smlib's run() catches a solidcore panic and replaces the kernel (kept: the app does not die
natively), returning Error::Internal; the kernel turned it into KernelError::Failed and the
engine into Other, so hostile_params_never_panic and the fuzzer no longer saw kernel panics.
Now the kernel's wrapper errors go through KernelError::smlib, which keeps Internal; a body the
wrapper could not take because of a panic is Internal too. A feature whose evaluation hit it
records the result as internal (not cached: the kernel was replaced), and the command that
made it, or any command whose kernel call returns it, fails with EngineError::Internal under
its own id. Crash reports mark a panic inside a solidcore call recovered (kernel::guarded), and
the crash test's kernel mode panics inside a solidcore call.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
The kernel crash test's command must fail with EngineError::Internal naming
diagnostics.crash_test, as a panic in the command itself does: the panic detectors
(hostile_params_never_panic, the fuzzer) look for Internal. Today it is EngineError::Other.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
solidcore's algorithms recurse (subdivision, intersection, filleting) and its README asks for
8 MiB, a native main thread's stack; Rust starts other threads with 2 MiB, and only the web
build set more. kernel::KERNEL_STACK is 8 MiB and kernel::kernel_thread a thread builder with it.
The kernel's parallel map runs on a pool of its own with that stack (on a worker of a pool
already, as in map_with_local_pools, it stays on that pool as before), the one-thread mesh pools
and map_on_threads' helpers have it, and so do the engine's interference worker and the
evaluator's plan-mesh thread. A test recurses 5 MiB deep on each.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
While a dump folder was set, every call serialised each input body before it ran, so the rig's
timings (taken with a folder set) measured the serialiser as much as the kernel. The callers now
hand run_on the bodies the call reads; they keep them until its result is known, and only a
failing call serialises them, into the folder as before. A test counts the serialisations: an
answered volume and union none, a failing call one per input.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
The check had become one-sided after the C++ kernel's removal: a case that started passing,
or a change in the count, went unnoticed. As on main, a case that now passes fails the test
until it is added to the list, one that no longer passes is a regression, and the count is 111.
At solidcore 8ddce9016 solidcore passes exactly the 111.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
The workflow triggers on the root Cargo.toml and Cargo.lock, and the plan gets their diff
(--pin-diff): a changed line that names solidcore's repository (Cargo.toml's git rev, Cargo.lock's
source) selects every area and the ladder. A path filter cannot tell a pin from another
dependency's change, so the plan reads the lines; another dependency's change runs no area.
The fillet rig is not triggered: the kernel rig's fillets shard runs the same cases.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
A solidcore bump changes only the root Cargo.toml and Cargo.lock, which neither PATHS-ALL nor
the workflow's trigger paths list, so it ran no rig. The tests want a change to the solidcore
lines of those files to run everything, another dependency's change in them to run nothing, and
the workflow to trigger on both files.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
A shell, offset or thicken solidcore's one construction refuses because a face's offset folds
(SolidError::ShellRefused, ShellRefusal::Fold) is Error::WallFolds with the largest wall it
names; the kernel says "the wall is thicker than a face curves, so the face's offset would
fold; the largest wall that works is 10 mm". A blend too large carries solidcore's NoRoomReason
as Error::TooLarge's RoomReason, worded in its message before the largest size that fits:
a collapsed offset (solidcore#2431) says the ball is as wide as the face it rolls on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
A thicken of solidcore#2350's parabolic sheet past its apex's radius of curvature is refused
by solidcore's one construction (ShellRefused, Fold, naming 10); the wrapper passed it on as
"solidcore: ..." with no words of its own. A round at the radius of the round it rolls on
beside the 851 part's arc wall is refused as a collapsed offset (NoRoomReason::OffsetCollapses,
solidcore#2431); the message named the size but not why. Both fail until the reasons are
carried through.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
At solidcore 8ddce9016 three equal convex rounds at a vertex of three planes are the blend
corner's Ball (fillet_blend_corner). The wrapper set a setback on the corner after the corners
were made, so a set-back vertex was already a Ball (NByTwo) and its setback was refused
(CornerRefused): a_setback_corner_left_open_returns_its_hole (both) and setback_corner's
closed form. solidcore's AddSetBackVertex, called before the corners are made as the fork's capi
round_edges does, keeps a set-back vertex out of the Ball, and its corner is set back.
The corner a setback is set on may also be a blend corner whose event is the Setback, as the
capi takes it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
At solidcore 8ddce9016 a Ball corner is a BallPatch (fillet_blend_corner) and a round between
NURBS faces a ProceduralBlend (fillet_round_as_definition). The arrays' B-spline part of a face
had neither, so their export fell to the approximation and failed ("an approximated face surface
that is not a B-spline"): the rounded cube, the variable and curvature rounds and the
all-edges box in the smlib tests, and the Ball-corner and round cases of the rig.
