The dihedral range and mesh.quality's scaled Jacobian were both right:
they are computed by the same function on the same final corners (tet4
and tet10 give identical numbers), but a dihedral range does not bound
the corner scaled Jacobian. A needle has dihedral angles near 60 and 90
degrees and a nearly flat apex corner. Before 417c0c88 sphere-on-plate
at size 4 had dihedrals 36.3-141.6 and 106 needles below SJ 0.2 (min
0.095); now 36.0-137.6 and min SJ 0.351.
The mesher's report now carries min_scaled_jacobian so it cannot read
as all-good while needles remain.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
The improvement objective was the smallest dihedral sine, which misses
needles: every dihedral angle fine while a corner is nearly flat, so
the corpus meshes had scaled Jacobians down to 0.01-0.07. The objective
is now the smaller of the dihedral sine and the scaled Jacobian at the
worst corner (the measure analysts check, Knupp 2001), weighted so a
Jacobian of 0.3 (the skill's limit) counts as a 30 degree dihedral
angle (the schedule's target): the Jacobian decides only near that
limit. The existing edge removal, multi-face removal, smoothing and
insertion then remove the needles.
Skill tet10 min SJ: sphere-on-plate 0.092 -> 0.335, filleted-bracket
0.021 -> 0.318 (both pass, nothing marked). Default-size tet10:
thin-ribs 0.240, shallow-groove 0.321, rounded-box 0.311.
The strong-relaxation test relaxes below 0.45: nothing is left below
0.3. Literature note in docs/meshing/literature.md.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
SolidCraft lists a solid's faces in a different order from run to run
(the rounded box, for one), which reordered the triangle soup and the
face tags with every coordinate the same; welding keeps the first copy
it meets and every later stage visits ids in order, so the tet mesh
changed between runs. CanonicalSoup sorts the triangles by their
coordinates, numbers vertices by first use and face tags by first
appearance, and maps the tags back; mesh.tet.generate and the corpus
use it.
Tests: the washer's tet10 mesh has a fixed fingerprint (identical on
arm64 and x86-64 builds); a reordered, renumbered input gives the same
mesh; heavy: rounded-box and filleted-bracket mesh bit for bit the same
on repeated runs and with 1 and 8 threads.
The bracket oracle cases stay unfrozen: SolidCraft's tessellation
itself differs in the last bits between arm64 and x86-64. The coarse
case moves to 203.6 MPa with the canonical order; its band is 5%.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01UhhZMBTviDQxZvwnyP89fo
Segment-to-segment penalty contact between element faces: each secondary
face and each main face facing it are projected onto the secondary face's
plane, split into triangles, clipped (Sutherland–Hodgman) and integrated
with a three-point rule per sub-triangle (Puso–Laursen segmentation, the
penalty at the integration points as Fischer–Wriggers), so a uniform
pressure gives the consistent nodal forces on both meshes and the contact
patch test passes on non-matching meshes. Every contact runs both ways at
half the penalty: pairs with each surface as the secondary, general and
self contact face against face. A bucket grid of face boxes every
increment, faces integrated in parallel and their forces added in face
order: the same numbers on any thread count.
The penalty per area ε = s γ min(m/A)/Δt² comes from the lumped masses,
each node's directional area and the stable increment, bounded by
Gershgorin's row sums so contact never shortens the increment. Coulomb
friction with exponential decay by a regularised return map at the
secondary nodes; rigid walls take Coulomb impulses. Contact work and
friction dissipation join the energy balance; each contact reports its
peak force, first and last contact, deepest penetration, friction and the
nodes adjusted at the start (initial penetrations adjusted strain-free or
refused with the node, face and depth); a cap on contact faces.
