applies to pull requests only. On main, nightly and Run workflow the test build and
the shards run regardless of the repository checks, so the action matrix and property tests
that hang off Build tests keep their signal. Doc tests (main and nightly only now) no longer
wait for the checks.
- tests/robustness/ turns on the goldens, so a change to goldens.json runs the seed tests.
- tests/oracle/ no longer selects the kernel, which never reads it.
- The workflow header and CONTRIBUTING describe what main and pull requests now run.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
A change that merged main twice in a row ran the full set on every other push: the change before
the second merge was itself a carried run, its shards skipped, so it had no passing shards to
carry. The script now walks down the first parents through consecutive merges of main that kept
the change's patch (each equal to the next), at most 10, to the run where the shards ran.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
The guidance told contributors and agents to keep their branch current with every upstream merge
and to re-run CI after integrating. Over three days that meant 1,241 merges of main into pull
requests and 730 re-runs: about 45,000 of 79,800 ci runner-minutes, on runners that capped how
fast everyone's checks ran. CI already tests each pull request merged with the latest main, and
a real failure fails again on the same commit. Now: merge main in only when GitHub reports a
conflict, re-run only a job runner trouble cancelled, and maintainers add approved-for-ci when a
pull request is ready to merge and merge approved ones in groups by area.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Merging main into a pull request (Merge branch 'main', or GitHub's Update branch) restarted the
whole run. Over three days 1,241 such merges started 966 runs and about 21,000 runner-minutes, and
872 of the merges brought in nothing that touched the pull request's files. Select checks now
recognises the case (scripts/ci/main_merge.py): HEAD is a merge whose other parent is on main, the
change's own patch is unchanged (git patch-id), and a ci.yml run of the change before the merge
passed with its test shards. Then only the repository checks run, the ci job names the passing run
in its summary, and main tests the merged result after it lands. Anything else, or any doubt,
runs the full set as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Doc tests move to main and nightly, beside the reference tests and the Windows jobs: on pull
requests they were the only failing job 3 times in about 7,900 runner-minutes over three days.
Build tests, the test shards and the doc tests now start only after the format and repository
checks pass; 212 attempts in those three days ran their heavy jobs beside a failed cheap check.
Also corrects the header, which still said pull requests run the Windows jobs and the reference
tests.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
- Pushes to main use the same code/no-code rule as pull requests (`touches_code` in Select
checks): a push changing only documentation, packaging, other workflows or the bots' scripts
runs the repository checks only. With the merge queue off, every push to main ran the full suite
and the Windows jobs, docs-only ones included.
- affected.py: test data under tests/ selects the packages that read it (robustness, hole matrix,
projection, taper: solidcraft-engine; oracle: the CLI, the kernel, xtask; action matrix: the
CLI, the engine, xtask), and a vendored crate the packages that use it (truck: the kernel and up;
egui-wgpu: the UI and the apps), instead of the whole workspace. tests/tools and unknown vendored
crates still select everything.
- Docker (the web app image) skips pushes touching only docs, test data or bot scripts; Pages skips
test data and bot scripts; Generated files runs only when the manifests, lockfile, command
registry, CLI or xtask change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
* inspect.clearance: pairs whose bodies did not change are not measured again
Each clearance check measured every close pair again, though between two
checks usually only a few bodies change or move. Results are now kept per pair,
keyed by both bodies' geometry and placement and the check's settings; a pair
whose two bodies are unchanged is taken from that cache. With 40 overlapping
parts, a check after moving one of them took 0.8 s instead of 23 s (741 of 780
pairs reused). Pairs that could not be worked out are tried again each time.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* inspect.clearance: boxes and placed keys exactly as before the reuse
The pair boxes are taken over the mesh's triangles with finite corners, as a
distance shape's are, so pruning is unchanged. A placed body's identity now
includes its model body's own frame, so two placements are never mistaken for
one another.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* inspect.clearance: a large body is meshed once for all its pairs
Each pair that touches cuts the larger body down to the pair's region with a
kernel boolean, and that boolean meshed the whole body twice: once for its
checks and once inside the intersection. A sheet with 600 faces paid about
2.5 s per pair for meshes that were the same every time.
A body can now keep its meshes for reuse (Body::keeping_meshes); its booleans
take the meshes from there, and the intersection takes its shell meshes
from there too (truck-shapeops: and_with_history_meshed). Clearance uses such
copies for one check. Results are unchanged to the bit. On a 3 x 2 m sheet
with bends and 300 holes, the clip took 0.23 s per pair instead of 2.6 s, and
24 rivets took 5.6 s instead of 16.5 s.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #2535
* claim: #2535
* Sketch: preview all slot sides and caps while choosing width
Fix missing temporary slot geometry for all five tools and paint every preview curve with a shared sampling budget. Preserve the arc slot construction guide. Add geometry, painting, invalid-input and interactive typed-width regression coverage.
* Sketch: honor typed slot widths below the pointer minimum
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Extrude: a profile edge short of a face is not grown through the skin beyond it (#1669)
A cut or join profile edge that lies on a face of the target is grown off it by 0.5 % of the body, so that the tool's wall does not lie in the face. Whether it lies on a face was judged half a growth step either side of it: empty just outside, material just inside. An edge 0.25 mm short of a 3.75 mm wall's far face reads the same, and the growth cut through the skin it should leave (55 050 mm³ instead of 56 130 on the test's wall). Now the space just past the edge, 1e-4 of the body out, must already be as the operation wants too.
In #1669, the pleat's window built directly at the far end of a case opened through the wall at 3.75 and 4 mm, while its mirror, built without the growth, rightly left the skin and then failed to join the loft through it. Built either way, the window now leaves the skin.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Regression tests: the issue's pleats mirror and build at the far end (#1669)
The issue's battery case at 3.5 mm: the mirror of the window and the pleat builds (it needed #1670's meshing of twisted faces), and at 3 mm the same window and pleat built directly at the far end of the finished case join it (they needed #1667's end rounds).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Far-end pleat test: the direct build matches the mirror's volume
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Erikas <erkexzcx@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
A loop made only of arcs took the all-arcs shortcut in `offset_loop`, which
moves each arc by its radius and keeps its start and sweep angles: exact for
arcs on one circle or meeting tangentially, but where arcs of different
circles meet at a corner (a circle with lobes) the moved arcs no longer
met, and the sides were built through the wrong points without an error.
`arc_corners` moves each such corner to where the two moved circles cross,
nearest the old corner (`circles_cross`, closed form), and cuts both arcs
there. Circles that no longer cross are refused ("not supported yet"), an
arc the taper turns around closes the profile. Between the sections the
corner is a straight ruling, as for every side in `taper_loop`: on the test
profile (2.5 mm at 45°) it stays within 0.015 mm of the exact corner.
Test `tapered_extrude_of_arcs_meeting_at_corners`: the volume against the
integral of the union of the shrunk circles, 18905.42 mm³ against 18907.67
(was 18975.45).
Co-authored-by: Mariusz Pacek <pacekm90@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Snapshot: draw a see-through appearance see-through, and give the CLI --body, --tool and --section
A headless picture dropped a body's opacity: the scene carried only a colour, so a design with
a translucent appearance was drawn opaque by `solidcraft-cli snapshot`, by MCP screenshots and
in a design's thumbnail, although the window has always drawn it as glass. There was then no
way to look at the body a cut would remove inside the material it meets: only a volume to
measure and reason about, which hides exactly the errors a picture makes obvious (a tool that
stops short, or lands in the wrong place).
- `SceneMesh` carries the body's opacity; `view::build_scene` takes it from the design
(`model_body_opacity`), and a selected body stays solid so the selection colour reads.
- The GPU capture routes a see-through body into the scene's glass at its own alpha, as the
viewport does; the CPU rasterizer draws the solid bodies first (they own the depth buffer),
then the see-through ones blended over them, far sides before near sides and without writing
depth, so what lies inside shows and a line between the two is not hidden.
- `snapshot --body NAME:opacity=0.3,color=#rrggbb`, `--tool NAME` (that body solid in the tool
colour, the rest see-through) and `--section z=27` run the commands the app itself runs
(`appearance.assign`, `inspect.section`): the options are shortcuts, not a second renderer.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* Tests: measure the tool colour as it is drawn through a see-through target
The tool-colour filter was derived on paper and demanded a red more than twice either other
channel, which the target in front of it tints away (measured: (200, 108, 85) through opacity
0.3); and the colour check painted the tool, which a solid target hides.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fmt
* Snapshot options: strip the colour's # rather than slicing it; README
Slicing `value[1..]` on an input-derived string is what the house rules ask us not to write,
even where a length check has already made it safe. The error a failed assignment reports now
names the body rather than claiming it came from --body, since --tool reaches it too.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* --tool: assign the bodies around the tool in groups, not one command at a time
Each appearance.assign is an undo step with a document snapshot behind it, so one per body
made --tool cost grow with the size of the design for no reason: the bodies around the tool
all get the same opacity, and the ones that share a colour can be named in one call. A design
whose bodies all wear the default grey now costs two commands whatever its size.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* Rasterizer: say where its glass and the GPU's differ
Splitting each see-through body by facing orders its own two layers; it does not sort across
bodies, which the GPU does. Worth stating rather than leaving a reader to find it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
`joint.as_built` with `at` an origin on A or B keeps that origin on its
side (found again by its face's name on each solve) and holds the frame
where it is now on the other, so the joint follows the face or hole as
the design changes. Without `at`, or with an origin on another part,
both sides hold their frames as before.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Fillet fit: a round counts only what lies between its edge and its arc
The fit check refused a round when any other face of the body reached into
the wedge between its edge and its two setback lines. A round changes only
the part of that wedge between the edge and its arc: a face on the ball's
side of the arc is left as it is. An outside round concentric with the
inside round of a 3 mm wall (R18 over R15) was refused as too large
(largest 14.21 mm) although it keeps the wall 3 mm thick.
The face check now tests a round's fan of prisms from its edge to its arc
(Frame::pieces, already used for blends that cross), after the wedge as a
quick filter. Chamfers keep the wedge, which is what they change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01L15TLCjpXsQdp8ovC4ZacH
* Fillet fit: a curved face clears a round's thin end pieces only by a chord (#324)
The fan test shrank each piece by the curved faces' tolerance; the thin pieces at the
arc's ends vanished, so a hole wall 0.3 mm into a round beside it was missed and the
round failed to build instead of offering the radius that fits (issue_324_tests).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #3001
* fix(ui): Convert Mesh asks for the mesh body and its tolerance (#3001)
It had no dialog, so the toolbar ran mesh.convert with no body and it always
reported "select a mesh body". A dialog now takes the mesh body (a selected one
fills it) and the tolerance.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2943
* fix(update): release-note links open in the browser (#2943)
The notes kept a link's text and dropped its target, so the blue text could not
be clicked. Links keep their web address, and each paragraph or item is
followed by its links as buttons that open them, as the agent chat shows links.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2823
* fix(view): lines thinner than a pixel no longer vanish without multisampling (#2823)
A line quad narrower than a pixel covers no pixel centre when it falls between
two rows, so the grid's 0.7 px lines came and went as the view zoomed where the
viewport draws without 4x MSAA (some browsers and devices). Such lines are now
drawn a pixel wide, faded by the width they lack.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2700
* Two-sided extrudes: the second distance goes the other way from the first side, either way round (#2700)
extent_sides ignored `direction` when an extrude had `distance2`, so through all with
`direction: negative` and a second distance went the positive way. The second distance now goes
opposite the first side wherever that went (direction: negative, a negative distance, a To
Object below the sketch). A pending format step makes older designs' reversed two-sided
extrudes positive, which keeps their shape.
The Extrude dialog's Two Sides shows Distance 2 for every extent type (it was hidden for All and
To Object but still sent as 10 mm) and offers Reverse; the second arrow points the other way
from the first.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* The two-sided arrows test checks each tip against the sketch, not just their order
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2592
* fix(agent-chat): the design bridge is named solidcraft-app (#2592)
A user's own MCP entry named `solidcraft` (as docs/mcp.md suggests) took the
bridge's place in agents that keep their own entry on a name clash, so the chat
worked on a headless session instead of the open design.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
On Windows, Git's default core.autocrlf checks the icon masters and the
palette out with CRLF, and `cargo xtask icons` failed four ways:
- the accepted-icon hashes (lint_icons.py and the review page's
review_state.master_hash) were taken over the files' bytes, so every
accepted icon read as changed; both now use master_hash, which hashes
the text with CRLF read as LF;
- icon_palette.revision hashed the bytes while save_colours hashed the
text it read, so every palette save was refused as stale; revision now
reads the text too;
- the tools and their tests read and wrote text in the locale's encoding
(cp1252 can't write U+2194); they now say utf-8;
- the icon inventory listed UI files with backslashes; it writes /.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Bill of materials: part properties on components and assembly.bom
A component definition now carries part properties (part number,
description, revision, other tools' keys) and free text properties,
saved with the component and shared by its occurrences.
component.metadata and component.metadata.set read and change them in
one undo step; Paste New copies them; parts.insert numbers standard
parts from their standard and size, and parts.resize renumbers them.
assembly.bom counts the placed components (flat: parts at any depth
through their subassemblies; top: direct children), grouped by part
number or by definition, sorted by part number, and writes CSV (fixed,
quoted columns) or JSON. The app gets a Bill of Materials window on the
INSPECT panel that lists, edits and exports the rows.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Bill of materials: name a merged row's equal definitions once
Two standard parts with the same size are separate definitions with the
same name; their row now reads "SHCS M6x20" instead of repeating it.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Window widths test: close the Bill of Materials window between commands
The test closes the floating windows a command opens before starting
the next one; the Bill of Materials window was left open and measured
against the next dialogs' limits.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Bill of materials: sub-parts in their part's row, and part properties without re-evaluating
assembly.bom takes group_children: true. A part with a part number keeps
what is placed inside it (a bolt's washer and nut) in its own row, listed
under it as children (items 1.1, 1.2, with each and count) instead of rows
of their own; bolts with different contents stay different rows, and
total_occurrences is the same either way. The window has a switch for it
and shows the children under their part; the CSV lists them after their row.
Changing part properties re-evaluated the model, re-seated every joint and
rebuilt the placed model: 361 ms per change with 30 jointed bolts. A change
that only touches components' metadata and properties now takes the same
path as a render-only change (proved on the whole Document), also for its
undo and redo: 0.6 ms.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Bill of materials: look the structure up once, not per subassembly
Deciding whether a subassembly is a part (it has bodies of its own) scanned
every body of the model for each subassembly, and each body looked up its
feature by a linear search of the timeline: subassemblies × bodies × features.
assembly.bom now indexes the placements, the body owners
(Document::body_owners: one pass over the bodies with the timeline in a map,
the same rule as model_body_component) and the numbered components once per
call. 600 bodies and features with 60 subassemblies: 8.1 ms → 1.3 ms; the
old cost grows with the cube of the design's size.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Centralise tool modifiers with classic defaults and preview-safe Ctrl.
Route sketch/selection chords through a shared resolver. Keep Alt snap
suppress and dual toggle as defaults; optional Shift add-only and Ctrl
snap mapping. Skip preview rollback while a manipulator drag is active.
* Address #1616 review: classic snap suppress, single viewport owner, docs.
Default suppress to Ctrl/Cmd or Alt; drop navigation set_suppressed; remove unused label_key; add preview rollback gate test; document sketch modifiers.
* fix(#1616): correct Modifiers primary in tests; simplify rollback fixture.
* Treat Ctrl as primary off macOS, and rustfmt the modifier files.
* kernel: trim keeps picked regions, mutual trim, curves swept along a direction (#767)
Trim moves to surface_trim.rs: TrimKeep::Side (the old behaviour) or
Regions (the connected pieces the points pick, parted where the tool runs
along an edge). mutual_trim cuts two one-face surfaces by each other's
surface, extending one that stops short when asked. SplitTool::Curves takes
a sweep direction; a closed curve's field is signed by inside/outside
instead of extending its end segments.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* kernel: classify display edges of a face with a closed-edge hole without a panic
A hole cut by a closed curve is a loop of one edge that starts and ends at
one vertex; mapping the face rebuilt that edge through a check that
panics on it (the guard caught it and dropped every smooth flag of the
body). Such faces are now left out of the ruled-surface test.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* surface.trim: keep several regions, trim by projected sketch curves, mutual trim (#767)
keep takes a list of points (the regions they pick); curves: {sketch,
curves?, direction?} sweeps sketch curves onto the surface (along the
surface's normal: not supported yet); with + keep_with trims two
surfaces by each other, extend? meets a short one, stitch? sews the
kept pieces. Trimmed faces carry their lineage, so they are named
after the faces they are pieces of. New fields noted in a pending format file.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* Trim dialog: sketch curves as the tool, several keep points, mutual trim (#767)
The tool input takes sketch curves too; keep points are places (several
on one face add up); another surface with its keep points makes a mutual
trim with Extend to meet and Stitch toggles. Edit Feature brings it all
back. Labels in every locale catalog.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* ROADMAP: M8 trim by projected curves, several regions and mutual trim (#767)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
---------
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Sketch points placed by expressions over the design's parameters
A point typed as an expression ("W", "H / 2") in sketch.point, sketch.line and
the other point arguments is fixed where its expressions come to, and moves
there again before each solve when the parameters change: an outline that
follows its parameters needs no dimensions and no solve per change. The
expressions are saved with the point (one pending design-format change), listed
with the sketch's other expressions, and renamed with their parameters; one
that cannot be worked out is refused.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Sketch point expressions: accurate params doc, out-of-range regression test
The sketch.line params said every point argument takes expressions; only those that
also take a point id do (the rectangle corners, an arc's through point and the like
are plain [x, y]). PointExprs' comment called its expressions mm. A test covers a
parameter that puts a placed point out of range and a design file whose point
expression overflows: both are errors, not crashes, and the sketch recovers.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
assets/locales/ belongs to no crate, so affected.py selected the full
workspace (every crate, all four test shards) for a catalog-only PR. Only
solidcraft-ui-egui embeds the catalogs and xtask checks them; map the folder
to those two packages and their dependents.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The window listed (and so evaluated) every parameter and re-checked the add row's
expression on every frame, and laid out every row. It now lists once per design
revision, caches the add row's check, and lays out only the rows in view.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The chat in a window of its own takes the files dropped on it and shows the drop hint while files are dragged over it; a window floating inside the app counts as the chat's place for the app's drop handling.
A PDF goes whole to agents that take embedded documents; to the others, or past 12 MB, the message names it with its type, size and place. Any file that is not a picture, text or CAD file is attached by reference instead of refused. Chips show the type's extension; the file dialog offers All files.
Files copied in the file manager attach as files, a picture on the clipboard attaches as a picture, text pastes as text. A picture-only or files-only paste is seen through the V key coming up (egui-winit sends nothing when the clipboard has no text); macOS sends no key-up while Cmd is held, so there only pastes that come with text are seen and + > Paste a picture still works.
* CI: heavy checks run after review (approved-for-ci label)
Pull requests without the approved-for-ci label run Select checks, the
format and repository checks and a quick cargo check. The label runs the
full suite, and every push after that does too. Other labels start no work
and never cancel a run. Main, merge groups, nightly and Run workflow are
unchanged.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* CI: approved fork PRs run heavy jobs on self-hosted-fast in private directories
An approved-for-ci label on a fork pull request sends Clippy, Build tests,
the test shards and Doc tests to the trusted runners, with private copies
of the build directory, Cargo home and rustup home that are deleted after
the job. Only the labeling event qualifies; later pushes use the old fork
routing. The skipped run of an unrelated label names its ci job differently.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* Revert "CI: approved fork PRs run heavy jobs on self-hosted-fast in private directories"
This reverts commit 439f2f7dab.
* CI: name the ci job differently on an ignored label's skipped run
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
An orbit drag now applies its whole turn so far to the camera it began
with, instead of adding each pointer step to the camera of the step
before. The view then depends only on where the pointer is: dragging
past straight up or down and back no longer loses the pitch that the
clamp swallowed, and returning to the starting point brings back the
starting view. The pivot is still picked once, when the drag begins,
and kept still on screen by the rigid turn about it.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2544
* S command search and the command line: names, places and matching in the current language (#2544)
The S box showed each command's English label and toolbar place in every language, and found
only what was typed in English; the command line matched English only too. Rows, places and
tooltips now go through the label lookup, and the query matches the translated label, tab and
panel as well as the English ones and the id. The pin tooltips get catalog entries.
A new sweep (i18n_painted_tests) runs every language, reads the text each frame paints, and
fails on English painted where the language has a translation, skipping the design's own names
and key names, so a painter call that skips the lookup can't come back unnoticed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* The painted-text sweep also covers every Preferences page and four right-click menus
Each surface must show (the Preferences title in that language, or new text for a menu), so the
sweep can't pass by painting nothing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* S box rows: labels stay 'static; the place's separator is drawn, level in every language
A font's "›" (and one centred on the row) sits on the baseline beside Japanese or Chinese words,
where it reads as a comma. The separator is now a drawn chevron (glyphs::breadcrumb) centred on
the ink of the panel's name. search() returns the place as (tab, panel).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* A new design's name, Untitled, shows translated in the browser too, like the other default names
The tab and the save prompt translated "Untitled" through their own special case; the browser's
top row showed it in English. It is now one of the default names the label layer translates
(Locale::default_name, with Sketch1, Body2, Component3), so every place that shows a given name
shows it translated, and the two special cases go. The painted-text sweep no longer skips a design
name that the language translates, and splits a tab's dirty mark off, so it catches a place that
shows the raw name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2424
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
* Sketch Offset: constrain to the ends of an offset spline (#2424)
An associated offset spline hid all of its generated fit nodes, its two
end points included, so nothing snapped to them, and a curve drawn from
one shared the node and was refused ("invalid offset output point roles").
- An open offset's end points are shown and snapped to again; its
interior fit nodes stay private.
- After every sketch edit, a curve that took an offset node gets its own
point there, coincident with the node: the offset keeps sole ownership
and the curve follows the offset when its source moves.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
* Scenario: lines snap to and close onto an offset spline's ends (#2424)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SBN24y2TcqRcm2zn8Gkj27
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Clearance: check a shared volume against the bodies and their union
An intersection can come out well formed and still wrong: one of a
mirrored pair of imported parts shared three times the volume of its
twin. The shared volume is now compared with vol(a) + vol(b) - vol(a ∪ b),
on the same (cut down) bodies; when they differ by more than 1 % (plus the
bodies' own measuring error), the pair is reported unchecked with both
numbers. Where the union gives no answer, the intersection stands.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Clearance: let the union cross-check allow for the bodies' measuring error
vol(a) + vol(b) - vol(a ∪ b) is a difference of measured volumes and
carries the error of all three, so correct small overlaps on large bodies
came back unchecked (0.96 mm³ where the union implied -0.75). measure
keeps a face's fine tessellation when it is within 1 % of one with a
5 times coarser chord, so a measured volume can be off by up to 0.25 %:
the allowance is now 1 % of the larger of the two answers plus 0.25 % of
the three volumes. Past it, a union that implies nothing (within that
noise of zero, or below it) while the shared part is at most a tenth of
the smaller body cannot tell, and the shared part stands. A shared part
of 412 mm³ where bodies of a few thousand imply 141 is still caught.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Clearance: tighten the union check's measuring allowance to 1e-4
The old note cited measure's 1 % gate, which compares face areas, not
volumes, and its 0.25 % was about 100 times the disagreement the bodies'
measures actually show. The allowance is now justified from volumes: a
curved face's chord shortfall cancels between a body and the union except
where the chord changes, which stays near 2e-5 of the three volumes; 1e-4
leaves five times that. A 40 mm roller pressed 0.2 mm into a 2 m plate
exercises it: with no allowance the pair is left unchecked (7.52 mm³
measured against 8.16 implied), with it the overlap is reported.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* geom: UnitVec3; a face's analytic axes and normals are unit by type
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* geom: UnitVec3 adds to and subtracts from a Vec3
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* kernel, doc, engine: cylinder axes and measured normals held as UnitVec3
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* kernel, engine: functions that take a unit axis or direction say so by type
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* doc: a face's normal goes to Plane::from_normal as is
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* kernel, doc: main's new cylinder helpers take the face's axis as a UnitVec3
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #2474, #2494
* fix(ui): construct tools ask for their picks instead of failing (#2474, #2494)
The 13 construct tools without a dialog of their own ran their command with no
parameters from the toolbar and reported that `a` (or `face`, `edge`...) is
required. One pick dialog now serves them: labelled inputs per tool,
preselection fills them, and each pick becomes the construction reference the
command takes; Point Along Path keeps its position.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Reword comments, docs and package descriptions so they describe SolidCraft's own behaviour instead of comparing it to another product. The trademark notice, the commercial-CAD rule and the navigation preset name stay as they were.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
A round origin's end is kept by the name of the face across the rim. On
a face in pieces (a sheet split by a groove or a bend) that name lists
the faces round the piece, and when those change (a hole nearby moves
or goes) the name matched neither rim, so the end fell back to the
nearer end of the stored pick: a pick near one end of a thin wall that
the design moved past mid-wall turned a part through 180 degrees. The
end whose rim matches the name better now wins: the face itself, a face
merged into it, then a piece of it, then the same face by its stem.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This reverts commit cac4bccea8. Orbiting
about the point under the cursor goes back to 0.15.0's pivot choice
(the picked face or vertex point, an edge's midpoint, or the visible
geometry's centre over empty space), matching 0.16.1.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
the format and repository checks job still took a self-hosted-fast runner for every same-repo
pull request. It now follows the same rule, and the notes at the top say so.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Pull requests that can change a robustness golden check every seed but the 17 slowest, in the
same nextest run as the other tests instead of a second step after it (16-30 minutes on its own).
The slow seeds (about four fifths of the goldens' time) still run in the merge queue, on main and
nightly. A failed golden still prints how to re-record it, and junit.xml now holds the whole run.
- The slowest tests start first (nextest priority), so a shard no longer ends with one of them
running alone.
- tspline::tests::hostile_tsplines_never_panic joins the heavy tests pull requests skip (up to
12 minutes on a GitHub runner; the same kind of check as hostile_params_never_panic).
- The ci profile builds dependencies without line tables: the test archive pull requests' shards
download goes from 3.4 GB to 2.2 GB.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
On main and in the merge queue every test shard ran on the self-hosted machine but waited for
the GitHub-hosted Build tests job and then spent about 12.5 minutes downloading its 3.2 GB
archive (four shards at once over one connection). There, each shard now builds the tests in
the runner's warm build directory (the existing build-tests one) and runs its slice directly;
Build tests runs only where the shards run on GitHub's or the untrusted runners.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Same-repo pull requests ran their heavy jobs on the three fast runners too, and the merge
queue's test shards waited behind them for hours. Pull requests now use the GitHub-hosted path.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Orbiting about the point under the cursor (or the origin) turned the camera
about its target and then panned to put the pivot back. The pan works at the
target's depth with the parallel-view scale, so in a perspective view the
pivot slid off its pixel and the model swung about the camera.
The camera now turns rigidly about the pivot: Camera::orbit_about keeps the
pivot's coordinates in the camera's frame and rebuilds the target from the
new orientation, so the pivot stays exactly on its pixel in both projections.
The Shift snap spring and view.navigate's pivot orbit use the same.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* UI: the Interference dialog and the continuity legend's Gap are translatable
The Interference dialog's title, its Bodies input, the Compute and Compute
again button, the checked-bodies count, "could not be checked" and the
clearance error were English in every language. They now come from the
catalogs (en, ja, zh, de, ko, fr).
The continuity legend's Gap (an opening between faces) and the sheet-metal
hem's gap shared the message "Gap", translated as the hem's gap (French
"Jeu"). The hem keeps those translations under ui.sheet_gap; the legend
gets ui.continuity_gap, and both read their message by id.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01L15TLCjpXsQdp8ovC4ZacH
* Polish: the sheet-metal gap is ui.sheet_gap too
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01L15TLCjpXsQdp8ovC4ZacH
* Every catalog gives the continuity legend's Gap the word it showed before
The catalogs without their own ui.continuity_gap would have fallen back to
English in the legend; they now keep the translation the shared Gap gave
them. French and German keep separate words.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01L15TLCjpXsQdp8ovC4ZacH
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Browser sidebar: drag its right edge to size it, and remember the width
The left browser sidebar was fixed at 250 px. Its right edge is now a drag handle: the width follows the pointer, is clamped to 170-620 px, and is kept in the preferences (browser_width) so it survives a restart and Reset Defaults returns it to 250. The few pixels of grip sit inside the panel's own margin, so the tree rows and the scroll bar keep their input.
* Browser: float the tree over the workspace on a translucent card
Preferences -> General gains "Parts browser -> Float it over the workspace": the tree then draws on a card over the 3D view instead of taking a column, so the workspace keeps the whole window width. The card hugs the tree, scrolls past 70 % of the view's height, and uses the remembered browser_width. The canvas keeps every point the card does not cover, and turning the preference off docks the browser again.
In a CJK UI, the primary CJK face goes first in Proportional, so digits, units and
symbols in mixed labels share the CJK text's metrics and baseline. A Korean UI puts
the Hangul face first. The user's ~/Library/Fonts is scanned for a CJK font on macOS.
Rebased onto main as one commit (main had added the Thai fallback).
Co-authored-by: Mithril-3d <145893541+Mithril-3d@users.noreply.github.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* claim: #2089
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3MF import: read objects in other model parts (production extension) (#2089)
Slicers save multi-part projects with the 3MF Production extension: the
root model's objects are components with `p:path` naming another model
part of the package, which holds the meshes. The reader skipped those
components and build items, so such files did not open at all.
Components and build items now keep the part they name, and the parts
objects name are read once each (looked up through an index of the zip
entries), with the number of part names and their total size capped.
Each part's unit is applied as it is read, so parts in different units
fit together. Cycle detection keys on (part, object), so parts that
place each other in a loop are refused.
Tests: components and build items in another part, part names in any
case, a part in centimetres under a root in millimetres, and hostile
packages (missing part, a path out of the package, an empty path, an
unknown object, a part that is no model, a loop between parts, more
part names than a package can hold). The 3MF mutation fuzz also runs
on a production-extension package.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* 3MF import: count every placement toward the budget (#2089)
An object that only places other objects added nothing to the triangle
budget, so a few objects each placing hundreds of components multiplied
at every level up to the depth cap (300 components, 3 levels: about 1 s;
deeper never finished). That was already true for one model part and now
reaches across parts. Every placement now costs at least one unit, so
such a file stops at the budget.
Test: six levels of 300 components over an empty object (300^6
placements) stop with a budget error in well under a second; without
the change the test does not finish.
Suggested in review by @jyviko.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Tyler Chiriboga <18270301+Tyler-Chiriboga@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
The control port opens before the first frame, so a single ui.inspect sometimes returned no viewport and the check failed within seconds. The probe now polls ui.inspect until the viewport exists, for up to 60 s, and fails with the last reply if it never does.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Trunk formats the content hash without leading zeros, so it is sometimes 15 hex characters. The immutable-cache rule and the smoke test's wasm match now accept 8 to 16.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
The kernel will move from truck to solidcore (#1597). The tests that must
catch regressions across that switch are the ones that don't depend on the
kernel: the reference tests and the robustness goldens. cargo-mutants checks
whether they would notice a change in the kernel's behaviour.
- apps/solidcraft-cli/tests/oracle_cases.rs: the reference tests as a cargo
test (ignored; CI keeps replaying them through cargo xtask oracle), run
over one process per core.
- .cargo/mutants.toml: mutants are judged by that test and the pull-request
sample of the robustness goldens, not by the kernel's own unit tests.
- A mutants profile (ci with incremental builds): about 15 s to rebuild after
a kernel change instead of about 3 minutes.
- docs/mutation-testing.md: how to run it, its cost, two cargo-mutants 27.1
problems and the pilot's results (15% of sampled kernel mutants caught).
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Joints: a face or circle origin keeps its turn as the face tilts
A face- or circle-centre joint origin worked out its frame's x from the normal alone
(`Vec3::any_perp`), which switches axis where the normal crosses 0.9 along X: a part on a face
tilted past about 64° by a parameter spun a right angle about the face. New origins keep the x
they have when picked (`x`, component coordinates) and each solve lays it onto the face as it is
now, so the part tilts with the face without spinning. Origins saved without it are placed as
before.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Joints: origins on an edge of a face, at its middle or a corner
`joint.origin` and `joint.create` take `edge: [x,y,z]` (a point on an edge of a planar face),
with `on` (which face; default the larger planar face beside it) and `corner` (the end near that
point). The origin sits at the edge's middle (z out of the face, y into it) or at the corner (x
and y along its two edges into the face, right-handed whichever way the edges run). The edge, the
face and the corner's other edge are kept by name and found again on every solve, so a part set
40 mm in from a panel's corner stays there when the panel changes size; a name that finds nothing
leaves the joint out with the reason.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
The trusted runners' guard refuses issue and pull_request_target events by design (they carry a
write token and handle untrusted input), so the bots failed there. They go back to GitHub's runners.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
They check out main's own scripts, never pull request code, and were holding dozens of
GitHub-hosted runner slots that pull request builds need.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#2247 changed 16 robustness seeds (bodies N -> 1, and some seeds then build
different steps; seed 54's new shell fails on #2247's parent too). Recorded
on Linux.
Designs saved before #2247 that combined a split-off piece by name (tools
Posts, Posts (1)) failed on the missing piece and left both posts off the
case. A combine now skips a missing `X (n)` when `X` is a tool too or the
join's target, with a warning. Format note for the meaning change.
On the self-hosted-fast runners /proc/meminfo isn't readable, so awk exited 2 and failed every
job the sizing step runs in (Doc tests on #2509 and #2510, on minisforum-fast-2 and -3) before
compiling anything. Sizing is an optimisation: without a readable memory figure (meminfo or a
numeric cgroup limit) the script now sets nothing and Cargo keeps its default. Also drops
`set -e`, so no other probe can fail the job.
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#2278 dropped the golden sample from pull requests, and two geometry PRs then
broke main together. Pull requests that touch the kernel, engine, doc, sketch
or geom crates now run every seed, split over the shards, with the command to
re-record in the failure output.
Fork PRs whose number ends in 7, 8 or 9 stay on the untrusted pool; the rest
build and test on GitHub's runners, which have free slots.
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Build tests on the untrusted runners (fork PRs, gVisor on shared boxes) fail intermittently with
rustc killed mid-crate ("exit status: 2", no diagnostic) on the biggest crates: solidcraft-engine,
ui-egui. Nothing limits Cargo's parallelism, so it starts one rustc per core the sandbox reports.
scripts/ci/build-jobs.sh sets CARGO_BUILD_JOBS from the memory available (and the cgroup limit,
where lower) at 3 GB per job, capped at nproc, and logs the numbers it saw.
Claude-Session: https://claude.ai/code/session_01QWhMwdqJYQKQq8RCLSL1LT
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#2064 (multi-profile extrusion) merges coplanar faces the boolean used to
leave split. Seeds 67 and 94 change face/edge counts at the same volume;
seed 13 has four fewer edges before its random fillet, which then picks
different edges by index. Every step still passes the validity and
volume-direction checks. Pull request CI runs every 8th seed only, so
main's full run was the first to see these.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #1209
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011YdMi9BDdjWZ8twfeBHAXz
* Prism boolean: a curved edge at one level crosses the walls there only (#1209)
The prism boolean sews its result slab by slab and keeps, for each section segment, the operand
wall it was read beside; its history is built from those tags. `Stations::new` took every curved
edge between two walls as crossing them at every level, so where two pieces of one cylinder are
stacked (an arm's round end under a boss of that radius) the circle between them cleared the
wall tags of every slab. `prism_lineage::sewn` then gave up, the join or cut came out with no
history, and every face of the body was named from geometry (`exact: false`).
A curve whose control points lie at one level (a B-spline, or a NURBS curve with positive
weights, which lies in their convex hull) is at that level: it crosses the walls there only.
Any other curve still spans every level.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011YdMi9BDdjWZ8twfeBHAXz
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Document the maintainer-confirmed workflow: open a draft PR to establish ownership, then comment /claim to mark the issue as taken. Align AGENTS.md and CONTRIBUTING.md and distinguish the existing takeover policy from the bot's automatic cleanup deadlines.
Signed-off-by: Karl Von Richter<karl@embeint.com>
Fork Build tests jobs were hitting the 45-minute limit while uploading their archive, after
the build had finished, and every one had to start over.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Add complete Italian UI localization
Translate all 1509 English catalog entries originally, register Italiano, guard complete coverage and exercise Italian labels and UI rendering.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
* i18n: Italian is a partial catalog
it.json lacks 19 messages main has gained since this branch began, so as a
complete catalog it would fail every_complete_catalog_carries_every_english_message.
Like the other community languages, Italian is now partial: its gaps fall
back to English and new labels don't require an Italian string. The docs
line goes back to main's wording, with Italian among the partial catalogs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Ben Herbruck <bherbruck@herbrucks.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The trusted runs' test archive is still built on GitHub's ubuntu-24.04
runners; minisforum's newer glibc runs it unchanged, so shards 3-4 can join
1-2 on the self-hosted runners.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Fork pull requests now run on the untrusted self-hosted runners, which start
every job empty. Let them restore main's rust-cache like GitHub's runners do
(save-if stays main only, so a fork job never writes a cache).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Brings assets/locales/es.json back to full coverage of en.json: 60 new
messages translated (motion study, CAM, DXF export progress, sweep rails,
threads, grid follows the view) plus 13 terminology revisions (Loft,
Boss, timeline marker, viewport, ellipsis), all following the glossary.
It also unifies the snap label, as asked in the review of this pull
request: ui.snap (Snap) and cmd.sketch_snap (Snap Inference) now share a
single name, "Referencia a objetos".
Built on top of current main, so the three messages upstream had already
translated in es.json (Direction, In plane, Normal to sketch) are kept.
Co-authored-by: carlymx <24634598+carlymx@users.noreply.github.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
The Clearance report, Explode Text, the continuity legend, the Grid follows
the view preference and inspect.silhouette, so fr.json covers en.json again
(1518/1518).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* feat(geom): surfaces through curve networks, and spline spaces to build them
Adds spline_space (clamped B-spline spaces, Hermite interpolation, exact
embedding by Bézier degree elevation and knot insertion, a monotone cubic
for reparameterising) and network (loft, Coons and Gordon surfaces through
curves in one or two directions, with G1/G2 cross-boundary conditions
against a neighbouring face). Part of #763.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* feat(kernel): boundary surfaces from sketch curves, 3D curves and edges
kernel::boundary_network turns sketch chains, 3D sketch polylines and edge chains
into network curves, gives G1 frames of the faces along open edges
(EdgeFaces), builds the B-spline face (sides split at chain vertices so a
join shares the body's edges) and joins it into a surface body by stitching.