Both are their nets, exactly. A ball patch also goes out as SURFACE_BALL_PATCH, its sphere's
centre and radius in FACE_DEFINITION_REALS, as the record's BallPatch reads it, and comes back
as solidcore's BallPatch on that net: the rounded cube's eight corners stay on their spheres
through the record and back.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
solidcore main at 8ddce9016 adds cone_unbounded_parameterized_as_nurb,
ellipse_open_arc_keeps_end_angle (solidcore#2422), fillet_round_as_definition,
fillet_round_on_round_as_definition, fillet_overflow_crease_along_edge,
plane_first_torus_section_exact, keep_exact_pcurves and shell_one_construction
(solidcore#2367), and drops fillet_extended_rail_end_at_side_edge_crossing.
Every one is on. fillet_keep_features is no longer held back: solidcore#2374
fixes the round beside a boss it refused (solidcore#2370).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
A block's front top edge rounded r 3 beside a boss joined on its top fails SM_ERR_INVALID_INPUT
with it on; off it builds the closed form, 6099.225530 mm³, as the C++ fork does. Found by
ui-egui's opening_a_large_design_never_stalls_the_window.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SdcgEu4Lr3moCCisdZShKz
solidcore's fit is the largest size less a margin (34.99 where the L bracket's bend builds 35,
4.29 where #3435's ring builds 4.3), so the feature's search keeps naming the size, and solidcore's
is the answer only where the search finds none. The refusal keeps its words, "larger than the
faces beside its edges allow", with the edge after them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SdcgEu4Lr3moCCisdZShKz
solidcore sizes a blend it refuses as too large (the session's fit, SolidError::FilletRefused under
fillet_gn_section_generator). The smlib crate gives it as Error::TooLarge with the size, what it
sizes and a point halfway along its edge (an equal chamfer's ball radius turned back into its
distance); the kernel words it as the fillet feature's search does (largest_fit_note), and the
feature runs no search when the refusal names it. The 3x2 mixed corner whose rails cross, which
solidcore refuses (fillet_refuse_mixed_corner_crossing_rails), is not supported yet, citing
solidcore#2256.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SdcgEu4Lr3moCCisdZShKz
With it on, 11 rig cases that pass with it off are refused as breaking through the part, and
190-fillet-disc-rim-past-its-2mm-wall fails status 1001. A fix under the same flag is in progress
on solidcore.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SdcgEu4Lr3moCCisdZShKz
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360).
- b2fd08e85 jyviko: Profiler: no smlib category left in its tests and help
- d49a7eac0 jyviko: Known-failing tests: the convex chamfer that would grow the part passes
- d55fedd56 jyviko: smlib export test: a rounded cube's corners checked on solidcore's spheres
- 017a2a607 jyviko: Known-failing tests: the fillet into an earlier round passes on solidcore
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
2 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360).
- 19db2d7c0 jyviko: Docs and comments: solidcore is the kernel
- f1ee045de jyviko: smlib: rustfmt after the C++ kernel's removal
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
1 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360).
- 7ad1d3e12 jyviko: diagnostics.kernel takes a typed contract, as main's other diagnostics commands do
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- b8fa0d8aa brett19: Known failing on solidcore after the f916a8d sync; blend_fit in the entry-point guard
- 5a93e6c4d brett19: Make the C++ SMLib kernel opt-in at build time (feature smlib, off by default)
- 29a148abd brett19: CI: build without the C++ SMLib kernel except where a job needs it
- e30cfd2ac brett19: Tests and the rig for a default build without the C++ kernel
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 4c1eed960 brett19: Tabbed plate: the rim and one corner arc round together (solidcore#2223)
- dd00ff989 brett19: Native wire fixture: a box edge round's cylinder keeps its own frame
- 584075662 brett19: Known failing on solidcore after the smlib merge: SMLib 5e8c9c1's new blend (solidcore#2273) and main's new tests
- 8af1fdf5a brett19: Known failing on SMLib (the .smlib list): as the smlib branch lists them at f916a8d
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- f9fc68167 brett19: solidcore: a curvature fillet is the equal-area Gⁿ section; mixed wedges refused
- d52cf6b14 brett19: Known failures at solidcore adf26ddbc: curvature and variable rounds built
- 5ec92b0ba brett19: Move the solidcore pin to 2dbd88b6e
- 92618ac4c brett19: Turn on solidcore's new fixes at 2dbd88b6e
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- e30e983c3 brett19: Move the solidcore pin to 4d3ce6264
- c8351719e brett19: Move the solidcore pin to adf26ddbc
- d33b6eb30 brett19: Turn on fillet_gn_section_generator
- 3b4adeb73 brett19: solidcore: a variable round is built again, as the union of its balls
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 9f64cce54 brett19: solidcore: a variable-radius fillet is refused until solidcore builds the union of its balls
- bd4dc081d brett19: Known failures on solidcore after the merge: the refused curvature and variable rounds, develop's new cases
