Long whole-job solves ran silently, so there was no way to tell a slow
engine from a hung one until the timeout fired. --progress hands each
solve a JobProgressLog that prints [job/engine] lines for start, finish
(or failure/timeout), every plate commit, and candidate evaluations
throttled to one line per 2 s so parallel runs stay readable.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Engines had to reverse-engineer the benchmark validator: Opus55 assumed a
0.01 arc tolerance (the validator uses 0.001), Gpt6Astra added hand-tuned
paddings and copied the validator's check order, Qwen picked its chord
tolerance to stay under a constant it could not reference.
NestTolerances publishes the validator's arc tolerance, the Clipper grid
and SafeClearanceMargin (with its derivation). NestLayoutCheck moves the
benchmark NestValidator's checks into OpenNest.Engine as a public API
(Clears for a part pair, Violations for a whole result); NestValidator is
now a thin wrapper. Verdicts are unchanged: tests compare ordered
violation lists against a frozen copy of the old validator, and a
tangent-disc stress test covers 432 pairs at the safe margin.
Co-Authored-By: Codex <noreply@openai.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Engines optimized guesses at the benchmark cost: Opus55 re-implemented
salvage credit, Qwen used plate area per part area, Gpt6Astra ignored
salvage. NestJobCost moves StockLadder's EstimateNetArea into a public
home (net sheet area, unplaced-part penalty, whole-result Evaluate) and
the benchmark and StockLadder now call it. Scores are unchanged: tests pin
it against a frozen copy of the old computation and real benchmark runs.
Bounds still include marks, as before, so scores do not move.
Co-Authored-By: Codex <noreply@openai.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every nesting-geometry consumer filtered only rapids, so scribe/etch
moves counted as part material. An etch tick that ends a hair outside
the outline (PEP bend ticks start on the notch edge) made the part
"open geometry leaving the material region": the job validator threw
and every built-in engine plus Gpt6Astra crashed on real PEP jobs
(PT75, drawing 4980 A01 PT77). Marks are only on the surface, so they
should never affect placement, collision, area, or validation.
- SpecialLayers.IsMaterial excludes Rapid and Scribe; used by drawing
area, canonical angle, part collision, PartGeometry, plate perimeter,
best-fit/pair evaluation, rotation analysis, GPU evaluators, and both
validators. Timing, display, splitting and posts still see marks.
- ConvertGeometry also maps the saved SCRIBE layer name to Scribe, so
programs rebuilt from stored entities keep their marks.
- NestReader repairs older files (e.g. PepNestExport output) whose
programs saved etch as cut moves while source entities kept SCRIBE.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Conservative flattening circumscribes arcs, so at 0.01 the validator was up
to 0.01 too strict along curves: PEP's P260626-03 layout, exactly 0.25 apart
along an arc, failed with a 0.004 sliver. At 0.001 the worst error on either
side is 0.001. Validating the 26-job benchmark set with Opus55 went from
about 2 s to about 5 s.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Clipper validator flagged valid Opus55 and PEP layouts (P260805-03,
P260626-03). Two causes:
- Arc.ToPoints(circumscribe) scales every vertex out by 1/cos(step/2),
endpoints included, so a 0.03125 corner fillet flattened at 0.01 poked
0.013 past the straight edges it meets. ClipperBridge now flattens itself:
circumscribed arcs keep their endpoints on the arc and put interior
vertices on tangent intersections, with the segment count chosen so the
outward error stays within the tolerance. Arc.ToPoints is unchanged for
its other callers.
- The conservative padding made a layout exactly at the spacing fail.
NestValidator now uses OffsetForValidation: the same conservative
flattening, round joins at a tenth of the tolerance, no padding. Its only
leniency is that join chord error at convex corners.
With both, Opus55 is valid on all 26 benchmark jobs (25 before the
Clipper migration; the old failure was a spike artifact).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Per-entity offsetting left spikes and inverted loops wherever a feature is
narrower than the spacing (1.nest), and RemoveSelfIntersections only
caught proper crossings. The callers that already flatten to polygons now
take a single Clipper region offset instead:
- PolygonHelper (BestFit) and PartBoundary use the conservative mode, which
keeps their never-under-estimate guarantee. PartBoundary also keeps holes
that appear when a perimeter curls back on itself.
- NestValidator offsets perimeter and cutouts in one region; Clipper drops
collapsed cutouts, so the collapsed-or-flipped heuristic goes away.
- CutOff.IntersectPerimeter offsets through the bridge. The old
OffsetEntity(Left) grew CW perimeters but shrank CCW ones, so with the
plate's perimeter cache a cut-off ran through the part; slots narrower
than twice the clearance now close up instead of leaving a gap.
- GetOffsetPartLines (3 overloads) and the AddOffset* helpers had no
callers and are removed, as is EntityView's never-defined DRAW_OFFSET
block.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The benchmark is about to be used as the objective for LLM-designed
engines, and several gaps would have rewarded the wrong behavior:
- Ranking was utilization-first, so dropping awkward parts raised the
score. Rank valid > fully placed > cost > plates, where cost is
salvage-credited sheet area plus a largest-sheet penalty per unplaced
part; placing a part is never scored worse than omitting it.
- Salvage rate was ignored in scoring; cost now uses EstimateNetArea,
recomputed from job geometry rather than trusted from the engine.
- Rotation constraints were never validated. Add RotationPolicy.Allows
(shared with NestJobPlacementValidator) and check every placement.
- Returned sheets were trusted, so an engine could loosen spacing or
invent a size. Sheets must now match offered stock.
- Part-in-part placements were flagged as overlaps; spacing now accounts
for cutouts, with an X-sorted sweep to prune distant pairs.
- Summary averaged per-job percentages; it now sums areas and cost.