G2 is refused for now (not supported yet). Part of #763.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(kernel): second derivatives of B-spline surfaces in the vendored truck-geometry
BSplineSurface::uuder and vvder multiplied the second differences by p
instead of p·(p − 1), as BSplineCurve::der2 does, so every surface of
degree p ≥ 3 (NURBS too, through its homogeneous surface) reported
1/(p − 1) of its second derivative: a cubic surface's curvature read as
half. This is why a G2 boundary surface on a cylinder measured half the
curvature it carries (#763). New tests hold both to finite differences
for degrees 2 to 5.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* feat(kernel): G2 boundary surfaces against a body's face
The network always carried the face's curvature across a G2 edge; the
refusal came from measuring it with truck's halved surface second
derivatives (fixed in the previous commit). EdgeFaces now reads the
face's second fundamental form from the exact second derivatives
instead of differencing the normal. A new test holds the curvature
across a cylinder's edge to the cylinder's within 0.1%, also with a
sheared far curve; G3 is refused (a network holds up to G2).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* test(kernel): a boundary patch with straight corners measures its own area
A stadium's arcs meet its lines at 180°, so the Coons patch is singular
at its corners. From such a corner the reference mesher's boundary walk
ran onto the B-spline's extrapolation past its domain (a planar spline
passes through every point of its plane there), and the face measured
289 mm² for its 158.5 (#763). The vendored mesher's bounded projection
(nearest_projection, #2043) keeps the walk on the domain; this holds it
there, by the faces, the measuring mesh and the display mesh. The
display path (#1653) was never affected.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* feat(geom,kernel): three-sided boundary surfaces (one side collapsed to a point)
A network curve may be a PointCurve: the apex where the two sides beside
it meet. The curves through it cross it at the point itself, its row of
the net collapses there, and the kernel's face leaves that side out of
its loop (three edges, one vertex at the apex). One point per patch, as
a side; refused elsewhere, when it moves, or with a continuity
condition. Tests: a triangle and a quarter disc filled flat with their
analytic areas (40 mm², 25π), by geom and by the kernel's faces and
meshes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* feat(doc): boundary surfaces through curve networks and loops of three or four
FeatureKind::SurfaceNetwork saves a network as `u` and `v` curves, each
a sketch chain (planar or 3D), a chain of a body's edges with an
optional G1/G2 continuity, or a point (the apex of a three-sided
patch), and an `operation` that can join the surface into the body of
its edges. A SurfaceBoundary of three or four sketch curves, refused
until now, builds as the flat Coons patch of its loop (three collapse
the side opposite the first curve). The format change is one pending
file; feature-walk goldens record the new variant's sketches, bodies
and points.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* feat(engine): surface.boundary through loops of three or four and curve networks
The command keeps its loop form (two to four curves of one sketch; three
and four now build the flat patch of the loop, crown refused there) and
gains a network form: `u` and `v`, each curve a sketch curve id, a
{sketch?, curves} chain (planar or 3D), a {body, edges, continuity?}
edge chain (g1/g2 with the face beside it) or a {point} for a
three-sided patch, with `operation` new or join. Edits may move between
the two forms (timeline.redefine accepts the command now that it has a
dialog).
Tests: analytic areas of loop patches and of a hyperbolic paraboloid
bounded by four 3D lines (every mesh vertex on z = xy/10), a loft
across two sketches, G1 and G2 measured across a cylinder's edge with
inspect.continuity (angle < 0.1°; curvature jump 1e-16 with G2 against
0.027 with G1), 26 hostile networks refused without panicking, and u/v
added to hostile_params_never_panic.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* feat(ui): Boundary Surface dialog
Curves and Cross curves take sketch curves and edges (picks that meet
end to end in one sketch or body are one chain); Continuity and
Operation (New Body or Join). Two to four sketch curves of one sketch
with nothing across are the loop patch; anything else is a network.
Edit Feature brings either back (per-side orders, crown and stations
the dialog does not show are kept). Strings in en, ja and zh.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* docs: ROADMAP M8 and the bridge patch's notes for boundary networks (#763)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* test(oracle): a boundary surface through a loop, thickened (#763)
Case 83: four sketch curves (a square's three sides and a half circle
for its top) make one boundary surface, thickened 2 mm; analytic values
from tools/oracle-analytic/cases_surfaces.py (V = 2A, area 2A + 2P,
centroid by moments). Recipes gain boundary_surface, thicken and remove.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
* refactor: boundary surfaces, review cleanups
Project onto the face once per derivative evaluation (the frame is
passed to natural() and reused by scale()), keep each track's
finite-difference step instead of rescanning its table, export the
kernel's edge-pick and chain caps and the default station count for
the engine, doc and dialog to share, map TooLarge spline errors to
KernelError::TooLarge, reuse the engine's operation parser, and skip
the dialog's sketch lookup when the picks are no loop.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
---------
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
While a sketch is open its grid is drawn on the sketch's own plane, centred
near the view target and snapped to the major step, whatever the Grid switch,
the sketch's Sketch Grid option or Grid follows the view say. Headless window
pictures follow. Outside a sketch nothing changes.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Lists and Packaging lint each took their own runner slot on every PR push. The list checks now run
in the existing checks job; Lists stays on main. Packaging lint skips PRs that only touch icons.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With Windows off pull requests, GitHub's 20 runners sat idle while every
fork job queued for the self-hosted pool. Select checks, repository
checks, Clippy, doc tests and the summary go back to GitHub's runners;
Build tests and the test shards stay on the isolated self-hosted pool.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Pull requests now run the repository checks, Clippy, the affected
crates' tests in 2 shards, and the doc tests. Windows, the reference
tests and the robustness goldens run on pushes to main and in the
nightly, where a break is fixed forward. This is the split Bevy, Zed
and Ruff use, and it cuts each pull request's CI to about a third.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Blacksmith isn't provisioning runners: its dashboard says the account
"requires additional verification", so every job labelled blacksmith-*
stays queued with no runner. Fork PRs' Linux jobs go back to the
self-hosted-untrusted pool (#2249/#2252) and the rest to GitHub's runners
until Blacksmith verifies the account; then #2237 can be re-applied.
This reverts commit dc40f051b1.
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
On a busy CI machine (several fork jobs under gVisor at once) a web
offline launch took over 90 s and the icon review server took over ~2 s
to answer. Launches now get 240 s and one retry; the review server gets
up to 60 s to start.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Fork pull requests, the nightly run, Run workflow, the test archive and
test shards 3-4 move from ubuntu-24.04 to blacksmith-<N>vcpu-ubuntu-2404
(16 vCPU for Clippy and the test archive, 8 for test shards, doc tests,
reference tests and cross-compiles). Trusted runs keep the self-hosted
routing. Blacksmith runners count as self-hosted to GitHub, so steps now
tell our own machines apart by runner name.
TEMPORARY: this branch routes its own run like a fork pull request.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Every fork run waited in GitHub's shared queue just to start. The gate
and summary jobs now run on the isolated runners as well; the local
output stays false for forks so later jobs never pick the trusted pool.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Fork PRs were waiting hours for GitHub's shared runners. Their Linux
jobs (checks, Clippy, Build tests, the test shards, Doc tests, Reference
tests) now run on the ephemeral, network-jailed self-hosted-untrusted
runners, without the trusted runners' warm build directories.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Show each dimension's parameter name while creating or editing it, and let
users insert another driving dimension at the expression cursor or selection.
Use the available editor width and keep Apply aligned with the right edge.
Add a tools dropdown with Measure, named user parameter expressions, and
access to the existing parameter table for adding and editing parameters.
Preserve the unfinished dimension expression while either window is open.
Cover reference insertion, dependent updates, parameter selection, dialog
round trips, applying and cancellation with UI integration tests. Update the
parameter guide.
Signed-off-by: Karl Von Richter<karl@embeint.com>
The toolbar gives up room in small steps (labels shorten, then icons only, then panels hand buttons to their drop-down one at a time, then trailing panels go to the » menu) instead of collapsing whole panels, so medium-width windows no longer lose icons or leave a blank gap.
Preferences > Graphics grew a row (Grid follows the view, #2035), so the Colour vision
control now sits below the fold of the fixed-size page; the scenario scrolls to it.
The page itself already scrolls for real users.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Pull requests from branches of this repository share the work with the self-hosted runners like main does; fork pull requests stay on GitHub's runners. Select checks and ci run on a light self-hosted runner; the repository checks, Clippy, doc tests, reference tests and test shards 1-2 on the fast ones. The test archive is still built on GitHub's runners (glibc).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* docs: ribbon UX redesign proposal (mockups and notes)
Seven mockup images and a short description of the proposed ribbon shell, with floating and
compact layout options. Pictures only; no app changes. The sketch variant with tab locks is
left out.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* docs: update the UX mockups to the current command set
Adds the Sketch Pattern panel, Perforate, new joints and motion commands, Fair and Match Surface,
and the Render workspace; the workspace switcher now matches the app (Design and Render available,
Animation, Simulation, Manufacture and Drawing later). Command map covers all 389 commands.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* docs: draw the UX mockups with the reviewed icon SVGs and accents
Ribbon, browser, history and navigation icons now come from the icon system in #853, with the
Sketch tab in the Sketch green and constraints in red.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* docs: address review on the UX proposal README
Describe the current toolbar in our own terms, and list only Design and Render as the workspaces in
scope.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
The merge queue runs the full suite on exactly the commit it then puts on
main, and main's push run then ran the full suite again on the same files.
Select checks now asks the Actions API whether a passing merge_group CI run
exists for the pushed commit; if so the push runs only the repository checks.
A push that skipped the queue, or an API that cannot answer, still runs
everything. The nightly still runs the whole suite.
Pull requests also run less: the reference tests only when solidcraft-cli (or
anything it depends on) is affected, and doc tests only for the affected
packages that have a library. The ci check requires the reference tests only
when they were selected.
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#1583 turned the grid to the upright origin plane in side views for
everyone. That is now Preferences > Grid follows the view, off by
default; headless pictures keep the ground grid.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#1752 listed cs as complete, but it lacks messages added to en.json since,
which failed the completeness test on main. Missing messages fall back
to English until the catalog catches up.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The icon check from #853 requires every command to have one; #1900 added
inspect.clearance without, which failed Format and repository checks on
main and every PR.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#853 recorded seeds 13, 71 and 94 with the face and edge counts its own branch built
(40/101, 19/50, 27/63). With main's kernel they still build as they did before #853
(41/103, 20/52, 26/61), so every full goldens run has failed since it landed. This
records them again; #853's seed 20 change and the other 93 seeds are unchanged.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Clippy, the doc tests and test shards 1-2 of pushes to main and merge-queue
groups run on the self-hosted runner, alongside GitHub's runners for
everything else. Pull requests, the nightly run and Run workflow stay on
GitHub's runners. Clippy and the doc tests keep warm build directories
outside the checkout (scripts/ci/warm-target.sh).
The test archive is still built on GitHub's runners: the self-hosted
machine has a newer glibc, so binaries built there don't start on
ubuntu-24.04.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* claim: #1801
* Insert SVG: the file picker opens on SVG files (#1801)
Every Open went through one host picker whose filters start with designs, CAD
files and meshes, so Insert SVG (and Insert DXF) in a sketch opened on STEP,
IGES and meshes, and the SVG file had to be found under another filter.
The host's pick_open now takes the extensions the command reads and offers
them as the picker's first, default filter. Insert SVG and Insert DXF pass
their type; every other Open passes none and keeps the list it had. The
browser host ignores them (it shows its own file panel).
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
Replace the incorrect trunkrs.dev landing-page link with the trunk-rs/trunk GitHub repository and add the project's documentation website so contributors can find the web build tool and its setup instructions.
Signed-off-by: Karl Von Richter<karl@embeint.com>
Only the Windows jobs still wait for GitHub's runners on main and in the
merge queue. The Vulkan packages are preinstalled on the self-hosted
runner (no sudo), so that step runs only on GitHub-hosted runners.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Main's CI has failed since 14:47 on `robustness_goldens::all_3`: seed 55 (a
filleted box cut by spheres and patterned) used to end with an invalid body and
now passes the validity check, and its volume moved by 1.3e-5 relative.
The outcome changed between 2ae21143 (main's last green run) and 7f9074d6,
where the kernel meshing change is #1670 (twisted ruled faces divided to the
chord tolerance). The volume changed between 8a055afd and 31b2ddc0, where
#1753 changed truck-meshalgo's triangulation (the faces of an edge share its
points exactly). The kernel's validity and volume checks measure those meshes.
Pull requests run every 8th seed, so only main's full run saw seed 55.
Only seed 55 is recorded, with the values main's CI printed on Linux.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LGeCa5jWWdroMWvXxN3LLM
SolidCraft read the HID device directly, which bypasses spacenavd and
everything set in spnavcfg (swapped and inverted axes, sensitivity, dead
zones). When the daemon's socket is there it is read instead, with a fixed
32-byte record decoder and no library; without it the hidraw path is
unchanged. The daemon sends a record only when values change, so its
readings are held until the next one rather than going stale.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A constraint is applied with its first-picked entities held and the second
moving; when the second can't move, a first one moves instead. That fallback
was skipped as soon as ANY first entity was locked, so a free construction
line between two fixed lines was never tried, and Symmetry was refused with
"neither entity can move".
Each first entity that can move is now tried on its own, after the second.
Nothing changes where the second can move.
- motion.plot: joint values and a point's path sampled as the study plays (closed loops
following), with velocities and accelerations by finite differences, per step or per
second with a duration.
- motion.interference: the pairs of parts that pass through each other along a study, with
the first step, the step ranges, the depth and the shared volume; occurrences, against
or contact sets choose the pairs; bounded samples and work.
- Motion Study dialog: a Study box opens a saved study; a chart of a joint value or a
picked point with a cursor at the step shown (click to move there); Check collisions
with the steps shaded on the chart. ui.start motion.study takes study.
## Spanish (Spain) Localization 🇪🇸
This contribution adds a Spanish (Spain) translation for SolidCraft.
The translation is based on the English localization file (`en.json`) and uses terminology commonly found in professional CAD software such as Autodesk Fusion 360, SolidWorks, and FreeCAD.
### What's included
* Spanish (Spain) localization file (`es.json`).
* Translation of menus, commands, tools, settings, and interface messages.
* CAD-specific terminology adapted for Spanish-speaking users in Spain.
* Original JSON keys and dynamic placeholders preserved.
### Notes
* The translation has not yet been tested within the SolidCraft application.
* The Spanish language may need to be registered in the language selector.
* Feedback and corrections from native Spanish speakers are welcome.
I hope this contribution helps make SolidCraft more accessible to Spanish-speaking users.
Thank you for developing and maintaining this open-source CAD project!
Seed 31: #1581 refuses a fillet selection with an earlier round's smooth seam instead of re-rounding another nearby edge. Seed 83: #1316 was recorded before its branch took in #1495, whose shell keeps the wall beside a concave edge.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Pushes to main skip boolean_props (random inputs); the nightly property-tests job runs it. The three shrunk cases are ignored tests with their issues (#1647, #1648, #1649), and proptest's failure file is named outright.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Comment /claim on an issue to be assigned and get the in progress label; /unclaim releases it. A daily sweep frees claims with no PR after 72 h or with a PR idle for 14 days.
A scale across a revolution's axis keeps the face a surface of revolution and is written as one; any other refusal from the writer is an export error, not a panic.
Check gh pr list --search "#N", then open a draft titled "<summary> (#N)" with an empty claim commit; the lowest-numbered open PR for an issue wins, and a draft with no pushes for 72 hours may be taken over after a comment. Agents on a shared machine check the load before a big build.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The search-order test wrote an extensionless codex-acp, which Windows
discovery rightly skips; it now writes the first name this platform
looks for (codex-acp.exe on Windows, codex-acp elsewhere).
Fixes#1554
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The workspace switcher's menu closed on the press (a press outside the
menu) and the click reopened it; the menu now keeps the button that
opened it, a press there is not outside, and its click toggles. A
panel's label toggles its command list instead of always opening it.
Fixes#1522
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Message catalogs sorted one message per line (cargo xtask lists [--fix], checked in cargo xtask ci and the Lists workflow); the language list one per line, English first then by id; THIRD-PARTY-LICENSES.txt and docs/features.md rewritten on main by the Generated files workflow instead of in PRs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Grok Build is added as one more entry in the agent discovery list, over ACP, like the others. Thanks @cfaulkingham.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A PR that saves something new adds crates/doc/src/format/pending/<name>.rs (a note, and an upgrade step if older designs must be rewritten). The build writes the next format number while changes are pending and lists them in format_pending; scripts/fold-format-pending.sh fixes them as v<N>/ at release time. Formats 23 and 24 move into v23/ and v24/.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Clippy, the test archive and the test shards cover the packages a PR changes plus their dependents. Main, nightly and manual runs stay full.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
An Edge Flange feature can set the shape and size of the relief at the corners its flanges make, in place of the sheet metal rule's. Design format 24.
Co-authored-by: NOWN-DEV <NOWN-DEV@users.noreply.github.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Adds File > Export > Current View as DXF: the visible bodies projected from the current camera onto 2D LINE/LWPOLYLINE entities, with hidden edges omitted or dashed.
Co-authored-by: cronos <19868002+crono2250@users.noreply.github.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
FreeCAD .FCStd and Rhino .3dm files open, insert and drop like STEP and IGES:
cadmpeg (Apache-2.0, pinned =0.6.1, native builds only) detects the format,
decodes it under its service resource caps, validates the model and writes it
as STEP AP214, which becomes an Import feature in a new component named after
the file, with the up and names options.
Only the fcstd, rhino and step codecs are compiled in (no default features).
ACIS is left out: cadmpeg's sat codec is a ShapeManager stream decoder.
native_cad.rs is the single importer seam. Self-made FreeCAD and Rhino box
fixtures, hostile-input, ZIP-bomb and cap tests, engine tests for the
component path; the browser build reports these formats as not supported.
A note under design format 23 says an Import's file can name these formats.
Co-authored-by: Jacob Oursland <jacob.oursland@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Thin extrude caps are certified by the curves the wall follows, so a sketch attached to a cap survives switching between solid and thin (#1519).
Co-Authored-By: Philipp Bischoff <PhilippBischoff0@users.noreply.github.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Motion links can carry a coupling (gear pair, rack and pinion, screw) whose ratio follows radii, picked round faces or pitch. Design format 23.
Co-authored-by: SychPL <SychPL@users.noreply.github.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The web app reads a 3D mouse through the Gamepad API and drives the camera through navigation::device, so it follows each page's navigation capabilities (#1336).
Co-authored-by: CamWight <CamWight@users.noreply.github.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Adds an Interface font setting that leads the proportional family, with the default face and CJK/Hangul fallbacks kept behind it (#1290).
Co-authored-by: DuBearte <DuBearte@users.noreply.github.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A field the definition doesn't have is refused with the fields it does have, instead of reporting success and changing nothing (#1491).
Co-Authored-By: Erikas <erkexzcx@users.noreply.github.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
glTF files open, insert and drop like 3MF/STL/OBJ: a component named after the file, the node tree as nested rigid components, repeated meshes as occurrences, metres to mm, Y up turned to Z up by default (up: z keeps axes), materials as PBR appearances. Bounded, hostile-input tested reader; no new persisted data. Closes#1505.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Copy, cut and paste sketch geometry (Ctrl+C/X/V, context menu)
Sketch geometry had no copy or paste: Ctrl+C / Ctrl+X / Ctrl+V arrive as
clipboard events that nothing handled, and the sketch context menu had no
Copy, Cut or Paste.
- sketch.copy / sketch.cut / sketch.paste: lines, arcs, circles, ellipses,
splines, conics, points and text. Constraints and dimensions wholly among
the copied entities come along (each dimension gets its own parameter
with the same expression); ones that reach outside, fixes and pattern or
offset relations are dropped. Paste goes centred at `at`, by
`translate`, or offset from the original (same sketch) / at the same
coordinates (another sketch, on its plane). One undo step; the pasted
geometry is selected.
- Ctrl+C / Ctrl+X / Ctrl+V in a sketch (paste at the cursor), and Cut,
Copy and Paste in the sketch context menu. Outside a sketch Ctrl+C /
Ctrl+V copy and paste the selected timeline features. A copy also puts a
line of text on the system clipboard so the next Ctrl+V is delivered.
Fixes#1480
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Keep a pasted text's height and angle expressions, drop its place's
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Checked locally on main merged with this: cargo check --workspace and the ui-egui tests (the only failures are the local-only TexturesDelta ones that main has too).
The product preview asked the viewport executor for the full physical
viewport, so a maximised 4K window (~8 MP) was refused against the 4 MP
limit on every frame. The preview now draws at the largest size of the
same aspect the executor admits (RenderLimits::fit, scratch included)
and filters it up into the viewport through the existing offscreen
blit. Viewports within the limits still draw at full size.
PNG setups past the image limits now say what the limits are and the
largest size at that aspect, and the image panel offers to use it.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A sketch text's height, place (x and y) and angle can each be an expression over the design's parameters, and the text follows parameter changes, undo and reopening. Closes#1362.
Co-authored-by: NOWN-DEV <NOWN-DEV@users.noreply.github.com>
The flat pattern now keeps a 2D view of its own, driven by navigation::handle with orbit off,
so pan, zoom to the cursor and reverse zoom follow the navigation preset. Input over it no
longer reaches the Design camera, and Fit and Home frame the pattern. Closes#1483.
* Chamfer: refuse a result that grew a convex edge's part (or shrank a concave one's)
A chamfer of a curved edge that runs off a face it skims (robustness seed 57's
ball rim) built a part bigger than before. Checked from the two faces' normals
at the edge; rounds and unclear angles are left to the wedge bound.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Action matrix: the sheet metal tray's relief corners refuse a 0.25 mm chamfer
Each is a 2.5 mm concave edge (the corner of a corner relief, between two
planes at right angles): a chamfer there fills at most 0.25²/2 × 2.5 = 0.08
mm³. What was built took 0.95 mm³ away; the chamfer guard now refuses it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Robustness goldens: seed 57's wrong-way chamfer is now refused (#1390)
On main with #391, seed 57's chamfer grew the part; this refuses it, so the run goes on
to its later steps and ends ok.
---------
Co-authored-by: Phanoob <phanoob@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* Agent chat prototype: drive an ACP agent the user already has from a side panel
A desktop-only chat panel (off by default) that starts a coding agent over the
Agent Client Protocol and gives it SolidCraft's MCP server, bridged to the open
window through the token-protected control channel with the default folder
limits. It finds installed agents (Claude Code and Codex adapters, Gemini CLI,
Copilot, Qwen, OpenCode, goose) with version and login state, streams text,
tool calls and plans, shows the agent's permission requests, and when an agent
isn't signed in shows its own login methods without handling any credential.
Design notes and findings: docs/design/in-app-agent-chat.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* docs: say only what the end-to-end run showed
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat: gate, one undo per message, docking, process groups, discovery polish
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat: Settings → AI, several agents with a switcher, OpenAI-compatible endpoints, permission modes, toolbar button
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat: translations, tool cards, Add agent dialog fixes, test agent saves on its own
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat: third-party licences for the keychain crates; clippy
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat: badge text from the theme tokens
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Agent chat: Auto-allow model edits is the default permission mode
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A motion link's ratio is read in b's units per a's (mm per degree, degrees per mm) when an angle and a distance are linked; stored values stay per radian, so saved designs are unchanged. Follow-up to #1415.
Co-authored-by: SychPL <SychPL@users.noreply.github.com>
Adds a bounded normalization path for PNG16 and ICC-bearing JPEG texture inputs, producing float RGBA with recorded color and orientation decisions.
Co-authored-by: 0xlillith <0xlillith@users.noreply.github.com>
Adds a read-only review window to Validate Physical Material Data: paste a schema-1 card, see its known properties in SI units, unknown ones marked, with stated conditions and source claims. Nothing is assigned or saved.
Co-authored-by: 0xlillith <0xlillith@users.noreply.github.com>
* Render preview uses the shared camera navigation
The Render workspace's product preview had its own hard-coded camera
controls. Camera input now lives in one component, navigation::handle,
which the modelling viewport and the Render preview both call (NavCaps:
orbit, pick). Render gets the navigation preset, Orbit around, zoom to the
cursor, the middle double-click and the axis view or navigation cube.
Flat pages can call it with orbit off. A test rejects pages that read the
wheel or move the camera themselves.
Fixes#1454
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Render panel hint says the camera moves as in Design
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The continuity core (kernel bridge: Hermite control rows reproducing a support surface's position, tangent, curvature and change of curvature, G0 to G3), surface.blend between two edge chains (#764), and a crown for fitted patches. Existing patches and fillets are unchanged by default (crown 0 is the plain patch bit for bit). SurfaceBlend and Patch.crown are noted under format 22.
A point within the point tolerance of a boundary curve's end is that end in the boolean, whatever its parameter, so text cuts no longer leave unmatched sliver vertices (#391). Robustness goldens re-recorded for seeds 0, 34, 39, 41 and 57 with reasons.
Every other package (AppImage, deb, rpm, tar, Flatpak, Snap, MSI, portable zip, FreeBSD) already
puts solidcraft-cli next to the app; the macOS bundle only had it in a separate zip. The agent
chat starts it as its agent's MCP server and looks for it beside the running program.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch: the first dimension scales a sketch drawn by eye (#1365)
Lines drawn by eye have no size of their own until a dimension gives them one, but the first
dimension kept every other drawn length and distorted the sketch to fit (a free construction
line through the origin turned instead of the shape moving). The dimension tool now scales
the whole sketch about its origin to the first length dimension, as long as nothing in the
sketch fixes a size yet: no other driving length dimension, nothing fixed, projected or
patterned. Every geometric constraint and angle holds on the scaled sketch. Later
dimensions size what they measure as before; sketch.dimension does this only with
scale_sketch: true, so scripts are unchanged.
Fixes#1365
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Scenario s01: dimension the box's height too, now that the first dimension scales the sketch
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Material-card validation now names the exact condition fields missing for each property (missing_condition_fields), alongside the existing unspecified_conditions.
Co-authored-by: 0xlillith <0xlillith@users.noreply.github.com>
New frame::add_outlined adds a button that draws its own outline with the theme's outline width set to the same width in every state, so its padding no longer drops a pixel on hover. Used by the dialogs' Select chip, Sketch Text's Path chip and the bug report's Open GitHub draft button. New dialog_chip_hover test. Fixes the second point of #1335.
startup_steps_survive_a_rebase_and_running_ends_startup reads the open journal through a clone of its own handle (read_journal_file), as read_live_journal does, instead of opening a second handle that Windows' mandatory lock refuses. Test-only. Fixes#1448.
New Body extrusions and Offset Face previews show successful intermediate distances from the same setup while a distance drag runs, then compute the latest one (tool previews keep streaming as before; errors still need exact inputs). The drag arrow and value box are painted from this frame's pointer, removing a frame of lag. Drag regression tests and a perf note added.
List menus open below-right of the cursor and flip to end at the cursor where they don't fit (shared context_menu::place / show for viewport, browser, toolbar and Select Other). Right-clicking a plane in the viewport, a construction plane row or an origin plane row (new menu with Create Sketch and Show/Hide) offers Create Sketch on that plane. Fixes#1354.
AppRun installs nothing when an appimagekit_*.desktop entry for the same AppImage exists, and removes the entry it wrote earlier. An empty ~/.local/share/solidcraft/no-desktop-integration turns the integration off for good; SOLIDCRAFT_NO_DESKTOP_INTEGRATION=1 still works. Getting-started guide documents it. Fixes#1399.
scenario::scratch_dir builds the harness folder under solidcraft_engine::io::vfs::temp_dir() (long names), so a short 8.3 TEMP on Windows no longer trips the short-name refusal before the scope checks the scenarios test. Fixes#1385.
joint.motion_link offset is read in the linked value's units (degrees or mm, or an expression); links are still stored in radians, so saved designs open unchanged. A new joint whose values fall outside its limits starts at the nearer limit, stores it, and says so in warnings. Loop dof also counts at a nearby pose, so a part at a dead point of its loop keeps its mobility. The mechanisms skill is updated to match.
New command inspect.compare {a, b, tolerance?}: measures of both and their differences, two-way surface deviation from a bounding-volume tree, regions beyond the tolerance clustered by side (only_in_a / only_in_b) with integrated volumes and faces, a verdict, optional face colouring as one undo step, and warnings for open meshes. Geometry in solidcraft-geom (compare.rs), command in the engine. Closes#1363.
"2,5 mm" is 2.5 mm; inside a function's parentheses the comma still separates the arguments. Noted under design format 22: an expression typed with a decimal comma is saved as typed, so 0.12.0 asks for an update instead of failing to evaluate it.
A curved chamfer chain with straight planar ends can end across two earlier planar bevel faces, shortening their shared seam to where the new chamfer plane crosses it; picks on one tangent chain are applied once. Fixes#1303.
The mesher samples edges through a fallible evaluation, so an intersection curve whose projection fails to converge no longer panics; a boolean result with an edge that does not evaluate is refused with an error. Fixes#1252.
* Imports become components: a STEP assembly's tree as nested components (#1386)
Every imported file (STEP, IGES, 3MF, STL, OBJ: inserted, dropped, opened as a
design, or file.insert_step / file.insert_mesh / doc.open from scripts, the CLI
and MCP) now comes in as a new component named after the file, with its Import
feature inside it. A STEP assembly's products become nested components; a part
placed more than once is one component with an occurrence per placement, and
the Import feature skips the file's later copies (`skip`, format 21 note). The
contents are held as one rigid group (rigid joints, glyphs hidden); moving an
inner part is refused until component.unlock ("Unlock Contents") releases it.
Y up turns the import's component. Replies report component, component_name
and occurrence. Undo is one step; designs with loose imports are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Keep an import's rigid group hidden when the window loads the file in the background
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Import instances' skip list is format 22: format 21 ships in 0.12.0
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Catalog entries for Unlock Contents
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* View-dependent display refinement: bodies seen up close are drawn from finer display plans (#1320)
Each frame the viewport works out every shown body's screen-space error (the display
quality's chord tolerance in pixels where the body is nearest the eye) and picks a
refinement step past the chosen display quality, each halving the chord, with hysteresis.
Steps are planned off the UI thread and swapped in when ready, within a triangle budget
and a cache size. Faces (GPU plan or CPU mesh), edges, silhouettes and highlights of a
refined body all come from the refined plan, so a highlight stays on the drawn triangles.
Planned bodies are chunked by content on the GPU, so a refinement re-evaluates only its
own chunk. Display only: the reference mesh, the B-rep and exports never change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Display refinement: only collect finished plans while the view stands still, time the choice, keep the web planning one step a frame
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Format
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Keep the Display quality note as it was (the longer one pushed the Colour vision list off the Graphics page)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Web app: read a STEP file a piece per frame, with progress and Cancel (#1376)
The kernel's STEP reader walks the file first and converts its solids afterwards,
and StepReading exposes those stages: parse and walk, then convert a solid (or a
batch) per step, then finish into the shared import cache. The web app's file
load drives it from the frame loop for about 30 ms a frame, shows the same
progress card and Cancel as the desktop, and then runs the insert or open, which
finds the file already read. The desktop path is unchanged in behaviour.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Web app: make the imported bodies' meshes a few per frame too (#1376)
After the read, the insert or open runs, and then the bodies' display meshes are
made a frame's worth at a time ("Preparing the view"), as the desktop worker
warms them, instead of all in the first frame that draws the design.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
component.insert_design (a saved part as a linked component) had no button, so an
assembly of separately saved parts looked impossible. It is now on the ASSEMBLE panel
of every tab and in the File menu; it asks for the file, and a STEP or mesh file picked
there is inserted as Insert STEP or Insert Mesh would.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Windows: resize border from the first frame, DPI-correct window moves (#1381)
- The egui resize zones were laid out after the start page's early return, so a launch without
a design had no resize edges at all. They now run on every page.
- Windows: the main window's procedure answers WM_NCHITTEST with the resize edges and corners,
Windows' own border width at the window's DPI (8 px at 100 %), so the OS shows its resize
cursors and runs its own resize (snap included) from the first message; no egui zones there.
- Windows: WM_GETDPISCALEDSIZE returns the size winit then applies (same logical size), so the
rect the OS suggests while the title bar is dragged onto a monitor with another scale is the
rect the window gets.
- Zone sizes and hit test shared in titlebar.rs and unit-tested; Windows CI checks the hit test
right after launch and a real edge drag.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Windows resize smoke: find the main window by its title
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Windows resize smoke: probe the maximized window at the screen's edge
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Measure now reads the picked item's exact shape (cylindrical faces, circular and arc edges,
sketch circles, arcs and lines, profiles, planes, points):
- A circle, hole or arc shows Ø, R and its length (circumference, or arc length and angle).
- A slot (a sketch profile, or a half-cylinder end of a slot in a solid) shows its width,
centre-to-centre length and overall length.
- Two items where one is round measure from the centre or axis by default (axis to axis,
axis to plane, edge or point), with the closest and farthest distances listed too; the
panel switches between centre, min and max, and the canvas line follows.
- inspect.measure and MCP `measure` (new `items` argument) add diameter, radius,
center_distance, min and max; existing fields are unchanged.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Fillet/chamfer: point the drag arrow into the corner on inside edges (#1384)
The radius/distance arrow always pointed against the faces' outward
bisector, which is into the body on an outside edge but away from the
corner on an inside one, where the blend adds material. The arrow and
the spline radius handles now take each edge's convexity into account,
so dragging out along the arrow grows the value on both kinds of edge,
and each edge of a mixed chain follows its own.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Tests: no unreachable! in the blend arrow helper
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Preferences ▸ Navigation ▸ Orbit around now offers View centre (the default, as before), Origin and Point under the cursor. A saved orbit_cursor preference migrates to the cursor option, and the control channel still accepts the old key.
Thanks to @kuraneko1 for this.
Co-authored-by: kuraneko1 <98832587+kuraneko1@users.noreply.github.com>
The groove's facets came in sweep order, and the BSP takes each list's first
polygon as its splitter, so the tree was as deep as the list: quadratic time and
memory. An M8 x 60 mm thread ran out of tree depth after 12.9 GB. A fixed shuffle
gives a shallow tree and the same result (volume and faces equal to 1e-9).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
fr.json covers the messages main added since #1224 (render workspace, design properties, display quality, surface untrim, joint drive stops, Save as Mesh), with the same terminology conventions.
A dimension to a line, or between two parallel lines, drew its extension lines from the
measured end point and its foot on the other line, so they ran the whole length of the
lines to wherever that end happened to be. They now start from the nearest point of each
line to the dimension line, and disappear where the dimension line already crosses it.
Fixes#1361
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Converted sheets take the rule's corner relief between walls and its bend relief where a bend ends beside more of the fixed face, and a wall on part of an edge bends only where it stands. Noted under format 21; nothing new is saved.
A new face-based Extrude Join (and outward Press Pull) joins the picked face's body by default and stores that scope, so an edge-touching neighbour is no longer tried first or absorbed. Explicit targets and participants still win. Related to #226.
The Joint and As-built Joint dialogs explain why picks make no joint and offer Create Components from Bodies when every pick is on one component; joint.create and joint.as_built suggest component.from_bodies. Labels in all catalogs. Fixes#1239.
Offscreen GPU test devices start the backend the app would choose (WGPU_BACKEND, else one hardware backend on Windows) instead of every backend at once. Test-only. Fixes#1330.
The Midpoint Line tool previews the line from the cursor's mirror image through the middle to the cursor, which is what the second click commits. Fixes#1358.
Fixes#1187. New parameters.move {name, before|after} reorders user parameters (the design's own params order, so it is saved with no format change, and undoable); the Parameters dialog's user rows get up/down buttons. Label in every catalog. I merged main in. Checked on main + this branch: cargo check --workspace, the engine parameters tests, the UI parameter_order test, and docs/features.md regenerates unchanged.
Closes#1284. sheet.export_dxf writes layers by role (OUTLINE, CUT, TEXT or ETCH, dashed BEND_UP/BEND_DOWN, optional BEND_NOTES, POCKET_<side>_Z…_MM and PERFORATION), inner contours before the outline; text: cut|etch, bend_notes and note_height, per_body files. Letters are derived from the timeline, nothing new is saved (format unchanged). I merged main in (the export's parameter list keeps main's pocket layer names). Checked on main + this branch: cargo check --workspace, the engine cmd::sheet tests (160) and the io DXF tests.
Fixes#1167. While a sketch is edited, a body that a rolled-off, unsuppressed Remove deletes is not shown (unless the sketch is built on it); the history stays rolled back, only the view changes. Drawing, picking, screenshots, thumbnails and the browser's Bodies folder agree; document.inspect reports removed_later. Checked on main + this branch: cargo check --workspace, the engine sketch_edit_removed/occurrences tests and the UI viewport/browser tests.
Part of #771 and #772. solid.sweep and surface.sweep take orientation (perpendicular, parallel or a direction), twist (up to 4 turns) and end scale (0.01 to 100), applied about the profile's normal through the path's start; default options run the old sweep, so existing sweeps are unchanged. New fields are optional, left out when unset and noted under format 21; twist and scale are model parameters. Both sweep dialogs get Orientation, Twist and End Scale rows; labels in every catalog. Checked on main + this branch: cargo check --workspace, kernel/engine/UI sweep tests, the doc suite and the robustness goldens.
One docked Save as mesh dialog replaces Export STL / 3MF / 3MF with face tags, with an optional Send to slicer that lists the slicers installed on this machine. Engine: file.save_mesh and file.slicers; the Convert Mesh dialog goes (mesh.convert and the MeshConvert feature stay, so existing designs open). Strings in all catalogs; screenshots in the PR. Checked on main + this branch: cargo check --workspace, the engine slicer/mesh/file tests, the UI catalog and menu tests and scenario s14 (the one local colour-token failure is epaint's TexturesDelta panic, which main shows on this machine too).
Fixes#1075. sketch.constraint.symmetry takes a centre point (the origin or any sketch point) as well as a line; the Symmetry tool's third pick may be the origin point or the sketch's own X/Y origin axis. Checked on main + this branch: cargo check --workspace, the engine cmd::sketch tests (257) and the UI symmetry pick tests.
* Optional supersampling (SSAA) for the viewport, adaptive and within the texture limit (#1288)
Preferences ▸ Graphics ▸ Supersampling: Off (default), 1.5× or 2×. The frame is drawn into
textures of the viewport's own kind (format, depth/stencil, MSAA) at that multiple of its pixels,
over transparent black, then filtered down into the viewport with four bilinear taps (the exact
2 × 2 box at 2×) and composited premultiplied over the background. Lines keep their on-screen
width; silhouettes and edges come out smoother.
Adaptive (on by default): the view is drawn at full speed while the camera moves and
supersampled once it has held still for a quarter of a second.
The supersampled size stays within the device's largest texture side (#869) and a 384 MiB
budget for colour and depth (MSAA samples included); under 1.25× is not worth it and the frame
is drawn as usual.