Verified (crates/explicit/tests/contact.rs, tests/oracle/explicit-*):
colliding bars (2L/c +1.2%, ρcAv₀/2, KE kept to 0.005%, contact energy
back to 8e-5 of it), blocks sliding to a stop on a rigid wall and on a
plate under constant and decaying friction (0.005–0.2%), the patch test
(σ_zz uniform to 0.06%), two spheres colliding against Hertz (peak +2.3%,
duration −0.4%; the oracle case waits for a graded sphere mesher), a free
body untouched by general contact, determinism across threads. Oracle
measures gain `over` (ratios). Existing fixtures do not move.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
The bracket's tet mesh is not bit-identical across platforms (#115): its fillet peak is 213.31 MPa
on macOS and 212.42 on Linux CI, which turned the oracle job red after #110. Those two quantities
are no longer frozen; the page names them mesh-dependent instead of printing one platform's value.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
The end-to-end demo met a raw-CDT sliver that the validity gate accepted (exactly positive) and
the solver refused. Now:
- One validity criterion, `fieldcraft_model::admissible_jacobian` (normalized Jacobian above
1e-12), used by the implicit solver (fem `positive_inverse`), the explicit tet4/hex8
kinematics and the mesh gate, which proves it on the whole element (tet4: the solvers' own
Jacobian; curved kinds: a Bernstein floor), so whatever passes the gate passes the solvers.
- CDT rounding repaired by flips on the rounded mesh (Diazzi et al. 2023 §5.1) instead of the
±2-ulp nudge.
- `fieldcraft-mesh::tet`: a mutable tetrahedral complex whose every change is a checked
replacement (positive and admissible tetrahedra, faces matched with opposite orientation) with
an exact undo log; edge removal (Klincsek), multi-face removal, Delaunay insertion with
boundary faces, edge collapse; a curvature- and thickness-graded size field; constrained
Delaunay refinement with encroachment splits; ODT smoothing; the Klingner–Shewchuk schedule;
boundary vertices moved onto the exact B-rep surfaces at the end; tet10 with mid-edge nodes on
the geometry.
- `mesh.tet.generate` gains size, order, quality, quality_target, curvature_angle, grading,
max_passes, max_points and reports the CDT's and the mesh's quality; `solve.probe` takes a set
and reports its largest von Mises stress.
- Oracle cases filleted-bracket-fillet-stress(-coarse): the fillet's peak von Mises stress on
tet10 meshes of 2k and 13.5k elements, 213.3 MPa (a four-level study stays within 1.8%); the
CDT cases keep the raw tetrahedralization.
- Corpus quality columns: all 20 valid parts mesh and pass the gate.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
The corpus runs mesh.hex.octree's mesher on every part at 1/25 of its diagonal (the size of
the structured hex column) and reports hexahedra, the smallest and mean scaled Jacobian, the
mesh volume against SolidCraft's and the time; a part it cannot mesh gets the reason. The
column caps the elements at 500 000 so a part whose grid explodes is refused in seconds.
Loaded parts carry their bounding-box diagonal for the stages. docs/meshing/corpus.md
regenerated: 15 of 21 parts mesh, including filleted-bracket, rounded-box, sphere-on-plate
and the analytic sphere.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
filleted-bracket-blocks splits the bracket with SolidCraft's Split Body
into three slabs across its width, the fillet corner halved along its
diagonal and a four-cell O-grid round each hole; the corpus runner meshes
partitioned parts as one conforming block mesh. A heavy test clamps the
upright, loads the foot and checks hex8 and hex20 converge (hex20 tip
deflection within 0.1 % of a fine tet10 solution). Roadmap M5:
conforming interfaces done.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
Crates fieldcraft-*, binaries fieldcraft and fieldcraft-cli, the .fieldcraft file (format tag
fieldcraft/1, manifest fieldcraft.json), FIELDCRAFT_* environment variables, the MCP server
name, egui widget ids (fc_*), the app icon document FieldCraft.icon and its generated family,
docs and licences. No compatibility shim: no files of the earlier name exist.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
Main's oracle job failed at 8cba3a7: scordelis-lo-roof-solid u_z moved from
-0.3014029226397275 (blessed before #50) beyond its 1e-9 band. The case is a
thin shell (R/t = 100) of one layer of hex20R solids, ill-conditioned, so
rounding alone moves u_z by about 1e-9 relative: on macOS #50's multithreaded
supernodal factorisation moved it by 8.2e-10 (to -0.3014029223929), #52's
interpolation order by a further 1.3e-10 (to -0.3014029223541), and Linux
CI by slightly more, past the band. #52 kept the lattice geometry the same.
The result itself is unchanged at the 1e-9 level (0.33% from the
reference).