- c6fa7e563 brett19: Move the solidcore pin to c9a06dc7c
- de9a654db brett19: Turn on solidcore's new fixes at c9a06dc7c
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 02de15f90 brett19: Move the solidcore pin to 9fb13f145
- 09b646005 brett19: Turn on solidcore's new fixes at 9fb13f145
- c494a18f9 brett19: Rig: the reimported peg's rounds fail as the peg's do
- e9ac0e606 brett19: solidcore: a curvature fillet is refused until solidcore has the equal-area Gⁿ section
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- a1b4119c3 brett19: Turn on geometric_g2_continuity: 2992's slowdown is gone
- f2ce343a0 brett19: Known failures after the new fixes: four tests and a rig case pass
- b57c24f7f brett19: Known-failing tests: the tabbed plate cites solidcore#2223
- 0aff35705 brett19: Merge fix: solidcore reads planar sheets without SMLib, so their edge reader is always built
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- fb851d9c2 brett19: Known-failing tests: main's four tests that fail on solidcore as on C++
- 6de10a7fb brett19: Model cache test: say why it reaches the C++ kernel
- 84bca5969 brett19: Move the solidcore pin to c931c9641
- e1b76bdfc brett19: Turn on solidcore's new fixes at c931c9641
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 15c5cd5d5 brett19: Known-failing tests: cite the solidcore issues now filed
- 462738b6a brett19: Known failing after the smlib merge: the r5 test-peg corner on solidcore, main's C++ test lines
- 4a98a1f8a brett19: Model cache: a saved model is its kernel's
- 9c2388ef8 brett19: Geometry kernel strings in Vietnamese, now a complete catalog
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
1 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- afb55d806 brett19: One kernel flag: solidcore, the default, or smlib; routed is gone
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
2 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 3908128a8 brett19: Known-failing tests: refreshed from the default run at solidcore de85a3e00
- cd94f78f3 brett19: Rig: --route is gone; a run has one kernel
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 75229f2b9 brett19: Offset tests: a sphere moved to its centre leaves nothing again (solidcore#2092 fixed by #2096)
- 6221b7f4b brett19: Rig: solidcore known-failing lists at pin de85a3e00
- b04e363e7 brett19: Rig: #316's renamed shell cases pass on solidcore, off its known-failing list
- c786db6bf brett19: A blend's pick measures close edges on their curves, not their polylines
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 197cd4f43 brett19: Booleans: the copies no longer take an index
- 0f843c295 brett19: Pin solidcore at main de85a3e00 (#2080, #2081, #2086, #2095, #2096, #2097, #2102, #2103)
- 0b48548af brett19: Turn on solidcore's new fixes and take the rational-span flag off HELD_BACK
- d0c95f0dc brett19: Chamfer: second offsets through SetFilletParameters (solidcore#2095)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 1bf9c8665 brett19: HELD_BACK: geometric_g2_continuity is a fourfold slowdown on 2992, not a hang
- 126a281bb brett19: Pin solidcore at main c093bc5a3 (#2054, #2060, #2061, #2065, #2066, #2068, #2069, #2072, #2076)
- 5d59d42d8 brett19: Turn on gap_samples_non_bspline_v_stride, gap_array_append_updates_longest, imprint_skips_collapsed_definition
- 1c90e60aa brett19: Working copies by solidcore's brep_copy, not our own key domains (#2047 fixed by #2055)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 32b26888c brett19: Turn on offset_keep_cavity_regions and offset_four_face_vertex_split
- 38849d7b0 brett19: Hold geometric_g2_continuity: with the rational-span flag off, 2992 and 3368 don't end
- 5d7dca1a9 brett19: Offset tests: both kernels refuse a top moved -9; the sphere to its centre waits on solidcore
- da4db59a3 brett19: HELD_BACK: geometric_g2_continuity's reason is 2992 alone
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 2b1a63d3f brett19: Pin solidcore at main 2898e1197 (#2053)
- 9337aff4c brett19: Every solidcore fix on unless it appears broken (brett19's rule)
- 2cd566aae brett19: Hold bspline_trim_keeps_rational_end_spans: it fails 3078's check
- 9df34988c brett19: Pin solidcore at main bb6b65706 (#2049, #2051, #2055)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- eff6416d4 brett19: Pin solidcore at main f75e9d2e8 (#2023, #2024, #2034-#2040, #2043-#2045)
- 15e3a61da brett19: Pin 8: decide the new solidcore flags against the fork at 26cc451
- 6a0166853 brett19: A round mitred to one that overflows goes first with it (fork 9ba2b12); flags() and warnings
- d4eb66f25 brett19: Turn on offset_splits_corner_vertices (fork 2e1bb8c's behaviour)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ
4 commits, squashed; the originals stay reachable from branch jyviko/merge-main-develop-solidcore (#11360). Its changes conflict with main as it is, so this wave keeps the previous files; the next wave that merges cleanly holds them.
- 8ad306e69 brett19: boolean2d through boolean_2d, typed edge-vanishes, typed put-back refusal
- b6285e525 brett19: Overflowing rounds go first again (solidcore#1980)
- 2647b12e5 brett19: Hold back fillet_refuse_crossing_rails: not in the fork
- 0b627d637 brett19: Restore all_solid's loop: an unused index slipped in
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DtL5M41G8Nua7HvuFPZkmJ