Labels in en/de/ja/ko/zh; `ui.prefs` `supersampling` and `supersampling_adaptive`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Supersampling: factor and adaptive on one row; Off reads Off (1×)
The Graphics page grew past the colour vision menu in the preferences scenario, and a second
Off on the page took its click.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
solidcraft-cli snapshot takes --camera JSON and --view with a saved view name (through view.apply, as the MCP screenshot does), plus --no-grid and --no-axes; the MCP screenshot and ui.render take grid/axes, the headless render skips a hidden origin axis, and the app's ui.render follows the Grid and Origin switches. Closes#1245.
A Convert to Sheet Metal bend point can now lie on a bend modelled with its radius (a cylinder tangent to two flat faces) or in the middle of a chamfer; either becomes one bend at the faces' virtual sharp, with the rule's radius. Holes and engravings carry over. Bend points on edges convert as before. Recorded under format 21. Tests in cmd::sheet::convert_tests; the sheet-metal guide is updated.
Drive Joints opens on the joint picked in the browser, else the first joint that moves, and won't start on a rigid joint. ui.start takes params: joint.drive opens on a joint (id or name) at given values, driving nothing until OK; a command or param it doesn't take is refused.
file.insert_step and doc.open on a STEP file take `up` (turn a Y-up file so its up is +Z) and `names` (names for the imported bodies). The MCP tool and the app's control requests pass them through; an import without them works as before. Tests in the kernel's STEP reader, the engine, MCP and the file-load UI; docs/step-import.md and docs/mcp.md describe both.
A solidcraft-mechanisms agent skill with a worked example that a test replays through the MCP server (the actuator follows the hinge with the stated stroke and the drive stops at the rod's limit); the skills index and docs/mcp.md list it.
While Draft, Accessibility or Curvature Map analysis is on, a legend at the bottom left of the viewport says what each colour means and which values it stands for, with the theme's analysis colours and stripes. Labels translated in en/de/ja/ko/zh. Fixes#973.
Without a 3DxWare profile, Linux now uses the driver's object-mode mapping (physical Ry and Rz swapped and reversed), so a tilt rolls and a twist spins. Windows and macOS are unchanged; a profile still takes precedence. Fixes#1172.
The bridged MCP server starts when no app is listening, says so in its instructions and in each tool error, and connects on the first call after the app starts. A batch that could not be sent says the app is not running. docs/mcp.md describes it.
snapshot starts from --in (a design or STEP file) with the script on top, as run does, and refuses options it doesn't take; an unknown command id names the closest registered ids.
A control-channel method that reads the orbit camera, sets any of target/yaw/pitch/roll/distance, and fits listed occurrences (or everything) keeping the direction. Its reply passed back sets the same view again; a refused field changes nothing.
joint.drive answers moved, and a reason (with unconnected) when the joint can move nothing; motion.play answers moved and unmoved; joint.list answers floating chains. joint.drive values come back in joint.list's shape (deg/mm). quiet: true trims joint answers and motion.play poses. A single value keeps a joint's other values. Drive Joints shows the floating case (drive.unconnected, all locales).
component.list {compact: true} gives counts of bodies, features, sketches and construction planes instead of their names, lists occurrences without identity transforms and children as occurrence ids; bodies: true keeps body names. The default form is unchanged.
* Preferences ▸ Graphics ▸ Display quality: Coarse, Normal, Fine, Ultra (#1287)
A viewing preference that sets how finely the viewport tessellates curved faces and edges:
chord tolerance relative to each body's size (3e-3, 1e-3, 3e-4, 1e-4) and the most a facet may
turn (30, 20, 12, 7 degrees). Normal is the tolerance the viewport has always used.
Bodies keep a display plan per level, so switching is a re-plan the first time and instant
after; the reference mesh the model reads (picks, holes, measuring) and exports are unchanged.
Away from Normal the model scene builds edges, and faces on the CPU path, from the plan at that
level, so silhouettes and edges get finer with the faces and a highlight still lies on the
displayed triangles.
The preference is saved with the others, set over the control channel (`ui.prefs`
`display_quality`), labelled in en/de/ja/ko/zh. `solidcraft-cli bench-open --display LEVEL
[--edit NAME EXPR]` times the plans at a level and after a parameter edit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Highlighted edges follow the display quality too
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A regression test: an inside flange on a 100 x 60 plate turned by every whole degree from 0 to 89 builds and keeps the analytic volume. At 62 degrees the trimmed corner used to land on a border of the sew's vertex grid and the flange failed to close.
A fillet or chamfer on a cylindrical boss's bottom arc now builds with a closed display mesh: the seam-extension containment ray counts a shared curved-boundary endpoint once. Merging the split faces of a three-edge arc corner rebuilds the face on the original input face's full surface instead of a piece's restricted patch, and runs under the kernel's panic guard. Fixes#1262 and #1263.
Co-authored-by: Karl-VonRichter <20079846+Karl-VonRichter@users.noreply.github.com>
The empty design shown at launch, New on a design that is still empty, file.new and the blank design left when the last tab closes all take the Units of new designs preference. Opened files and the sample part keep their own units. Fixes#1205.
Co-authored-by: raythurman2386 <raythurman2386@users.noreply.github.com>
Section Analysis clipped the origin and construction planes and axes along with the model. They now draw from their own unclipped lists, with the same depth and blending; bodies, edges, sketches and highlights are clipped as before. Fixes#1229.
Co-authored-by: DuBearte <DuBearte@users.noreply.github.com>
A symmetric solid extrude of 10 mm is now 10 mm thick, 5 mm each side of the sketch plane, and switching from One Side to Symmetric keeps the thickness entered. Under format 21 the extent records its convention; designs saved without it keep the per-side reading, so their geometry and parameters don't change. Fixes#1236.
Co-authored-by: Karl-VonRichter <Karl-VonRichter@users.noreply.github.com>
A round or chamfer whose edge chain runs on across a curved seam (the end circle of a lofted tube whose axis bends) is now built: the section solve keeps its contact on the seam through a plane taken afresh at the seam's nearest point at each step. Straight seams are unchanged. Robustness seed 41 is re-recorded: its second fillet, refused before, now builds. Fixes#1213.
Co-authored-by: gvocale <gvocale@users.noreply.github.com>
A fillet or chamfer that consumes both seams of a side, leaving its old boundary and the fitted contact on the same curve, now returns TooLarge ("the blend would consume a neighbouring face") instead of a zero-area face that doesn't mesh. Genuine two-edge faces are kept. Fixes#1113.
Co-authored-by: Karl-VonRichter <Karl-VonRichter@users.noreply.github.com>
A modelled thread as other CAD systems write it to STEP has cylinder faces trimmed by helices
that turn several times around the shaft, and cylinder faces bounded by two circles with no
seam edge. The display planner refused any face whose boundary winds around its surface, so
the whole body fell back to the reference mesher, which drew those faces with triangles across
the solid.
The planner now cuts such a face along a parameter line into one period: every loop and its
copies a period apart are clipped to the strip, and the parts of the cut line inside the face
join the pieces into closed loops, with the seam nodes made once for both sides. The cut face
is meshed by Delaunay with sag refinement, and the sag check runs once more after shaping.
Bodies with a face trimmed by more than one turn take their reference mesh (picking, measuring,
the CPU viewport) from the plan too.
Tests: an ISO M10x1.5 and M12x1.75 thread built from helices, as made and after a STEP round
trip: the display mesh stays within the tolerance of its surfaces and the thread's volume, and
has no degenerate, sliver or flipped triangles.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Action matrix regressions of #1255: knuckle rounds, thin holes, a 45° flange hole, an axis after an edit
- Curve blends: past the end of a curved edge the run of sections along the end
tangent wrapped round the ring of centres; it is shortened while it still
crosses the end face (the hinge knuckle's cap rounds at 0.728 mm).
- Boolean retry guard: each body's volume is measured finer than its smallest
radius, and a volume out of bounds is measured again ten times finer before
it refuses (a 0.133 mm drill measured -30272 mm³; a plastic part's coarse
mesh was off by more than the slack).
- Sheet metal: a drill cap's certificate allows the evaluation noise its fitted
tilt shows (a hole square to a 45° flange).
- Construction geometry on a named face uses the face found by name, not a
second pick at its nearest point (an axis through a corner round after a
parameter moved it).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Retry guard: measure the target, tool and result at one tolerance
Measured at different tolerances, the chords' error on the faces they share
didn't cancel: a blind hole in a long D-profile bar measured 20 mm³ more than
the bar.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Action matrix baseline from this branch's full run: 37 actions pass again, none fewer
Blessed with cargo xtask action-matrix --from DIR --bless from the 16 shards of
run 38673859774.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The three walks over a feature (expressions, references, geometry) are generated from FeatureKind by #[derive(Walk)] in a new proc-macro crate, solidcraft-derive, from #[walk(...)] attributes written next to each field. The hand-written walks in params.rs, feature_refs.rs and feature_geometry.rs are gone; Ruled Surface and Extend have their attributes, and the three golden walk records are unchanged.
Co-authored-by: jijinbei <jijinbei@users.noreply.github.com>
A corner whose two curves end on separate points held together by a Coincident constraint (a line drawn to where an arc starts) is welded into one shared point before Fillet or a Chamfer rounds or cuts it, so it no longer reports that the corner must join two lines. Only a partner point held by nothing but that constraint is welded. Fixes#1208.
Co-authored-by: DuBearte <DuBearte@users.noreply.github.com>
* Title bar: inactive tabs without outlines, File menu, Help on the right
Inactive tabs go back to the plain bar colour (the Home tab's readable
hover stays). The separate Home button and theme toggle are gone: Home is
a tab and the theme is in Preferences. The folder icon becomes a File text
menu, Save moves into it (Ctrl+S unchanged), and Help sits on the right
next to Report a bug. Tabs start right after File, Undo and Redo.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Title bar: File button hover uses the app bar hover token
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
An opt-in Touchpad navigation preset: Shift + pointer motion pans, Alt/Option + motion orbits, vertical scroll zooms, all without a mouse button. Existing presets are unchanged; view.navigate gains continuous motion for the control channel and scenarios.
Extend works on any surface: natural (continuing the face's own surface), tangent (G1 run-off) or curvature (G2 run-off), by a distance, up to a plane or face, or back to the surface's bounds, with neighbouring edges lengthened to meet. Untrim and Extend share face parameter-loop helpers. The new SurfaceExtend fields are noted under format 21. Closes#760.
A design note for SolidCraft in the browser: tiered storage, a SharedWorker job broker, GPU dispatch and in-browser agent control through the same engine commands, with the generic pieces in craft-libs. Linked from the MCP and web getting-started docs.
A loft section with a hole is refused with a clear message instead of being lofted as if filled, in both section-to-section and section-to-point lofts. Fixes#1184.
The labelled Report a bug button takes its translated label's width, so longer translations no longer run over the Preferences gear; narrow bars still fall back to the compact icon. Fixes#1145.
Browser rows that are names someone gave (bodies, sketches, components, joints, views...) show as typed instead of being translated as UI labels; default names like Sketch1 are still translated. Fixes#1228.
Ruled Surface: strips off an edge chain in normal, tangent, angle or direction mode, exact planes/cylinders/cones where possible, mitred or round corners, stitched into one surface body. SurfaceRuled is noted under format 21. Closes#762.
* Smart Extrude: direction from the picked side, Cut into a body stays a Cut (#1257)
- A sketch profile (or a thin extrude's curve) extrudes toward the side it was
picked from: from the bottom view it goes down. Body faces still go outward.
- The automatic operation is latched per side of the profile: a drag that starts
into a body is a Cut and stays one past the far side; a Join does not become a
Cut when it reaches another body. Passing through zero decides afresh for the
new side.
- An explicit operation, Direction or Reverse ends the automatic choices.
- auto_operation also looks just past the profile, so a long cut through a body
reads as a Cut from the start (the engine's operation: "auto" too).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Scenario s07: the XZ profile seen from behind now extrudes toward the camera
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#1164 joins a body that sits flush on a rounded rim into it. In seed 13 the
rectangular pattern's second box (x 5..7.5) sits flush on the half-ring's top
face with a 27 mm2 contact patch. It used to stay a separate shell (one body,
three shells); now it joins the ring (area -54 mm2, twice the patch). The
fillet edge and hole position the generator picks next come from the new
geometry, so the end state differs: 3 bodies, 2647.95 mm3 (was 4, 2300.91).
Every step stays a sound solid, and no other seed moves.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Action matrix: nightly shards cut by measured cost, heavy designs in slices, and a determinism check
The full matrix ran as four fixed quarters of the corpus and two of them hit the 60-minute
job limit. Now action_matrix_shard runs shard k of n (ACTION_MATRIX_SHARD=k/n), planned from
tests/action_matrix/costs.json: designs that take more than five minutes are split into slices,
and the pieces go largest first to the least loaded shard. The nightly runs 12 shards and uploads
each one's outcomes and timings; cargo xtask action-matrix --from DIR --bless records them in the
baseline and costs.json without running the matrix again. action_matrix_determinism runs a sample
twice and lists any action whose outcome changed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Action matrix: costs measured on runners, 10 shards, determinism on the lighter designs
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Action matrix: baseline and costs from a full run on main, 16 shards, timeouts listed
The baseline is regenerated from the 10 shards' results of a full run on 01b1b705
(cargo xtask action-matrix --from DIR --bless): 8 new designs, 302 actions that pass now,
names that changed, and 40 actions that passed before and fail now, each listed in #1255
(22 fillets refused since #949, 4 hinge blends since #1101, 9 on the loft cut, 5 single).
costs.json now holds every design's measured time, and the nightly runs 16 shards of about
17 minutes. An action that passes in the baseline and times out is listed but not judged,
and stays passing when blessed: two runs of the same build differed only on 10 actions that
took 50-84 s against the 90 s limit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Action matrix: slice heavy designs by a hash of each action's name
Slicing by index gave some slices only fillets on oracle 41 (28 slices, a multiple of the
actions per edge), three times the planned time; two shards ran past their limit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The preview builds only the tool; the boolean runs once on OK, off the window thread, with a Computing label while it takes a while. Follow-up to #1020.
Lands the Render workspace (R3), bounded preview-image jobs with cancellation and atomic PNG export (R4a), and opaque PBR looks as one change.
- Render workspace: an overlay-free product preview on the installed GPU device. Returning to Design restores its camera. render.scene and render.settings.apply are added.
- Image jobs: render.preflight/start/status/cancel/export run bounded CPU preview jobs. Export captures the job, overwrite choice and document, and checks the caller's file scope before writing.
- Opaque PBR: looks can carry metal fraction, roughness and dielectric IOR. The product viewport and preview PNG use them; geometry and mass are unchanged. The v1 catalog export refuses PBR rather than dropping it.
- Render setups and PBR looks are saved under design format 21. Labels are in en/de/ja/ko/zh/ru.
Supersedes #706 and #800. Refs #620, #673, #756, #1004.
Contributed by @0xlillith.
Co-authored-by: lillith <35337584+0xlillith@users.noreply.github.com>
Adds a French catalog as a partial locale (missing strings fall back to English, like Russian), registered in the language list, with tests and attribution.
A design carries metadata (format 21): project, part number, revision, author, organization, description and license, plus other tools' data under reverse-DNS keys, kept as written. Edited in the Design Properties window (Document Settings), set by document.metadata.set, reported by inspect_design, and written into STEP, IGES and 3MF where the format has a place for them. Translated, with tests.
The AMD FirePro D300/D500/D700 Metal driver hangs on the tessellation compute shader and macOS resets the GPU, freezing the desktop. Bodies are built on the CPU on that device, and SOLIDCRAFT_CPU_TESS=1 forces the CPU path on any GPU. The check is a tested predicate, and the troubleshooting guide documents the variable.
A crash inside a graphics driver left only "an unclean exit between
commands": no backtrace, no module, no hint that it was at startup.
- The crash journal now keeps the startup step the app is in (window,
graphics instance, surface, adapter request, adapter chosen, device
creation, renderer and viewport shader compiles, surface configure,
first frame and its present, running) and the renderer as soon as the
adapter is picked. egui-wgpu tells the app about its steps through a
small hook. An unclean exit before the first frame names the step.
- Windows: an unhandled-exception filter (src/native_crash.rs, the one
Windows-only module allowed unsafe) writes the exception code, the
faulting module and offset, and a minidump next to the journal; the
next start puts them into the report.
- The Windows CI job runs a child that faults on purpose and checks the
report.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Colour vision modes (Protan, Deutan, Tritan) in Preferences > Graphics, with palettes, shader and icon colours adapted per mode, translations and tests.
render.settings.set saves validated named render setups (camera, output size, exposure, environment) as one undo step, and render.settings.get reads them back without applying a camera. When validated render settings are the only change, the engine keeps the evaluated model instead of rebuilding it. Setups are saved under design format 21; format-20 designs read as having none. Fixes#658; refs #620.
Contributed by @0xlillith.
Co-authored-by: lillith <35337584+0xlillith@users.noreply.github.com>
Publishing a non-prerelease release runs winget.yml, which opens the
winget-pkgs pull request for bherbruck.SolidCraft with Komac (pinned
2.16.0 binary, SHA-256 checked) from the three versioned MSIs. It skips
cleanly without the WINGET_TOKEN secret or before the package's first
version is in winget-pkgs. docs/releasing.md has the one-time token
setup; the README's Download section gains the winget install line.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Windows: start Direct3D 12 first, and Vulkan only without a D3D12 GPU
wgpu started Vulkan, D3D12 and OpenGL together and preferred Vulkan, so
every Windows start loaded all installed Vulkan drivers. An old Intel
Vulkan driver (igvk64.dll) crashed 0.10.1 at startup inside the driver.
The app now probes D3D12, then Vulkan, then OpenGL, one backend at a
time, and starts only the first with a hardware adapter (WARP when none
has one). WGPU_BACKEND still wins; graphics.json's "backend" sets the
order by hand; the software renderer (safe mode, the retry) uses D3D12
alone. A new Windows CI job starts the real app under several settings.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: a test filter YAML parses
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: the startup smoke test writes graphics.json where the app reads it
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* CI: run the WARP shader tests one at a time
WARP's shader JIT (d3d10warp.dll) crashed twice in 124 runs of the gpu
tests with four test threads, with a null read inside its own optimizer
while several devices compiled compute pipelines at once. Serial runs
did not crash.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Bounded in-memory PNG/JPEG texture decoding with explicit base-colour/data-map roles; RGBA8 pixels are published only after a complete decode. Input only: no material assignment, GPU shading or persistence. Refs #900.
Contributed by @0xlillith.
Co-authored-by: lillith <35337584+0xlillith@users.noreply.github.com>
Trimmed faces are cut into four-sided patches (squares around holes, butterflies for round faces, motorcycle traces from concave corners) and each is meshed as a Coons grid, with a cut grid as the fallback; cells are at most 2:1 along straight directions. Folded blends are refused rather than overlapped. The README lists display meshes and the gpu crate.
When projecting a face's boundary loop onto its surface fails from an unhinted first point, the walk is retried once from the first point an unhinted search does find; a walk that succeeds is unchanged. A fillet strip on an extended surface now meshes, so the rim cut in the reported case builds in one boolean.
The screenshot controls, the submission steps and the Open GitHub draft button stay outside the scrolling review; the clipboard fallback is explained before a long draft opens; text fields leave a pixel of room so the window no longer widens while idle at 140% scale. Instructions in en/de/ja/ko/zh.
A large design showed nothing but a progress bar until it had rebuilt completely. While
it rebuilds, the evaluation now offers the model built so far (placed as the design
places it) every 1.5 s to a window that asks, and the open card draws it as a small
picture on a thread of its own, the way it shows a saved preview. The bodies' meshes
made for it are the ones the finished model shows, so little is done twice. A design
that opens within the first interval shows none.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A primitive's `targets` live in the feature's participants beside its kind; renaming a body now renames them too, so a box or cylinder cutting the renamed body no longer fails with unknown body.
* Viewport: keep the model scene in blocks and upload only what changed
Every change rebuilt the whole model scene (every body's triangles and edges) and
uploaded all of it again, so a design an agent builds one body at a time paid for the
whole model on every command. The scene is now built in blocks of a few bodies, cut
where a body's own key says, and each block's bytes and GPU buffers are kept while the
block comes back unchanged. Body boxes (fit, picking, ground shadow) are kept per mesh.
Rendering is unchanged: the bicycle from bench-agent opens to a byte-identical
screenshot.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Browser: find an occurrence's placed bodies by name, not by search per row
Each body row of a placed component searched the placement's bodies for its own, every
frame: quadratic in the component's bodies (720 for the bench-agent bicycle's hardware).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The reader converts each item once, but every instance made its own display mesh
(tessellation, seam and tangent flags), the most expensive step of opening a large
assembly. A later instance of an item now shows its first instance's mesh, placed by
the difference of the two placements. Only rigid placements share; split lines and
any operation drop the link. Measurements and export read each instance's own
geometry, so eval output is unchanged.
Generated assemblies: mesh phase 2,215 -> 139 ms (901 bodies) and 1,028 -> 85 ms
(5,401 bodies), with the same triangle counts.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
sheet.export_dxf writes an AutoCAD 2000 drawing with ELLIPSE and SPLINE entities for ellipses, splines and text outlines instead of chords, with curves: arcs for machines that read only lines and arcs. Perforation circles go through the same writer. Closes#1089.
Rectangular and circular sketch patterns stay linked to their source: editing the pattern's count, spacing or the source geometry updates the copies, and the copies keep their point and radius relations. Patterns are saved under format 20; older designs keep their independent copies.
PathScope::check reports a missing folder inside the scope as NotFound with the nearest existing folder, instead of calling it outside the allowed folders. Paths outside the scope and dangling links are still refused. Fixes#1077.
Edges shorter than the model tolerance are dropped from the flat cells, and the refold sews touching corners, so a cut-out across a bend line no longer fails with a degenerate edge.
doc.open and the STEP/mesh imports run on the file loader's worker with the request's path scope, and a font named by file path in a design is checked against the scope before it is read.
A design built over MCP is many small commands, and each one did work over the
whole design, so building or showing a 1,300-body design took minutes:
- moved components were transformed and meshed again after every change; world_state
now reuses each placed body whose definition and placement are unchanged
- next_d_name searched the design once per candidate number
- Document::eval evaluated every parameter for every expression; now only when the
expression names one
- param_rows' used_by was quadratic; the desktop no longer gathers completions on every
change while no text box has focus
solidcraft-cli bench-agent builds a generated ~1,300-body bicycle command by command
and times it (241 s -> 16 s here). Eval output is unchanged apart from timings.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Open and Save use parented AsyncFileDialog sheets; their answers run on the UI thread in a later frame, so resizing the window while a picker is open no longer re-enters winit and aborts. Editors, tab changes, Close/Quit and model-changing control requests wait for the answer; the local library Open and Export in the Appearance dialog use the same flow. Fixes#1088.
Ruled B-spline sheets that are circular cylinders or cones are written as CYLINDRICAL_SURFACE and CONICAL_SURFACE, so other readers no longer bound the part by sheets that reach past its faces.
Startup no longer stops at a bad settings file or a missing GPU: a settings file that fails to load is set aside, a failed GPU start retries with the software renderer, and a safe mode starts with defaults. Adds a troubleshooting page to the guide.
The engine's headless scene takes a palette; ui.render and the CPU viewport build it from the window's theme tokens, so a dark window renders a dark picture. CLI snapshots, MCP screenshots and Home thumbnails keep the default colours.
Moves the per-feature geometry handling (transforms, face points, profile selections, point expressions) into one geometry walk in feature_geometry.rs, with a golden record of what each feature kind moves. sheet.perforate holds no world geometry and joins the no-op arm.
Where an Import's stored labels pair one for one with its numbered built bodies (X, X (2), X (3)), component.paste_new stores the copy under the bodies' names, so the copied features after it find their bodies. No format change.
sketch.text and sketch.edit_text make the curve a text follows construction in the same undo step, so it no longer cuts the letters into pieces; the reply names the curve as guide.
A line end snapped onto a curve within the horizontal/vertical snap slides along the curve to where the axis crosses it; the preview, value boxes and click share one placement (#1121).
The feature menu of the bottom timeline and the side history gets Copy (the selection, or the right-clicked feature) and Paste (at the marker). Ctrl+click in the bottom timeline toggles a feature in the selection (#1034).
timeline.paste keeps a cylinder Emboss's persistent face name, remapped onto the pasted features and bodies, so the copy embosses the pasted cylinder instead of failing (#1035).
Inactive document tabs get a faint outline so neighbouring tabs stay apart, and the Home tab uses the same hover tint as the design tabs so its label stays readable in the light theme. Fixes#1115.
Adds sheet.perforate: round holes on a staggered or square grid over the flat sheet, kept clear of edges, cut-outs and bend zones. The holes go to the flat pattern, the DXF and the measured volume rather than the B-rep, so a panel can carry many thousands of them.
Adds local scalar appearance and material libraries: a catalog in the io crate with safe writes, engine commands to list and apply library entries, and a library dialog in the UI, with docs.
* Refuse fillets and chamfers that run past their faces or cross each other (#1080)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Blend fit: chamfers keep their old acceptance of crossed loops; skip the volume bound along tangent chains
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Blend fit: one selection per check, a coarse volume pass first; record the refused oversize blends in the action matrix baseline
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Thin Extrude builds walls of a set thickness along sketch curves, open chains and closed loops, with the wall centred, inside, outside or on both sides at two thicknesses. Saved under format 20 (thin on an Extrude); an extrude without thin saves as before. Closes#721.
Part 2 of #876: one walk over the references a feature holds (the sketches it's built from, the features it names, the bodies it works on) replaces three hand-written matches, with a golden record covering every variant.
Where SolidCraft can replace itself safely, the Update window gets Update now:
- AppImage: the new AppImage is downloaded next to the old one, verified and renamed over
it; Restart now starts it as SolidCraft exits.
- Windows MSI: the verified .msi starts (msiexec /i … /passive) once SolidCraft has closed,
after the usual questions about unsaved designs.
- Windows portable zip / standalone exe: the running solidcraft.exe is renamed to .old, the
new one takes its name (solidcraft-cli.exe too), Restart now; .old goes at the next start.
- macOS: the verified .dmg is opened (the app is ad-hoc signed and not notarized, so it isn't
replaced behind the user's back).
- Flatpak, Snap, deb, rpm, tar.gz: never replaced; the window shows the command.
Every package's SHA-256 must match the release's SHA256SUMS.txt (HTTPS, same release); a
mismatch deletes the download and changes nothing. With PUBLIC_KEY set in update/install.rs,
SHA256SUMS.txt.minisig must also verify (minisign-verify, MIT) with a trusted comment naming
the release. release.yml signs the list when the UPDATE_SIGNING_KEY secret exists and fails
when the app expects signatures but the secret is missing; docs/releasing.md has the steps.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Pieces that share a name are numbered by where their centres lie. Two pieces lying almost level along the sort axis no longer swap numbers between runs because of noise in their centres' last bits, so a saved name keeps pointing at the same piece. Closes#569.
* Straight edges stay one edge after cuts (#1049)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Robustness goldens: cuts join straight edges along untouched walls (#1049)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fmt
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Robustness goldens note: what each changed seed shows when replayed on both builds
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* GPU tessellation: a pipeline the device cannot build falls back to CPU meshing instead of crashing (#1098)
Build the tessellation compute pipeline inside error scopes; TessGpu::new returns an error,
and the viewport logs it and meshes bodies on the CPU. A Windows CI job builds the pipeline
with D3D12's FXC on the WARP adapter.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Tessellation shader: basis functions in one-dimensional local arrays, so D3D12's FXC compiles it (#1098)
FXC has to unroll loops that write an array inside a struct (Basis.n0/n1) or a 2-D array
(ndu) by a counter, and cannot with run-time trip counts (X3511). A2.2 basis functions in a
local 1-D array returned whole; derivatives from the degree p-1 functions; the pole step
offsets computed rather than indexed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
At start, and at most once a day, a background thread asks GitHub for the latest stable
release (one HTTPS GET to releases/latest, nothing else sent; 15 s timeout; silent when
offline; GitHub's rate limit respected). When it is newer than this build, an Update button
appears next to Report a bug. It opens a window with the version, the release notes and
Download (the asset that fits this installation), Skip this version and Remind me later.
- Versions compare as semver; a nightly counts as the release it was built after, so it is
offered only a newer stable release.
- The installation is detected (AppImage, Flatpak, Snap from the store or a file, deb, rpm,
tar.gz, Windows MSI/portable/exe, macOS app); Download picks the fixed-name asset.
Development builds and the Flatpak (no network permission) never check.
- Preferences > General > Updates: Check for updates automatically (on) and Check now.
- HTTP via ureq + rustls with the operating system's certificates (desktop app only).
- docs: Updates and privacy page.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A renderer-independent OpaqueSurface (colour, metallic fraction, roughness, dielectric IOR) with bounded inputs, an original GGX/Smith/Schlick reference response with Lambert diffuse, and exact piecewise sRGB8 decoding. Numerical contract only: no document format, command or UI change.
Loading a design raises next_id above the largest feature, component and occurrence id, so a missing or stale value can't hand a new item an id already in use, and duplicate ids are rejected as a damaged design. Id allocation uses checked_add, so an exhausted id space is an error instead of a wrap or a panic.
Part of #1016 (points 1, 2 and 4). nekomario28's #1061 reached the same fix independently; thanks to both.
Co-authored-by: nekomario28 <nekomario28@users.noreply.github.com>
Bodies imported with the same label come in as X, X (2), X (3). Renaming one of them no longer renumbers the others: body.rename now pins the current names of the feature's other bodies when it stores the new one, so later renames, materials, appearances and features still find them.
A tapered extrude now offsets spline and conic curves as well as lines and arcs, so sketch text and any other profile can take a taper, and the join of tapered letters onto a face finds its boolean result.
A chamfer of a hole corner, where the mitre extension runs into the wall, now takes the local blend and comes out as one bevel instead of a notched cut; where that blend can't build the edge, the cutters that used to succeed are kept (#814).
With a Chinese UI on Windows and macOS, the first CJK fallback was a Japanese font without the simplified characters, so one word mixed glyphs from two faces with different line heights and they sat on different baselines. For zh the Chinese face now joins the fallback chain before the Japanese one, and the chain is rebuilt on the frame the UI language changes; other languages keep the previous order.
sketch.text and sketch.edit_text took at, height, angle and text with a silent default when a value had the wrong kind, so a height of "H" made 5 mm text and at: ["H", 25] put it at the origin. An input that is given (not null) now has to be what it names, or the command is refused with the input's name, and a null at no longer passes as the origin.
The live preview of a join, cut or intersect now sends the tool body's scene as soon as the tool is built, then runs the feature for its error while the dialog keeps its spinner. OK pressed while the preview is still computing the same commands waits for it instead of computing the feature a second time on the UI thread; the session shares its evaluation cache with the preview's scratch copy.
macOS hands files opened from Finder, the Dock or open -a to the application delegate's application:openURLs:, which winit's delegate lacks, so AppKit showed "SolidCraft cannot open files in the SolidCraft design format". open_documents adds that method to the delegate's class before launch finishes and queues the paths for the UI, which opens them like files named on the command line.
This is the one place SolidCraft's own code uses unsafe: the app crate denies unsafe_code instead of forbidding it, and only the two macOS-only functions in open_documents.rs allow it, with the reason documented there and in AGENTS.md. Adapted from VectorCraft (see ATTRIBUTION.md).
Closes#997.
A planar union no longer leaves a sliver where an edge crosses its neighbour, so a sheet whose outline runs on past a flange's edge with a slight bend unfolds again.
The command search asked its text field for focus on every frame. On Wayland that restarted the text edit's held key, so the key kept repeating in other windows after the palette closed. Focus is now requested once, when the box opens, with a test that the field doesn't take focus back while the box stays open.
Fixes#724.
timeline.export_script (and solidcraft-cli script) writes a design's history as a command script that solidcraft-cli eval replays into the same design, so a bug that only shows on one design becomes a one-file repro. timeline.add restores one stored feature with its id, names and references, through the same checks a design file gets when it opens.
Closes#886. SychPL's #1078 approached the same issue independently; thanks to both.
Co-authored-by: SychPL <SychPL@users.noreply.github.com>
On Windows the temporary folder often has an 8.3 short name, which the path scope refuses. vfs::long_names spells folders SolidCraft derives from the environment with long names, and the MCP scratch folder and the recovery folder use it, so screenshot without a path and the recovery commands work there again.
Pasting an Import together with later features that name its bodies now points those features at the pasted copies, including bodies whose names carry a number such as X (2).
Since the model's mesh went back to the reference mesher (#940), the GPU viewport drew
opaque bodies from their display plans while hover and selection lit faces with the
reference mesh's triangles. Two triangulations of the same face at equal depth fought, so
a hovered face shimmered in wedges instead of lighting whole.
Faces lit over a body drawn from its plan now use the plan's own triangles
(ModelBody::plan_mesh, evaluated once on the CPU), and the highlight pass gets a small
depth bias toward the eye so f32 rounding between the GPU and CPU evaluations of the same
triangles cannot speckle it either.
Test: every face of a box with a pocket and a round boss, hovered and rendered offscreen
through the viewport's real pipelines on the GPU path, lights the same pixels as on the
CPU path (skipped without a GPU adapter).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The template still carried the 0.9 notes, and nightlies publish it as is. Keep the introduction, the upgrade notes that hold for every version and the install instructions; the release-specific part is written for each release.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* feat: validate physical material cards with sources and conditions
* feat: query directional elastic modulus from physical material cards
* fix: register physical material validation labels in complete locales
* fix: register directional elasticity labels in complete locales
* Add German and Korean labels for the material property commands
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* fix: a modelled thread that would cut nothing is refused (#995)
An M5 modelled thread in a 5 mm hole cut nothing into the nut: the
internal groove runs from the minor diameter out to the major one, so
with the hole as wide as the thread it lay entirely in the hole, and the
faceted cut only left the wall faceted with slivers ("the thread goes the
wrong way"). The fit check takes a 5 mm hole as an M5 hole.
- thread_groove measures how deep the groove cuts into the material
(hole wall out to the major radius, or shaft surface in to the root);
under a quarter of the basic depth 5H/8 the modelled thread is refused
with the measured depth, the tap drill (as solid.hole drills it) or
shaft size to use, and the size that fits the face, e.g. "this 5.00 mm
hole is too wide for a modelled M5 thread: it would cut 0.00 mm of the
thread's 0.43 mm depth; drill 4.2 mm for M5, or thread this hole M6".
- Cosmetic threads and their fit check are unchanged.
* test: pin the cut-nothing limit for coarse and fine pitch; show the hole size with show_mm
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
The dialog_docked scenario check and the ui_layout width test looked up
the dialog window by the id "sc_dialog", but dialogs are now registered
under ("sc_dialog", <title>). Every lookup failed with "no command
dialog", so the test had to stay ignored and could no longer catch an
oversized dialog.
Read the rect the dialog publishes each frame instead (viewport
dialog_rect in the check, "dialog:outer" in the test), and re-enable
every_dialog_opens_docked_within_its_width_limits. The exploded view
dialog it was ignored for now opens 365 px wide, inside its limits.
Claude-Session: https://claude.ai/code/session_01VbBYY5zfMXgq939AF9Vren
Co-authored-by: PhilippBischoff0 <PhilippBischoff0@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Convert to Sheet Metal without bend edges refused every plate whose bounding-box diagonal reached a thousand times its thickness ("no outline on that face"), and just under that limit a hole smaller than the mesh's sag fell out of the outline and the DXF. The outline's tolerance is now at most a tenth of the thickness: a thin plate is meshed again at that tolerance. Nothing new is stored. Tests: long thin strips, sheets thousands of times longer than thick, and a strip keeping holes down to 1 mm.
inputs_mut, inputs and expressions are now built on one exhaustive walk (walk_exprs / walk_exprs_mut) instead of two hand-written matches over FeatureKind. inputs() no longer clones the feature, plane chains are walked with the same nesting cap for inputs as for expressions, and a new variant without a walk arm does not compile. A golden record of every variant's expressions guards the walk.
docs/mcp.md gains a "Camera and close-up recipe" section: the orbit camera's conventions (radians, eye direction for yaw and pitch, orthographic half-height, which side inspect.section keeps) and a seven-step recipe for a close-up of one region of a large STEP model. Documentation only.
* Fillets and chamfers meeting at corners on curved faces, on top of #935
What #935 and #926 leave of #851 and the matrix:
- Corner build for blends meeting on faces of any surface (curvecorner.rs):
the knife-edge tip's round with the arc's or the back edge's, edges
picked in any order, a round ending in an earlier round of its radius
(as a feature of its own), and chamfers meeting at the arc's corner.
Its end curves are fitted by chord length: a uniform parameter
through points clustered at the ends doubled back on itself there,
and the display planner refused the face.
- A straight edge ending at a curved face (the back edge into the
knife-edge tip): rolled, or cut by the face in the planar corner
build (blend_corner.rs).
- A rolling ball turning about a rounded corner of its own radius
(a pivot: a piece of sphere or cone).
- Seams that are an earlier blend's straight contact (fitted curves)
count as straight.
- No spline across a joint where the strip's curvature jumps (a line
into a small round): each strip ends on the joint's section.
- Sections at a chain's vertices on the faces of the edge from it
alone.
- Every new fillet path passes #949's check for openings.
Tests measure volumes through the display plan, as #935's do.
Action matrix: 51 more fillets and chamfers pass (sweep, tangent arc
chain, sheet-metal hem, bracket holes), none fewer. Robustness seed 33
recorded (its last fillet builds differently; display and measured
volumes now agree).
Co-Authored-By: bherbruck <brennen.herbruck@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015bV2ntwLfUHV4JnCMPSyRi
* Corner blend tests: scale, seed 33's volumes, tighter engine bounds
From the review of #854:
- seed_33_display_plan_and_measure_agree: seed 33's bodies measure
the same through the display plan and the measure (2e-5), which the
golden's re-record rests on.
- the_arc_rounded_after_the_back_edge_matches_both_together: the order
that still leaves 0.05 mm³ at the tip, as an ignored test.
- corner_blends_hold_at_any_scale: the cases this branch adds on the
#851 part, at 0.01x and 100x, take off what they take at full size
times k³ (within 5e-6, checked to 1e-4).
- The engine tests on the part as drawn measure through the display
plan; bounds are about twice what each reads (2e-4 to 3e-3), and the
front corner's sets are checked against the kernel's CSG grid.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015bV2ntwLfUHV4JnCMPSyRi
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
press_pull_on_the_enclosure_lid_is_prompt took 61-87 s of wall time on loaded runners against a 60 s bound, for about 9 s of CPU. It now checks the operation's CPU time (under 300 s) and how far it grows the peak memory (under 1 GB; the runaway it guards against passed 24 GB), from /proc/self where the system has it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The sampled pi-case hole on Body1/F13:side:l7@F14.1 became a real hole with #952 (7.5 s instead of 170 ms); beside the heavy tests on main's push runs it passed the matrix's 90 s action limit in its edit round, reported as the design not loading. The sample now takes every CPU in nextest so nothing runs beside it, and the edit round writes a start line so a hang there is reported as <action>+edit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Adds the read-only material.properties.validate command: schema_version 1 engineering material cards in SI units are checked for numerical consistency and elastic stability, with derived isotropic moduli, missing conditions and sources needing review. Nothing is assigned or saved.