A quantity may now set `fixture_relative` (with a required
`fixture_reason`) for a wider fixture band; Scordelis-Lo uses 1e-7 and is
re-blessed. Moved-fixture messages print full precision.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* mesh: one transfinite interpolation for mapped hex and mesh.block; stable oracle report
mesh.block's lattice is now built from the structured hex mesher's
transfinite code: faces by quad::transfinite_grid, the interior by the new
quad::transfinite_volume (Gordon-Thiel, blended by the block's graded
parameters), which hex::mapped now calls too. 2D patches are the
transfinite_grid of their sides. Edge curves return their corners exactly.
Oracle: a zero-reference quantity (rounding noise) keeps its absolute
tolerance as its fixture band, and docs/oracle.md shows such noise and
relative errors inside the fixture band as bounds, so the report no longer
changes with the order of floating-point operations (it failed the
'docs/oracle.md is current' check on main after #50). Scordelis-Lo moved by
1.4e-9 relative (an ill-conditioned thin shell of solids: sub-ulp node
changes) and is re-blessed.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* simplify: shared face filling and lattice index; faces blended like the volume; exact ends for every edge
quad::transfinite_faces fills a block's six faces (with a per-face hook
that hex::mapped uses to project onto its B-rep faces) and
quad::lattice_index fixes the lattice layout for all three users.
transfinite_grid_blended blends faces by the block's graded parameters, so
faces and interior use one parametrisation and mesh.block reproduces its
earlier Gordon-Hall lattice (the Scordelis-Lo fixture is back to its
blessed value). Straight edges also return their corners exactly.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
---------
Co-authored-by: jyviko <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
* Structured hex meshing: multi-level extrude and revolve, one-to-one sweep, mapped blocks (M5)
modelcraft_mesh::hex::hex_mesh meshes a face-tagged boundary surface with
hexahedra by the first structured method that fits: a straight or
rotational multi-level sweep (imprinted cross-section arrangement, cuts at
reflex corners, mapped/annulus/free regions), a one-to-one sweep through
rings of linking faces (least-squares affine layers with residual
correction), or a mapped logical cube (trivariate transfinite
interpolation). Interval assignment couples shared curves; boundary quads
carry B-rep face tags; the certified validity gate, a tiling check and a
fidelity check guard the output; hex20 mid-edge nodes go onto the faces.
Shape::face_surfaces exposes the analytic face surfaces; the corpus runner
gets a structured hex column.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* mesh.hex.generate: structured hex meshing as a command
Meshes one body, every body of the open design or an imported surface,
replaces the mesh, tags element faces with the B-rep faces (so sets by
geometry resolve) and reports per body the method, counts and quality.
The quality pass certifies each element once and feeds the validity gate
(quality::validity_gate_checks). Shape::size gives the cached body size
for defaults instead of a full measure.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Hex: disk O-grids, convex interval cost, early size guard, hex20 corpus column
Smooth single-loop sections (cylinders, tori, vases) get a butterfly O-grid
instead of the free mesher; interval assignment raises counts by the
squared relative error so equal sides share increments evenly; a mesh far
beyond the element limit is refused before any work and mapped limits are
no longer swallowed. The corpus reports hex8 and hex20 quality.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Catalogue: structured hex commands ready (mesh.hex.sweep, mesh.hex.mapped, mesh.hex.generate)
mesh.hex.sweep and mesh.hex.mapped run mesh.hex.generate with the sweeping
methods or the mapped one; mesh.hex.generate is the sweep entry's related
command. The planned M6 octree entry moves to mesh.hex.octree (its icon
renamed with it). docs/features.md regenerated.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Docs: structured hex literature (B8), architecture, roadmap, parity; corpus hex column
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Hex: simplify review fixes
Prepare each sweep frame's cross-section once (arrangement, cuts, fill)
and reuse it across refinement retries; candidate frames come from one
setup pass. Edge triangles come from the feature edges, layer counts go
through size::edge_divisions, sweep residual weights are computed once
per node, flux shape functions live on the stack, and the arrangement's
exact side test and distances serve the region probe. The engine takes
the boundary options mesh.tet.generate takes and names unnamed faces by
index like it; the corpus hex20 run reuses the winning method; the
mapped entry lists mesh.hex.generate too; the octree icon is redrawn
under its new id.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Regenerate docs/features.md and docs/meshing/corpus.md after rebasing
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Regenerate docs/features.md and docs/meshing/corpus.md after rebasing onto main
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Regenerate docs/features.md and docs/meshing/corpus.md after rebasing onto main