- --spacing and sheet sizes parsed with the current culture.
- Warn when .nest jobs offer only their original sheet sizes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Benchmark jobs could only come from .nest files. A JSON manifest now lists
DXF files with quantities (plus sheet sizes, spacing, edge spacing, quadrant
and per-part allowRotation), imported through CadImporter. DXF paths resolve
relative to the manifest; sheet sizes are required from the manifest or
--sheet-sizes and are read in the DXFs' own units. Folder scans pick up
*.nest and *.manifest.json, and invalid manifests fail loudly.
BenchmarkRunner now runs (job x engine) solves concurrently, capped by
--parallel N (CLI default 3; --parallel 1 is sequential). Results are written
by index so report order is unchanged. Concurrent solves compete for cores,
so Time(ms) is only clean at --parallel 1; the run prints a note when N > 1.
Also fixes --output for manifest jobs, which tried to read the manifest as a
.nest to copy metadata from.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Repo-wide sweep with the pinned CSharpier 1.3.0 tool. Whitespace and
line-wrapping only; OpenNest.Engine.Tests (109) and OpenNest.IO.Tests
pass after reformat, full solution builds 0 errors.
Added .csharpierignore so csproj/config XML keeps its existing layout
(CSharpier's XML wrapping churns attributes with zero benefit).
Formatting is now enforceable: dotnet csharpier check . passes.
Add OpenNest.IO/Bending/BendRepair: opt-in repair of unambiguous paired
ETCH/SCRIBE bend ticks, bounded to <=3.175 mm endpoint movement with
explicit source units. Cut geometry is never modified.
- CadImportOptions.BendRepair configures it; CadImportResult exposes
per-bend BendRepairReports; CadImporter/Dxf wire it into import.
- Console: --repair-bends-mm <limit> --cad-units inches|mm prints
per-bend reports for newly imported DXFs.
- New OpenNest.IO.Tests project (net8.0, synthetic DXFs, 30 tests)
covering bend detection and repair, added to the solution.
- Update README.md and CLAUDE.md for the new pipeline and build/test
instructions.
Previously each job fixed one plate size and ran a single Nest() call,
which doesn't reflect the actual problem: a real job is fulfilled
across however many plates are needed, drawn from a pool of standard
sheet sizes, not forced onto one fixed sheet.
NestEngineBase.Nest() has no way to pick its own plate's size - it
fills whatever Plate it's given - so size selection now lives in the
harness itself, applied identically to every engine:
- BenchmarkJob carries the full candidate size pool (CandidateSizes)
instead of one fixed PlateSize; one job per file, not one per size.
- BenchmarkRunner drives a loop: while items remain, pick the smallest
candidate size that fits the largest still-unplaced drawing (reusing
the codebase's own MultiPlateNester.CreatePlate/FitsBounds), build a
fresh plate of that size, and run one Nest() call to fill it. Repeat
until everything is placed, no candidate size fits what's left, or a
safety cap (40 plates) is hit.
- NestValidator now validates bounds/spacing per plate but the
quantity cap once globally across all plates, since that limit
belongs to the whole order, not any one sheet.
- JobResult/Report report PlatesUsed and a per-size breakdown instead
of a single-plate bounding-box compactness metric; utilization is
now aggregated across every plate the engine used. Ranking keeps the
same rule (utilization first), with fewer plates as the tie-break
when both are fully placed and tied - the natural multi-plate
analogue of the old single-plate compactness tie-break.
Smoke-tested against the synthetic sample across 5 candidate sizes:
correctly builds one job, picks the smallest fitting size, uses
however many plates each engine needs (1-2 here), and still catches
StripNestEngine's pre-existing out-of-bounds bug.
Collision.HasOverlap (and Part.Intersects, which shares the same
underlying algorithm) was observed to return false negatives on real,
complex production geometry: a layout with a combined placed area over
7x the plate's work area passed the polygon-based spacing/overlap
check with zero flagged pairs. This is a pre-existing gap in
OpenNest.Core's Collision detection, not something introduced here,
but it let an obviously-invalid layout score as "valid".
ValidateAreaBudget adds a hard mathematical backstop that does not
depend on Collision at all: non-overlapping parts confined to the work
area can never have a combined area greater than the work area itself.
Also switch WorldPolygon from Shape.ToPolygon() (default up to 1000
segments per arc) to ToPolygonWithTolerance(0.01), matching the
convention already used elsewhere in the codebase (e.g.
BestFit.PolygonHelper) - arc-heavy real parts were producing
thousands-of-vertex polygons for a simple spacing check.
Loads any .nest file (or folder of them) via NestReader and nests every
drawing with quantity > 0 using each registered NestEngineBase, so it
works sight-unseen against arbitrary real jobs without any hardcoded
geometry. Optionally sweeps a fixed --sheet-sizes list instead of each
file's own plate size.
- BenchmarkJob/JobLoader build immutable job specs; a fresh Plate and
NestItem list is created per (job, engine) run so state never leaks
between engines or jobs.
- NestValidator rejects a layout if any part falls outside the work
area, any two parts are closer than PartSpacing (checked via each
part's own world-space polygon inflated by the spacing, so it holds
for arbitrary concave/holed geometry, not just bounding boxes), or a
drawing gets more parts than requested.
- Scoring matches Plate.Utilization() (placed area / full sheet area);
ties among fully-placed layouts break on the smaller used bounding
box (more usable remnant).
- Report prints a per-job ranked breakdown plus a per-engine summary
(wins, avg utilization, time), and can write a flat CSV.
Verified end-to-end against a synthetic .nest file (not committed)
against the four built-in engines; caught a genuine out-of-work-area
bug in StripNestEngine in the process.
ajandClaude Opus 5.5