Thanks @0xlillith.
Fixes#971. At the node limit the offset certificate splits the fitted curve where it meets the exact offset midpoint, so curves with different parameter speeds certify within the same 0.002 mm tolerance. Associated offset fit nodes are hidden from display, picking, box selection and snapping until Break Association.
Thanks @Karl-VonRichter.
#970's lenient retry of a face's parameter boundary searches the whole surface as a last resort; on robustness seed 20 it found an edge's points across a cylinder's seam, half a turn apart, and the Cut failed. A lenient boundary whose points jump half the parameter range or more is refused, and the face is left undivided as before (#970's elbows step at most a fifth of the range). Seed 81's volume moved by 0.1 mm3 with #952's hole direction; recorded.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Fixes#928. Extrude To / From Object and Revolve To Object now resolve their target in the feature's own component frame, so a face in another component whose occurrence is moved is found where it is shown instead of the extrude stopping at a flat plane through the picked point. The point is named after the nearest face, a named point keeps its named face when another component touches it there, and such features re-evaluate when a placement changes, so they follow the target when it moves later.
Co-authored-by: SychPL <SychPL@users.noreply.github.com>
Fillet gets a Curvature (G2) radius type (type: "curvature", alias "g2"): a quintic blend whose size is the equivalent circular setback, for convex and concave straight edges between planar faces with a perpendicular planar end face at each end. Unsupported cases are refused rather than falling back to a circular fillet, and a blend that would cross a loop in its end face is refused. Persisted as a fillet style under format 19; labels in en/de/ja/ko/zh.
Co-authored-by: midasdf <midasdf@users.noreply.github.com>
Touching sews now split 4-use straight edges on extruded walls (along their rulings), not only on planes. Faces round such an edge are ordered by the way they leave it and, where they leave it tangent, by their curvature. Sheets bound the material wedges as before, except where that would make a face touch itself (a hole touching the outline from inside); edges ending there pair the empty wedges instead, so each loop gets its own vertex. Coincident faces are still refused, plain sew still refuses vertex contact, and shell_groups is unchanged.
A Ø10 hole tangent inside a stepped Ø40/Ø20 boss now cuts exactly instead of failing with "wire is not simple". Robustness seeds 42 and 51 are recorded: their touching hole patterns in stepped bodies now cut exactly instead of by a tool nudge.
Co-authored-by: froglemonade2 <froglemonade2@users.noreply.github.com>
The CPU renderer's line depth test allowed a fixed share of the scene's depth range (2e-4), which on large designs or close views let edges a millimetre or more behind thin sheet show through it. Each pixel now keeps the screen depth slope of the surface it shows, and a line passes where it is no further behind than that slope spans across the line's width, plus f32 rounding. Section caps use their plane's slope.
Edges lying on steep faces now draw their full width; edges behind a face stay hidden. Only crates/render/src/raster.rs changes; the GPU viewport is unchanged.
Co-authored-by: NOWN-DEV <NOWN-DEV@users.noreply.github.com>
* Boolean nudge tried both ways: no sliver where a cut's end touches a round side (#897, #827)
Robustness seed 14 cuts a cylinder whose end lies exactly on the apex of the
plate's arc side. Since the display tessellation plans (#805) the generator sees
the arc's true extent, so the cut's base lands on -10 instead of -10.000000000000002
and the end is tangent to the arc. The direct boolean fails there and the
last-resort tool nudge pushed the end into the arc, leaving a planar sliver a
tenth of a millimetre wide; shell then picked that sliver and refused it. The
13 October merges were not the cause (the bisect lands on #805).
The nudge is now tried both ways and the simpler result (fewer faces) is kept,
so the opposite nudge parts the touching faces instead. The same change joins
the hinge barrel in #827, whose top lies tangent to the lid's underside where
the mounting leaf ends. Each result still passes the same checks as before.
Goldens: seeds 14, 26 and 85 now build to the end, 31 and 78 come out with
fewer faces. Seeds 6, 43 and 73 still stop at shell: their joins now succeed and
shell reaches corners it can't offset (a cusp where a box face touches an arc,
coplanar faces facing opposite ways), which is unchanged.
Fixes#897Fixes#827
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Nudge the other way only when the first nudge fails or leaves a sliver
Trying both nudges and keeping the one with fewer faces could prefer a result
that keeps two pieces joined by a hair over one that separates them (the
centred coplanar revolve cut). The opposite nudge now runs only when the first
one failed or left a face narrower than 1/200 of the part, and it is taken
when it works with fewer of those.
Goldens re-recorded on current main: seeds 14, 26 and 85 build to the end,
31 and 78 with fewer faces. The other changed seeds are main's own since its
last full record, identical with and without this change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Nudge the other way only for slivers the nudge made
On the sensor enclosure's lid the first nudge left thin faces the operands
already had (thin walls, small rounds), so the opposite nudge ran a second
boolean for nothing and the press/pull took 79 s against main's 55 s. The
opposite nudge now runs only when the first one failed or left more slivers
than the operands brought; the press/pull test takes 36 s against 42 s on main
(same machine, one after the other).
Goldens re-recorded on top of #940.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Goldens on current main: seeds 31, 43 and 78 with the two-way nudge
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Keep the hinge and tangent-cut tests beside the #871 fillet test
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Goldens: seeds 33 and 91 as main's fillet changes build them
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Goldens: keep main's #950 record
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The state probe opened every cache it listed; when a worker deleted one in between, the open made it again, empty, and the next wait timed out. The probe now reads each cache with caches.match over the builds' files and marks, which never makes a cache (3 failures in 20 runs before, 0 in 30 after). The worker only ever starts a complete build.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Trimming the base at an edge flange's setback snaps corners within tolerance onto the trim line and merges away the zero-width run a ledge as tall as the setback left along it, so the ledge goes with the strip as a lower one does. Splitting outline edges at other cells' corners now skips points within the model's tolerance of an end, so a ledge a little taller keeps its sliver and the bend runs under it, also off the axes. Both used to fail with "sew: a degenerate edge". Nothing new is stored.
A hole drilled along a panel, into its rim, has no place in the flat pattern
as a cut-out, so every such hole was refused. It is now kept as a bore of that
panel, in flat coordinates, and cut only from that panel's material when the
folded part and its flat pattern are built, so the sheet still unfolds and
folds back with it. A hole that would run into a bend, or cross a panel at an
angle, is still refused. The flat pattern lists the bores.
The action matrix worker takes ACTION_MATRIX_ONLY (just these actions) and
ACTION_MATRIX_UNFOLD (unfold and refold every sheet after each action) for
rechecking one group by hand.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Holes on round faces from the app, one hole per pick, and Flip for holes at sketch points (#967)
The Hole dialog takes any face: a hole on a round face drills against the face's normal where it
was picked, and its arrow shows that. A round face split at its seam is picked whole, but makes
one hole where it was clicked. Holes at sketch points get Flip (saved as flip in the hole's points,
under format 18), in the command and in the dialog.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Holes at sketch points show their arrow: into the part, or away with Flip (#967)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Sketch Mirror, rectangular and circular Pattern, and Offset now show a green live preview that follows the selection and settings; OK commits, Cancel leaves the sketch unchanged. The shared dialogs show the current selection step and source counts, sources can be toggled with Ctrl/Cmd-click, cleared and reselected.
Offsets keep editable driving distance dimensions (One Side, Two Sides, Symmetric) that update the derived curves without moving the source, with undo/redo and save/reopen. Format 19: older files keep their fixed offsets.
New dialog labels in en, zh, ja, de and ko.
Fixes#696. Fixes#517.
Cutting one lofted round tube from another failed in truck's divide_face ("parameter boundary") for some section counts: the intersection points lie on the loft's seam, the previous point's parameter can sit a hair outside the range, and the Newton search from there fails.
The boundary is built as before; only when that returns None is it built again leniently, starting each search from the hint clamped into the parameter range and, as a last resort, searching the whole surface. Nothing that succeeds today changes. The boolean-retry guards are untouched.
Fixes 5 of the 6 failing hollow-elbow cases in #968 (new test crates/engine/src/tests/loft_cut.rs, volumes within 0.3% of exact). 90 degrees / 14 sections still fails in a later step and stays open in #968.
The plate check compared the face outline's area (from the tessellation, so each hole was a polygon of chords) with the exact volume, and refused plates with a few small holes. Where the outline runs along a cylinder about the plate's normal it is now an arc of that circle; coaxial pieces of one radius are one wall, so a hole split into two, three or four faces is one circle. The check passes with its tolerance unchanged, and the flat pattern and its DXF carry one CIRCLE per hole (ARC for rounded corners and slot ends).
A round blank now converts too, so unfold_cut_refold_and_convert checks the refusal with a block thicker at one end instead of a rod.
* Hole matrix: placement, types, extents, edits and 0.9.0 designs, with today's failures listed
Each case checks the drill goes into the material, the bore sits on its axis, a blind hole stops
where it should, the removed volume against its analytic value, the face count and that every
body is still one closed valid solid. Failing cases are listed in tests/hole_matrix/known_failures.txt
as a ratchet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* A hole on a face drills into the material, also in older designs (#950)
Without a direction, solid.hole took the direction from any face whose infinite plane passed
within ten tolerances of the point, the last one found winning: the far side of a thin plate or a
face of another body at the same height could set it, and the wrong direction was saved. It now
uses the face nearest the point, against its outward normal (exact for planes, cylinders and
spheres, so holes on curved faces work without a direction). When a design is evaluated, a face
hole whose saved direction leaves the face it starts on is mirrored into it, so designs saved
with the wrong direction drill into the part.
Fixes#950.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hole at a point on no face (a planar face's centroid outside it) keeps the plane it lies in
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Record the hole direction fix in the goldens and the action matrix baseline (#950)
Seeds 7, 21, 81 and 85 drill their holes straight into their faces now. In the action matrix,
holes on four faces that used to go the wrong way now build, and two whose correct direction
meets a cut the kernel refuses are marked failing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Goldens: seeds 7, 21 and 85 drill their holes straight in (#950)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Fillet: refuse a result with openings in its surface (#871)
A fillet on #871's part (radius 1 on a window's edges and on the sides of a
block 1.26 mm away) came back closed as topology with every face meshing, but
the rounds overlapped and the end faces' boundary loops crossed, so the model
showed holes. The constant-radius fillet now checks its result before taking it:
the body's mesh, welded within a millionth of its size, must have no edge
without as many triangles running the other way along it, except for cracks
within the mesh tolerance where fitted curves meet. A result that fails is
refused, so the corner build gets its try and the radius search treats it as
too large.
Body::open_mesh_edges is the check. Kernel tests cover the #871 part (refused
at radius 1, closed with a round on every edge at 0.5), and an engine test runs
the fillet from the report and leaves the part as it was. All 96 robustness
seeds build the same with and without it.
Fixes#871
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Check fillets for crossed face boundaries, not free mesh edges
Free edges in the mesh also come from cracks where fitted curves meet (corner
patches, rounds ending on curved edges), so the mesh test refused good fillets.
The check is now on the cause of #871: a planar face whose boundary loops cross
each other or themselves has no inside. Body::crossed_face_loops samples each
planar face's loops within 1/1000 of the size and counts proper crossings
(touching within a millionth doesn't count).
The pyramid apex blend and the loft fillets build again. Ten action-matrix fillets
stay refused, recorded in the baseline: r = 3 on the 2 mm rim of oracle 09's
shelled box (the round runs past the inner wall; the result was 240 mm3 lighter
than any round can make it), and the six seam-edge fillets on oracle 60's bolt
holes, whose boundary jumps across the hole.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
PathScope stores its roots canonicalised (\\?\ on Windows, /private/var on macOS), so the default-folders test now canonicalises scratch_dir() before looking it up, like the recovery folder on the line above.
The same fix came in from @Karl-VonRichter in #969, found on macOS.
Co-authored-by: Karl-VonRichter <20079846+Karl-VonRichter@users.noreply.github.com>
- A fold's zone clipped across the root of a tab folded before on the same
line kept a sliver of no width there, which made the zone look like it
crossed a slanted edge. The one real piece is the zone now (#638).
- When the tabs a fold line spans can't take the side beyond them along,
because it joins the side that stays through other material (a tab on the
inside edge of an L, the fixed point on the far leg), the tabs' own ends fold
instead, the same part as with the fixed point on the near leg (#639).
Co-authored-by: jonaswre <3731681+jonaswre@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Name a side drawn on several sketch curves by all of them
An extrude named each side wall after the one sketch curve its profile
segment lies on. A wall along two coincident lines (squares drawn edge to
edge) or along collinear lines merged into one profile segment had no
single curve, so it was named from geometry and could not carry a sketch
attachment. One sketch extruded up and down and joined (the logo) refused
a sketch on every one of its side walls.
Such a side is now named by every curve it lies on, in sketch order: the
first is its name, the others aliases. The set is fixed by the sketch, so
the name is exact. The refusal message no longer blames merged faces.
Adds the reported logo design and a synthetic up-and-down join to the
action-matrix designs, and a regression test that sketches and extrudes on
every flat face, fresh and after save and reopen.
Fixes#954
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Action matrix: record the up-and-down join designs
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
- A new cut was clipped against the panel with every one of its cut-outs, so
each hole of a perforation cost the square of all the holes before it and the
second cut of 500 ran out of the flat budget. A tool is now clipped against
only the cut-outs whose boxes meet its own (the same result).
- The planar booleans' point-in-loop test now skips loops whose box is clear of
the point, before the costly on-the-loop test; with a thousand holes that test
was most of the time.
The 1000-hole panel now takes about 15 s instead of failing.
Co-authored-by: jonaswre <3731681+jonaswre@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Fillets and chamfers ending on a boss's pieced wall or beside it, on a face that runs along its seam twice, and loops the loop blend refuses
- A straight edge whose end meets one curved surface in several faces (a cylinder a join left
in pieces) ends on that surface: the section is cut where it crosses each seam, which is
shortened there; where no seam is crossed (the wall missing inside the other body) the
section lies on the bared surface and the rest of the side edge becomes a seam.
- Two end faces that aren't both planes meeting in lines: each face takes the section's points
on it, cut where it crosses the edge between them. Not where an end face runs on smoothly
from a side (an earlier round): the corner build takes those.
- Face boundaries close over several connector edges.
- A face on a periodic surface running along its seam twice (as a STEP file gives it) is
rebuilt the same way (#691).
- A curved chain ending square on a face that runs on past its side extends that face's seam.
- A smooth chain stops before an edge between faces that meet smoothly.
- A loop the loop blend refuses for its shape is blended edge by edge; refusals about its size
stand.
- Action matrix: ACTION_MATRIX_ACTIONS and ACTION_MATRIX_VALIDITY for quick local runs; the
fillets and chamfers that pass now recorded.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
On a large thin sheet the probes around the fixed point stepped clean through
the sheet (twice the mesh tolerance of a 465 mm part is more than its 0.8 mm
thickness), so a point exactly on the face touched no material and the fold
was refused. The probes now also step a quarter of the thickness.
Co-authored-by: jonaswre <3731681+jonaswre@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#921's grid lines through the vertices that split a ruling edge also went into the meshes Boolean operations intersect, which changed seven robustness seeds (seed 2's sphere cut failed). Deflection::rulings now asks for them only in the kernel's own tessellation (display, export, measurement). The seeds #936 (rejected instead of unsupported), #946 (26 and 85 build, others fail differently) and #921's display side (about 1e-6) moved are recorded with their reasons.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
#820 fixed examples/bracket.json's fillet pick, which lay far from every edge and rounded the outer corner instead of the inner one. This re-records examples/bracket only: its edges carry the fillet's new names, and three chamfers beside and across the inner fillet are refused (the blend is larger than the neighbouring faces).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Rounded Rectangle sketch tool (#891)
A new SKETCH > CREATE command, sketch.rectangle.rounded, in the Rectangle
family: two opposite corners and a corner radius make a 2-point rectangle
whose four corners are filleted alike, as one undo step. The radius is one
parameter (one radius dimension, the other arcs equal to it), so changing it
changes all four corners. While drawing, Width, Height and Radius boxes take
typed values; typed width and height dimension the overall size between
opposite sides. Without a typed radius the corners get a fifth of the shorter
side.
Fixes#891
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Rounded Rectangle in the Simplified Chinese catalog
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Each section now starts at the rotation whose vertices line up best with the lead section's, vertex for vertex, instead of at the one vertex closest in direction to the lead's first. Fixes the twisted loft of matching non-convex sections; regression test loft_pairs_the_vertices_of_matching_non_convex_sections.
Session::replace_design keeps auto_project when file.new, doc.open and doc.recover swap in another design, and the UI keeps it across tabs, adopt and the last-tab close. Regression tests in engine (sketch_project) and ui-egui (documents).
Rolling-ball blends now handle knife edges, where the two faces have opposite normals; an edge that runs out along an end face tangent to one side; and rounds that end in an earlier round on curved faces, picked together or one after another. The three edges at a vertex of smooth faces, rounded to one radius, end in a piece of sphere. When the given order of up to four edges is refused, or leaves a face boundary off its surface, other orders are tried.
Closes#851.
Co-authored-by: lasselian <22126420+lasselian@users.noreply.github.com>
Adds a complete Simplified Chinese (zh) catalog covering every English message, a font fallback for Chinese text, and `cargo xtask i18n --check`, which checks that the complete catalogs (ja, zh) carry every English message and that every command label is translatable.
Co-authored-by: Noly1990 <20500931+Noly1990@users.noreply.github.com>
Packs the scene's planned bodies into one GPU upload and dispatch, and evaluates intersection edges by seeded Newton refinement from a fitted B-spline through exactly solved leader points, falling back to the previous evaluation near tangency. Intersection nodes carry their faces' parameters. This changes speed, not what the app draws.
Part 4 of 4 of #303.
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Thicken now moves faces through the same mover as Offset Surface. Spheres, cones, tori and B-spline faces thicken as well as planes and cylinders, and a thickness that would fold a curved face over is refused. Where thickened faces meet at a sharp edge, a wall closes the gap and corners are closed with flat patches. Thickening into an inside corner is still refused with a clear message.
Closes#765.
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
Offsets picked faces along their normals into a new surface body: planes and cylinders move exactly, other faces get a fitted B-spline within tolerance, and offsets that would fold over are refused. Adds the SurfaceOffset feature (noted under format 18), the surface.offset engine command and an Offset Surface dialog.
Closes#758.
Co-authored-by: jyviko <2501648+jyviko@users.noreply.github.com>
A section stayed on with nothing in the browser to switch it off; only the
viewport's context menu had Remove Section. While there is one, the browser
now has an Analysis folder listing it. Its eye (and the folder's) hides the
section and shows it again on the same plane; a double-click or Edit Section
Analysis opens the dialog, and Delete removes it. Browser rows with an eye
publish an `eye:<label>` handle for automation.
Fixes#822
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Closing a socket with unread input resets the connection, and the reset could reach the client before the refusal did (requests_run_only_after_the_token failed 17 times in 100). The server now ends its side and drains, briefly and up to 64 KiB, what the client had sent before closing: 0 in 100.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A through-all hole's tool is twice the largest body long, and the boolean's tolerances
followed that length: a 6 mm hole through a 920 mm part was traced at nearly 1 mm, wider than
the hole, and the cut came out empty or failed. When every other strategy has failed, the
boolean is now asked again (unshifted, then once shifted) at tolerances scaled to where the
bodies meet: the overlap of their boxes and the thinner operand's thickness. Its result is
checked as the others are and held to the retry guard. It runs last, so whatever an earlier
strategy answered is kept as it was, and only for single-shell operands (bodies in several
pieces are retried a piece at a time, each with this pass of its own).
The sampled shortcut also gave up on a slender tool lying across its box (a long hole drilled at
an angle fills little of it): it now samples past the first 2000 points until enough land
inside, so a tool that misses a body leaves that body as it is. Bodies that had enough samples
in the first 2000 are sampled exactly as before.
Baseline: the 90 actions that newly pass, rechecked one by one with the final build.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A design with components and joints exports as a URDF robot description (file.export to .urdf, the MCP export tool, solidcraft-cli run --out, File > Export URDF...). It writes one binary STL per component in metres, links with mass, centre of mass and inertia from material density, and maps rigid, revolute, slider, cylindrical, pin-slot, planar and ball joints to URDF joints or short chains. Loops, broken joints and loose parts are reported as warnings. Options: mesh_prefix, effort, velocity. Docs in docs/urdf.md.
Each wall of a surface built from sketch curves (Surface Extrude, Revolve, Sweep and Loft) is now named after the curve it comes from, exactly rather than from geometry, so references to those faces survive edits.
The browser build fetched its Japanese fallback font only for a Japanese UI,
so a Japanese design name, file name or typed text showed as tofu with the UI
in English. After drawing each frame the web shell now looks through the
painted text for kana, CJK ideographs and full-width forms while no CJK font
is loaded, and fetches the same font once on the first hit. A failed fetch
with an English UI leaves the language alone and is not retried in a loop.
Sessions that never show such text download nothing extra.
Fixes#720
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Toolbar drop-downs show all their rows after a shorter list (#874)
A panel's drop-down kept its size by an id made from the panel name alone,
so PLASTIC's MODIFY reused the one-row size of MESH's MODIFY and showed two
rows with a scroll bar. The id now includes the tab, and the rows' scroll area
may grow past the size the area remembered (up to the room below it), which
also covers the same panel collapsed and expanded.
Fixes#874
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Keep the Sketch Create menu's own sizing
Its rows sit in egui menus (also inside the overflow menu), where a minimum
height moved the rows between frames. Only the panel drop-downs need it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Moving a window between monitors of different scale can briefly ask for a
surface larger than the device's largest texture (11512 x 6808 against 8192 in
the report), and Surface::configure failed validation, which wgpu turns into a
panic. The surface is now clamped to the device limit with its aspect ratio
kept, and egui draws the whole window at the matching smaller scale, so the
frame is only softer until the size settles. The device also asks for the
adapter's own texture limit (16384 on most desktop GPUs) instead of 8192, and
the resize path and an unsupported surface log instead of unwrapping.
Fixes#869
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Surface Loft lofts open or closed curve chains, one section per sketch in pick order, into surface bodies. Adds the SurfaceLoft feature (format 18 note), the surface.loft command, a dialog with Edit Feature support, and kernel and engine tests.
Part of #757.
* Fillet and chamfer: refuse a smooth seam as having no corner to round
Picking the seams where R10 rounds run into the faces beside them, and asking for an R1
fillet there, failed with "not supported yet: blending a smooth edge (... both faces next
to the edge must be planar)". That reads like a missing feature, but the faces meet
tangentially at those edges and there is nothing to round.
An edge whose two faces are tangent along it is now refused as an invalid pick that names
the edge and asks for the sharp edges.
Fixes#896.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Robustness goldens: seed 72's smooth-seam fillet is now a rejected pick, not unsupported
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
#825 chose the faces to split by boxing each face, dividing its curved edges down to the contact tolerance; on the sensor enclosure's lid that division never finished (press/pull and extrude on 18 faces grew past 24 GB and timed out in the action matrix, on CI as here). The boxes now come from the edges' control points, with no division; a face whose extent can't be found counts as near. A test holds the lid's press/pull to a time budget.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Surface Sweep carries open curves and closed profile boundaries along a sketch path into surface bodies, with smooth walls along curved point paths. Adds the SurfaceSweep feature (format 18 note), the surface.sweep command, a dialog with Edit Feature support, and kernel and engine tests.
Part of #757.
Walls swept along a spline have flat B-spline tops and bottoms, so booleans between them went to the general intersection, which fails on coincident spline faces. Such faces are now put on true planes, and two prisms of the same height and direction are decided in 2D with their spline outlines fitted by arcs within a micron. The result is kept only when its volume is plausible; otherwise the general boolean takes over. The free-form flag is cleared by a drop guard so an unwinding boolean can't leave it set.
Fixes#844.
The exact surface search could return an arbitrary parameter at a collapsed NURBS boundary, which corrupted the trimmed triangulation domain and lost volume on fine meshes of a fillet tapering to zero. Search results in the boundary projection are now validated, preferring a bounded parameter at a pole and snapping to a finite endpoint only when its transverse tangent has collapsed and the residual improves. Adds four regressions for decreasing chord tolerances, pole parameters, unbounded planar support and provisional preview.
Fixes#796.
Body::with_planar_sources compared every planar face of a result with every unsourced planar face of its inputs, so a boolean on an import with a few thousand facets ran out of its comparison budget before it started. The lists are now sorted by normal x and each face only compares with entries inside a narrow normal window found by binary search; the exact same_support test and the work budget are unchanged. Adds an engine test joining a pin onto a 4000-sided imported prism.
Surface Revolve turns open curves and closed profile boundaries about a sketch axis into surface bodies, with cone apexes and sphere poles where a chain ends on the axis, and a closed shell made solid. Adds the SurfaceRevolve feature (format 18 note), the surface.revolve command, a dialog with Edit Feature support and analytic kernel and engine tests.
Part of #757.
A lip picked on a face that isn't a shelled body's rim, such as the Sample Plate's top
face with its two holes and the boss standing on it, ran its band round the largest inner
edge. That was the boss's foot, so the band stood against the boss's wall and the join
failed with a boolean error about curved faces.
The band now goes round the face's opening (an inner edge with no material just over the
face inside it). A rim face that has a smaller hole in it still works. A face whose inner
edges are only where other features stand on it, or whose openings are alike in size (a
plate's holes), is refused with a plain message that names the rim face it needs.
Fixes#892.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Generates N parts of four primitive bodies, each moved into a placed component. The test (12 parts) checks the open, the world placement and the saved round trip; the ignored bench_generated_assembly prints parse, evaluation inside and outside the features, mesh warm-up and Measure.
A boolean that cuts a pocket into the top of a cylinder (Shell does this) splits the
outer wall's straight seam edges at the pocket's floor. The wall's parameter grid has no
division along its rulings, so the triangulation fanned out from the two split vertices
to the far rims and its triangles cut through the part: the shelled Bolted Flange sample
drew and exported its rim cut off (mesh volume 29201 mm3 for a 40796 mm3 body).
Give the grid a line through each vertex that splits a boundary segment running along a
ruling, in a direction that needs no division of its own and within the grid budget.
Fixes#819.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The app's control protocol gets engine.batch: the steps run in one request (so one frame), each as engine.execute runs it, stopping at the first failure and rolling back to the checkpoint by default. The MCP batch tool sends it to a running app (with the control token), is never resent after a lost reply, and falls back to step-by-step when the app refuses (agent cursor shown, or a step loads a file). A batch may take up to an hour to answer.
find_close_pair's budget now counts the grid's own work too (each cell a triangle is filed under or visits), not only the pair tests, so a pathological pair of meshes can't spend unbounded time building and walking the grid. Adds tests for find_close_pair and the budget.
face_vertices lists each vertex once per face (first-met order) and normal_near measures each distance once; seam flags, merged topology and cone signatures are unchanged because the dropped repeats were identical entries.
Hole centres from a sketch now drill into the material on whichever side of the sketch plane it lies: material the spot starts in wins, then the nearest body ahead, else against the sketch normal as before. A sketch on a part's underside or below it now drills up into the part. Covers counterbores, through-all, edge and corner centres, patterns and reopening.
Delete Face takes heal (default true, as before). With heal: false the picked faces are removed and the gap stays open, leaving the remaining faces as a surface body (kernel remove_faces); deleting every face is refused. Saved only when false, noted under format 18.
A body transfer works out the moved bodies first and puts them in place in the evaluator's own copy of the model, instead of copying every body of the model for each transfer.
Display meshes after opening, the face names read when a feature finds a face by name, and inspect.measure over all bodies are now worked out per body on plain scoped threads (solidcraft_kernel::for_each_on_threads / map_on_threads), in body order and with the same results; the browser build stays serial.
After each feature the bodies before it are indexed by name once (first match first, as the linear searches were), feature components come from a map built once per evaluation, and Body::same compares shared face sources, paint and split lines by pointer first.
Body::retained_bytes counts face and edge uses on the borrowed boundaries (the same numbers, checked by a test), and each model body keeps the footprint measured when a feature made or changed it, so the evaluation cache's budget no longer walks every body's topology after every feature.
NameCell::same now refers to its source body and takes the source's names on first request, instead of naming every face of every moved or placed body up front. Names are unchanged.
Where two boundaries of a projected face family touch at a single vertex, an even number of boundary edges meet there. The walk now stays on the face it came along (or takes the first edge in geometric order), so every loop closes; an odd count is still refused as a real branch. Adds a test with two squares meeting at a corner.
A sketch text gets an optional style: bold and italic faces (a slanted fallback for italic), left/center/right alignment, letter and line spacing, and a path curve of the same sketch for the baseline to follow (reverse takes the other side). Edit Text gets B/I toggles, alignment, spacing fields and a Path chip, all live. The style is saved only when not plain and is noted under format 18.
solidcraft --control PORT writes a random per-launch token to a user-only file (<data>/control/PORT.token; 0700/0600 on Unix, icacls on Windows; SOLIDCRAFT_CONTROL_TOKEN overrides) and refuses any connection whose first request lacks it. mcp --connect, the demo and packaging smoke clients send it. MCP execute/batch/open/save/export/screenshot are annotated destructive and open-world, new_design destructive. Docs: control-protocol.md, mcp.md, AGENTS.md, pending release note in releasing.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Only one run per concurrency group can wait. On a busy main each push replaced the waiting run, so the nightly and Run workflow runs on main were cancelled before they started. The group now includes the event.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Renaming a body used to replace every string in later features equal to its old name, including sketch text, plane and feature names and expressions. Now only values under keys that name bodies (body, bodies, target, targets, tools) at any depth are rewritten, with a test that a sketch text spelling the body's name stays while a move follows the rename.
Corrects the modeling skill: an edge of the face a shell removes is rounded after the shell, a slot along Y is dimensioned horizontally, long shallow arcs measure to about 0.1%, and a boolean retry warning can hide wrong geometry when the volume does not match.
A hole drilled square to a flange's panel, not only the base's, is now a flat cut-out of the sheet: it unfolds with its panel to where the bend allowance puts it, shows in the flat pattern and DXF as a circle, and survives unfold and refold. A pocket tool spans the cut-outs already inside it (a counterbore's hole), so the order they are made in no longer matters. Fixes#560.
Main has failed the full robustness goldens since #838, #839, #841, #855 and #856 landed together. Records the eleven seeds that now build differently (2, 6, 14, 22, 36, 43, 50, 63, 71, 73, 94), with a note. Shell now refusing seed 14, which used to build, is tracked in #897.
When an extrude cuts with profiles whose outlines stay more than 1 µm apart (the letters of a text), their tools are apart too, so they are joined and cut from the body in one boolean instead of one per profile. A ten-letter text cut drops from about 13 s to 1.5 s with the same result. Sheet metal targets and touching profiles keep cutting tool by tool. Tapered extrudes are unaffected.
Agent tool calls (MCP, and bridged control requests carrying allow_paths) reach files only in the open design's folder (strictly inside home or the temporary folder, not in a hidden folder, and never its hidden files), the app's recovery folder, the server's private scratch folder and each --allow-path; --allow-any-path lifts the limit. Paths are resolved first, so .. and symlinks that lead out (including a save through a link, #845) are refused. On Windows, UNC, device, NT, verbatim non-drive, drive-relative, alternate-stream, reserved-device and 8.3 short-name paths are refused before any file system call, and the resolved path must be on a local drive. doc.recovery_discard and doc.recover {discard} delete only real recovery entries. Docs: mcp.md, control-protocol.md, README.
tests/robustness/goldens.json records what each of seeds 0-95 builds: its steps, how it ended (ok, or the kind and the start of the error it was refused with) and, for the design it left, the bodies' total volume and area (within a relative 1e-6) and their face, edge and shell counts. tests::robustness_goldens fails on any change; ROBUSTNESS_GOLDENS_BLESS="why" records one, with the reason and the seeds it moved. Pull requests that touch crates/kernel run every 8th seed; pushes to main and the nightly run every seed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Bumps the workspace to 0.9.0 and writes the release notes. Minor bump: designs are now format 17, and there are many new features.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A Sketch Palette toggle (sketch.options show_construction) that hides construction geometry from drawing and picking; saved with the sketch's view options under format 17.
Co-authored-by: William Dunn <william.dunn@quasarke.com>
A coincident face is pushed apart only when the space it moves through holds nothing that changes the result; otherwise the boolean falls through to the exact strategies.
Co-authored-by: NOWN-DEV <257771410+NOWN-DEV@users.noreply.github.com>
The partial-contact imprint (#704, #751) split every face in the contact plane against every tool face, one call per face, and measured all the tool's cylinders for each pair: the sensor enclosure's Case shell went from 6.6 s to 76 s and timed out in the CI matrix. Only faces near the contact are split now, each surface once, in one call: 7.0 s.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Loading a design (from a worker's start or restart to its first output) gets 300 s; each action keeps 90 s. oracle/41-tube-radial-holes (about 23 s here) timed out loading on a loaded CI runner, and reloads after a hung action were charged to the next action.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
sketch.pattern.rectangular gets a tool window with both quantities and spacings and an optional direction, instead of a click flow with a fixed count.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
surface.extrude sweeps open curve chains and profile boundaries along the sketch normal into a surface body (walls only, no caps). Saved under format 17.
Refs #757
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Per-face target side lengths, the caller's facet angle, refined grid lines, and fixes for underflowing coordinates and straight edges ending exactly on their vertex. Nothing drawn changes yet.
Refs #303
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A mesh over the conversion budget is refused with its triangle (or vertex) count, the limit, and that only constant-section parts convert.
Refs #346
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
STL (binary and ASCII) now welds each body's export mesh with the same tolerance as 3MF, so filleted and drilled bodies are closed. The weld grid's neighbour lookup no longer overflows for bodies far from the origin.
Fixes#784
Follow-up to #718: a face place snaps only to sketch points that are shown, and the face under the cursor is looked up once per pick.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Adds oracle case 81 (a box edge rounded at a clamped cubic B-spline radius) with analytic expected values, and passes spline fillets through recipes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Icons for the surface toolset, and its planned commands registered and disabled with the issue that tracks each one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Sketch Move gets a toolbar placement in Sketch > Modify and a Move/Copy tool window with X and Y distances, an angle, an optional centre point and a Create Copy flag, replacing the context-menu route that ran the command with no parameters.
A command dialog opened while sketching no longer hides the Sketch Palette: the dialog docks at the left of the viewport and the palette keeps the right. A view too narrow for both keeps the dialog alone.
Right-clicking a curve or point of a sketch that isn't being edited now opens that sketch's own menu, with Edit Sketch, instead of the body menu. Solid geometry in the selection keeps the body menu.
Sketch Mirror now opens a tool window with Entities and Mirror line inputs, so both picks stay visible and editable until OK, replacing the old click-then-right-click flow.
Completes the Japanese catalog (all commands, panels, dialogs and preferences) and routes the remaining hard-coded UI labels through the translation lookup. English stays the default.
The feature history can sit in a panel on the left, with folders that hold one unbroken run of features, drag and drop as one undo step, and a resizable divider. Timeline folders save under format 17.
26 baseline actions that fail on main: 17 are the retry guard's refusals of geometry that used to delete a body or fill past an edge (#644), 9 vary between runs (#707). Load timeouts left as they were.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A blend no longer re-rolls an edge it already refused unless a later blend came near it, so long curved chains stop paying for the same refusal over and over.
truck's face and edge IDs are their addresses, so walking maps keyed by them followed heap layout, and the ribbed lid's side-wall hole passed or failed by chance. The planar boolean's rebuild and glue now take truck's connected components in the shell's order, and the planar boolean's own maps are ordered. Healing's union-find keeps the lesser root and takes a run's ends in order of use. The face classification after an intersection orders each component's faces by the shell first. The lid now builds the same body on every run, with a test that builds it six times in one process. Its 0.5 mm side-wall holes, and one sensor enclosure hole of the same kind, now fail every time and are marked failing in the baseline.
Fixes#707.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A sketch's link to a projected edge now keeps the edge's persistent name and finds it by that name when the body changes, instead of taking whichever edge ends up nearest its point.
The Sketch Palette's line types are three drawn icons (solid, one gap, gap with a dot) instead of text glyphs, and the construction orange is now chosen by the caller (`fixed_accent` on the toolbar) rather than hardcoded in the icons.
Sketch Scale now opens a tool window with the entities, an optional base point and a typed factor, instead of running the command with no parameters and raising an engine error.
Adds `fit: taper` to Fillet: where a constant radius doesn't fit along a straight edge, it follows the room there as a smooth spline, down to zero where a face closes to a point. Taper to Fit is in the Fillet dialog, and the feature's warning says how far it tapered.
Outside a sketch, a sketch tool's shortcut opens Create Sketch on the selected plane or planar face (or asks for one), then starts that tool in the new sketch.
`SolveApp::start` now honours a command's enabled flag, so shortcuts, the command palette and the control channel can't start a disabled command, and Move is disabled while a sketch is open.
The MCP server instructions now point to `body_topology` for finding edge points (there is no `list_edges` tool), with a test that every tool the instructions name exists.
Robustness seed 16 failed in most runs: the imprint split a block's side along both halves of a cylinder's round side, and the second split re-traced the first circle, sometimes leaving a sliver the glue could not match. The imprint now splits along each surface once (10/10 runs pass, 3/12 before).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Adds a Feature Options section to the Sketch Palette with the running tool's modes as icon buttons, so you can switch between e.g. circle or rectangle variants mid-tool.
A cut or an intersection between prisms (a plate and the tools through it) is now decided in 2D before either operand is meshed. A body's size and volume are measured once, a placement reads its orientation from its matrix, the evaluation cache is bounded to 64 MiB, and STEP import copies fewer shells. Hole plates evaluate 20-30% faster with less memory, and models are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Convert to Sheet Metal now takes a solid or a shelled body by its bend edges (`bends`): the faces reached from the fixed face across those edges become the sheet, keeping the part's envelope, with the flat pattern by the rule's K-factor and radius. Other edges between sheet faces are cut open. The result works with edge flanges, cuts, Unfold/Refold, the flat pattern and DXF. A conversion without `bends` is the plate conversion as before. The optional `bends` field is noted under format 17. Fixes#647.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Extrude and Revolve from a body face take `face_body` to say whose face is meant where faces of several bodies meet flush at the picked point. Without it, an ambiguous pick now names the bodies that meet there instead of saying there is no face. Fixes#703.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The mouse wheel over a dialog's list, a drop-down or a menu that hangs over the model now scrolls only that list; the view zooms only when the pointer is really over the model. Fixes#678.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Box, Cylinder, Sphere, Torus and Coil dialogs now start their sizes in the design's units with the unit written out ("20 mm", "20 in"), like the other dialogs. OK makes the same sizes as before. Fixes#630.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A Single hole picked on or near a sketch point that lies on the face (a circle's centre) now goes to that point, as a Multiple pick does. Picks away from sketch points, or near one that isn't on the face under the cursor, stay where they were clicked. Fixes#657.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Dragging a text outline now moves the whole editable text, and Move/Copy shifts or rotates it as one piece (a copy stays independently editable). Content, font, height and outline identities are kept; a drag is one undo step and Esc restores the start.
