boolean2d goes through api_brep::boolean_2d, SM_ERR_OFFSET_EDGE_VANISHES through
is_offset_edge_vanishes (solidcore#2016). A mitred round that cannot be put back
(solidcore#2027) is Error::Unsupported. The tests held for solidcore#1462, #1973,
#1980 and #1981 run again. copy_of stays as it is: results of offset_faces on a
brep_copy (NamingScheme::Notify) fail check_brep.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
On: fillet_corner_face_curve_halves, brep_edit_statuses_pass_on,
fillet_run_on_bspline_trims, offset_tangent_gap_statuses and ssi_alignment_statuses
(both soft like the fork since solidcore#2017). The other new flags are held back,
each with the fork's source as the reason.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
The fork places a recognized plane's origin after squaring its axes (0286682, in SMLIB_REV;
SmPlane.cpp:2836-2852 at 2fc375f), and solidcore#2004 brings the same behind this flag. It closes
sweeps/4305's Bad Gap and moves sweeps/5037 to SMLib's volume.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
`shell_brep_simple`, `offset_brep_simple` and `local_operation_simple` carry the same options and
the source's attributes, so the 120 lines of key domains and notification stores in core/offset.rs
go (gap-audit D2). The held-flags comment now says what each flag's fork code does. Ignored tests cite
solidcore#1980, #1462, #1973 and #1981 instead of local files.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
Reading the fork at SMLIB_REV finds `perspective_sphere_offsets` is there (709c5be: the cull projects
each sphere's own radius from its centre), so it goes on. The new
`gap_samples_use_trimmed_spans` and `twist_simple_duw_sign` are held: upstream's defects are still in
the fork.
The tabbed plate and #1997's rounded box are compared with the C++ kernel. `sweeps::List` is visible
to the crate for that.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
On, their fork commits in SMLIB_REV: `fillet_variable_span_follows_fillet` (4438859),
`simple_coincidence_crossing_is_point` (c2e3d9d), and from solidcore#1775 the statuses whose fork
passes do what the flag does: `boolean_temporary_transform_statuses` (e7ebc8d),
`primitive_path_curve_statuses`, `planar_path_sweep_statuses` and
`planar_path_profile_placement_statuses` (c8f6a31), `fillet_trim_status` (9b7cd48). The last two
were held as having no fork commit, which was wrong.
Held: `fillet_corner_statuses`, `fillet_edge_curve_domain_statuses`,
`offset_tangent_gap_statuses`, `ssi_alignment_statuses` (the fork carries on softly where these
refuse) and `translation_sweep_regularize` (no fork commit).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
`face_tight_box_searches_trim` goes on now that solidcore#1961 brings fork 326de73.
`ssi_low_angle_start_on_probation` goes on: fork 30d25bf was rebased to 1f5c0f7, in SMLIB_REV.
Three port fixes and the new `extent3d_contains_both_segment_endpoints` stay held: the defects are
still in the fork's source at 2fc375f.
`measure::check` no longer matches `check_brep`'s operation string. solidcore does not raise that
error (a Caller body without keys is checked under arena identities), so the arm was dead; the
refusal is decided on the typed `NamingScheme` alone, and a test holds it there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
On: fillet_mitre_gap_patch (fork 2223108), fillet_run_on_past_far_end_only (7224896), good_drop_result_distances_in_order (eaa998a). Held back: geometric_g2_continuity and solidcore's new flags with no fork commit in SMLIB_REV.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
The ordering (fork 241f7ba) is ported but off (OVERFLOW_GOES_FIRST): solidcore's cliff overflow loses the attributes of target edges its subtraction replaces, so a shortened picked edge reads as removed and is silently not rounded (2 rig cases wrong). Tests: the capi's rounds_at_a_concave_face_corner_are_mitred; overflow_goes_first (ignored, gives 6107.529369565 with it on); two ignored solidcore-only repros.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
L1: GetWireEdges uses the process kernel, not a fresh default one.
L3: a face whose surface the B-rep's list does not hold is refused, as smlib_flat.cpp:1520.
L4: a hyperbola or parabola without its NURB makes one on a copy, as GetOrCreateGwNurbPointer.
L5: every rejection of the C++ Validate is in ours, read line by line; a mutation test against
the C++ kernel holds it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
Flags on: fillet_cliff_overflow, fillet_3x2_tangent_side_edge_roles, offset_extension_as_itself, fillet_mitre_concave_face_corners (fork commits in SMLIB_REV 2fc375f). Held back: the port's own fixes with no fork commit. Ports capi blend_call / overflow_first (fork 241f7ba) onto the hooks of solidcore #1929, with blend_dropping routed. From ws/cliff-1905: blend_failure on NoRoomReason, fillet_over_a_cliff un-ignored, a_cliff_overflow_the_fork_refuses_has_no_room.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
Closed edges including moved()'s open.is_empty() branch, a curved face split across its
seam, union_all of three, merges with several origins on each side, sheet offset, a
curved sheet's thicken and a shell of a cylinder match C++. Two tests are ignored for
solidcore differences written up apart: boolean_2d carries no attributes, and a flat
sheet's thicken loses the edge tags on the sheet's side.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
smlib_history.cpp:246 counts any failure of SmApiEdgeClosestPoint as off the piece, so
the face is Modified. The port passed on an error that was no status, which failed a
boolean that had succeeded. An edge without its curve shows it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
Found comparing drafts while checking M2: the capi's status (smlib_shim.h:82) answers
SM_ERR_OFFSET_EDGE_VANISHES (1043) as SMLIB_ERR_UNSUPPORTED; the port passed on Status(1043).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
Port the capi's moved_faces (smlib_ops.cpp:156-165, :1078), its mark on the picked
faces as solidcore#1875/#1877 lay out, since a surface has no identity apart from
its owning face. The reason it was skipped (solidcore#1462) was stale. No draft
found reaches SMLIB_ERR_NO_EFFECT on C++, so the parity test holds the same answer,
volume and counts over the nearest cases, and a unit test holds the reading.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
C++ -> arrays -> solidcore -> arrays -> C++ over a filleted box, a holed plate,
a swept B-spline, an ellipse edge, a sealed void, sheets, wire edges with a point
shell, two lumps and a shelled rounded box. The reference is the C++ kernel's own
crossing, which itself moves trims by about 1e-7 on a shelled box; tolerances and
EXTENT are excepted, and the test says why. No port difference found.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
face_tight_box_searches_trim lacks fork 326de73 and good_drop_result_distances_in_order lacks
eaa998a's tight convergence test. With them on, 15 thread cases fail on their tight box and the
peg's round is refused; both agree with SMLib eaa998a with them off.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
Port of the capi's close_onto_shared_edges in core/merge.rs: edges still free after the manifold
stitch are glued at the same tolerance to the edges they lie on, whatever uses those
(allow_spine_edges), on a copy. Faces then between two material regions go. The copy is kept when
no edge has more than two faces. A gluing that leaves fewer material regions than it found is
refused with Error::Unsupported.
Tests in core/merge_tests.rs mirror capi/tests/sheets.cpp, on solidcore and on C++: a biquadratic
cover over a 10 mm cube sews into a closed solid of 6 faces and 1100 mm^3; a bicubic cover lying
on the top is refused; a flap off a top edge stays beside the cube, 7 faces, 1000 mm^3, 700 mm^2.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MNXLXkZLQ6T7TSi4wswcmk
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>