The shipped wasm was built from a vendored acadrust 0.4.1 carrying local
changes in two files (read_mesh array bounds, OLE2FRAME corner decoding),
and the published read_mesh patch did not even apply to the pristine
crate, so the notice could not reproduce the source compiled into the
wasm (MPL-2.0 3.2). acadrust 0.6.5 already does both on its own: the
road-surface MESH and the C0060203 OLE2FRAME come out identical to the
patched build, so the parser now uses the crate exactly as crates.io
serves it.
The notices name 0.6.5, its upstream commit and the crate SHA-256 (equal
to the Cargo.lock checksum the build used), the stale patch file is
removed, and the page links to the crate source instead.
Verified on 22 samples, loading each with the old and the new wasm in
Chrome: all load, 17 render pixel-identical, and the 5 that differ are
0.4.1 misreads that 0.6.5 corrects. The wrapper changes this needed live
in hmwebviewer rust/dwg-wasm (branch chore/acadrust-0.6.5).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Mirrors hmwebviewer's rust/dwg-wasm change into the sample: the wasm
artifact plus the parser wrapper, which now goes through parse_dwg_json
and JSON.parse instead of the JsValue-returning parse_dwg.
Building the whole result as a serde_json::Value tree pushed the live
wasm heap near 2.9GB for a 19MB drawing, and the allocator's cost grows
with that heap, so parsing went quadratic — 354s, with the tab frozen
throughout. Streaming one entity at a time keeps it linear: 2.5s.
The same build also fixes three bugs that drew stray 4km grey shapes
where two CF-PATT SOLID hatches should be. Max hatch span 4655 -> 110.
Adds two docs recording both investigations, in the format of the
existing ones: measurements, the hypotheses that were ruled out, the
fixes, regression checks, and the MPL-2.0 position (acadrust is
unmodified, so no new source-disclosure obligation arises).
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