With nothing selected, Extrude and Revolve no longer pre-pick the last sketch's profile when a feature already builds from it, so OK no longer cuts into a part nobody picked. Fixes#684.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Sheet-metal cut-outs and base flange outlines drawn with ellipses, splines or text stay exact curves in the flat pattern and the unfolded body; the folded body follows them with tangent arcs within 0.001 mm (#659).
The nearest point on a triangle with a zero-length edge (a cone's tip) is no longer NaN, so an axis or tangent plane on a drill tip works again (#567).
Adds render.backends, a metadata-only command that lists the render backends, their executors and profiles, with strict request resolution in the render crate (docs/render-backends.md).
A cut built slab by slab (the exact prism boolean) sewed the result from horizontal
slabs and only joins merged the coplanar pieces back. So a floor vent cut through the
Raspberry Pi case split the long wall into strips at the ports' tops and bottoms
(87 faces instead of 79 at 32 mm), and at 24 mm the end opening split the end wall.
- Prism booleans merge coplanar neighbours for every operation (faces only: edges,
curves and surfaces stay exact), keeping each face's construction source where all
its pieces share it, Split Face lines and regularization protection. Painted or
protected operands keep their authored pieces as before.
- The prism history now tells apart a wall's several edges with its caps (a wall
with openings meets each opening's floor), so cuts beside existing openings keep
exact history and faces keep their names.
- Tests: the Pi case's faces pinned after every feature at 24, at 24 changed to 32,
and built at 32 (each outer side one face); random openings and bosses on a
shelled case never split another side; every corpus design rebuilds a parameter
edit like a fresh build and reverts exactly.
- Action matrix baseline: demo/pi-case re-blessed (fewer, merged faces; no
regressions), demo/pi-case-0.8 recorded, and 88 new passes on
regressions/lid_ribs_flush_with_window.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A clover of same-size rectangles drawn around a shared corner was refused as over-constrained.
Constraints that remove no degree of freedom but still hold (zero residual) are now kept and shown dimmed as redundant. Conflicting ones are still refused. A redundant driving dimension is refused with "already determined by ...". Typed sizes become driven (reference) dimensions. Drawing no longer adds Horizontal/Vertical to a side that is already implied through two coincident corners. The solver reports redundant constraints using a Jacobian row-dependence test.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two L's extruded up and down from one 2 x 2 grid now join into one solid (the planar rebuild gives each fan of faces around a pinched vertex its own vertex copy), and a pad whose end lies half past the edge of the face it stands on fuses exactly: contact faces are imprinted with each other's outline and the common faces glued away, still held to the retry guard. Face splitting handles a line crossing an edge twice; glue matches differently parametrised curves and keeps several pieces apart. Robustness seed 16 now fuses with the exact volume.
Fixes#679, fixes#674.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The ratchet has no flaky marker, so regressions/lid_ribs_flush_with_window Lid/F11:+y face.hole is marked as failing until the boolean is made deterministic (#707).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An edge flange picked on an edge that cut-outs have partly removed is built along what is left of that edge (the stretch nearest the pick), so each side of a notch can get its own flange.
0.8.0 shipped format 16. Lineage names of a piece qualified by the faces around it, and sheet face names of a rim (rim.<edge>) or a bend end (rim.bend0/1), now come under format 17. Format 16 designs need no transformation, and the version stamp makes 0.8.0 ask for an update.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Home, Fit, the orbit pivot and the scene's depth range counted sketches the view doesn't draw (ones used by an extrude or revolve, ones in hidden components), so after a body moved away from its sketch Home framed the empty spot it came from. They now use the world-space bounds of what is shown now (view::visible_bounds / view::sketch_drawn, viewport::view_bounds). The pick-box cache holds its world state instead of its address. A middle-button double-click goes Home in every preset (Preferences › Navigation: Home, Fit or Off).
Fixes#689
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
As soon as a pick settles the dimension (one line, circle or arc, or the second point or line of a distance or angle), its value box takes the keyboard with the measured value selected. The text follows the cursor while typing and Enter places it there (or in its default place), as one undo step. A click before typing pins the text and keeps the box focused, a click on another entity makes a distance or angle, Tab hands back to placing first, Esc cancels. Letters go into the box, not to tool shortcuts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Sheet metal rims and bend ends attach by name (<panel>:rim.<edge>, <flange>:rim.bend0/1), as do face profiles from sheet faces. A plain or counterbored Hole square to a panel is machined into the flat pattern. Boss, Lip, Rest and Snap Fit keep the body's history and name their own faces. Hole tools name their faces by the part they cut, and pattern copies as copies. A pattern's repeated fillets keep the body's history. Loft sides are named by their section curves. A piece Offset Face moved off its face's plane attaches by its own qualified name. Sides of a profile grown to overlap a flush body keep their sketch line's name, so walls a Join merged take a sketch (#681). Face sketch 75% -> 90%, offset plane 75% -> 90%, press-pull/extrude 64% -> 85% in the action matrix; baseline blessed.
Fixes#494, fixes#495, fixes#560, fixes#561, fixes#563, fixes#564, fixes#681.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The loading card was centred on its height, so Cancel moved when the progress bar or the saved preview appeared, and a click could miss it. That's what made the Cancel test flaky on slow runners. The card is now anchored at its top, Cancel sits in the heading row and the progress bar is always shown. Progress::hold_at lets the test hold a load at a feature boundary until it's cancelled.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Main was red since #577 and #644 landed together. Seed 16's pattern copy rests half past the block's edge; the join used to add 72 mm³ past the edge silently and is now correctly refused. The test records that; proper fusion is tracked separately.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The `pi` story of examples/demo/enclosure_ui_mcp.py keeps the 0.6 flow and adds the rim
rounded in one pick (a tangent chain), "Pi 4" engraved in the floor in DejaVu Sans Bold, and a
PETG material defined in the design, with its mass in Properties, before the height change.
The 0.6 story stays as `pi-0.6` (it built demo/pi-case). The center rectangle now comes from
the Rectangle drop-down. The finished case is in the action matrix corpus as
demo/pi-case-0.8.
docs/screenshots/demo-0.8.gif (9.3 MB) and demo-0.8.mp4 (9.9 MB) replace the 0.6 files: 41 s
at 9x playback.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The Pages deploy failed while the latest published release (0.7.0) still carries ai.storyteller.solvecraft: the bundle check only accepted the new id and exited without a message. Accept either id (both architectures must agree), mark the old refs end-of-life only once the new app is in the repository, show the published id in the install instructions, and name the failing command on any error.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Joining ribs flush with a lid's window left a zero-thickness slit far from the rib (the planar fallback glued back only one of the faces its pieces shared) and displaced the tool on the other joins. Gluing now joins every shared face pair, coincident-face separation checks there is material behind a pushed face for every join, and any join or cut that needed a retry must be closed, keep volume monotonic and leave the part unchanged outside the tool's box, or it is refused with an error. Feature command replies carry the feature's warning.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The loading box was a full-window modal whose backdrop covered the custom title bar. It is now a light overlay under the title bar, so dragging, snapping, resizing and the window buttons keep working during open, import and recovery. The timeline rebuild shows feature N of M, and Cancel or Esc stops it at the next feature (solidcraft_doc::progress). Recovery restore now runs off the window thread too. A control or MCP doc.open or import is answered once the file is in and doesn't block frames; edits sent meanwhile wait their turn. The web build still opens in place.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A single edge's blend no longer needs a plane square to the edge at each end. Straight edges: each rail ends where it reaches the end face's surface, the side edges there are shortened or continued along their line or circle, and the blend ends where its cylinder or chamfer plane meets that surface (a fitted curve checked against both). A chamfer may end across two faces. Curved edges: at a slanted or curved end the rolling ball runs on past the end and the strip is cut back where it meets that face. A straight edge running smoothly into a bend or rounded corner is blended along its whole smooth chain. Rolling-ball fixes: smooth chains follow edges stored against their curve, the ball touches only the faces of the edges it rolls along, the seam normal at a joint comes from the face there, and each strip is fitted with neighbouring sections at its own spacing.
Action matrix: edge.fillet 83.6% -> 93.0%, edge.chamfer 82.6% -> 93.3% (697 new passes, no regressions); reference tests 76/76.
Fixes#556
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Select Other keeps at most 20 rows per kind, nearest first, so a dense spot (a modelled thread, an imported mesh) no longer lists thousands of rows. The title shows how many were under the cursor in all, and ui.selectOther reports it as total.
component.insert_design adds a component holding another design's solid bodies, linked to its file (stored relative to the assembly once saved). component.update_links reads the linked designs again, component.break_link keeps the bodies and forgets the file, and component.links lists them. Links are saved under format 16; an older build keeps the bodies and drops the link.
Fixes#533
An edge flange whose edge ends at an inside corner of the base now trims only the strip under its own edge, so legs and stems past that corner stay. Where the bend ends beside more of the base, the rule's bend relief (relief_width, relief_depth) is cut into the base, or a note says why it was left out.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Fixes#396. A union that would come back as shells touching over an area is rejected and the face-in-face case is fused exactly; line and point contact stay as before.
Body::new no longer turns a long thin solid (a through-all hole's tool) inside out on the strength of a coarse mesh volume: it checks again on a fine mesh first. 79 more face.hole actions pass in the action matrix; the remaining no-result cuts stay under #559.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
MCP execute accepts verbose: false too, using the same compact reply as batch (#599), and the MCP docs gain a section on compact replies.
Co-authored-by: lillith <35337584+0xlillith@users.noreply.github.com>
Fixes#602. Collinear-edge merging, piece naming and the dialog preview each work in one pass instead of quadratic loops; the issue's example goes from 86 s to 5.1 s.
Fillet fit: reduce lowers the radius only where it doesn't fit: picked edges are grouped where they can limit each other (sharing a face or vertex), and each group gets the requested radius where it fits, otherwise the largest that does. Refs #293.
Co-authored-by: Iraklis Kourtis <jyviko@gmail.com>
A flat face that no construction certificate names on its own now attaches by its exact history name (a new lineage face certificate): cut walls and floors, chamfer faces, each pattern or mirror copy, side walls a Join merged, and the pieces a later feature split a face into. Construction certificates still come first; a seed keeps its certificate when it is patterned. Two flush extrudes of one outline joined end to end keep their exact history in the prism boolean. Face sketch 52% -> 71%, offset plane 52% -> 71%, press-pull/extrude from face 41% -> 60% in the action matrix; baseline blessed.
Fixes#491, fixes#492, fixes#493, fixes#557.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
New view.apply command restores a named view, a standard view or any camera (view state, not an undo step); the desktop viewport follows it and the browser's Restore uses it. The MCP screenshot tool takes a named view or a camera. Un-ignores the #528 regression test. Fixes#528.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
The Flatpak app id, desktop/metainfo/MIME file names, icon name and the window's Wayland app_id / X11 class are now ai.storyteller.solidcraft, the id the ArtCraft launcher expects.
Existing Flatpak installs move over: the Pages repo marks app/ai.storyteller.solvecraft end-of-life with a rebase to the new id, so flatpak update offers the switch and Flatpak carries ~/.var/app along. A new install next to the old one copies settings and autosaves on first launch through a read-only grant on the old data folder. The metainfo declares <replaces>/<provides>, the desktop entry X-Flatpak-RenamedFrom, and AppImages drop the menu entry an older AppImage installed. deb/rpm upgrades remove the old-named files.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A save with cache: true also stores a small bounded PNG thumbnail (240 × 150, at most 256 KiB, keyed to the exact document) made from cached meshes only. Home uses it when its local picture is missing, without rebuilding the design; ordinary saves and downloads still omit it. Closes#303.
Co-authored-by: lillith <35337584+0xlillith@users.noreply.github.com>
A point dimensioned from linked geometry (projected, included or intersected) keeps its side when a parameter moves the link's source past it: the solve reads sides and turns from the stored state instead of from beside the moved line, both on recompute and when editing the sketch.
Co-authored-by: NOWN-DEV <257771410+NOWN-DEV@users.noreply.github.com>
Box, Cylinder, Sphere, Torus and Coil go where a face or vertex was picked before starting them: centred on the point, standing on a planar face (a box on an axis-aligned face sits on it; cylinders and coils take the face normal as axis). The dialog shows the position with a button to place at the origin instead. Fixes#580.
Co-authored-by: lasselian <lassesorensen80@gmail.com>
Sketch Create groups its drawing tools under Line, Rectangle, Circle, Arc, Slot, Polygon and Spline. Each family's main button starts the part's last-used variant and its caret lists the others; choices are per open part, kept across sketch sessions and tab switches, and never touch the design or undo history.
Co-authored-by: Karl Von Richter <karl@von-richter.net>
A design can define its own physical materials (material.create, material.edit, material.delete): name, density and an optional look. They work everywhere built-in materials do: assignment, mass properties, colours and configuration cells. Stored in user_materials, an additive field under format 16. Un-ignores the #554 regression test. Fixes#532.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
An edge flange on another flange's far edge (a return) runs along that edge from end to end, following the parent's closed or cut-back corner, and returns of one feature close toward each other at convex base corners, mitred the rule's gap apart. Bends stop square where the mitre allows. Fixes#593.
Co-authored-by: NOWN-DEV <257771410+NOWN-DEV@users.noreply.github.com>
The bend table, flat pattern and DXF list a bend only where its centre line has material: a bend that cut-outs (or pockets that together go through the sheet) shorten keeps what is left, and one they split gets a row per piece, all with the bend's number.
Co-authored-by: NOWN-DEV <257771410+NOWN-DEV@users.noreply.github.com>
MCP batch takes verbose: false for compact replies: feature steps leave out the full body list and sketch steps the ids they repeat outside created. Feature replies now also list changed_bodies and removed_bodies next to created_bodies, so a compact reply still shows every effect of a step. Un-ignores the #530 regression test. Fixes#530.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
A Box cut, or a Cylinder cut whose axis is the sheet's normal, machines a sheet metal body like the extrude cut of its footprint: through the sheet it's a cut-out (a cylinder an exact round one in the flat pattern and DXF), part way it's a pocket. Primitives at other angles or across a bend are still refused. Un-ignores the #541 regression test. Fixes#541.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
Faces that a sheet fold or unfold moves keep their names: points on a moved panel are taken back through the panel they lie on to where the sheet held them before the feature, and named from the faces there. Un-ignores the #535 regression test. Fixes#535.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
Tapered extrudes accept profiles that mix lines and tangent arcs (slots, D-profiles, rounded rectangles): each arc keeps its centre and angles and its radius moves with the taper, so its side stays an exact cone; lines are offset and mitred. An arc meeting another edge at a corner is still refused. Un-ignores the #546 regression test. Fixes#546.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
Draft extends the picked faces to every face on the same surface, so one pick on a cylinder's side (stored as two halves) drafts the whole round wall into a frustum. Un-ignores the #547 regression test and adds a kernel test for a round hole. Fixes#547.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
An extrude cut or join only extends its tool past an end that lies on a target face, and by less than half the way to the next face level, so a tool can no longer cut through a thin floor beyond an end that touches no face. Un-ignores the #545 and #544 regression tests. Fixes#545.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
Inscribed, circumscribed and edge polygon tools preview their actual outline (side count and orientation) while drawing, with the current line type, instead of only the support circle. Preview work is bounded. Fixes#520.
Co-authored-by: lillith <35337584+0xlillith@users.noreply.github.com>
sheet.fold tests its fixed point against the whole sheet body, so a point far above or below the sheet is refused atomically, while a point on any face (including a flange's leg) still works. A stored fold whose point ends up in the air recomputes, keeps its side and warns. Un-ignores the #371 regression test. Fixes#371.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
sheet.flat_pattern reports every cut-out exactly in a new cutout_shapes list, parallel to cutouts: a round hole as a circle, other cut-outs as their lines, arcs and (for other curves) polylines. Pockets get the same. The DXF writer and the JSON share one circle test. Un-ignores the #542 regression test. Fixes#542.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
A scheduled/manual CI job scans locked dependencies for advisories, bans and unknown sources with cargo-deny 0.20.2 (never on pull requests, contents: read, no secrets). The first scan flagged quick-xml 0.38.4 (RUSTSEC-2026-0194/0195), so the workspace moves to quick-xml 0.41. Fixes#468.
Co-authored-by: lillith <35337584+0xlillith@users.noreply.github.com>
Hidden bodies, occurrence placements, sketches and origin items are saved with the design and come back when it is reopened. Visibility is still not an undo step, but hiding or showing something now counts as an unsaved change. Additive field under format 16. Fixes#531.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
Construction axes, points and planes resolve their picks against the model as their component sees it (in the root, every placement, pasted instances too), and a Hole picked on another component's body cuts that body. components-arm 355 -> 502/594 and joints-hinge 304/390 -> 371/408 in the action matrix.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Construction (X) keeps the active sketch tool, its picked points and inline values while switching line type, and drawing previews and hover show the same dashed pattern as committed construction and centre lines. Fixes#584 and #581.
Co-authored-by: lillith <35337584+0xlillith@users.noreply.github.com>
sheet.flange refuses the sheet values its rule owns (thickness, K-factor, bend radius, reliefs, hem gap) instead of silently ignoring them, and says to use a rule. Un-ignores the #570 regression test. Fixes#570.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
A stored fold whose fixed point drifts off the sheet after a parameter change keeps the side the point lies on and shows a warning instead of breaking the timeline. A new fold still refuses a fixed point off the sheet, atomically. Un-ignores the #370 regression test. Fixes#370.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
A copy added by raising a body pattern's count after a rename is numbered after the highest stored copy, so it no longer takes the number the rename freed. Un-ignores the #372 regression test and adds one for undo/redo and save/reopen. Fixes#372.
Co-authored-by: jonaswre <jonas.wrede@beyondit.gmbh>
Refs #303. file.save / file.save_as can store a warm native geometry cache (cache: true, off by default) keyed to the document hash, app version and adapter revision. Opening shows a bounded preview from it until the rebuilt session replaces it. Any cache failure counts as a miss.
ui.at {plane} now returns the middle of the origin plane's square, or else the middle of a quarter that a click actually picks, so scripted clicks on XY no longer start a sketch on XZ from iso-like views.
apps/solidcraft-cli/tests/action_matrix.rs tries sketch, offset/angle/mid planes, press pull, extrude from a face, hole, draft, shell, fillet, chamfer, axes, points, planes, projection, move, copy, mirror and pattern on the final bodies of 104 designs (the old design files, the examples, the 76 reference recipes, the enclosure demos, the Pi case built through the UI, eight typical parts), through the commands and picks the desktop sends, each design in its own process, with an edit round on a sample of passes. tests/action_matrix/baseline.json is a ratchet: an action that passed there and fails now fails the test; newly passing ones are reported for cargo xtask action-matrix --bless. Every CI run checks a fixed sample; the nightly Action matrix job runs the whole matrix in four parts. 30,567 of 35,576 actions pass (85.9%); results and causes in docs/action-matrix.md, one ignored reproduction per open cause (#556-#568).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Fixes#374. Feature replies gain created_bodies: the bodies the new feature made, apart from the ones it changed or that were already there. Documented in the execute tool description and docs/mcp.md.
Fixes#478. Desktop Ctrl/Cmd +/-/0 now update the saved UI scale instead of flashing back, and the Preferences scale slider applies the new layout on release (range 50-250%, matching the preference validation). Addresses the drag jitter in #272.
Fixes#527. Bodies are paired by a sort-and-sweep over their bounding boxes, and the cap moves from 200 bodies to 20,000 overlapping pairs per call (reported as truncated / skipped_pairs).
sketch.text and sketch.edit_text take an optional font: an installed family or a .ttf/.otf/.ttc file. Edit Text gets a Font field. The font is kept with the text under format 16; no fonts are bundled, and the font search is bounded.
Fixes#513
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
solid.move takes copy: true to keep the source bodies and transform a new copy of each, with a Create Copy box in the Move dialog. Designs without it save and load exactly as before.
Display meshes also bound the angle one facet turns, so bends and small holes on a large body are drawn round instead of as chamfers or squares. Bounded per face, and only the display mesh changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Sheet Metal Rules get a Corner Relief Shape (notch, round or rectangular), sized by the rule's corner relief, cut where two bends meet at a corner of the base. Notch, the default, keeps existing designs as they were. File format 16.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Text outlines count as drawn geometry when picking profiles, and "all profiles" leaves out the counters of letters (the holes of A, B, 8), so text extrudes as letters with their holes whether or not other curves are drawn next to it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When a body that was already cut into pieces is cut again, the new pieces are numbered past the names already taken ("Bar (2)"), instead of reusing "Bar (1)".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A SolidCraft-owned portable B-rep schema (geometry schema 2 with analytic curves and surfaces, explicit frames and ranges, fitted intersection curves) and a versioned body envelope (SCBREP03) with its own metadata records, a bounded binary codec, a validator with resource budgets, and an adapter that refuses what it can't convert exactly. Pinned by a box-and-fillet golden fixture. Not yet written by application saves. Refs #303.
When sample points say two bodies are apart or one is inside the other, the facetings are now checked for surfaces passing through each other; if they do, the boolean is computed instead of shortcut. If it still can't be intersected, the sampled answer is kept as a flagged, unverified result, and Interference lists such pairs as "could not be checked" with the reason. Fixes#442 and #443.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A hole positioned on a face now follows that face along the hole's own axis instead of sliding to the face's nearest point. When the face no longer reaches the position, the hole stays where it was placed and the feature says so.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When adjacent profiles are merged for one extrude, collinear lines on either side of where a shared boundary ended are joined back into one edge, so the body has no split faces or extra vertices for later booleans to trip on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two line segments that lie on one another within the tolerance, a hair from parallel, no longer report a crossing somewhere along their overlap; the ends of the overlap split the edges instead. This stops booleans from leaving sub-tolerance edges or stray vertices.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every Options row puts its control in one right-aligned column, Look At is a single label with an icon button, Options folds with a disclosure triangle, and the toggles use plain nouns (with dedicated ids for Profile and Points so they don't collide with other labels).
The Linetype buttons now read "Draw as {linetype}" with nothing selected and "Convert the selection to {linetype}" with a selection, in all four locales.
Bump the workspace to 0.7.0 and write the 0.7.0 release notes. Minor bump: the design format moved to 15 (sketches on cylinder caps and sheet metal panels), so 0.6 can't open 0.7 saves, and new features landed (Select Other, per-edge flanges, sketches on caps and sheet metal faces).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A profile traced over a face sketch's projected boundary left the extrude's caps and sides without an exact planar source, so a sketch or offset plane on the new top face was refused. The drawn curve is now the edge's source where it covers a projection. Clearer fixed-plane message; regression tests in the engine, UI harness and design corpus, plus a corpus-wide flat-face attachment check (known gaps under #491-#495).
Fixes#489
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Hold the left button still (long-press on touch) or right-click › Select Other… to list everything under the cursor: sketch points and curves ("Line 3", "Point (arc 2 centre)"), vertices, edges, faces, bodies, profiles, planes and axes, filtered like a click and by the open dialog's active input. Hovering a row lights it up; clicking selects it (Shift/Ctrl add) or feeds the dialog. A repeat click on the same spot with nothing changed selects the next thing there; a double-click keeps its own actions.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Replace the 0.2.1 enclosure recording with a 40 s take of the current desktop UI building a Raspberry Pi 4 case: a sketch dimensioned with the board's parameters, extrude, fillets and shell, an M2.5 standoff patterned on the 58 x 49 mm hole rectangle, port cut-outs, floor vents, a colour, then height 24 -> 32 mm in Parameters. The demo script gains the pi story (now the default).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Profiles revolved together merge into one region that keeps a vertex where
each shared edge met the axis. With the axis edge in several pieces the loop
skipped the builders for a profile with an edge on the axis, and the generic
sweep turned each piece into a degenerate face that booleans failed on.
Runs of collinear axis lines are merged into one first.
Fixes#487
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix: name an untitled design after its file on save and open
Design tabs all read "Untitled" because a design's name only ever came
from file.new. Saving a design still under the default name now names it
after the file (bracket.solidcraft -> "bracket"), and opening a file whose
stored name is the default shows its file name. Names chosen by the user
are kept. The rename is part of the save, so it is not an undo step and
does not leave the design marked as changed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HWLPLy7ug7URSK5WybjRNa
* test: expect the file's name after saving an untitled design
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HWLPLy7ug7URSK5WybjRNa
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The live preview also evaluates each command as a new body, to show the tool of a join or cut
(as_new_body sets every `operation` to "new"). A combine has no new-body form: its operation is
between bodies that already exist, and solid.combine rejects "new" ("operation must be join, cut
or intersect"). That error became the preview's error even when the combine itself worked, and
since OK is greyed while the preview has an error, a combine could not be applied from its
dialog at all. as_new_body now leaves solid.combine alone, so its preview shows the result.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
`fit: reduce` acts only on refusals marked too large. Three kinds weren't:
the corner build's contact lines leaving their faces or running backwards,
and a curved chain's contact reaching past the end of the seam it trims. Two
more were hidden: a later edge's too-large refusal lost to an earlier edge's
other error, and the corner build's lost to the local blend's in `fillet`.
The messages are unchanged.
Refs #293
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Moved in by more than its radius, a round collapses and its neighbours meet
along a new edge. That edge was always built as a straight line keyed by its
ends, so on a disc whose circular edge is rounded the four arcs where the floor
meets the side became one chord used by four faces, and the shell failed with
"This shell is not oriented and closed" (#319).
The new edge now runs where the edge's face meets the face across the round
(the round's far side, crossed as a strip), through the edge's middle moved
onto both: a line or a circular arc, checked to lie on both moved faces, and
refused otherwise. Twins are the new edges with the same ends through the same
middle.
Fixes#319
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(ui): draw sketch constraint glyphs in red
Traditional CAD convention marks constraint glyphs in red. glyph_edge
and glyph_text in both themes now use red values. glyph_bg is
unchanged. The same glyph box switches to the error colour when a
constraint fails. The new reds are chosen to stay distinct from that
error colour. They are also distinct from the blue used for selection
and hover. See the pull request body for the before and after RGB
values and the contrast reasoning.
* fix(ui): match constraint toolbar icons to the red glyph colour
The CONSTRAINTS panel buttons and their drop-down rows drew their accent
strokes in the normal sketch tool green. They now use glyph_text, the same
red the matching constraint glyph box uses in the sketch view. A new
panel_accent helper in toolbar.rs picks the colour: the constraint red for
the CONSTRAINTS panel, the existing sketch green for every other sketch
tool, and the plain accent everywhere else. No other toolbar styling
changed.
* Add a one-surface fill of closed 3D loops to geom
A closed loop of curves that does not lie in a plane has no exact surface;
the usual answer is one smooth face fitted over it. fill_loop does that:
- the loop's best-fit plane (Newell normal), with a frame chosen from the
loop itself so the result does not depend on where the loop starts, its
direction, or where the part sits
- the loop seen along that normal must be simple and must not fold or run
along the normal; otherwise it is refused ("not supported yet"), never
filled with wrong geometry
- one non-rational bicubic B-spline height field over the loop's box,
fitted by least squares to the loop's samples (The NURBS Book 9.4.1,
9.4.3, eq. 9.62/9.65) with a thin-plate smoothness term
lambda*A*int(h_uu^2 + 2 h_uv^2 + h_vv^2), A one grid cell's area, so the
fit is the same at any scale; only planes are reproduced exactly
- span counts per direction (long narrow loops fit), refined until the
loop is within tolerance, else refused with the deviation reached
bspline: bisected span search, basis functions on a stack array, one
clamping helper shared by curves and surfaces, one binomial helper.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
* Patch fills non-planar loops of edges with one smooth face (#250)
surface.patch with edges refused any loop that is not planar ("not
supported yet: patching a loop that is not planar"). It now fills it:
- planar loops still get one exact planar face, but planarity is certified
conservatively: lines by their ends, B-spline/NURBS edges by their control
points, intersection curves by their samples lying on a support plane
within tolerance; a loop that can't be certified is filled instead
- other loops get one face from geom's fill_loop, bounded by the body's own
chained edges, whole (no split): its history is one generated face whose
edges are the body's, and it stitches back into a closed solid
- intersection-curve edges are sampled by a seeded, guarded double
projection shared by the planarity test and the fill (plane_fit)
- the face tolerance is 10x the planarity tolerance (documented: later
operations take the face to contain its edges within it)
- edge picking runs inside the guard; the geom->truck conversion is shared
with the T-spline sheets (spline_conv); the Newell plane fit is shared with
STEP import
Tests: 3- to 8-sided, concave (dart, arrow, plus), flat-corner, two-edge and
imported rims fill as one face and stitch closed (volumes against an
independent integral); finite mesh normals, same result for any pick order
or placement; non-height-field loops and too-fine detail are refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
* Test the Patch dialog filling a non-planar loop of picked edges
The Patch dialog's only test used a sketch profile. Picking the four
curved rim edges of a box cut by a sideways cylinder now makes one
surface.patch command whose result is a single smooth face over the rim.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A sketch on a flat face of a sheet metal body was refused ("this face has no exact planar
source"), so the only way was a plane frozen in space. When the sheet changes (a flange's
angle, the thickness), the frozen sketch stays behind: a cut from it fails on the slanted
flange or leaves a blind pocket in the thicker sheet.
`sketch.create` with a face of a sheet metal body now attaches the sketch to the panel the
face belongs to. It keeps the face's name, `sheet:<body>:<panel>:<top|bottom>` (the panel is
the base or a flange's feature tag), and takes the panel's exact plane from its flat region
and fold each time the model is evaluated, so the sketch follows the sheet through parameter
changes, unfold and refold. Its coordinates are the flat pattern's (the bottom face's y
mirrored). A face that is gone is reported as a lost attachment, never replaced by the
nearest face. The attachment follows a renamed or copied body, shows as `sheet_face` in the
command result and as its own status in Redefine Sketch Plane. `fixed: true` still freezes.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Sketch on a cylinder primitive's caps (#420)
A solid.cylinder's caps carried no construction certificate, so a sketch
on its top or bottom was refused unless fixed:true (#420's crossed
cylinders, built from primitives).
- kernel: FaceSource::CylinderCap { producer, high, plane };
Body::with_cylinder_certificate turns a cylinder's two extrude caps into
certified caps. kernel::cylinder's output is unchanged for its other
callers.
- doc: FaceCertificate::CylinderCap { high }, named F<id>:top / F<id>:bottom
as the cylinder's face names already are; New Body and Join cylinders
are certified, shells carry it as their planar source.
- format 15: designs can store the new certificate kind; version 14
designs need no change, and older builds ask for an update instead of
meeting an unknown kind.
Tests: top and bottom attachments follow a height and base edit and a
reopen; crossed joined cylinders keep both caps attachable; a shelled
cylinder's floor keeps its cap identity; malformed certificate JSON is
refused.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YDDPmNdRoBgkHcLeUDKVfb
* Format 15 also covers sketch_axis links (#448)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YDDPmNdRoBgkHcLeUDKVfb
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Project, Intersect, Include 3D Geometry and Text resolved the active sketch's
plane without the model, so on a construction plane that is built from geometry
(perpendicular to a plane through a sketch line, through three points, midway
between two faces, tangent, along a path) they stopped with "this plane is built
from the model: evaluate first". Evaluation resolves the same plane against the
model before the sketch, in its component's frame; the sketch commands now do
the same, in one place, for adding a link, refreshing the links on an edit,
snapping a point onto an edge and the internal face profiles.
Tests: a sketch on a plane through a sloped sketch line projects a point and a
line of that sketch, follows a parameter that tilts the plane (also after reopen
and after another edit), holds text, intersects a body and includes an edge; the
same on a three-point plane and a midplane.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A left press with Ctrl (or Cmd) held, no Shift or Alt, on empty space orbits
for that press, unless the preset binds Ctrl+left itself. It only orbits
once the pointer moves: a Ctrl+click still adds to the selection, and
Ctrl+drag on sketch geometry still drags it without snapping. Works with a
tool active, without adding geometry. Ctrl+box selection from empty space
becomes the orbit; Shift+box selection still adds.
Preferences › Navigation and the UI tour mention it. Test drives every
preset through real pointer events.
Claude-Session: https://claude.ai/code/session_01YDDPmNdRoBgkHcLeUDKVfb
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The README called the solver damped Gauss–Newton. It now places what the
constraints fix by construction (loci intersections and rigid-cluster
decomposition, after Owen 1991 and Fudos–Hoffmann 1997) and runs
Levenberg–Marquardt on the rest, with the rank of its Jacobian giving the
remaining degrees of freedom. The constraint list also gains the ones it was
missing (point on curve, curvature, point-to-line, arc sweep, arc length, slot
length), and live drags are described.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With "Orbit around: Point under the cursor", a drag that starts over
empty space used to pivot on the point under the cursor at the target's
depth, so an orbit begun near the edge of the view swung the model
around an off-screen point. It now pivots on the centre of the visible
bodies and sketches (the same bounds Fit frames), or on the view centre
when nothing is visible. A drag that starts on the model is unchanged.
Claude-Session: https://claude.ai/code/session_01HWLPLy7ug7URSK5WybjRNa
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* docs: add OpenGraph and Twitter card tags to the book and web app
Links to the docs site and the browser app now unfurl with a preview card.
The book gets the tags through theme/head.hbs (og:title follows the page
title); the web app's index.html carries its own, pointing at /app/. Both
use the social preview image, now served from the book at /solidcraft/og.png.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HWLPLy7ug7URSK5WybjRNa
* docs: list the social preview image in ATTRIBUTION.md
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HWLPLy7ug7URSK5WybjRNa
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
One edge flange feature gave all its flanges one height and one angle. A
part whose flanges lean differently took one feature per angle, and
flanges of separate features don't close toward each other, so its
corners stayed open.
- sheet.flange (edge): an entry of `edges` is a point [x, y, z] as before,
or {point, angle?, height?} whose angle and height (expressions, checked
like the feature's) replace the feature's for that edge. Bad input is
refused with the edge named. Before, the object form was read for its
point and the rest was ignored without a word.
- The feature stores its edges as FlangeEdge. An edge without values of
its own is written as the plain point it always was, so designs without
them are read and written as before; an unknown key in an edge is
refused, not dropped.
- The edges' values are feature inputs of their own (Edge angle, Edge
height, after the feature's), so parameters, renames and
parameters.change reach them. timeline.edit names the edges again when
it changes them, so each value stays with its edge.
- SheetBody::add_flanges_each trims every edge by its own setback and
adds the flanges. The corners between them close as between equal
flanges: the closing already works from the two flanges it joins (each
one's fold and leg), so the flat pattern follows by hand from the
angles, the heights and the rule (the tests derive it).
- The Flange dialog shows a row per picked edge with an angle and a
height box (empty: the dialog's value) once more than one edge is
picked, and brings them back on Edit Feature.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
`sketch.constraint.symmetry` (line), `parallel`, `perpendicular`,
`collinear`, `tangent` and `coincident` (point on curve) now take `x` /
`y` for the sketch's own axes, matching what `solid.revolve` means by
lowercase x|y. A real curve with that id still wins.
An axis becomes a linked construction line on it: a new
`LinkSource::SketchAxis` resolved in sketch coordinates, so it stays on
the sketch's axis on any plane, is shared by every constraint on that
axis, is left out of profiles and is reported in `created`. Equal,
concentric, midpoint and horizontal/vertical keep refusing axes, since
the axis line's finite length or fixed direction would make them
meaningless.
Claude-Session: https://claude.ai/code/session_01YDDPmNdRoBgkHcLeUDKVfb
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A Union dropped every construction certificate, so after any join no face
could take a durable sketch: sketch.create failed with "this face has no
exact planar source" even on a face the join never touched (#420, and the
"CREATE SKETCH: operation failed" reports #344, #195).
- kernel: map planar sources through a Union from both operands, with the
same exact-support rule Cut uses. Coplanar supports of the two operands
carry distinct identities, so a face merged from both stays without one.
- doc: Join extrude and Join box tools are certified like new bodies; a
cap moved off its designed height (into the target, or past a curved To
face) is not certified. Join producers are accepted when certifying.
Tests: both operands' faces attach and follow an upstream edit and a
reopen; a merged coplanar face is refused and fixed:true works; a face a
join covers whole is lost without moving to the new roof. Tests that used
a joined face as their "uncertified" example now use a merged face; the
trimmed-face case in a_sketch_on_a_face_stays_on_that_face now stays
attached to the surviving piece.
Claude-Session: https://claude.ai/code/session_01YDDPmNdRoBgkHcLeUDKVfb
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
kernel::sew refused a solid with a thin planar face away from the origin: a slab
30 x 2 mm and 0.00013 mm thick with a corner at (-300, -200, 5), turned 0.45 rad
about x, failed with "sew: face 1: cannot attach a plane to a wire that is not on
one plane", though its faces are flat to 1e-14 mm.
truck's plane fit sums the cross products of a boundary's control points about a
centre and takes the points' sum for that centre (it divides by the number of wires,
not of points): eight times as far from the origin as a rectangle of four lines. The
sum is the same about any centre, but its rounding grows with the square of the
centre's distance, and on a thin face, whose normal is short, it turns the normal far
enough for truck's own 1e-6 flatness test to fail: truck finds the slab's four thin
faces 3.0e-7 to 2.1e-6 mm out of the planes it fitted.
A planar face keeps truck's plane wherever truck gives one. Where truck refuses, the
same fit is made on a copy of the boundary moved so that the mean of its control
points is the origin, and the plane is moved back; a boundary that leaves its plane
or has no area is refused as before.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A body of several pieces (a plate and the islands a cut leaves inside its
holes, such as letter counters) was written as one BREP_WITH_VOIDS whose
other shells were voids. Export now writes one solid per lump, each with
its face colours; a real void stays a void of its lump.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A full revolve of a profile with more than one edge on the axis (a spherical shell's half ring:
both radial lines lie on it) took the generic sweep, built in place. Truck tessellates the
degenerate faces those edges sweep reliably only with the axis through the world origin, so on a
sketch plane off it the body measured about a third too large (a shell from r 18.5 to 40, 10 mm
above XY: 324 339 mm³ instead of 241 561), and booleans with it failed or removed the wrong side.