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Regenerate docs/features.md and docs/meshing/corpus.md after rebasing onto main
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Regenerate docs/features.md after rebasing onto main
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Regenerate docs/features.md after rebasing onto main
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* Regenerate docs/features.md after rebasing onto main
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
---------
Co-authored-by: jyviko <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Weld and face-to-face check move from the mesh.block command to
modelcraft_mesh::join (bounding-box-limited weld, grid-binned hanging-node
test), for node merging and imports to reuse. mesh.block and
mesh.tet.generate share fresh_labels. The block map evaluates each edge
curve once per lattice parameter; vector helpers come from vec3; tet10 and
hex20 mid-side nodes follow ElementKind::edges. Session::run_script returns
each step's output, which the oracle reads instead of its own loop.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
The oracle binary (modelcraft-testkit) runs each case's recipe through the
command engine, measures its quantities by JSON pointers into step outputs
or commands, compares them with the cited references and the frozen
fixtures, and writes docs/oracle.md. 20 theory cases (patch tests of every
solid element, Timoshenko cantilevers, Lame, Kirsch, beam frequencies,
torsion and bending through couplings, explicit waves and impact, CDT
invariants), 8 published benchmarks (MacNeal-Harder patch, straight and
curved cantilevers, thick cylinder, Scordelis-Lo with solids, NAFEMS LE1,
LE10, FV32) and 10 pending cases with citations.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* test: hostile-params test runs under small session work caps
Session gains Limits, the defaults of mesh.tet.generate's max_steiner and
max_tets (production values unchanged; a parameter still overrides them;
file.open keeps them). The registry-wide hostile-params test runs on
sessions with small caps, so the needle-triangle tessellation it meshes is
refused at once instead of meshing to the 50k Steiner cap: the engine test
binary drops from ~350 s to ~20 s locally.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* ci: move heavy tests to a heavy job on main, nightly and on demand
Four slow tests are #[ignore = "heavy: ..."] (every corpus recipe, the
cantilever convergence study, the explicit-vs-modal cantilever period, the
generic-predicate rational cross-check); fast representative tests of each
area stay in the PR gate. cargo xtask test --heavy runs every ignored test
(benchmarks included) and the doc tests; cargo xtask test and ci run the
rest without the doc tests. The new heavy CI job runs it on pushes to main,
nightly and on workflow_dispatch. Branch jobs restore main's rust cache and
no longer save their own. AGENTS.md documents the rule.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* test: the three slow LBM verification cases run in the heavy job
The Stokes-drag sphere array, the staircase-pipe convergence study and the
developing-channel case are heavy (21 s, 11 s, 6 s alone); plane
Poiseuille, Couette, Taylor-Green, reproducibility and the stability
limits stay in the PR gate.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
---------
Co-authored-by: jyviko <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
* mesh: boundary surface preparation (weld, manifold and orientation checks, exact self-intersection, feature edges)
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* engine: mesh.surface.check on an open design's body or an imported surface
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* engine: struct-init the test session
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* surface: weld at an absolute distance too; the engine welds CAD tessellations at twice SolidCraft's linear tolerance
Faces of a SolidCraft tessellation carry their own copies of shared boundary points, which agree
only to within the kernel's tolerance (up to ~2e-6 mm on the test corpus).
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* docs: surface preparation in the literature survey, architecture and roadmap
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* surface: simplify after review
- The weld distance travels with the mesh (SurfaceMesh::weld_distance, set by Shape::tessellate),
so every consumer welds a tessellation correctly without opting in.
- Exact tests are infallible on predicates::orient2d_projected (shared with generic_orient2d), and
compute each vertex's plane side once per pair with an early reject.
- One Box3 for the weld, the BVH and shells; a Half struct for edges; the weld's two lookups share
one loop, with chained grid cells; cached normals for feature edges; shared parameter helpers.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
* corpus: boundary surface preparation as a corpus stage
20 of 21 parts are valid boundaries with their expected genus; cross-cylinders' tessellation has
cracks and flipped triangles where its two intersection curves meet.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01NmAttHhUZTokgPeEJia8sm
---------
Co-authored-by: jyviko <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
