Commit Graph
13 Commits
Author SHA1 Message Date
Irakli Kourtis 8cdefd4cd5 test: reference cases for variable, chord-length and tapered fillets (#776)
Four analytic reference cases for variable, to-zero, chord-length and taper-to-fit fillets, with the formulas in tools/oracle-analytic.
2026-10-13 12:35:16 -04:00
bherbruckandClaude Opus 5.5 2ebceb6918 Sheet metal: sew the folded-material guard per piece so adjacent flanges cut (#296) (#355)
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>
2026-10-12 14:27:40 -04:00
e12758ff91 Public reference tests: analytic and FreeCAD expected values (#310)
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>
2026-10-12 10:43:43 -04:00
bherbruck af536a7907 Rename displayed product to SolidCraft and add header bug reporting 2026-10-11 16:16:48 -04:00
bherbruckandClaude Opus 5.5 d10312b7fd Command ids are SolveCraft's own: 173 ids that were another product's internal command ids become dotted, lower-case ids (<area>.<thing>[.<variant>]: solid.extrude, sketch.rectangle.two_point, sheet.flange, joint.drive, ...), in the registry, toolbar, dialogs, shortcuts, palette, control channel, MCP, examples, scenarios and docs
The old ids live only in crates/engine/src/legacy_ids.rs (the legacy migration table): find_command maps them, so old scripts, recipes and control calls keep working, and saved shortcuts, recent and pinned commands are migrated on load (designs store no command ids). Parity maps the observed catalog ids to ours through xtask/data/fusion-ids.tsv. The agent rule now reads: our own dotted id, never another product's. Cargo.toml's repository points at github.com/bherbruck/solvecraft.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-09 12:17:23 +00:00
bherbruckandClaude Opus 5.5 f50d743c55 docs/oracle.md: 73/76 (35 passes; 50 area exempt)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AS5Tn3xieewUSg3uxrMVdL
2026-10-08 11:37:33 +00:00
bherbruckandClaude Opus 5.5 a8d54e0a3f docs/oracle.md: no-oracle section
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AS5Tn3xieewUSg3uxrMVdL
2026-10-08 05:50:50 +00:00
bherbruckandClaude Opus 5.5 20aa3ed4e1 Oracle: seam-normalised topology. When body count, volume and area pass but the raw counts differ, Fusion's faces/edges/vertices are normalised from its face, edge and vertex lists as SolveCraft's merged counts are (closed-surface patches split by seams count once, seam edges dropped, two-edge vertices joined) and compared exactly, with both raw counts in the note (56 and 61 pass)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AS5Tn3xieewUSg3uxrMVdL
2026-10-08 05:27:34 +00:00
bherbruckandClaude Opus 5.5 3229fd9740 Kernel: booleans with coincident planar faces push the faces apart exactly (perpendicular neighbours stretched or shrunk), union cuts the slab beyond the plane
Oracle: 29/29.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AS5Tn3xieewUSg3uxrMVdL
2026-10-07 22:41:29 +00:00
bherbruckandClaude Opus 5.5 f5c2c725c8 Extrude/revolve "all profiles" means the drawn ones: auto-projected face edges only count when the drawn curves make no profile
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AS5Tn3xieewUSg3uxrMVdL
2026-10-07 21:58:18 +00:00
bherbruckandClaude Opus 5.5 431b87c2b9 M1: hole feature (simple, drilled, counterbore, countersink), cone classification and apex counting
Oracle: 19/29 parts match Fusion.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-07 19:22:27 +00:00
bherbruckandClaude Opus 5.5 b85ccd62cf M1: oracle recipe coverage (arcs, refs-based constraints/dimensions, extents), taper and through-all extrudes, curve splitting in profiles, region merging, angle dimensions by sector, smoother tangency solve, merged topology for revolves
Oracle: 13/29 parts match Fusion.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-07 18:19:28 +00:00
bherbruckandClaude Opus 5.5 0f3e7701a7 M0: docs (README, CLAUDE/AGENTS, ROADMAP, control protocol, NOTICE, attribution), example, parity and oracle reports
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-07 17:58:32 +00:00