Such profiles are now swept about an axis through the origin and translated into place, as
revolve_touching_axis already does for profiles with one edge on the axis.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Minor bump: the SolidCraft rename (#389) and the design container format (#327) both change
the files people work with; builds 0.5 and older can't open designs saved by 0.6. Release
notes cover the new format, what's new and the upgrade from SolveCraft.
Claude-Session: https://claude.ai/code/session_01HWLPLy7ug7URSK5WybjRNa
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Pattern copies of a post joined to a body already holding the original post had
their tops in its plane, so every join first pushed faces apart (about 190 ms,
then discarded). Faces now coincide only where their boxes meet, the push-apart
retry runs once per boolean instead of twice, and splitting prism outlines skips
the sampled arc projection for points off the arc's circle.
Sensor enclosure evaluation: 3.07 s -> 1.63 s, same bodies.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The planar arrangement solved a line and a circle, or two circles, for their two
crossings even where the curves only touch. At a tangent contact the two roots are
rounding noise spread along the curves by its square root: a 1e-12 mm error in the
centre's distance from the line puts them 2e-6 mm either side of the contact on a
2 mm arc, well inside the tolerance of both curves; when the noise falls the other
way there is no root at all. Cutting a rounded corner (an arc tangent to the two
edges it ends on) out of a plate with planar_regions_boolean then ran that far along
the plate's edge past the arc's end before turning onto the arc: a 200 x 37 plate
turned by 0.25 degrees came back with six edges, one of them 3.45e-6 mm long.
Curves that touch within the arrangement's tolerance now meet at the one point their
centres give: the foot of the circle's centre on the line, or the point on the line
of centres of two circles tangent inside or outside. Circles of nearly one radius
about nearly one centre keep the general formula.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A cut-out made in the flat pattern could cross a bend only with edges
running along the bend or square to it: a round hole or a slanted slot
over a bend line failed to fold ("not supported yet: curved cut-outs
across a bend", "... cut-out edges running across a bend at an angle").
Sheet: a flat edge in a bend zone now folds whichever way it runs: along
the bend to a line, across it to an arc, and any other way to a curve on
the bend's cylinder (a slanted line to a helix). The curve is carried as
a chain of cubic Hermite pieces fitted within 1e-7 mm on both faces of
the sheet, with breaks that depend on the flat edge only, so the top
face, the bottom face and the wall share them. The wall between the two
chains is a ruled surface with radial rulings: the cut-out's wall stays
square to the sheet. The bend's own faces stay exact cylinders, now sized
to the whole boundary of their cell (a curved edge may bulge past its
corners).
Kernel: such an edge has no line, arc or conic form, so the sewing
interface gets EdgeSpec::Chain (a chain of cubic Bézier pieces: a cubic
B-spline with triple inner knots, its ends on the shared vertices) and
SurfSpec::Ruled (two chains joined by straight lines, so a wall built
from the chains holds them exactly). The sheet module's rev_edge becomes
EdgeSpec::reversed, so the new variant is handled where edges are
reversed.
The test for "along" and "across" is in millimetres (1e-6 mm at the
edge's ends) instead of relative 1e-9, so coordinates rounded by a script
fold as drawn.
Tests. Kernel: a box with one wall bulging as a cubic (chains of two
pieces at the bottom and the top, the wall ruled between them) sews to a
valid solid of the analytic volume, 27.6 (measured 27.59965); chains with
the wrong number of control points, a NaN or too many pieces, and a ruled
wall between chains of different piece counts, are refused. Sheet: a slot
turned 30° and a round hole across 72°, 90° and 108° bends fold to valid
solids that lose the cut-out's flat area times the thickness (K = 0.5:
400 and 100.53 mm³, measured within 0.02); a slanted flat edge folds to
its exact helix within 1e-7 mm (measured 1.9e-8) on both faces; lines
along and across the bend keep their line and arc. Engine:
fold_through_cut_outs now expects a round hole across a fold to fold, and
to unfold to the same round hole; the per-bend machining guard (#296)
sews a bend that carries a round cut-out, so a hole beside it cuts and a
hole in the bend is still refused.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
The flat-pattern point budget (16 384) counted every cut-out by its 0.001 mm
chords, arcs included, so a plate with 240 round 4 mm holes (about 100 chords
each, 480 exact arcs) was refused with "sheet metal flat geometry exceeds the
bounded processing budget" when the holes were cut in the unfolded sheet. The
planar booleans work on exact lines and arcs whole, so those now count once
each; a sampled curve still counts its chords. A second limit keeps the
chords of everything, arcs included, at 65 536, the same figure the planar
boolean applies to the sampling it does itself, so the work stays bounded.
planar_region no longer samples an exact loop only to discard the result: the
complexity check right above it already covers the same budget.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A panel label too wide for its panel (beside its caret) wraps between words
onto a second line; a single word wider than the panel stays whole. Each line
is centred, and the label's hover area grows to hold both lines. A panel's
buttons are centred in the panel, so a one-button panel such as Finish Sketch
lines its icon up with its label.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* fix: fillet fit review follow-ups
From the review of #388 (merged before these landed):
- The Fillet dialog's Applied Radius showed nothing when editing a fillet
that was already reduced: the preview found the radius by diffing the
states before and after. The engine's preview now keeps the command
replies, and the dialog reads the radius of the feature they name.
- Pattern copies of a reduced fillet try the radius the original was
reduced to first, instead of each searching again (up to 16 fillets).
- fit: reduce with a chord or variable radius is refused instead of being
silently ignored.
- One helper appends a note to a feature's warning; too_large is imported
like unsupported in the loop and curve blends.
Refs #293
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
* Use the solidcraft_doc crate name after the rename
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HWLPLy7ug7URSK5WybjRNa
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
- frame::heading draws a panel's title row (heading_bg) or a section heading
(subheading_bg) on a band darker than the dialog, in both themes. Every
docked command dialog, the Sketch Palette and Redefine Sketch Plane use it.
- Sketch Palette: "Options" and a new "Sketch Status" heading are section
bands; the status line sits under its own heading.
- Sketch Palette: Finish Sketch no longer centres in the leftover height, so
the palette ends one line below it, one line after the status.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
object_height, curved_target and curved_target_any took the first solid body with any face
within its pick tolerance of the To Object point. That tolerance is 5 % of the body's size,
so a large body beside a small target took picks made on the target: a ridge extruded up
to a post 2 mm above a plate treated the plate as the target, nothing trimmed the tool, and
it ran past the post (to x 9.08 instead of -2.31).
They now share face_at: the nearest face over every solid body, still within that body's
tolerance. appearance gains nearest_face (the face and its distance) and pick_tolerance,
which face_index_at now uses as well.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(doc): extrude and revolve in a moved component from another component's sketch
A feature built from a sketch of another component read the sketch's plane as if it were in
the feature's own component. When the two components are placed differently (a moved or
jointed part), the tool landed transformed along with the feature's component: a join on a
part turned half a turn about Z, from a root sketch, came out on the far side of the origin.
Extrude and Revolve now see the sketch through both placements, as sweeps along 3D paths and
sketches on another component's face already do (sketch_in_feature_frame, using
profile_in_frame). The sheet-metal rotated-frames test built its cap witness from a sketch of
the moved component while the root was active, relying on the old reading; it now builds the
witness in the sketch's own component, which is what it measures against.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test: To Object extrude in a moved component from a root sketch
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* style: rustfmt
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Edge flanges meeting at a corner of the base were closed only at exactly
90°; at any other angle they stayed open (a V at 72°) and the rule's gap
was not used.
The flanges of one feature now close at any convex base corner and any
bend angle: each panel end is cut square to the sheet on a line in the
flat (slanted off 90°, stored as ext plus a new per-end slope) so the
edges nearest each other, on the inner faces, run parallel the rule's gap
apart about the plane bisecting the inner faces. The construction reads
the two flanges' own angles, so it holds for unequal ones as well. At 90°
it is the extension the old code gave (reference case 69-sm-tray is
unchanged). A corner that can't close (a gap under 0.01 mm, flanges
bending opposite ways, ends that would overlap in the flat pattern) stays
open, and the flange feature reports a warning that says where and why.
A flange of a later feature leaves the corner as it was (reference case
75-sm-bracket-holes: the earlier flange keeps its length).
Tested at 72° and 90° against a hand derivation, on a trapezoid, and
every 5° from 45° to 135° on a rectangle and a trapezoid (gap exact, one
flat region without hairline edges, one valid shell holding the flat
material).
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With sketch curves selected, the palette's Normal / Construction / Centerline
buttons now show the line type of the selection (each type of a mixed
selection); with nothing selected they show the drawing mode, as before.
Previously no button was highlighted while curves were selected.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A parameter change can break a feature without failing the command that
made it; parameters.change reports it only in its `errors`. `run` then
exited 0 and exported the bodies as they were before the failing feature:
a shrunk sheet-metal tray whose unfold failed came out folded, without
the holes and cut-outs made after the unfold.
`run` now checks the evaluated timeline before writing anything. Any
feature with an error fails the run with status 1, lists the failing
features on stderr and writes none of the --out and --save files, as a
failing command already does. --allow-errors writes them anyway. `eval`
is unchanged and still reports the errors in its timeline.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Snap package: build, test and attach a strict snap to every release
packaging/linux/snap/snapcraft.yaml (core24, strict, gnome extension) repackages
the release's Linux tarball like the Flatpak does. The release workflow builds it
for amd64 and arm64, installs it with --dangerous, models a box in the confined
app on headless X11 and Wayland, and attaches solidcraft-<version>-linux-<arch>.snap
plus the version-less copy. When SNAPCRAFT_STORE_CREDENTIALS is set, nightlies go
to edge and release drafts to candidate, and snap-store.yml promotes candidate to
stable when a release is published. The README gets a Snap badge and install line.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
* snap smoke: stop the app's whole process group, refuse a busy control port
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
* snap smoke: quit the app cleanly between backends
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Badges and one-line instructions for Windows, macOS, Linux and the web app,
right after the tagline and demo. The links use releases/latest/download with
version-less file names, which release.yml now attaches (and checksums)
alongside the versioned installers. The Flatpak note and the name/compatibility
section move into the book's Getting started page.
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Every panic (in a command, in the kernel, anywhere else) and every unclean
exit (abort, stack overflow, kill) leaves a report in a local crash folder:
message, file:line, a trimmed backtrace, version, commit, system and renderer,
the command and recent commands, and when possible the design and replay.json
that reproduce it (solvecraft-cli crashes replay <id>).
- Cost: nothing until the app installs the hook (tests, CLI and web never
do); then one small journal line per top-level command of the design on
screen, taken back with a truncate when the command changed nothing.
- Unclean exits: each run holds a lock on its journal; a journal nobody holds
at the next start belongs to an app that died. Autosaves restart the
journal from the autosaved design.
- Bounded: a crash's signature (place + message, numbers ignored) is counted,
not rewritten; at most 5 new reports per run, 20 reports, 50 MB, 90 days;
recovered kernel panics are pruned first.
- Agents: diagnostics.crash_reports / crash_report / crash_reports.clear
commands, the MCP crash_reports tool, solvecraft-cli crashes, solvecraft
--crash-log; diagnostics.crash_test (with SOLVECRAFT_CRASH_TEST) checks it.
- People: the Crash Reports window opens after a crash (Help > Crash
Reports…) with a reviewed issue draft per report; nothing is sent.
- Release builds keep line tables for file:line in backtraces.
Tests: child processes that really panic in a command, in the kernel, on a
thread, and abort; dedup counts, limits, hostile folders, journal parsing,
the UI window and draft, the MCP tool. docs/crash-reports.md.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
kernel::sew builds an arc edge, and the wall an extruded arc sweeps, with truck's
three-point arc (circumcentre, then a turn about it). On a very flat arc that
construction ends off its own vertex: a 12 mm chord with a 0.08 um sagitta (a 225 m
radius) ends about 1e-6 mm short of it, and the planar face holding the arc, 1238 mm
from the origin, fails with "cannot attach a plane to a wire that is not on one plane".
truck's plane fit takes the sum of the wire's control points as its centre, so a gap
of that size is scaled by a distance of the order of the coordinates and tilts the
estimated normal beyond the fit's 1e-6 flatness tolerance.
Arcs whose sagitta is under 1% of the half chord are now built as their exact rational
quadratic (apex 2/(1-t^2) along the bow, weight (1-t^2)/(1+t^2), t the sagitta over the
half chord), for edges and for the extruded walls on them. Deeper arcs keep truck's
construction.
Co-authored-by: NOWN-DEV <257771410+NOWN-DEV@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* feat(kernel): typed TooLarge refusal and fillet_max search
Blend refusals caused only by the size (larger than the neighbouring faces,
taller than the walls, longer than the next edge, larger than a corner
radius) now return KernelError::TooLarge, displayed as before. fillet_max
finds the largest radius below a request at which the fillet of the edges
works, to three significant digits, by building candidates; it is an explicit
call (up to 16 fillets), not run on every refusal.
A corner edge that doesn't run square to the edge is now reported as such
instead of as a blend larger than the neighbouring faces.
Refs #293
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
* feat: fillet fit option reduces a radius too large for its edges
solid.fillet takes fit: exact (the default, as before: a radius too large
for the geometry around the edges is refused) or reduce: the radius is
reduced to the largest that works for all the picked edges (kernel
fillet_max, three significant digits, every candidate built). The feature
stores the requested radius, so it refits on every evaluation; when it
reduces, the feature's warning says so and the command returns
radius_applied. Pattern copies of the fillet use the same rule. The file
format writes fit only when it is reduce.
The Fillet dialog gets a Radius Fit row (Exact / Reduce to Fit, constant
radius) and, when the preview reduced the radius, an Applied Radius line;
editing the feature keeps the choice.
One radius for all edges for now; per-edge radii and taper follow.
Refs #293
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Two-finger scroll pans the model instead of zooming; pinch-to-zoom
zooms toward the cursor and honors reverse zoom. Trackpad scrolls
arrive bracketed (Start..End, momentum included) while mouse wheels
send bare moves, so phases tell them apart: bracketed gestures pan,
bare vertical input zooms and bare horizontal input pans (thumb wheel).
Mouse line/page input keeps zooming, and ui.scroll now sends line
input so scripts and agents are unaffected. New Navigation preference
'Two-finger scroll pans' (default on) restores the old behaviour
when off.
One extrude cutting many circles from one sketch still cut the plate once per
circle, each boolean rebuilding and re-meshing a plate grown by every hole before
it (32 circles 2 s, 128 circles 60 s). The #261 fix joined clear tools with
kernel::union_apart only in pattern replay; apply_traced now does it for every
cut with 8 or more tools (patterns included), keeping each piece's face lineage.
* Release under the SolidCraft name: binaries, assets and installers
The desktop binary is now `solidcraft` and the CLI `solidcraft-cli` (the two app packages
are renamed; their folders move with the rest of the rename later). Every release asset is
named solidcraft-<ver>-… / SolidCraft-<ver>-windows-x64.exe, and the installers show the
SolidCraft product name.
Kept so existing installs keep updating (docs/compatibility.md):
- `solvecraft-cli` alias binary in every package, forwarding to solidcraft-cli with a
deprecation note on stderr (tests/legacy_alias.rs).
- Linux app ID ai.storyteller.solvecraft (Flatpak ID, desktop entry, Wayland app_id).
- MSI UpgradeCode, so the MSI upgrades in place.
- deb Replaces/Conflicts/Provides and rpm Obsoletes/Provides solvecraft.
- An old-named .zsync copy so earlier AppImages find updates; a legacy .flatpakrepo copy.
- .solvecraft file associations; SOLVECRAFT_* fallbacks in the packaging scripts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
* MSI: register the legacy .solvecraft extension without a duplicate open verb
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Tangent hole contacts: every symmetric contact gives the same faces (#307)
Three things made the four symmetric contacts of reference case 42 differ:
- The 2D prism boolean's ray-parity test miscounted a ray grazing the bottom
of a patterned hole, whose two halves are read back with centres a few ulps
apart: one half counted the touch, the other missed it. The loop chaining
then failed and those cuts fell back to the general boolean, which treats
circles tangent within its tolerance as overlapping and splits the hole.
The ray is now moved off vertices and arc extremes by far less than any
modelling tolerance.
- The prism result's volume check used a coarser mesh than the operands, so
a plate with many holes read too large and correct results were rejected.
- The merged topology did not join a hole's two pieces across the line where
two tangent holes touch (four faces meet there).
Case 42 now gives 39 faces (33 holes, 6 planes); its expected.json pins it.
Fixes#307
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
* Kernel test: tangent hole pinch no longer splits the wall (#307)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A rectangular or circular pattern of a fillet or chamfer repeated it at the
points picked when the feature was made. The feature itself finds its edges
by name on every evaluation, but a parameter that moves the edges (a taller
pocket) leaves the old points on other geometry, and the copies then
blended the wrong edges or failed ("the faces at the edge ends must be
planar"). The model state now records, per fillet or chamfer, the edges it
blended in this evaluation (part of the state's digest), and the pattern
repeats those.
Fixes#219
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A straight-edge fillet required each end of the edge to meet exactly one more
face, square to the edge. Now an end may meet one planar face at any angle
(the round ends in it along an exact ellipse, the cross-section slid along
the edge onto the plane) or two planar faces meeting at the end vertex: the
round ends across both and cuts back the edge between them where its
cylinder crosses it (Vida, Martin and Varady 1994: trimlines joined across
existing faces). The cylinder is swept to cover both trimmed ends. Square
ends are built as before.
This rounds the mitres where two chamfers meet at a plate's corner, and a
concave edge where a post stands on a plate (the round's end caps extend
the post's end faces); both were refused.
Fixes#215
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Sketch Dimension refused a dimension the sketch couldn't take, with
"that dimension would over-constrain the sketch; add it as a driven
(reference) dimension with driven: true", which the tool gave no way to
do. It now places it as a driven dimension instead: while placing, the
preview shows it in parentheses in the driven colour, and the click adds
it at once (it has no parameter and no value to edit), with a status line
saying why. Other refusals, such as an unsupported pick, are reported as
before, since the driven dimension is refused too.
Refs #320
Co-authored-by: Vitali M. <6042958+vitalibr@users.noreply.github.com>
A panel drop-down taller than the window ran past its bottom edge, so the
last rows (Sketch → Create → Intersect and below) could not be reached. The
rows now scroll inside the window, and a row reached with the arrow keys
scrolls into view. Row ids no longer depend on the enclosing Ui, so the
keyboard focus still finds them inside the scroll area.
Since 4e595de a partial-depth cut that ends inside a cylinder runs through the
exact stacked-section path, which leaves 0.3 mm wide strips of the cylinder wall
beside the cut. Their tessellation is one triangle wide, too few distinct
samples for the cylinder fit, so the merged topology reported them as a sphere
and a freeform face. Where the mesh fit finds no plane or cylinder, use the
face's exact surface (plane, cylinder or sphere) instead.
Fixes#308
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Projecting a cylinder's wall or the face of a round along a circular edge
failed with "face projection has an open or invalid boundary": these
surfaces are stored in halves joined by hidden seams, and the projection
dropped the seams but kept only the picked half. It now projects the whole
patch, the face as drawn, through the same surface_patch the viewport uses
to highlight it.
The guard that refuses machining through bends sewed every bend zone and
nonparallel panel into one solid. Two bends meeting at an open corner relief
share only an edge, so that sew failed with "nonmanifold edge contact" and any
cut on the base was refused. Sew and test each piece on its own instead.
Reference case 75-sm-bracket-holes now passes; its known_failure mark is removed.
Fixes#296
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Between two points, or on a line, that don't line up with an axis, the
Sketch Dimension tool only offered the distance between them: an auto
dimension ignored where its text went. Now the placement picks what it
measures: across from the two (between the perpendiculars through them)
the distance between them, as before; past that, above or below them the
horizontal distance, beside them the vertical one. With no text, an
explicit type, or two ends already on an axis, nothing changes.
The tool previews the dimension from the cursor, so the choice follows
the pointer while placing it.
Fixes#257
Co-authored-by: Vitali M. <6042958+vitalibr@users.noreply.github.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A sheet-metal tray built at 2000 x 2500 mm with an edge flange on all four
edges (72° or 90°) and unfolded failed to rebuild when W and H went to
400 x 300, 600 x 450, or W alone to 400: "planar region arrangement failed
or exceeded its bounded processing budget". The same tray drawn at 400 x 300
unfolded, and growing it worked.
The sketch solved again after the change leaves its corners ~1e-13 mm off,
so the flanges inherit a hair of tilt and their panels and bend zones meet
the trimmed base within 1e-13 mm instead of exactly. The flat pattern is a
balanced union of those pieces, and the arrangement chains its output loops
within 10 tol. In a later round, ray parity classified a bend zone's edge
midpoint that lay level with a corner whose two lines did not quite meet:
the ray slipped through the gap without a crossing, the zone's edges counted
as inside the other operand and were dropped, and the outline no longer
closed. A rectangle turned by 1e-14 rad failed the same way.
inside_loop now closes the gap where one line ends and the next begins with
a straight connector while it counts crossings; exactly closed joints are
unaffected. Joints with an arc keep the arc's own endpoint rule, which
treats ends within its angular tolerance as on the ray (a connector there
double counted at a circle split into two arcs).
Co-authored-by: NOWN-DEV <257771410+NOWN-DEV@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The space_mouse module was Windows-only (hidapi). On Linux it now finds the
multi-axis HID interface through /sys/class/hidraw and reads /dev/hidrawN with
plain std I/O, so there is no new dependency and no libudev needed.
A small report descriptor parser tells devices apart: a top-level Multi-Axis
Controller collection marks the 3D mouse, and report 1 declaring all six axes
selects the combined layout (X Y Z Rx Ry Rz in one report) that some homemade
devices use. Real split devices (report 1 translation, report 2 rotation) keep
their existing decoding. Windows reads the descriptor through hidapi and falls
back to the split layout if it is unavailable.
Co-authored-by: jochenhammes <15869208+jochenhammes@users.noreply.github.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
fill_triangles bounded the slab triangulator by edges x slabs, as if every
edge crossed every slab. A ring sampled finely at a close zoom has hundreds
of edges and slabs but only four edges in each slab, so the estimate ran a
hundred times over the hover highlight's budget and the fill was dropped,
leaving only the outline. Count the slabs each edge actually spans instead.
Region2::triangulate tested every edge against every slab, so its work was
the overestimate. It now sweeps upward with the edges crossing the current
slab, giving the same triangles in time proportional to them.
Fixes#164
Co-authored-by: Vitali M. <6042958+vitalibr@users.noreply.github.com>
Choosing 8x anti-aliasing in Preferences › Graphics on a device that
only allows 1x and 4x for the window's format made every later start
panic in wgpu's pipeline validation (`Sample count 8 is not supported
by format Bgra8Unorm on this device`), so the app could not be opened
again without editing graphics.json by hand.
- egui-wgpu (vendored): request TEXTURE_ADAPTER_SPECIFIC_FORMAT_FEATURES
when the adapter has it (WebGPU only guarantees 1x and 4x), and clamp
the sample count to the largest one the device supports for both the
colour target and the depth/stencil format. The supported counts are
exposed as `RenderState::msaa_supported`.
- At start, a saved count the device can't do goes back to the default
(4x, or the largest supported count below it), is written back, and a
"Settings were changed" notice says what changed.
- The Anti-aliasing list greys out unsupported counts, `ui.prefs`
rejects them, and an invalid saved `msaa` loads as the default.
- New messages translated in en, de, ja and ko.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Add B-spline basis, knot insertion and NURBS surfaces to geom
Freeform faces need local refinement and gap-free joins between patches
with different knot vectors, which tensor B-spline grids cannot give;
T-splines can, and the first user is an N-sided Boundary Fill of
non-planar loops. This lays the univariate groundwork they rest on, in
the L0 geometry crate so it stays independent of the B-rep kernel:
- Cox-de Boor basis of one function from its local knot vector, on
half-open spans with a left-limit side for the domain end, with
derivatives (NURBS Book eq. 2.5, 2.9)
- knot insertion into a single basis function (Sederberg et al. 2004
section 3) and Bezier extraction of one function by repeated insertion
- Boehm knot insertion for curves (NURBS Book A5.1), de Casteljau
evaluation and sub-segments
- a rational tensor NURBS surface with evaluation, knot insertion and
clamping, and a rational Bezier patch type
All input is checked (finite, monotone knots, positive weights, size
caps) and returns SplineError; nothing indexes unchecked.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
* Add T-spline surfaces: T-mesh, evaluation, AS check, refinement, extraction
Tensor B-spline faces refine only by whole rows and columns and cannot
join patches with different knot vectors without gaps. T-splines allow
T-junctions in the control net, which fixes both; the first user is an
N-sided Boundary Fill. This adds the maths only (no command, no change
in behaviour):
- T-mesh in index space with knot values per line, validated (outer
rectangle, rectangular faces, no empty lines, size caps); local knot
vectors read off rays from each vertex, rays leaving the mesh repeat
the boundary (Sederberg et al. 2003 section 3)
- rational T-spline evaluation and first derivatives for odd bi-degree,
with a complete (p-1)/2-line frame and the domain inside it
- T-junction extensions and the analysis-suitability check (Scott et al.
2012 section 2)
- local refinement that keeps the surface exactly: requested knot lines
are cut in, then every original blending function is carried as a
record and split (violation 1) or the mesh gains edges (violations 2
and 3) until each record is a blending function of the refined mesh
(Sederberg et al. 2004 section 3); analysis-suitable input stays
analysis-suitable by carrying conflicting T-junctions across faces
- NURBS <-> T-spline conversion for meshes without T-junctions, and
Bezier extraction over the rectangles no blending-function knot line
crosses
Measured: regular T-spline vs NURBS < 1e-12, round trip exact, partition
of unity with T-junctions < 1e-14, refinement deviation <= 1.4e-14 over
40 random refinements, Bezier patches vs surface < 1e-14.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
* Convert T-splines to sewn NURBS sheet bodies in the kernel
A T-spline is only useful to the modeller once it is a B-rep face set.
tspline_sheet builds an open shell: a mesh without T-junctions becomes
one clamped NURBS face; otherwise each Bezier element is a face, and
sides are split at the T-points where neighbours' corners land, so all
faces share vertices and edges and the sheet is one oriented shell
(V - E + F = 1). Every truck call runs under guard. No command yet;
Boundary Fill will be the first caller.
Tests: one face for a regular surface, sewn Bezier faces with
T-junctions, tessellated points on the surface to ~1e-9 mm, and hostile
control nets (coincident, 1e300, needle-thin) never panic.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Part 1 of splitting #251: the sketch crate only. A graph-constructive front
end (Owen 1991; Fudos & Hoffmann 1997) places what it can by ruler-and-compass
loci and cluster decomposition, Levenberg-Marquardt finishes and verifies the
rest, and the rank analysis stays authoritative for degrees of freedom. Solves
can follow a reference (the previous solution), so edits and drags keep their
branch instead of flipping.
New API: solve_with (hold, reference, analysis, budget) and
solve_after_value_edit; solve and solve_holding keep their signatures as
wrappers, so no caller changes here.
When the seed moves an arc's centre, the arc's unconstructed ends start carried
along (the arc translates) instead of left behind for LM to turn; found by the
reference test 20-slot-constrained.
Refs #203
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* feat(sketch): constructive solving ahead of the numeric solver
Part 1 of splitting #251: the sketch crate only. A graph-constructive front
end (Owen 1991; Fudos & Hoffmann 1997) places what it can by ruler-and-compass
loci and cluster decomposition, Levenberg-Marquardt finishes and verifies the
rest, and the rank analysis stays authoritative for degrees of freedom. Solves
can follow a reference (the previous solution), so edits and drags keep their
branch instead of flipping.
New API: solve_with (hold, reference, analysis, budget) and
solve_after_value_edit; solve and solve_holding keep their signatures as
wrappers, so no caller changes here.
When the seed moves an arc's centre, the arc's unconstructed ends start carried
along (the arc translates) instead of left behind for LM to turn; found by the
reference test 20-slot-constrained.
Refs #203
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
* feat(doc): keep a sketch's solved state across value edits
Part 2 of splitting #251 (builds on part 1, the sketch crate):
- Sketch edits solve for the document's values continuing from the stored
solved state (Document::solve_sketch), so a dimension edit keeps the
sketch's branch instead of re-solving from scratch.
- A value edit (parameters, a configuration, the units) that would make a
sketch that solved before fail or turn over is refused with the
conflicting constraints named; otherwise the solved sketches are written
back into the document (Model::adopt_solved_sketches), which changes what
saved designs store.
- Later features are keyed by a sketch's solved geometry (to a 1e-6 mm
quantum), so setting a value back reuses their earlier results.
- Solved sketches in a model state are shared (Arc) between states.
Refs #203
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
* feat(sketch): live drag that keeps its solution, ended by sketch.drag_end
Part 3 of splitting #251 (builds on part 2):
- sketch.drag follows a long move in short held steps, so the geometry keeps
its branch, and reports how far it got (reached, at).
- live: true runs one frame of an interactive drag: the sketch is solved
once and shown, later features keep their last result (marked stale)
until sketch.drag_end or any command that reads the model evaluates it
(CommandSpec::live, reads_model, doc_only).
- sketch.drag_end adds the constraint the drag let go on and evaluates the
model once.
- Undo and redo evaluate a new document once (execute does) instead of
twice.
- The viewport drives drags through live frames and drag_end; dimension
text drags start only off sketch geometry.
Refs #203
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Loop blends accept straight wall seams that are not lines
Filleting or chamfering the top outline of two overlapping cylinders whose
concave edges were rounded failed with "the wall seams must be straight":
the round's contact lines with the cylinders are straight NURBS curves, and
the loop blend only accepted seams stored as lines. A seam now counts as
straight when every sample lies on its chord; it is trimmed as a line as
before.
* Fillet and chamfer arrows point into the body
The radius arrow pointed out of the body along the edge's bisector, so
dragging it away from the edge grew the round. It now points into the
material, like the shell's thickness arrow: dragging toward the body grows
the radius or distance, dragging out of it shrinks it.
* Toolbar: no cap on pinned buttons (Fixes#323)
Every pinned command gets a toolbar button; `promoted()` is only the
default layout of a fresh or reset toolbar. A toolbar too narrow for every
button now collapses panels into their drop-down (one button that opens it,
listing every command, pinned ones included) from the right, then moves the
last panels into the » menu, instead of pushing a wide panel and everything
after it into ». Nothing is dropped.
Scenario s48 pins ten commands into SOLID > CREATE: all ten show at a wide
window, a narrow one collapses CREATE into its drop-down (all ten listed),
narrower still puts it in », and Reset Toolbar restores the defaults.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
* Narrow-toolbar test: panels collapse into their drop-down before »
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
* kernel: drop a needless borrow that fails clippy -D warnings on the current stable
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* test: reproduce #260 (Targeted cut deletes another body that the tool fully covers)
* Primitives honour `targets` for join/cut/intersect (#260)
solid.box/cylinder/sphere/torus had no targets field, so a `targets` list was
ignored and every body of the component took part: a targeted cut also cut a
non-target body, and deleted one it covered fully. Their `targets` now become
the feature's participants.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(kernel): name the zero-thickness rim when a shell can't open beside a round
Shelling a body through a face that a convex round meets tangentially, with
the round at least as large as the thickness, puts the cavity wall through
the rim where the round meets the open face. Both shell strategies then fail
and the first one's generic error ("a corner whose faces don't meet the same
way after the move") was reported. The shell now says why and what to change.
Successful shells are unchanged.
Found by fuzz seed 33, which reaches this case now that a disc's two edges can
be rounded to half its thickness.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
* fix(kernel): loop fillet as tall as its wall uses the wall up
Rounding a loop with a radius equal to the height of its walls was refused
("the blend is taller than the walls"): r = 2 on the top edge of a 2 mm
disc failed while 1.999 worked. When every wall seam ends exactly where the
wall-side rail would, the walls are consumed: their far edges become the
blend's rails and the walls and seams are removed. Walls used up only in
part, and radii past the limit, are still refused.
Refs #190
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* test: reproduce #259 (Join with explicit `targets` also merges a non-target body that the tool touches)
* Primitives honour `targets` for join/cut/intersect (#259)
solid.box/cylinder/sphere/torus had no targets field, so a `targets` list was
ignored and every body of the component took part: a join also merged a touched
non-target body, a cut also cut (or deleted) one. Their `targets` now become the
feature's participants.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Pieces a cut splits a body into are named after their faces since names
follow history (lineage::order_lumps), so they keep their body names when
an upstream edit moves the cut. The kernel's shell order named them
before, and the two orders can differ: a slot along X through a 40x30x10
block swaps Body1 and Body1 (1). A design saved by an older build stores
body names meaning the old order (fillet and chamfer bodies, targets,
participants, materials, appearances, joints), so it would act on the
other piece.
Each feature now says which order names its pieces
(Feature::shell_piece_order). The v13 to v14 step sets it on every saved
feature, so an older design evaluates as it was saved; features made from
now on, in new or old designs, name their pieces after their faces.
The corpus gains split-body-pieces, saved by v0.3.2: a fillet, a chamfer
and an appearance refer to the two pieces by body name. Without the
setting the two pieces swap names (the corpus test fails); with it they
stay as saved. Migration tests that compared an upgraded design with the
original now expect the setting on its features.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
The v13 to v14 step walked the whole design and marked any JSON key named
face, face_name or face_names, so any future field with one of those names
would be marked too. It now marks the fields that store a face or edge name
where the schema puts them: each feature's face_names and fillets' and
chamfers' edge_names, a sketch feature's links, joint snaps on a face in
joints and named origins, and face appearances. Face and edge references
(PlaneRef::Face, GeoRef::Face/Edge) can sit in many features, so they are
still found anywhere in a feature, by their tag and their picked point.
The test fixtures now have the shape real files have, and a new test checks
that same-named keys elsewhere are left alone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
The world-placement tests checked that the source body's mesh was built
before a placement adopted its names. Once names follow history they no
longer need the mesh, so that is a side effect's timing, not behaviour.
What the tests protect, a placement sharing its source's names, both
still assert through the public API: Arc::ptr_eq(face_names(placed),
face_names(source)).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
Why: every reference a user makes to a face or edge (fillet and chamfer
edges, sketches on faces, shell/draft/offset faces, joints, appearances)
is stored as a name and found again after any upstream edit. Those names
were re-derived by sampling points of each face and voting, and pieces
of a split face were numbered by where their centres lay. After an
upstream edit that splits, merges, mirrors or shrinks a face, a fillet
or sketch could silently re-attach to the wrong face, and nothing told
the user. The kernel now reports exact history for its operations, so
names follow it and fail loudly instead of guessing.
- crates/doc/src/lineage.rs: a feature's kernel histories composed into
one lineage over nameable operands (bodies before the feature, moved
copies, shell cavities, tools named by role: sketch curve, cap, box
side, imported face). Instrumented: extrude/revolve/sweep/loft tools,
join/cut/intersect, lumps after a cut, fillet/chamfer, shell, draft,
split face/body, pattern/mirror (bodies and features), combine,
boundary fill, move/scale, offset/delete face, thread, import.
- crates/doc/src/history_names.rs: names from lineage; split pieces told
apart by the faces only they touch (F2:top#[F2:side:l4,F4:side:l4]),
merged faces answer to every name (aliases), lump bodies ordered by
their faces' names instead of shell order.
- crates/doc/src/name_lookup.rs, piece_names.rs: references resolve by
exact name, alias or bounding faces, else nearest piece with a
warning; names worked out from geometry (operations without history)
are reported when only geometry decided.
- Format 14: names stored by older formats get a legacy mark and are
found as the build that saved them found them (some geometric names
name a face history names differently: a tube's outer wall was named
after its inner circle). New corpus file saved by 9a979cf.
- kernel: Body::lumps carries each piece's history over the whole.
- Adversarial tests: split, merge, mirrored twins, shrink to a sliver,
moving cut with body pieces, history-less bodies, certified sketch.
- docs/book/src/dev/naming.md describes the real mechanism and why.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
union_apart joins bodies whose boxes are clearly apart (many pattern tools,
#261) without any intersection work: their shells side by side. Each face
and edge of the result is its piece's own, unchanged, and the result now
says so (a history over the pieces), so a pattern's cut keeps the exact
names of every copy's faces.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
A cut of stepped prisms sews a wall over one segment through successive
slabs as one face, and the generated planar walls are then joined on
their common supports. Until now such a piece carried no history.
- prism_lineage: a sewn wall records what it is over in each slab it
spans (Spec::Wall { levels, slabs }), so a coalesced wall comes from
the operand walls under it in every slab
- the sewn piece's history is attached before the joining and carried
through it (Built::Traced); planar_regularize composes the history of
the body it joins with the join itself
- a composed face is Split only where record::mark_shared_split finds an
origin under several faces; an inner Split alone no longer makes one
- Body::same ignores history: an operation that leaves a body as it is
gives the same body, carrying that operation's history
The two boolean history tests that waited for this run again, and the
coalesced and regularized step cuts are covered.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
The far edge of a stepped slab (the operand cap edge where a segment's
wall ends) was recorded by testing whether each slab segment runs along
an ending cap's segment in 2D, a tolerance comparison of coordinates.
Matching by topology makes the far edge exact by construction instead:
when the wall's top is the ending level, nothing crosses it there, and it
meets the caps at that level in exactly one edge, that edge is over every
segment beside the wall, with no position compared.
A vertex walk up the wall's side edges from the segment's ends was tried
first and does not settle the cases that need it: a notched or stepped
wall continues past a segment's end (no side edge rises from it), and
walls split by earlier stations have mid-chain vertices with no side edge.
Whenever the walk succeeds, the wall is a rectangle over the segment and
the one-edge rule above already holds.
The positional match remains only where topology cannot say which edge
is over a segment: a wall that meets the ending caps in several edges (a
top split at earlier station vertices, or beside a wall crossing it at
that level) or that carries on past the level elsewhere. It is documented
in `Section::mark_far` and its pair budget no longer stops topological
marking. Far edges are also only marked on the slab that actually ends at
the level (a gap level no longer revisits an earlier slab).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
Chamfer is one of the most common features and is usually applied right
after a cut. A chamfer cuts with a prism cutter, and its history came out
None after four of the six common box cuts. Without exact history, any
reference to a chamfer face (or a face it trimmed) falls back to geometric
guessing in naming, which can silently re-attach to the wrong face.
Two gaps in the stepped-prism section tags caused it, both now closed by
bookkeeping at construction, never by sampling or geometric tie-breaks:
- Cap edge at the far end of a slab. A slab's section is read off the cap
where it starts and carried up to the cap where it ends; that cap's
outline can run along one wall in several collinear edges, so which one
a result edge lies on was unknown. When the stack is read, each slab
segment that runs end to end along one segment of an ending cap's
outline (beside the same wall) records that cap edge as `far`; the
lineage takes it when it bounds the wall and the result cap.
- Walls an earlier cut left in coplanar pieces. A segment carried through
a level kept the wall it was read beside, even where that wall gives
way to another piece at the level (or within the slab), so the tag named
the wrong face. `Stations` records the operand's edges across walls;
at each level a carried segment keeps its wall when nothing crosses it
there, moves to the one wall across it when it ends there, and is
otherwise unknown (no history). A wall crossed inside a slab is unknown
too, for simple and stepped prisms alike.
chamfer_after_box_cuts now covers all six cuts (notch, pocket, step,
slot, hole, bite), each verified against geometry and composed through
the cut over [block, tool]. The layered partial-depth cut test now
verifies its second cut, whose history was None before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
EdgeOrigin::Meet now names each side as a Side: the sorted, de-duplicated
origins of the face there. substitute maps a side to the union of the inner
faces' origins, so a meeting with a face an inner operation merged keeps its
lineage; meeting() derives sides from whole faces. Fillets after box cuts
that leave coplanar pieces now report exact history. Prism lineage meets a
cap of several origins and resolves wall junctions away from shared
vertices by the operand's one edge between the walls. bore_union's recut
reports history only for a one-face bore (pieces of a two-arc bore's recut
wall were given both arcs).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
Fillets (local, loop, rolling ball, corner, variable, chord), chamfers
(cutters, local corners, loops, two sides), shell, draft, offset faces,
split face/body, delete face, patch/copy/stitch/trim/extend and
transforms record their history over [body] while they build: faces
rebuilt from the body's, blends in place of edges (corner patches in
place of the edges at the corner), shell walls as offsets, edges as
pieces of the body's or where two faces meet. Multi-step operations
compose each step. The push-face offset fallback now has history, and
the bore union composes its narrowed tool.
A reflection reorders a body's edges, so transforms now match edges by
identity (mirrored copies reported the wrong edges).
Refs #136
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
ops::boolean's result carries its history over [a, b] whichever path made
it, recorded by the code that builds the faces:
- truck path (incl. jitter retries): shapeops history mapped through the
order-preserving copies and the cut's turned tool, then through healing
- heal records which input faces each healed face merges and which input
edges each joined or exact edge replaces (HealMap); opt-in, so heal_keep
callers pay nothing
- trivial shortcuts, the void, glue, faces pushed apart (push_faces records
its own history; the slab cut back on a union maps to the operands'
faces in that plane), the nudged retry, by_lumps and join composition
- prism booleans: section segments carry tags (operand cap edge, the wall
beside it, its vertices) through the 2D boolean; walls, caps and edges of
the extruded or sewn result follow from them
- planar and faceted polygon booleans: polygons carry their operand face;
faces join the pieces across edges, edges come from the pieces either side
History gains renumber/over (and consuming forms) for composition;
record gains regions/traced/build helpers. truck-shapeops
*_with_history now return the solid with an Option history. Paths that
cannot report (bore narrowing via offset, inexpressible lineage) leave the
result without history. Boolean results, tolerances and fallback order are
unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
One numbered loop type (crate::numbered) carries each segment's profile
numbers through orientation, snapping, circle re-seaming, re-sampling and
rotation; every builder orients its regions through it once. Builders that
make their own side faces pair each face with its provenance as it is made
(ruled walls by the wire edges each face holds, smooth lofts, coils and path
sweeps by segment), and cap_and_close appends the caps itself. Column
following is kept only for faces truck sweeps internally (rsweep, tsweep
along paths, cones, where an apex vertex stays put). Extrude derives face
sources and history in one topology pass. One attach entry point takes
operands with their edges already indexed.
Geometry, face order, face sources and recorded origins are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
Extrude, revolve (partial and full turn), tapered extrude, loft (incl. to a
point), sweep, path sweep, coil, primitives, lip, emboss and thicken attach
the exact history of the bodies they build: each face's profile segment, cap
or operand face, and each edge's operand edge, recorded by topology identity
inside the building call. Nothing is sampled; a body whose faces do not follow
the expected construction carries no history. Face sources are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
and_with_history / or_with_history return, with the solid, the operand face
each result face is a region of (and whether it is that face untouched) and,
for each result edge, the operand edge it is (a piece of) or the pair of
operand faces whose intersection made it. Faces and edges are recorded where
they are created (face division, intersection edges, split edges, edges
rebuilt on a moved vertex), resolved to operand terms at creation, so the
passes folding in extra shells compose without bookkeeping. and / or wrap
the same pass and drop the history; a missing record never fails them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
Bodies can now carry the exact history of the operation that made them:
for each result face its origins among the operation's operands (kept,
modified, split, merged, or generated from a profile segment, cap, edge or
offset) and for each edge the faces it separates and the operand edge or
face pair it came from. History is recorded by the operation, never
sampled, and checked against the body when attached.
- history.rs: the vocabulary (moved from provenance.rs), History::check,
identity, substitute (composes an operation with the one that made an
operand, including bodies made from a sketch) and face-source propagation.
- One edge order for a body (body::edge_order), shared by tessellation and
history, so history edge indices are the mesh's edge indices.
- Transforms report an identity history; placed_topology documents the
face/edge order guarantee it relies on.
- provenance.rs keeps the geometric trace for operations that do not yet
report history.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
A constant-radius fillet whose setback equals the width of a neighbouring
face was refused ("the blend is larger than the neighbouring faces"), so
r = 3 on the vertical edges of a 6 mm wide box failed while 2.999 worked.
The corner edge is now consumed: the rail ends at its far vertex and a side
face used up at both ends is removed, its remaining edge becoming the rail.
A face used up at one end only, and radii past the limit, are still refused.
Refs #199
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Shelling a body through a face that a convex round meets tangentially, with
the round at least as large as the thickness, puts the cavity wall through
the rim where the round meets the open face. Both shell strategies then fail
and the first one's generic error ("a corner whose faces don't meet the same
way after the move") was reported. The shell now says why and what to change.
Successful shells are unchanged.
Found by fuzz seed 33, which reaches this case now that a disc's two edges can
be rounded to half its thickness.
Claude-Session: https://claude.ai/code/session_01TWc1smHYwTSVrVc8SNvS44
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The Home page hides the open design, yet Save, Undo, Redo, File › Insert,
Save As and Export, and the design's keyboard shortcuts (Ctrl+S/Z/Y,
Delete, V, F2, F6, Esc, S and tool keys such as E) still acted on it:
Delete removed the selected body unseen, Save wrote a design the user
could not see, and tool keys left Home without being asked to.
On Home the top bar's Save, Undo and Redo are dimmed with a translated
"open a design first" tooltip, the File menu's design items are greyed
out, and only New, Open and Preferences keep their shortcuts. Everything
works as before once Home is closed.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* test: reproduce #267 (timeline.delete of a sketch leaves later features that use its body by name)
* fix: timeline.delete reports features that newly fail (#267)
A later feature that names a deleted body (a combine targeting a deleted extrude's body) is
listed under failing: [{feature, error}] in the reply instead of failing silently.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A line whose end lies behind the eye has no screen point: vs_line divided
by a non-positive w, which flipped the segment or dropped it whole. At a
low perspective angle the grid lines running along the view direction
reach behind the camera and vanished. Cut each segment at clip z = 0
before the screen-space expansion.
Fixes#182
Co-authored-by: Vitali M. <6042958+vitalibr@users.noreply.github.com>
The Home page returned from the frame before the Preferences window was
drawn, so the gear, Ctrl+, and Home's Preferences / Keyboard Shortcuts
links only set the flag and the dialog appeared after leaving Home.
Draw the Preferences window on Home too.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Snap 3D bridge drawing to points on visible sketch planes
* Keep cross-sketch point picks out of sketch tools
`pick()` returned every visible sketch's points while 3D drawing was on,
so `sketch_point_at` and the sketch tools could take another sketch's
point as an active-sketch local point plus id (flattening it or making a
bad coincident reference). `pick()` is local again; the bridge snap uses
`pick_3d` explicitly and hover mirrors it for feedback. `sketch_point_at`
also ignores foreign points defensively.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test(sketch3d): check local point scope with a sketch tool active
Main's Project preselection fix now offers other sketches' points to idle
selection, so assert the local scope where it still applies.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* style: rustfmt
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* test: reproduce #264 (Sheet metal: collinear tabs cannot be folded or flanged together)
* fix(sheet): fold and flange tabs that sit side by side on one line (#264)
A fold's bend zone and a flange's setback were cut with an infinite line, so
a second tab on the same line blocked a fold ('slanted or curved edge') and
two tab-end flanges in one feature shared one trimmed edge (nonmanifold).
When the zone falls apart into pieces, fold only the pieces the fold points
span, each as its own bend; trim a flange's setback only from the material
along its own edge, and attach each flange to the edge it was picked on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
- Remove xtask/data/fusion-catalog.tsv and xtask/data/fusion-ids.tsv.
- Remove the legacy command-id table (crates/engine/src/legacy_ids.rs) and its use in
find_command and in the shortcut, recents and pinned-command loaders. Only SolidCraft's
dotted ids name commands; any other id is an unknown command.
- Boundary Fill's id becomes solid.boundary_fill.
- Replace the parity report with SolidCraft's own feature checklist: `cargo xtask features`
writes docs/features.md from the command registry.
- docs/oracle.md, docs/step-import.md (corpus section) and docs/iges-import.md: reference
results are being rebuilt on analytic and open-source (FreeCAD) values; see #292.
`cargo xtask oracle` and `cargo xtask step-corpus [--iges]` write these placeholders.
- AGENTS.md, CLAUDE.md, CONTRIBUTING.md, README.md, ROADMAP.md and the book describe
SolidCraft as independent, with reference tests on analytic values and open-source tools,
and agents do not drive or extract data from commercial CAD products.
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* test: reproduce #261 (Rectangular pattern of a hole cut scales super-linearly)
* Patterns of many hole cuts scale near-linearly (#261)
Two quadratic costs:
- Meshing a face tested every point against every segment of every boundary
loop, so a plate face with hundreds of holes meshed in quadratic time. The
vendored truck-meshalgo inside test now skips loops whose parameter box is
clear of the point (they wind zero times around it).
- A pattern cut each copy in turn, rebuilding and re-checking an ever more
complex target per copy. Copies whose boxes are clear of each other are now
one tool of many pieces (kernel::union_apart, no intersection work), cut once.
512 holes now take ~7 s instead of not finishing. The test's volume tolerance is
widened from 1e-6 relative to 0.1 % of each hole: the measure meshes the bores
and reads each hole ~0.06 % small, on main too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hole-pattern test: a budget CI can meet; coarser apart-box meshes (#261)
CI runs every test binary at once and is about ten times slower than a
machine running this test alone (15 tests passed 60 s in the last run), so the
512-hole pattern (about 6 s alone) hit the 60 s timeout there. The budget now
only has to tell linear from the quadratic hang: 300 s timeout, 120 s budget.
union_apart meshes each copy coarsely (2 % of its size, box grown by twice that).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hole-pattern test checks the scaling, not a wall-clock time (#261)
CI runs every test binary at once: the 512-hole pattern (about 6 s alone, 10 s
CPU single-threaded) took 177 s there, so no fixed time is reliable. The test
now patterns 128 and then 512 holes and requires the second to take under 8x
the first (linear is 4x, measured 3.5-4.2x here; the old quadratic cost was 16x
or more), each with the volume check and a 900 s hang guard.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* test: reproduce #268 (Panic from truck-topology lib.rs:186 when joining struts that end on a bore wall)
* Sew refuses edges whose ends weld to one vertex (#268)
The sewer welds vertices on a 1e-6 grid but only rejected edges shorter than
1e-9, so a sliver edge between the two welded its ends into one vertex and
truck-topology panicked ("Two same vertices cannot construct an edge") inside
the guard. Such an edge is now a degenerate-edge error before any truck call,
and the boolean takes its usual fallback.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Create ko.json
* Update ko.json
* Update i18n.rs
* Fonts: add a Hangul fallback so Korean doesn't render as boxes
Only the first CJK font found was loaded, and on Windows and macOS that is
a Japanese font without Hangul (Meiryo, Hiragino). Load the first readable
font of each group: the existing CJK list, then Malgun Gothic (Windows),
Apple SD Gothic Neo (macOS) or Nanum Gothic (Linux). Also rustfmt the
Korean catalog test.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Reject ambiguous origin and construction plane names
* Reserve origin plane names for construction planes; suggest the other alias
A construction plane can no longer be created with or renamed to XY, XZ, YZ, Top, Front or
Right, so a plane argument can't become ambiguous in new designs. For older designs that
already hold such a plane, the ambiguity error now also names the origin plane's other
alias (XY for Top...). The existing tests build those legacy names directly.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* test: reproduce #265 (Renaming a pattern copy renumbers the remaining copies)
* fix: renaming a body keeps its sibling bodies' names (#265)
body.rename pinned only the names of the feature's bodies before the renamed one, so unpinned
pattern copies after it were renumbered into the freed name. Pin every sibling's current name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* test: reproduce #263 (sheet.fold accepts a fixed point that is not on the sheet)
* fix(sheet): refuse a fold's fixed point that lies off the sheet's material (#263)
A fixed point inside a cut-out or outside the base outline (in the panel's
thickness) silently picked a side. Refuse it with an error that names the
point; side-face picks and points on other panels keep working.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The oracle no longer uses Fusion measurements. tests/oracle/<case>/ now holds the 76 recipes
(our own format, with recorded measurements and tool-specific fields removed) and an
expected.json per case:
- 55 analytic cases: closed-form values from tools/oracle-analytic/ (stdlib Python, formulas
in each case's "method").
- 21 FreeCAD cases: the same recipe built with FreeCAD's public scripting API by
tools/freecad-oracle/ (an external program, not a dependency), with a reference.stl for 20.
solvecraft-cli oracle compares body count, volume, area, centre of mass, bbox, face counts
where the geometry fixes them, and, against reference.stl, the sampled Hausdorff distance and
the principal moments (spatial.rs from #301). cargo xtask oracle reads tests/oracle, writes
docs/oracle.md with the source of each value, and supports known_failure marks for tracked
bugs. CI runs it on every PR; a manual workflow regenerates the FreeCAD values. step-corpus,
seam normalisation and the through-all emulation, which only existed to compare with Fusion,
are removed.
75/76 pass; 75-sm-bracket-holes is a known failure (#296).
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
Co-authored-by: bherbruck <bherbruck@herbrucks.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Navigation presets are now a table of bindings (button, required modifier
keys, camera mode) instead of string matches. The Alias preset binds
Shift+Alt+left to orbit, Shift+Alt+middle to pan and Shift+Alt+right to
zoom. A left press with a binding's keys is latched until release, so that
drag neither box-selects nor drags sketch geometry; plain left drags behave
as before. Shift only requests snapping when the binding without Shift
still orbits.
The scenario harness sends Alt with ui.drag/ui.click, and camera checks can
assert panned/zoomed since the last camera step. The user guide notes Alt+drag
grabs by KDE and GNOME and how to free them.
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Toolbar drop-downs highlight the row under the pointer and the keyboard focus (#288)
A panel drop-down's command rows are full-width rows that take the theme's hover background
(icon, label and shortcut on it) when hovered or focused. Up/Down move the focus through the
enabled rows, Enter starts the command, Escape closes the drop-down. Rows publish
`dropdown:<id>` and the highlighted one `dropdown_hot:<id>`; scenario s46 checks pointer and
keyboard highlight.
Fixes#288
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
* Toolbar: drag commands from drop-downs onto the toolbar; pin icon; Customize Toolbar… (#289)
- A panel drop-down row drags onto the toolbar (a button or a panel's free space, with the
drop line) and is pinned there; the dragged command's icon follows the pointer.
- A toolbar button dragged off the toolbar, or back into the open drop-down, leaves it (its
icon shows a cross while releasing would remove it).
- A hovered drop-down row shows a pin at its end: click to pin or unpin.
- Right-click on the toolbar's empty space: Customize Toolbar… (a window listing the tab's
commands with check boxes, and Reset Toolbar) and Reset Toolbar; command menus have
Customize Toolbar… too.
- Still per workspace and tab, kept with the preferences. Toolbar buttons also publish
`button:<id>`; scenario s47 drags a row to the toolbar, unpins by dragging off and into the
drop-down, uses the pin icon and Customize Toolbar…, and checks it survives a restart.
Fixes#289
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
* Toolbar drop-down pin: a push pin drawn in code (upright and filled when pinned); control protocol lists the toolbar handles
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Appease X11 users with niri/wayland specific postfixup wrapper
* Linux: stay on native Wayland under niri in the app, not the Nix wrapper
Move the niri check from the flake's wrapper into apps/solvecraft/src/main.rs
next to the XWayland default, so it applies to every niri user, not only
Nix builds (flake.nix is back to main's, which also clears the conflict
with #210). SOLVECRAFT_WAYLAND=1 still keeps native Wayland, and =0 now
forces XWayland. docs/nix.md notes the trade-off: no file drag-and-drop
on native Wayland.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Fillets along a straight edge between two cylinders: the rolling-ball blend trims curved cap edges to its contacts
The edge where two overlapping extruded arcs meet is a straight line between two cylinders
ending on planar caps. The rolling-ball blend built its sections and surface, but only
straight seams could be shortened to the contacts, so a circular cap edge was refused with
"a curved edge where the blend ends". Curved seams now keep the part of their curve between
the new ends, found from the nearest of a few samples (a search from the far end of a long
arc can settle on the opposite side of the circle).
The ball's surface projections start from each face's last answer instead of a cold search,
falling back to one when the warm answer leaves the surface's parameters or is farther than
the last point: about 4.5x faster per fillet (370 -> 80 ms), so dragging the radius is usable.
Fixes#237
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Curve blend: refuse a closed curved seam where the blend ends
A seam that starts and ends at the same vertex (a full circle) has no
"between" two contacts, so trimming it built a broken edge. Return
"not supported yet" instead. The curved-seam trim moves into
trim_curved_seam so it can be tested directly.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
* test: reproduce #262 (sheet.fold: bend zone reaching the blank outline is reported as a slanted or curved edge)
* fix(sheet): name the bend zone reaching an outline corner instead of blaming a slanted edge (#262)
A fold whose bend zone contains a corner of the panel outline (a fold at a
tab's root) was reported as crossing a slanted or curved edge. Name the
corner and how far to move the fold line to clear it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Expose interactive editing and placement of sketch text
* Respect document units when editing sketch text
* Route Edit Text through the shared feature dialog
The sketch text editor is now a Kind::Text dialog opened through
Dialog::for_command (frame.rs docking, standard OK/Cancel, Enter/Esc).
Its fields are ordinary dialog fields: they take expressions, parameters
and units, and bare lengths are in the design's units. Untouched fields
keep their exact saved values. The dialog stays open across unrelated
design changes and closes only when its text is gone, the sketch ends,
or another command starts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Honor design units in sketch dimensions and measurements
* fix: apply document units to specialized length controls
* Accept explicit units in length fields; document design-unit dimension values
Length fields in a design-unit control now evaluate what is typed as a
length expression: a bare number stays in the design's units, while an
explicit unit ("10 mm", "0.5 in", "1 in + 5 mm") is converted.
Document that a bare numeric sketch.dimension value from scripts, the
control channel or MCP is read in the design's units.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
---------
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
- The dialog draws every page into a fixed, clipped 744x400 area, so no page
(or translation) changes the window width; descriptions wrap and tall pages
scroll inside it.
- OK / Cancel / Apply sit flush right in the bottom bar (right-to-left layout
within the fixed width); Restore Defaults stays left.
- The "settings" icon is a gear: a solid ring with six square-ish teeth and a
see-through centre hole, instead of the sun-like rays.
- Scenario s46 visits every page and checks the window keeps its width and
that Apply ends at the bottom bar's right edge (new `same_width` and
`right_edge` scenario checks).
Fixes#290
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
sheet.flange now lists name?/body_name?, solid.loft and solid.sweep list
targets?/name?/body_name?, and the primitives list name?/body_name? (all go
through add_feature, which reads them). A test pins the shared keys for
every body-creating feature.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
Params docs name other commands ("usable in sketch.delete"), so filtering by
an id returned 56 commands. The filter now looks only at id, label, tab and
panel; the tool description and docs/mcp.md say so.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
- Web: a failed CJK font fetch shows English for this session only; the
saved language preference is kept (CjkFontLoader, persisted on save).
- Web fetches the CJK font only when a CJK language (ja) is selected.
- fonts.rs: directory scans match CJK-specific names only (no "noto" or
"gothic"); CJK stays the last fallback after the Latin UI font.
- Revert unrelated clippy rewrites in component_clone.rs, coplanar.rs,
canvas.rs and issue_report.rs to match main.
- Tests: default English, fallback order, CJK-only matching, fetch only
for CJK, failed fetch keeps the saved language.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RjGw4gA5ZvKrU8MQP6sBgT
- Bundle BIZ UDPGothic Regular (SIL OFL 1.1) exclusively in solvecraft-web dist
- Remove external jsDelivr CDN fallback; fetch font solely from same-origin ./fonts/
- Fetch Japanese font on-demand only when UI language is set to Japanese
- Gracefully fall back to English UI on fetch failure to prevent tofu glyphs
- Update ATTRIBUTION.md, NOTICE, and THIRD-PARTY-LICENSES.txt
certify_nesting runs two planar booleans per pair of boundary loops, but
its work budget counted segments, so a mesh with many small holes could
trigger ~1M booleans. Refuse more than 50 loops (1225 pairs) with a
"too complex to convert" error and test a 400-hole plate.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With Object Type = Bodies, the Mirror, Rectangular Pattern and Circular
Pattern dialogs sent the feature that created each picked body
({"features": ["Extrude1"]}), so the copies replayed only that feature and
lost everything done to the body afterwards (a fillet, a chamfer, a cut).
They now send the bodies themselves ({"bodies": [...]}), which the commands
already support and the Path Pattern dialog already did. Picking Features
is unchanged.
Editing a body Mirror or Pattern from the timeline now reopens the dialog
with its bodies selected.
Copies made this way take the body-copy names ("Body1 (2)"), as the
commands already give; pattern_direction_edge now looks up "Body1 (3)".
Tests: UI scenarios mirror_body_keeps_fillets and pattern_body_keeps_fillets
(filleted cube, copy and edit through the dialogs); both fail without the
fix with the copies' sharp 8000 mm3 volume.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
construct.plane.offset now takes a planar face as its base
({"face": [x,y,z], body?, index?, path?}), captured with the same exact
face attachment a sketch on that face uses. The Offset Plane dialog accepts
planar faces as well as origin and construction planes.
A construction plane on a face is resolved through its face anchor when the
timeline is evaluated, so it follows the face when upstream features change
(and after reopening), instead of keeping the plane as it was when picked.
Tests: engine (top and side faces, follows a box resize, survives reopen,
hostile picks are refused without adding a feature) and a UI scenario that
picks a face in the dialog and builds a pad on the new plane.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Add aarch64-darwin to the flake systems so packages, apps and both
devShells are exposed on Apple Silicon Macs. The Wayland/X11/EGL
runtime libraries, the /run/opengl-driver search paths, the freedesktop
packaging and the wrapProgram post-fixup are Linux-only and stay that
way: on macOS winit talks to AppKit, wgpu renders through Metal, and
the GPU drivers ship with the OS, so the darwin package needs no
wrapper. x86_64-darwin is deliberately not added: current nixpkgs
unstable no longer evaluates rustc for that platform (26.11
deprecation), so the package would fail on toolchain evaluation.
docs/nix.md: platform list updated, macOS paragraph in the graphics
section, desktop-entry paragraph qualified as Linux-only.
A horizontal or vertical glyph sat beside its line on a fixed side, so a
dimension placed on that side drew its value box on top of the glyph.
A glyph that would overlap a dimension's value now goes to the other side
of its line, or steps along it when both sides are taken.
The scenario `glyphs` check also fails when a glyph overlaps a dimension
value; constraint_glyphs_clear_of_dimensions covers it.
Fixes#175
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
sketch.construction set `construction` but left `centerline` on, so the
curve stayed a centerline (drawn dash-dot, still marked as an axis) until
it was first switched back to normal. Making a curve construction now
clears its centerline flag, mirroring sketch.centerline.
Nothing on the load path checked a sketch's size, and every sketch in the
timeline is solved when a design opens. The solver sizes its dense matrices by
the connected component, so a 30 MB file with 200,000 chained points asked the
allocator for terabytes and the process aborted, which no panic guard can
intercept.
Reject an oversized sketch while the design is still raw JSON, so the typed
arrays are never built, and refuse a connected component with more variables
than fits in a bounded matrix: that component now reports a failed solve and
leaves its geometry alone. 10,000 chained points went from 9.6 GB and 25
seconds to a failed solve in well under a second.
Fixes#145
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
(cherry picked from commit 04abbf95c26f68c7d0a126ed9e2827049f593361)
The control channel read each request with no length cap, so a client sending
gigabytes without a newline made the server thread allocate until the process
was killed. Read lines against the 16 MiB cap the MCP server already uses, now
one shared constant, answer a longer one with an error and close; close a
connection after eight lines that are not JSON, serve at most sixteen
connections at once, and run at most sixty-four queued requests per frame so a
flood cannot starve drawing.
Every file command checked the size with the metadata and then read to the end.
A character device such as /dev/zero reports a length of 0, so the cap passed
and the read was unbounded; a FIFO blocked the calling thread, which over the
control channel is the thread drawing the window. vfs::read now takes the
caller's cap, refuses anything that is not a regular file before opening it,
re-checks on the handle to close the window between the two, and reads no more
than the cap plus the byte that proves it was exceeded. Callers pass the limits
they already had, and the recovery read and the CLI's own reads go through it
too. `solvecraft-cli eval /dev/zero` and a FIFO now fail at once instead of
never returning.
Fixes#147
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
(cherry picked from commit 4156e8acd11c90c435a54a3b0ccc59991db71b12)
Each of these asks the allocator for tens of gigabytes, or runs for hours, from
a file of a few kilobytes to a few megabytes. An allocation failure is an abort,
so no panic guard can report it and the user's unsaved work is gone.
- IGES composite curves (102) name their children by pointer, and a tree whose
every node names the same child twice fans out to 2^30 leaf visits under the
depth cap. Count the entities visited and the segments produced against one
budget, since a node that yields no segment trips no bound on the output, and
refuse a composite curve that contains itself.
- IGES directory entries could all name the same parameter range, each
materialising it as its own string and token list. Keep the total text
proportional to the parameter section the file carries.
- STEP knot multiplicities were expanded into one value each before the knot
count was compared with the control points: 20 million knots of multiplicity
64 built a 10 GB vector only to be rejected. Check the sum first.
- One STEP face could list any number of boundary loops, each cloning its edge
handles into a new wire that the face extent then samples. Cap the loops and
the edges they hold; 100,000 bounds went from 2.1 GB to an immediate error.
Also guard the IGES translation, which was the one importer stage outside every
panic guard, keep the vertex-sewing bucket key clear of the overflow its
neighbour lookup caused for coordinates near f64's range, and count how often a
loop passes each vertex once instead of once per edge.
Fixes#146
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
(cherry picked from commit c32792b71c76d74348ad5002d8d2da870937e58e)
3MF: flatten budgeted triangles but copied the whole vertex array for every
placement, so an object with 100,000 vertices and one triangle placed by 500
build items reached 6.8 GB resident, and 10,000 items (still a ~200 KB file)
asked for about 140 GB and aborted on the allocation. Copy only the vertices
the triangles reach, keeping their relative order so an ordinary mesh is
numbered as it was in the file.
DXF: every bound in convert was on the geometry that comes out, and a block
that emits none tripped none of them. Six self-inserts in one empty block fan
out to 6^16 visits, and a 176-byte file never returned. Count entities visited
and array cells placed against a work budget, skip the array loop for a block
with no geometry, and bound the array size on its own.
Fixes#144
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
(cherry picked from commit 3bf3b68a1340d9bdef8dcefd6c4fb5a7cae6eec4)
A construction plane built from model geometry resolves each of its sides
through the document again. That hop restarted the recursion depth at 0, so a
plane that is its own side never exceeded depth 1 and the 32-level cap never
fired: a ~600-byte design file recursed until the stack overflowed and the
process aborted, which no panic guard can intercept.
Carry the depth on the construction context so the geometry arms continue the
caller's count instead of restarting it.
Fixes#143
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
(cherry picked from commit 4bca1d49f27dfe9ef0140132746e1c7415daba8e)
A Join unioned its tools into the first target only, leaving other target
bodies the tool overlapped or touched as separate, interpenetrating bodies.
Each tool now merges with every target it meets (shared material or face
contact, tested on the operands' boundaries) into one body under the first
one's name; absorbed bodies are removed and their threads follow. A tool that
meets no target still joins the first target as a separate lump. Sheet metal
and mesh bodies are never absorbed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J53sR6gEaVt56soFDpTck8
Record ownership transfers in the timeline so earlier Join/Cut membership,
world placement, machining data and styles survive component moves, edits,
independent copying, undo and reopening. Preserve pending positions in history
snapshots and keep constructor transfers ordered during feature redefinition.
Retain shallow sheet pockets and exact planar intersection islands through
unfold/refold, with depth-aware flat previews and DXF export. Store explicit
3D wire coordinate spaces and migrate known legacy provenance without moving
saved points or analytic fits.
Infer unique eligible cylindrical Emboss targets, align Revolve manipulators
with actual extent geometry, and discard obsolete preview errors after input
correction. Certify constructed and immutable imported face styles using exact
source metadata while preserving legacy face names. Upgrade designs to format10
with bounded validation and geometry-preserving legacy migration.
Persist typed sketch dimensions and bounded ordered configuration rules in format 8. Use exact occurrence paths and source identity for editing, movement, appearance, and mass. Preserve sheet material regions through cutting, folding, and exact DXF export; retain nested cavities and material islands.
Validation: cargo xtask ci passed, including 368 engine tests, 178 UI tests, 68 UI scenarios, 136 kernel tests, and the 10-crate wasm check. Actual preview workflows and legacy format-7 origin-corner reopening were inspected.
description:Something in SolidCraft does the wrong thing
labels:[bug]
body:
- type:markdown
attributes:
value:|
Thanks for taking the time to report this! Inside the app, **Report an Issue** fills most of this in for you, with the command, its settings and your environment.
Before reporting a boolean or modelling failure, please try to reproduce it on the latest nightly or a build from current `main`, and say which build you used.
- type:textarea
id:what
attributes:
label:What happened
description:What did you expect, and what happened instead?
validations:
required:true
- type:textarea
id:steps
attributes:
label:Steps to reproduce
description:|
What you clicked or typed, in order. For a problem on a particular design, `solidcraft-cli script design.solidcraft --out repro.json` turns its history into a script that `solidcraft-cli eval repro.json` replays; attach it, trimmed to the failing step if you can.
Using an AI agent with SolidCraft's MCP server? The `solidcraft-reporting` skill in [`skills/`](https://github.com/tangency-labs/solidcraft/tree/main/skills) has it reproduce the problem in a fresh design and write these steps for you.
placeholder:|
1. Sketch a circle on the XY plane
2. Extrude it 10 mm
3. Fillet the top edge with 6 mm
validations:
required:true
- type:input
id:version
attributes:
label:SolidCraft version and OS
description:Help → About shows the version.
placeholder:0.15.0,Windows 11
validations:
required:true
- type:textarea
id:media
attributes:
label:Screenshots or recording
description:|
For anything you can see in the app, a screenshot or short recording of a small example helps the most. Please leave out private designs and personal details.
Describe the problem and resulting behavior. Link the issue when applicable.
Changes a default: no
<!-- Required line, replace "no" if this changes what a user gets out of the box (a default value, key binding, setting, tool behaviour):
Changes a default: yes — <what>, old behaviour kept behind preference <name>
Docs-only and bot PRs are exempt. -->
## Validation
What did you run, and what were the results? Include expected and measured values for geometry changes.
## Screenshots of the behavior (when observable)
For observable CAD, UI or model behavior, include actual descriptive screenshots or a short recording from a small reproduction. This includes kernel, geometry, assembly and backend effects that can be shown in the app. Coding agents should run the app and capture the evidence themselves, rather than asking a human for evidence they can produce. Show the actual failure/result, preferably before and after, with brief steps, expected/actual behavior and annotations where helpful.
Flowcharts and Mermaid diagrams cannot substitute for evidence of visible behavior. Use diagrams for architectural reasoning when actual screenshots cannot illustrate the point. Changes with no useful observable behavior may omit screenshot/video evidence. Exclude private designs and sensitive details; attach media or link to an accessible recording rather than committing generated review artifacts.
## Integration status
Record the upstream commit and PR status checked before handoff, any conflicts resolved, and validation after integration. Keep up with upstream merges and monitor this PR while the task is active; rebase before pushing and never force-push.
SolveCraft is a clean-room, open-source, pure-Rust parametric 3D CAD application in the style of
Autodesk Fusion: sketches with a constraint solver, a parametric timeline of features, a B-rep
kernel, and a desktop UI that agents can drive. Siblings with the same conventions live next to
it in `storytold/`: `cadcraft` (AutoCAD-style, the closest relative), `photocraft`, `vectorcraft`,
`gridcraft`, `designcraft`, `wordcraft`, `deckcraft`, `soundcraft`. Repos don't share code; when
code is adapted from a sibling (MIT OR Apache-2.0) it is noted in ATTRIBUTION.md.
SolidCraft is a clean-room, pure-Rust parametric 3D CAD app (sketch solver, feature timeline, B-rep kernel, an egui UI that agents drive). This file is a router: the hard rules, then the skill to read for your task. Humans: CONTRIBUTING.md.
## Start every session here
1. Read `plan/STATUS.md` (current milestone, next task), then the task in `plan/execution-plan.md`
and the relevant `plan/architecture.md` section. Behaviour reference: `plan/fusion/`
(`menu-tree.json` is Fusion's toolbar dump; `oracle/*/` are parts built in Fusion with their
recipes and measurements).
2. Kernel and library decisions: `plan/adr/0001-geometry-kernel.md` (what we may depend on, what
is rejected for licence reasons).
3. Follow the autonomous operation protocol (`plan/execution-plan.md` §6). Don't stop to ask
unless it lists the decision as the owner's.
## Start here
Work: ROADMAP.md, "Where help is wanted most" in CONTRIBUTING.md, issues labelled `help wanted`, `good first issue`, `needs-triage`; architecture in `docs/architecture.md`. Act when the code, docs and rules settle a question; ask only when it changes what users see. Local `plan/` notes are optional.
`plan/` is gitignored (local only).
## Hard rules (what enforces each)
- Independent: never copy another product's code, icons, text or templates; never drive or take data from commercial CAD. `skills/solidcraft-ground-rules/SKILL.md`.
- No GPL/LGPL/AGPL code, not even reading it (OpenCASCADE, FreeCAD, SolveSpace, CadQuery, LibreCAD, QCAD): `cargo xtask licences --check`.
- Every asset is original or openly licensed with a row in ATTRIBUTION.md; read `assets/AGENTS.md` before icon work: `cargo xtask assets`, `cargo xtask icons`.
- Never crash: no unwrap/expect/panic/unsafe in non-test code (unsafe allowlist in crash-safety), errors are `Result`, input is hostile: clippy `-D warnings`, `hostile_params_never_panic`.
- Everything is a command and programmatic calls never open dialogs: `skills/solidcraft-commands/SKILL.md`.
- One outer UI container, one camera path (`navigation::handle`), colours from theme tokens: `cargo xtask ui-containers`.
- Application code is Rust only and builds for wasm: `cargo xtask wasm`, `cargo xtask web-offline-test`.
- The product name is SolidCraft; the old name lives only in `docs/compatibility.md`.
- Sorted lists (message catalogs, `mod` lines, languages) stay sorted: `cargo xtask lists`. Never edit generated files (`docs/features.md`, `THIRD-PARTY-LICENSES.txt`) in a PR.
- A design-format change is ONE new file in `crates/doc/src/format/pending/`: `scripts/test-format-pending.sh`.
- A command has input and output schemas (will become required; `schema/unconverted.txt` only shrinks): `skills/solidcraft-commands/SKILL.md`.
- New UI strings go in `assets/locales/pending/<slug>.json`, ATTRIBUTION rows in `attribution/pending/<slug>.md`: `cargo xtask lists`.
- A changed default keeps the old behaviour behind a preference and fills the PR's `Changes a default:` line, and no second PR for an issue a lower-numbered one closes: `pr-checks.yml`.
- A regression test goes in the area's existing test file, never a per-bug file (#3226).
- Look at open PRs first, claim with a draft PR (lowest number wins), one topic per PR, never force-push `main`, no pushes just to catch up with `main`: `skills/solidcraft-contributing/SKILL.md`.
- Show observable changes with real screenshots or a recording, captured yourself.
- Run `cargo xtask ci` before every commit, without switching branches while it runs. `cargo xtask skills` keeps this file and `skills/` honest.
## Never crash
People trust SolveCraft with their designs; a crash loses their work. **This outranks feature work.**
- **No panics in non-test code:** no `unwrap()`, `expect()`, `panic!`, `unreachable!`, `todo!`,
`unimplemented!`; no `unsafe` (`unsafe_code = "forbid"`).
- **Errors are `Result<T, E>`** through the crate's error type and `?`. An unfinished feature
returns an error ("not supported yet: …"); it never panics.
- **Input-derived numbers are hostile** (command parameters, design files, control-channel
input): `get()` not `[i]`, finite checks, caps on sizes and loop counts.
- **Third-party kernel calls are guarded:** everything that calls `truck` goes through
`solvecraft_kernel::guard`, which turns a panic into `KernelError::Internal`.
- **Last-resort guard:** `Session::execute` runs every command under `catch_unwind`; an escaped
panic restores the design and reports `EngineError::Internal`.
- **Prove it:** every crash fix lands with a regression test (see `hostile_params_never_panic`).
SolveCraft is an open-source parametric 3D CAD app written in pure Rust, in the style of
Autodesk Fusion. You sketch on a plane, constrain the sketch, turn profiles into solids, and
change a parameter to watch the timeline rebuild. It runs as a desktop app (Windows, Linux), in
Agents: start from the router in [AGENTS.md](AGENTS.md), which points to a short skill in [`skills/`](skills/README.md) for each kind of work. This page stays the full human guide.
SolidCraft is an open-source parametric 3D CAD app written in pure Rust.
You sketch on a plane, constrain the sketch, turn profiles into solids, and
change a parameter to watch the timeline rebuild. It runs as a desktop app (Windows, macOS, Linux), in
the browser (wasm + WebGPU/WebGL2), headless from a CLI, and as an MCP server that AI agents can
model with.
@@ -10,27 +12,78 @@ Help is welcome, whether that's a one-line fix or owning a whole area. This page
build it, how the code is laid out, the few rules that aren't negotiable, and where help is
needed most.
Before choosing a task, check the [command inventory](docs/features.md) and
[roadmap](docs/roadmap-detail.md) for implemented capabilities and current priorities.
Questions, ideas or want to say what you're working on? Join the [Discord](https://discord.gg/FtU4NQNVDH).
## Kernel work is paused
We're evaluating a new geometry kernel. Until that's decided, please don't open issues, pull
requests or discussions about kernel work: booleans, fillets, chamfers, shell, offsets, sweeps and
lofts, threads and holes, STEP/IGES import and export, tessellation, or anything else that changes
`crates/kernel/`. The pinned issue on the tracker explains it. New ones get the `kernel-paused`
label (`.github/workflows/kernel-paused.yml`) and a maintainer answers them. Crashes and data loss
are still taken as issues, and everything outside the kernel carries on as normal.
## Get it running (about 10 minutes on a fast machine)
You need stable Rust 1.90 or newer. For the web
build you also need `rustup target add wasm32-unknown-unknown` and [`trunk`](https://trunkrs.dev).
You need stable Rust 1.95 or newer (required by egui/eframe 0.36). For the web
build you also need `rustup target add wasm32-unknown-unknown` and [`trunk`](https://github.com/trunk-rs/trunk)
`cmd::plastic::tests::enclosure_snap_fit_follows_resized_front_wall`, the `atomic_commands` contract tests
-`solidcraft-kernel`: `tests::boolean_identity_on_tangent_primitives`, `tspline::tests::hostile_tsplines_never_panic`, and the `boolean_props` property tests
(these run only in the nightly "Property tests" job: their inputs are random, so main skips them too)
**The action matrix** tries every ordinary action on every face, edge, vertex and body of a
corpus of designs against a checked-in ratchet ([docs/action-matrix.md](docs/action-matrix.md)).
A pull request that makes an action pass should record it with `cargo xtask action-matrix --bless`;
one that makes a passing action fail fails the sample every CI run checks.
**The reference tests** live in [`tests/oracle/`](tests/oracle/), each a `recipe.json`
in our own format with an `expected.json` beside it. SolidCraft rebuilds the recipe and compares
body count, volume, area, centre of mass and bounding box (and, where there is a `reference.stl`,
the surface itself). Expected values are either worked out in closed form
(`tools/oracle-analytic/`) or computed by building the same recipe in FreeCAD, run as a separate
program (`tools/freecad-oracle/`); each file says which. The generated
[reference test report](docs/oracle.md) has the current case count and results. The reference-test
CI job runs `cargo xtask oracle` during full validation, and on pull requests that reach
`solidcraft-cli`. A case that
fails because of a known bug names its issue in `expected.json` (`"known_failure": "#123"`);
never change an expected value to match SolidCraft's output.
## Changing the design file format
A design file says which format it is in (`"format": "solidcraft/N"`). If your change saves
something new in a design (a field, a feature type, a new kind of name), or makes older designs
mean something else, it needs a format change. **Add one file, and edit nothing shared:**
1. Create `crates/doc/src/format/pending/<what-it-is>.rs` (lowercase, dashes). Start it with a
`//!` note: what is saved, what older designs get without it, and that the last release asks
for an update instead of meeting it.
2. If older designs must be rewritten as they open, put `pub fn step(v: &mut Value) ->
crate::Result<()>` in the same file. Pending steps run in file-name order, after the older
formats' steps.
3. Don't touch `FORMAT`, `JSON_STEPS`, the history in `crates/doc/src/format.rs`, another
change's file, or any `v<N>/` folder: those are fixed once released.
That's all. The build counts the files: while any are pending it writes the next format number,
which no release reads, so a release refuses a design that has your change instead of losing it.
A file also lists the pending changes it has (`format_pending`), so a later development build
runs only the steps that file is missing. Two pull requests that each add a change merge without
a conflict. `crates/doc/src/format/pending/README.md` has an example.
**At release time** the maintainer runs `scripts/fold-format-pending.sh`, which moves the pending
files into `crates/doc/src/format/v<N>/` and fixes that number for good (see
[docs/releasing.md](docs/releasing.md)). Never add a file to a `v<N>/` folder afterwards.
## Lists many changes add to
Some files gain an item in many pull requests at once. They are kept so that two changes that
each add an item only meet when their items are neighbours, which spares everyone a conflict
and a fresh CI run:
- **Message catalogs** (`assets/locales/*.json`): one message per line, sorted by id. Add your
message anywhere and run `cargo xtask lists --fix`; `cargo xtask lists` (in `cargo xtask ci`
and the Lists workflow) fails on an unsorted catalog or an id that appears twice.
- **`mod` lines**: one per line, in the sorted group they belong to (`cargo fmt` keeps each group
sorted). Add a new module into the existing group, not as a new group at the end.
- **Commands without schemas** (`crates/engine/src/schema/unconverted.txt`): sorted, one id per
line, and it only shrinks. Converting a command removes its line; a new command is never added
to it.
- **Languages** (`crates/ui-egui/src/i18n.rs`): one per line, English first, then by id (a test
checks it).
- **Generated files**: `THIRD-PARTY-LICENSES.txt` and `docs/features.md` are rewritten on `main`
after each merge by the Generated files workflow (`cargo xtask licences`, `cargo xtask
features`). Don't regenerate or edit them in a pull request.
- **Design format**: one file per change, never an edit to a shared number (above).
- **New UI strings and attribution rows**: one new file per change, `assets/locales/pending/<slug>.json`
and `attribution/pending/<slug>.md`, instead of editing the catalogs or the ATTRIBUTION.md table;
`cargo xtask fold-pending` folds them on `main` (see the README in each folder).
## Sending a change
1. Work is tracked in [GitHub issues](https://github.com/bherbruck/solvecraft/issues), labelled
by area (`area:kernel`, `area:sketch`, `area:doc`, `area:ui`, `area:infra`). Before starting,
comment on the issue so it gets the `in progress` label and two people don't build the same
thing. For anything bigger than a bug fix that has no issue yet, open one first.
1. Work is tracked in [GitHub issues](https://github.com/tangency-labs/solidcraft/issues), labelled
by area (`area:kernel`, `area:sketch`, `area:doc`, `area:ui`, `area:infra`). For anything
bigger than a bug fix that has no issue yet, open one first. **Claim the issue with a draft
PR** before you start, so two people don't build the same thing:
- Check nobody has it: `gh pr list -R tangency-labs/solidcraft --state open --search "#N"`.
- Branch from the latest `main`, commit `git commit --allow-empty -m "claim: #N"`, push, and
open a **draft** PR titled `<summary> (#N)`, e.g. `Fillet: variable radius along a chain (#123)`.
- **The lowest-numbered open PR for an issue wins.** If yours lost, close it (or offer to help
on the other one).
- The bot records the claim when the PR is opened: it assigns you, adds `in progress` and
says so on the issue. It reads `#N` in the PR title and `Fixes #N` in the body. If someone
already holds the issue, or an older open PR references it, it says so on your PR instead.
- Want to stake a claim before you have a branch? Comment `/claim` on the issue (below).
- Push work in progress at least daily. A draft with no pushes for 72 hours may be taken over:
comment on it first, then open your own claim.
- Mark the PR ready for review when it is done.
2. Branch from `main` and keep the PR to one topic.
3. Run `cargo xtask ci` before pushing. It runs fmt, the asset, licence and layer checks,
clippy (`-D warnings`), then the tests with the wasm build check alongside. A PR has to pass it.
Heavy CI runs after review: a pull request gets the formatting and repository checks and a
quick compile (`cargo check`) when you push. Once a maintainer has reviewed it they add the
`approved-for-ci` label when the PR is ready to merge, which runs the full CI (build, tests,
goldens, Clippy); every push after that runs the full set again. Maintainers merge approved
PRs in groups by area, so PRs that touch the same code meet a conflict at most once.
4. Reference the issue (`Fixes #12`). In the PR, say what you tested and how. For geometry work, include the measured numbers (and
the expected ones).
Much of SolveCraft so far was written by AI coding agents (Claude) working in parallel, each
Commenting `/claim` assigns you, adds the `in progress` label, and replies with its automatic cleanup
deadline. It frees a claim with no linked PR after 72 hours, or a claim with a PR after 14 days
without pushes. These automatic cleanup deadlines are separate from the daily progress and
72-hour takeover policy above. A draft PR that references the issue claims it the same way.
`/unclaim` releases the bot claim early (only you or a maintainer can). Issues a maintainer has
taken are never freed automatically.
Help reviewers see what the words mean. For observable CAD, UI or model behavior, please
include actual descriptive screenshots or a short recording in the issue or PR. This evidence
is required for changes or bugs that can be demonstrated in the app, including kernel,
geometry, assembly and backend work. Coding agents should run a small reproduction and
capture the evidence themselves; please don't ask a human for evidence the agent can produce.
Show the actual failure or result, preferably before and after, with brief steps and
expected/actual behavior. An arrow or annotation can help identify the detail that matters.
Flowcharts and Mermaid diagrams cannot replace evidence of visible behavior. Diagrams are
useful for architectural reasoning when actual screenshots cannot illustrate the point.
Changes with no useful observable behavior do not need screenshot or video evidence. Use a
small example without private designs or sensitive details; attach the media or link to an
accessible recording rather than committing generated review artifacts.
To show a picture without committing it, attach the media to the PR (it goes to the PR's own
attachment storage and renders inline in the comment). A gist is the better home for a log or
another text file, and you link it; a gist does not render a picture inline. The review artifacts
stay out of the diff that way, and nothing has to be pushed to a branch of its own.
Attaching works in the browser for anyone who can comment: drag the file into a comment on the
pull request. A command-line agent without a browser session usually cannot attach a file (the
upload needs write access to this repository). Say so in the pull request, and the person running
the agent or a maintainer can attach it.
Keep an eye on your PR's status, but don't push to it only to catch up with `main` (no merge of
`main`, no Update branch button, no rebase for that alone): every push restarts the whole CI
run for everyone waiting behind it, and `main`'s own run catches what lands in between. When you
push a change anyway, or GitHub reports a conflict, rebase onto the latest `main` (push your
branch with `--force-with-lease`) or merge `main` in once, and resolve conflicts without losing
others' changes. Don't re-run CI on an unchanged commit either: a real failure fails again.
Re-run only a job that runner trouble cancelled, with "Re-run failed jobs". While a task remains
active, follow an open PR through its status changes; at handoff, clearly identify anything
still awaiting merge.
Much of SolidCraft so far was written by AI coding agents (Claude) working in parallel, each
owning one area, with a human setting direction and reporting bugs. Their working notes are in
[AGENTS.md](AGENTS.md). Human contributors are just as welcome. Use whatever tools you like, as
long as the change meets the rules above.
Two quick checks run on every pull request (`.github/workflows/pr-checks.yml`, not required): it
fails when a lower-numbered open PR by someone else already closes the same issue (the label
`not-a-duplicate` skips it), or when the description lacks the `Changes a default: no` / `Changes a default: yes — <what>, old behaviour kept behind preference <name>` line (docs-only and bot PRs are exempt).
## Labels and triage
The triage bot labels every new issue and PR (`.github/workflows/triage.yml`). It only adds labels;
correct them freely and it won't put them back.
- **PRs** get `area:*` from the paths they change (`crates/kernel` and `crates/geom` →
`area:kernel`, `crates/ui-egui` and the apps → `area:ui`, `.github/`, `scripts/`, `docs/` →
`area:infra`, and so on).
- **Issues** get `bug` (in-app reports and the bug template), `enhancement` (feature requests),
`priority:P1` for crashes, and an `area:*` from the title or the reported command. Until it has
a priority, an issue carries `needs-triage`; setting a priority removes it.
- **Maintainers** comment `/triage area:kernel P1` (any existing label; `P0`–`P3` for a
priority, which replaces the old one; `-label` removes one) to sort an issue or PR.
- **Short bug reports without steps to reproduce** get `needs-info` and one comment asking for
the steps (and the version, when it's missing). An app report that arrived empty is asked to
paste the report the app copied to the clipboard. The reporter's reply, or an edit that adds the
steps, removes the label; an issue that stays in `needs-info` for 14 days is closed, and can be
reopened any time. Feature requests, app reports with their details, and maintainers' issues
are never asked.
- `needs-triage` is the queue: [open issues with it](https://github.com/tangency-labs/solidcraft/issues?q=is%3Aopen+label%3Aneeds-triage).
## Where help is wanted most
See the [`help wanted`](https://github.com/bherbruck/solvecraft/labels/help%20wanted) and
[`good first issue`](https://github.com/bherbruck/solvecraft/labels/good%20first%20issue) labels.
**Kernel speed (slow fillets, booleans, STEP):** please don't rewrite the current kernel code for speed; it is being replaced by [solidcore](https://github.com/bherbruck/solidcore) ([progress](https://tangency-labs.github.io/solidcore/)). Report slow cases with a file under the `kernel-perf` label, and put kernel effort into solidcore instead. See [#1597](https://github.com/tangency-labs/solidcraft/issues/1597).
See the [`help wanted`](https://github.com/tangency-labs/solidcraft/labels/help%20wanted) and
[`good first issue`](https://github.com/tangency-labs/solidcraft/labels/good%20first%20issue) labels.
In order of how much they'd help users:
1. **Fillet and blend robustness.** Fillets are our own code (truck has no fillet operation).
@@ -145,15 +409,24 @@ In order of how much they'd help users:
faces, and fillets that consume neighbouring faces.
2. **Boolean robustness.** A random-design fuzzer currently gets about half of its designs
through without a kernel error. Each failing seed is a reproducible bug.
3.**Y-up / Z-up.** SolveCraft is Z-up only, so most STEP files (from SolidWorks, Onshape, or
Fusion's default) open lying on their backs. Fusion solves this with a "default modeling
orientation" preference.
4.**Imported appearances.** Colours from STEP files are lost when a feature changes the
imported body. They should become design appearances on import.
5.**Drawings, CAM and mesh tools.** Theseare out of scope for now (see
[ROADMAP.md](ROADMAP.md)), but anyone who wants to own one is welcome.
3. **Drawings, CAM and mesh tools.** These are out of scope for now (see
[ROADMAP.md](docs/roadmap-detail.md)), but anyone who wants to own one is welcome.
## Updating the generated feature list
After adding or renaming commands, run `cargo xtask features`. The registry is the
source of truth for `docs/features.md`; do not hand-edit the generated rows.
The generator sorts commands by ID and omits global and per-section counters,
so additions to unrelated areas no longer rewrite shared summary lines.
If two branches still conflict in this file (for example, adjacent additions in
the same area), first resolve the command registry and source conflicts, then run
`cargo xtask features` again and stage the regenerated file. Do not use Git's
`union` merge driver: it can silently retain stale or duplicate command rows.
The generator tests cover stable ordering and additions that leave existing lines
unchanged. The initial switch to sorted rows is a one-time larger diff.
## Licence
By contributing you agree that your contribution is licensed as MIT OR Apache-2.0, like the rest
of SolveCraft. SolveCraft isn't affiliated with Autodesk. Fusion is a trademark of Autodesk, Inc.
of SolidCraft. SolidCraft isn't affiliated with any CAD vendor; product names are trademarks of their owners (see NOTICE).
**Parametric 3D CAD; an open-source, [clean-room](CLAUDE.md) reimplementation of Autodesk
Fusion 360, built in pure Rust.**
<h1 align="center">SolidCraft</h1>
Sketch, constrain, and build solids with a parametric feature timeline. Change a dimension
and watch your model rebuild. Run it on your desktop, in your browser, or let AI agents
model through the built-in MCP server. Part of the storytold "craft" family.
<p align="center">
<strong>An open-source parametric 3D CAD application built in pure Rust.</strong><br />
Sketch, constrain, and build solids with a parametric feature timeline. Change a dimension
and watch your model rebuild. Run it on your desktop, in your browser, or let AI agents
model through the built-in MCP server.
</p>

<div align="center">
*An agent builds an electronics enclosure over MCP: sketches on faces, shell, a port cut, one screw boss patterned to four, then width 80 → 100 mm in the Parameters dialog and the timeline rebuilds. 3× speed; [full video](https://github.com/bherbruck/solvecraft/releases/download/v0.1.0/enclosure-demo-v3.mp4), script in [examples/demo](examples/demo).*
[![Download for Windows][badge-windows]][windows-msi]
[![Download for macOS][badge-macos]][macos-dmg]
[![Download for Linux][badge-linux]][linux-appimage]
[![Get it from the Snap Store][badge-snap]][snap-store]
[![Try it in your browser][badge-web]][web-app]
**Documentation:** the [SolveCraft book](https://bherbruck.github.io/solvecraft/) (user guide,
developer guide and command reference; sources in [docs/book](docs/book)), and the
[web app](https://bherbruck.github.io/solvecraft/app/) to try it in the browser.
</div>
**Status: early but usable for simple parts.** 76 of 76 reference parts built in Fusion
rebuild in SolveCraft with the same volume, area and topology ([docs/oracle.md](docs/oracle.md));
command parity with Fusion's Design workspace is tracked in [docs/parity.md](docs/parity.md).
Open work and known bugs are in [GitHub issues](https://github.com/bherbruck/solvecraft/issues),
and help is welcome: see [CONTRIBUTING.md](CONTRIBUTING.md).

*A Raspberry Pi 4 case from an empty design: a sketch dimensioned with the board's parameters, extrude, fillet, shell, the rim rounded in one pick, an M2.5 standoff patterned on the 58 × 49 mm hole rectangle, port cut-outs, vent slots, "Pi 4" engraved in DejaVu Sans Bold and a PETG material defined in the design, with its mass in Properties, then height 24 → 32 mm in Parameters and it all rebuilds. 6× playback with the idle pauses cut. [Full video](https://github.com/tangency-labs/solidcraft/raw/refs/heads/main/docs/screenshots/demo.mp4), recorded with the [demo script](examples/demo/record_ui.sh).*
## Download
- **Windows:** the [installer][windows-msi].
- **macOS:** the [disk image][macos-dmg] (Apple silicon and Intel).
- **Linux:** the [AppImage][linux-appimage], or `sudo snap install solidcraft` ([Snap Store][snap-store]).
- **Web:** [try it in your browser][web-app], no install.
ARM64 and 32-bit builds, portable zip, .deb, .rpm, Flatpak and the other packages are on the
[downloads page](docs/downloads.md). The downloads aren't signed yet: on Windows choose
**More info**, then **Run anyway** in SmartScreen; on macOS right-click SolidCraft and choose **Open**
the first time. On SolidCraft 0.11 to 0.14, download once by hand; see
Status as of **2026-10-10**. SolveCraft targets Fusion-style parametric modelling (sketch →
features → parametric timeline) as an open, pure-Rust, agent-drivable application.
The short version, for people who use SolidCraft. The detailed developer status (milestone
tables, limitations, measurements) is in [docs/roadmap-detail.md](docs/roadmap-detail.md).
## At a glance
## What SolidCraft is
| Question | Answer |
|---|---|
| Where are we? | **M0, M1, M13 done; M2, M5–M7, M9, M10, M14 in progress.** Sketch with solver and the full SKETCH create/modify set; extrude, revolve (two sides, symmetric, to object), fillet/chamfer (variable radius, chord length, a radius per edge, corner patches), holes, threads (cosmetic and modelled), patterns, mirror, shell (inside/outside/both), draft, loft, sweep, coil, split body and split face, delete face, construction planes, axes and points; persistent naming of faces and edges; parameters and configurations; components, joints, motion studies, exploded views; sheet metal and a plastic enclosure subset; STEP/IGES/3MF/STL/OBJ/DXF; autosave and crash recovery; desktop and browser (wasm) app, CLI, MCP server, oracle harness. |
headline counts in-scope tabs only and lists the deferred ones separately.
## In progress now
## Milestones
Each feature below is a [milestone](https://github.com/tangency-labs/solidcraft/milestones) holding
the issues for it, so the count shows how far along it is. Counts are as of 2026-10-25, one per
feature or request. Work merged on a feature branch (`drawings`, `render`) counts as built but
reaches the app only when that branch merges. They are rough, not a promise.
| # | Milestone | Status | Hours left |
|---|---|---|---|
| M0 | Vertical slice | done | — |
| M1 | Sketch depth: done — projection/intersect/include as linked geometry (auto-project on faces, lost-reference warnings, break link), full SKETCH create/modify set (slots, polygons, tangent arcs/circles, ellipse, fit/control splines, conic, text, trim/extend/break, fillet/chamfer, offset, mirror, patterns, move/scale, blend curve, centerline), constraints incl. curvature and polygon with over-constraint refusal, driven/arc-length/linear-diameter dimensions, AutoConstrain, DXF/SVG insert, 3D sketch curves (Project To Surface, Intersection Curve, Isoparametric Curve, Spun Profile), exact free-form profiles (B-spline/NURBS edges, any spline degree, curves cut by crossings), sweeps and pipes along 3D sketch curves, canvases and decals, curvature combs | done | 4 |
| M2 | Robust modelling: coplanar booleans in the kernel, holes (placed on faces and on sketch points), threads (cosmetic, then modelled), patterns, mirror, shell, draft, split, construction geometry | in progress | 16 |
| M3 | General fillets and chamfers (curved edges, chains, vertex blends) | planned | 30 |
| M4 | Files: STEP import (done: all 29 Fusion STEP files — [docs/step-import.md](docs/step-import.md)), 3MF export and 3MF/STL import as mesh bodies (done), DXF sketches, OBJ import | in progress | 6 |
| M5 | Persistent naming of faces and edges (plan/adr/0002): faces named from their history (sketch curve sides, caps, revolve sections, primitive sides, blends, pattern/mirror copies; split pieces numbered), edges by their faces; fillets/chamfers, sketches on faces, shell/draft/offset/replace faces, hole placements, joint origins and face appearances resolve by name first and warn when the entity split or vanished: done. sheet metal edges by base sketch line and flange side, shell inner faces, sweep and loft roles: done. Left: kernel-reported face history (exact through coincident surfaces) | in progress | 20 |
| M6 | UI depth: selection feedback and origin (done), timeline editing (done: hover and selection highlight modifier faces, including fillets and chamfers; edit feature, history marker drag, reorder with dependency checks, suppress, rename, delete with dependents, re-resolved references), right-docked dialogs (done), live previews for every feature dialog (done), keyboard-first commands with on-canvas value boxes and sketch inline dimensions (done), drag manipulators with automatic join/cut (done: extrude incl. two sides, fillet, shell, hole depth, move, pattern spacing, revolve rotator; taper handle open), dark theme (done), right-click menu with Repeat (done), measure tool (done), section analysis (done), selection filter (done), drag-solve of sketch points (done); appearances, preferences | in progress | 6 |
| M7 | Sweep, loft (incl. to a point), rib and web (to next / depth), emboss (planar faces), thread (cosmetic, and modelled M3–M20 outside and inside: the ISO profile screwed in as a faceted cut), coil (smooth B-spline helical walls), pipe, pattern on path, patterns of bodies and components, replace face (parallel), align, remove, feature copy/paste: done. Left: emboss on curved faces, exact (unfaceted) modelled threads | in progress | 12 |
| M8 | Surface workspace | deferred | — |
| M9 | Components and assemblies (Fusion-style): New Component / Create Components from Bodies, active component, per-component origin/bodies/sketches/timeline entries, occurrences with transforms, Copy/Paste vs Paste New, nested Browser tree with visibility and isolate; Ground, Move/Copy of occurrences, joints (rigid, revolute, slider, cylindrical, pin-slot, planar, ball) by joint origins with limits and Drive Joints, As-built Joint, Rigid Group, Contact Sets (driving and motion studies stop where bodies in a set would pass through each other), Motion Study (joint values keyed along a timeline, playback, export of occurrence positions as JSON/CSV), Exploded views (automatic or by hand, shown without changing placements); interference, per-component physical properties; STEP with assembly structure. Groundwork done: component tree, occurrences with transforms (instances, Paste New), world placement, active component as authoring frame, ground, Capture/Revert Position, moving bodies and sketches. UI: ASSEMBLE panel dialogs for New Component, Joint (snaps on faces, edges, circles and vertices; animated motion preview; offsets, flip, limits), As-built Joint, Joint Origin, Rigid Group, Drive Joints (live slider), Motion Link and Interference (overlaps shown red); joints in the timeline and the Browser | in progress | 32 |
| M10 | Sheet metal: flange (base, edge, contour), hem, unfold/refold, convert, flat pattern + DXF, rules: done (oracle 67-76). PLASTIC enclosure subset: Boss, Rib, Web, Lip/Groove, Snap Fit, Rest, Plastic Rules (library, edit, assign; rule draft and clearance as feature defaults): done, no Fusion oracle (licence), tested against hand-computed geometry. UI: SHEET METAL and PLASTIC tabs in Fusion's layout with dialogs (live preview, Edit Feature, pre-selection) for Flange, Hem, Unfold/Refold, Convert, Sheet Metal Rules, Create Flat Pattern (flat view with bend lines, bend table, DXF export), Boss, Lip/Groove, Snap Fit, Rest, Manage and Assign Plastic Rules; Fold (bend along a sketch line) | in progress | 36 |
| M14 | Configurations: a table of rows (parameter expressions and feature suppressions per row) switched with config.activate (undoable, all or nothing), kept through parameter renames and feature deletes (config.configure, config.table, config.column/row/activate). Left: configuration rules, per-row materials and appearances | in progress | 4 |
| M13 | Never lose work: atomic saves (temporary file, flush, rename; a crash at any moment leaves the old file or the new one) with the replaced version kept as `<file>.bak`; autosave of unsaved changes to the recovery folder every 5 minutes (preference) and after big operations, only when something changed; "Recover unsaved design?" on the next launch (`doc.recovery_list`, `doc.recover`, `doc.recovery_discard`); a kill-mid-save test (process killed while saving a 0.5 MB design 25 times). File format: `"format": "solvecraft/2"`, older formats upgraded step by step (`doc::format`), newer ones refused with a message, out-of-range numbers and broken placements refused; a corpus of designs saved by M0, components-era, joints-era and current builds (apps/solvecraft-cli/tests/corpus) that must open and evaluate as saved; damaged and hostile variants (truncation, bit flips, hostile values, removed keys, repeated features, deep nesting) never panic | done | 6 |
- **[Rendering & materials](https://github.com/tangency-labs/solidcraft/milestone/17)**: in progress, 9 of 34 pieces done, 3 of them built only on the `render` branch and in the app when it merges. The basic Render workspace is already in releases (0.22.0); the interface rework (#6281) and the material and light libraries are what is left.
- **[Drawings](https://github.com/tangency-labs/solidcraft/milestone/8)**: in progress, 3 of 21 pieces built (2 in releases, 1 on the `drawings` branch), and 3 more partly built there (sheets and borders, bill-of-materials types, the paper scene). Not in the app yet: it lands when the `drawings` branch merges. Today you can export the current view as a DXF.
- **[Sheet metal](https://github.com/tangency-labs/solidcraft/milestone/15)**: in progress, 1 of 30 requests done. Flanges, hems, unfold and flat patterns already work; the 30 are extensions, mostly the per-edge flange options (segments, miters, partial flanges, reliefs) and bend allowance.
- **[Sketching](https://github.com/tangency-labs/solidcraft/milestone/14)**: in progress, 12 of 50 pieces done. Dimension types, selection and snapping options.
- **[Solid modeling](https://github.com/tangency-labs/solidcraft/milestone/2)**: in progress, 9 of 50 pieces done. Blends, holes, threads and patterns.
- **[Assemblies & joints](https://github.com/tangency-labs/solidcraft/milestone/3)**: in progress, 4 of 19 pieces done. Joints, motion studies and exploded views already work; placement rules, linked sub-assemblies, fastener stacks and weldment corners are the pieces left.
- **[Import & export](https://github.com/tangency-labs/solidcraft/milestone/5)**: in progress, 7 of 39 pieces done. FreeCAD, SolidWorks and point-cloud import, and mesh export options.
- **[AI agents & MCP](https://github.com/tangency-labs/solidcraft/milestone/18)**: in progress, 2 of 8 pieces done. Agent chat and the MCP server.
- **[Interface, web app & plugins](https://github.com/tangency-labs/solidcraft/milestone/4)**: in progress, 31 of 96 pieces done. Ribbon redesign, navigation, themes, the web app and plugins.
- **[SMLib kernel migration](https://github.com/tangency-labs/solidcraft/milestone/1)**: in progress, 1 tracking issue open. The kernel is moving from truck to NVIDIA's SMLib (usd-brep, Apache-2.0) on the `kernel-smlib` branch; it is not in releases yet. Tracked in [#4012](https://github.com/tangency-labs/solidcraft/issues/4012).
## Known limitations
## Next
-Fillets and chamfers: constant, variable-radius and chord-length fillets; a radius per edge,
rounds next to earlier rounds of another radius, and smooth corner patches where three rounds
meet (a sphere for equal radii); straight edges (convex or concave) between planar faces with perpendicular
planar end faces (equal-distance chamfers also meet sloped planar end faces, including inner
basin rims within the rim width); convex fillets also meet sloped planar ends and reflex
corners around pocket rims, with unequal wall slopes and extensions confined to the face;
connected equal-distance chamfers also trim each other across unequal sloped rim walls;
whole smooth loops of a planar face, picked whole or by one edge (tangent
chain), including circles and walls at any angle (boss bases and tops, bores, pocket floors,
plate outlines), loops with sharp corners between straight edges (the blends meet in mitres), runs of
neighbouring edges along a planar face's loop (two top edges of a box: mitred, ending square;
with the edge between them too, the corner becomes a sphere octant),
rounded corners rounded again (sphere octants, tori at concave corners); every edge of a
convex planar body or of an extruded part; fillets inside patterns; closed chains of curved
edges between any two smooth sides (a branch pipe on a main pipe), by a rolling ball (the
blend a rational B-spline through exact arcs), closed or ending at planar faces square to the
edges.
- Booleans: prisms along one direction (plates with holes, slots, patterns of them, equal
sections stacked end to end) are combined exactly in 2D; bodies touching along one whole face
(a part and its mirror image) are stitched; coincident planar faces are pushed apart when
their neighbours are planes or cylinders. Other coincident curved configurations can still
fail. Fully internal voids are not supported. Spheres are built from six pole-free patches.
- Mass properties come from fine tessellation (curved faces within ~1e-4 relative); triangles
that cut through curved faces (trimmed faces after booleans) are split onto the surface first.
- STEP import (our own reader, crates/kernel/src/step_in; all 29 Fusion STEP files pass
`cargo xtask step-corpus`) reads solids (B-rep with analytic and B-spline geometry), units,
product names, colours and assemblies; offset surfaces and pcurve-only edges are not read yet,
and assembly components are flattened into bodies. Closed periodic faces (torus, B-spline
bands) are split in two for meshing; measure() uses a finer tolerance for small radii.
- Persistent naming (plan/adr/0002) names faces from their history and edges by their faces;
references resolve by name first and warn when the entity split or vanished. Kernel-reported
face history (exact through coincident surfaces) is not wired in yet.
- Modelled threads are cut as a faceted body (curved faces within 1/200 of the pitch); Extrude
and Revolve To Object on a curved face cut the tool back to the target's surface.
- Free-form curves cut at a tangent touch (no clean crossing) enter that profile as polylines.
Sweeps along 3D sketch curves follow their polyline (smooth walls when the path is smooth).
- Shell and offset faces work on bodies of planes, cylinders, cones and spheres (convex or not:
filleted boxes, bosses, domes, countersinks); draft on convex planar bodies and on walls (planes, cylinders → cones) between caps; lofts through three or more sections are smooth (B-spline sides through every section), two
sections ruled.
-**[Surfacing](https://github.com/tangency-labs/solidcraft/milestone/16)**: next, 6 of 14 pieces done. Patch, trim, blend and match exist; the rest of the surface toolset is planned.
- **[Manufacturing / CAM](https://github.com/tangency-labs/solidcraft/milestone/9)**: next, 1 pieces tracked, none done. There is no CAM in SolidCraft yet. Send to Slicer and a first toolpath slice are in review; proposals are [#1153](https://github.com/tangency-labs/solidcraft/issues/1153) and [#1621](https://github.com/tangency-labs/solidcraft/issues/1621).
## Kernel
Features are not tied to a release or a date. Work lands in the daily releases as it is finished.
Anything without a milestone, such as bug fixes, is not a feature and is not listed here.
truck (Apache-2.0) behind our own `kernel` crate, with our own blend operation and boolean
retry/validation; see `plan/adr/0001-geometry-kernel.md` (local) — summary in README.
## How releases work
- Releases ship daily, as numbered versions (v0.22.0, v0.22.1) with nightly pre-releases in
between. They are delivered through the in-app updater.
- SolidCraft tells you when a newer stable release exists and never updates without asking.
Where it can replace itself safely (AppImage, Windows installer and portable builds) there is
an **Update now** button; downloads are checked against the release's SHA-256 list. See
[updates and privacy](docs/book/src/guide/updates.md).
- Design files carry a format number. A new version opens designs from older versions, but an
older version refuses a design saved by a newer one (0.22.0 writes format 35, and 0.21.0 and
earlier will not open it). Each release's notes say when the format changed. Keep a copy of
an important design before you move to a new version.
## Licence
SolidCraft is licensed MIT OR Apache-2.0, at your option; see [LICENSE-MIT](LICENSE-MIT) and
[LICENSE-APACHE](LICENSE-APACHE). Versions already released stay available under those licences
"timing_scope":"wall time from render.start through completed PNG encoding; includes job preparation and readback, excludes CAD generation, preflight and initial CPU/GPU host setup; GPU setup timing includes adapter selection and pipelines when requested",
//! prints a table of the results; `--markdown` writes the issue's comment template filled in,
//! `--json` every task's rules, conversation and design before and after.
usestd::io::{BufRead,BufReader,Write};
usestd::net::{SocketAddr,TcpStream};
usestd::time::{Duration,Instant};
useserde_json::{Value,json};
/// Standard test v1, as #4851 states it. Changing a prompt or a rule makes it another test.
constTASKS: &str=r#"[
{"id": "01-cube", "prompt": "Make a 20 mm cube.", "check": {"bodies": 1, "volume": 8000, "bbox_size": [20, 20, 20]}},
{"id": "02-cylinder", "prompt": "Make a cylinder 30 mm in diameter and 50 mm tall.", "check": {"bodies": 1, "volume": 35342.92, "bbox_size": [30, 30, 50]}},
{"id": "03-block", "prompt": "Create a block 60 mm long, 40 mm wide and 10 mm thick.", "check": {"bodies": 1, "volume": 24000, "bbox_size": [10, 40, 60]}},
{"id": "04-sketch-extrude", "prompt": "Sketch a 50 by 30 mm rectangle on the XY plane and extrude it 15 mm.", "check": {"bodies": 1, "volume": 22500, "bbox_size": [15, 30, 50], "feature_kinds": ["Extrude"]}},
{"id": "05-plate-hole", "prompt": "Make a plate 80 x 60 x 10 mm with a 20 mm diameter hole straight through the middle.", "check": {"bodies": 1, "volume": 44858.41, "bbox_size": [10, 60, 80], "note": "80*60*10 - pi*10^2*10"}},
{"id": "06-fillet-box", "prompt": "Round all the edges of the box with a 3 mm radius.", "setup": [{"command": "solid.box", "params": {"length": 40, "width": 40, "height": 40}}], "check": {"bodies": 1, "volume": 63108.53, "tol": 0.003, "feature_kinds": ["Fillet"], "note": "all 12 edges of a 40 mm cube rounded r=3"}},
{"id": "07-move-body", "prompt": "Move the box 50 mm along the X axis.", "setup": [{"command": "solid.box", "params": {"length": 20, "width": 20, "height": 20}}], "check": {"bodies": 1, "volume": 8000, "bbox_min": [50, 0, 0]}},
{"id": "08-rename-body", "prompt": "Rename the body to Bracket.", "setup": [{"command": "solid.box", "params": {"length": 30, "width": 20, "height": 10}}], "check": {"bodies": 1, "volume": 6000, "body_name": "Bracket"}},
{"id": "09-volume-question", "prompt": "What is the volume of the part?", "setup": [{"command": "solid.box", "params": {"length": 30, "width": 20, "height": 10}}], "check": {"bodies": 1, "volume": 6000, "unchanged": true, "reply_number": "6[ ,. ]?000(?![0-9])|6(\\.0+)? ?cm"}},
{"id": "10-two-holes", "prompt": "Make a 100 x 50 x 5 mm plate and drill two 10 mm holes through it, each 15 mm in from the short ends and centred across the width.", "check": {"bodies": 1, "volume": 24214.6, "bbox_size": [5, 50, 100], "note": "100*50*5 - 2*pi*5^2*5"}},
{"id": "11-any-change", "prompt": "can you use the software to make a change to the current shape, any change at all", "setup": [{"command": "solid.box", "params": {"length": 30, "width": 20, "height": 10}}], "check": {"changed": true, "note": "verbatim user prompt from #4846; pass = the design changed"}}
]"#;
/// The agent the test adds. A fixed id replaces it on every run instead of adding another.
constAGENT_ID: &str="api-bench-assistant";
constAGENT_NAME: &str="bench";
/// Longest control reply read (the channel's own request limit).
constMAX_REPLY: usize=16<<20;
/// Most result data kept for `--json`.
constMAX_RESULTS: usize=64<<20;
/// Deadline of the requests that only set up or read (each is answered at once by the app).
| `naming-split-pieces` | 9a979cf | format 13 names: a slot splits the top; fillet, chamfer and offset face name its pieces by number (`F2:top#2`) |
| `move-bodies` | 00baf1d | `solid.move` before it could copy: two bodies, one translated in place, one translated and turned 90 deg about Z (no `copy` key anywhere in the file) |
| `split-body-pieces` | v0.3.2 (07e263d) | a slot along X cuts a block into two bodies; a fillet, a chamfer and an appearance refer to the pieces by body name (`Body1`, `Body1 (1)`) |
| `issue-489-stacked-join` | v0.6.0 (809ecf7), desktop | a 70 × 84 extrude, then a 70 × 62 strip joined on its top face, sketched partly over the face's projected edges (issue #489; expected values from main at 2880525) |
| `issue-616-lid-ribs` | v0.7.0 (a9879fa), CLI | a rounded lid with a skirt ring, a window and a slot cut through, then four ribs joined on the plate over the skirt, flush with the window's sides; v0.7.0 displaced the ribs and left a slit through the skirt and plate (issue #616; script `examples/regressions/lid_ribs_flush_with_window.json`, expected values from the fix) |
The scripts that made them are in `scripts/` (`issue-489-stacked-join` was drawn in the desktop
app and has none; `issue-616-lid-ribs` is made by its regression script). Every flat face of every final body must also take an attached sketch and an
offset plane; the faces that can't yet are listed with their issues in `design_corpus.rs`.
Never regenerate a file with a newer build: the point is that it was written by the old one. Add new files for new eras instead.
assert!(round.is_some(),"after kill {i} the design file is torn or missing");
seen.insert(round);
letbak=solvecraft_engine::io::backup_path(&path);
letbak=solidcraft_engine::io::backup_path(&path);
ifbak.exists(){
assert!(round_of(&bak).is_some(),"after kill {i} the backup is torn");
}
Loaded 100 of 5848 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.