#1519 stopped the sentinel blaming a backend for a run that had already
drawn for SENTINEL_PROOF_DELAY, on the reasoning that a backend which
kept the application alive that long had proved itself. That reasoning
does not hold for the failure this repository sees most on integrated
hardware.
Measured here on an Intel Iris Xe: with `--backend dx12` a large drawing
aborts the process 110 s in, long past the proof delay, so the sentinel
carried `survived=1` and the next launch went straight back to dx12 and
crashed again. The same binary on `--backend vulkan` opens the same
drawings with no out-of-memory report and no uncaptured errors at all,
so the backend really is the difference and moving off it really is the
cure. D3D12 is fine while the interface draws and only fails once the
whole scene uploads, which is why the clock is the wrong question.
The crash report from #1521 answers the right one. It records the
backend that was live and what the panic was, and it is named for the
process that wrote it, so it joins exactly to the `pid` the sentinel
already stores — no guessing from whichever report is newest, and no
new state to keep. A sentinel marked survived is now excused only when
that run left no report, or left one that has nothing to do with the
graphics device.
The test that matters is the pair, and both were run end to end:
killed from Task Manager after 40 s, no report -> dx12 kept
device crash after 110 s, report names wgpu -> dx12 skipped, Vulkan
`is_device_failure` is deliberately narrow: an indexing bug in drawing
code that happens to run on a frame must not cost the user a working
backend.
Co-Authored-By: Claude Opus 5 <[email protected]>
- V8 models are placed relative to their global origin (model header).
- A V8 model's extent is the union of its elements' stored ranges; a V7
model's text box is one text height tall.
- Files without a colour table use the standard DGN default table; colour
255 is read from the table's background entry.
- Colours from a file's colour table follow the background: near-white
colours swap with it, and over non-black backgrounds some shades drop a
step.
- DGN contrast pulls towards a pivot taken from the model's own colours,
after the fade, which is blended into the background first.
- Shared cells are read through their full 3D placement, so tilted
instances and 3D definitions are drawn in plan.
- Deleted V8 elements are skipped; lines along the model's extent stay in
the raster.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
While a grip drags, the edited entity is hidden from snapping and its
drag-start shape got a reference-only pass: the marker appeared on the
old edge's midpoint or corner, but the grip stayed at the raw cursor,
and the pass ran only with object snap tracking on. The reference pass
now always runs, still limited to discrete points (never nearest-style,
so the point cannot be pulled back onto the old shape) and skipping the
grip's own start point, and when no snap on the rest of the drawing wins
it places the grip too.
Fixes#1528
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Only the wire shaders read transparency_enable; hatch (solid, gradient
and pattern), text and 3D/fill-triangle fragments always drew the
entity's alpha, so turning Show Transparency off left them see-through.
Fixes#1529
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Booleans now cut a sphere wherever the other solid crosses it — box
edges and corners, planes through its centre, another sphere, a
cylinder on its axis — instead of refusing or keeping the wrong piece.
Reported in discussion #1499.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Every near-white colour turned black on a light background (and near-black
white on a dark one), including true colours. A 255,255,255 solid used as a
white mask became a black block. Now only colour 7 swaps: true colours and
ACI 255 resolve a hair off pure white/black and keep their value on screen
and in plots.
Discussion #1500
Co-Authored-By: Claude Opus 5.5 <[email protected]>
On Windows the drawing was held with an exclusive byte lock, which also
refuses reads, so AutoCAD / BricsCAD could not open it at all. The lease
now reads through a handle that shares reads but refuses write opens, with
a shared lock that refuses writes; the sidecar still keeps a second OCS
session out. Unix keeps its advisory lock, which never blocked readers.
Discussion #1416, issue #739
Co-Authored-By: Claude Opus 5.5 <[email protected]>
- Add transform_ellipse_geometry helper in scene::view::transform to compute exact principal semi-axes, normal, ratio, and shifted arc parameters under any affine transform (negative scales, reflections, negative rotations, non-uniform scaling, and translations).
- Use transform_ellipse_geometry in block_cache::transform_tangent for TangentGeom::PlanarEllipse so GPU analytical instances and curve endpoints match true WCS positions.
- Update entities::ellipse::apply_transform to use transform_ellipse_geometry for complete scaling, rotation, mirroring, and affine transformations.
- Clean up tangent parameters in to_render for Ellipse so start_param and end_param are consistently unwrapped.
- Add comprehensive unit tests in transform.rs, block_cache.rs, and ellipse.rs verifying negative rotations, negative scales, and distant origins.
Entity picks switch object snap off, but COINCIDENT (and the other
constraints that take a point) only recognise a point when the pick lands
exactly on it. Without snap there were no point markers and every pick
became the whole curve. Point-accepting entity picks now keep object snap
and hand the snapped point to the command.
Discussion #1526
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The palette installed the insert command directly and kept the previous
command's last point, so ORTHO locked the insertion point to an axis
through it. The command-start reset is now a shared helper that the
palette uses too.
Fixes#1525
Co-Authored-By: Claude Opus 5.5 <[email protected]>
flatten_plan 2,289 -> 1,098-1,144 us (-50..-52%); flatten_scene_apply 1,213 -> 789-813 us (-33..-35% vs baseline; -28.5% same-session O2-vs-O1, drift-corrected).
Pre-sizes collect/plan vectors and adds Scene::update_entities so FLATTEN issues one geometry bump per batch instead of per entity; single update_entity keeps its exact bump-per-call semantics.
Moves the flatten projection fns and command helpers into modules/draw/modify/flatten.rs (behavior unchanged) and adds flatten_plan / flatten_scene_apply to the COMMANDS suite.
Follow-up to 0bfd5dcc:
- A clipped block hatch kept its unclipped plane-local outline, which the
viewport plot projection prefers, so fills bled past the XCLIP boundary
on layouts. Rebuild it from the clipped outline.
- insert_spatial_filter / filter_handle run for every INSERT in each
render walk; resolve the xdictionary through the entity link or the
document index only, and leave the whole-object scan to set_insert_clip.
Fixes#1508
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Paste (and WBLOCK, which shares the kernel) kept the source boundary
handles, so the pasted hatch stayed tied to the original outline or to
unrelated handles in another drawing. The COPY remap now lives at the
document level and runs for both.
Refs #1370
Co-Authored-By: Claude Opus 5.5 <[email protected]>
A V2 plugin passed the version gate and then failed silently while
registering its ribbon, since the ribbon wire types have moved on. The
host now loads V3 and newer, so V2 plugins are rejected up front with a
version message. The V2 compatibility fixture crate goes too.
Fixes#1085Fixes#987
Co-Authored-By: Claude Opus 5.5 <[email protected]>
It moved the text a fixed drawing unit off the line and then another unit
up, wrong at any real scale. It now sets DIMTAD = Above on the dimension
and lets the automatic layout place the text.
Fixes#543
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The bundled DRAINAGE / DRAINAGE2 definitions use the RIGHT_ARROW shape
from qcadshp.shx, which had no substitute, so they drew as continuous
lines. Add it to the bundled shape set alongside BAT, ZIG and the rest.
Fixes#665
Co-Authored-By: Claude Opus 5.5 <[email protected]>
A running command only knew the entities that existed when it started;
committed results are now handed back to it, and OFFSET adds them to its
pick set.
Fixes#673
Co-Authored-By: Claude Opus 5.5 <[email protected]>
A fan only covers convex faces; architectural n-gons filled well outside
their outline (the sample's faces covered 63% extra area).
Fixes#680
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Only the displayed annotation context followed a transform, so other
scales reappeared at the old place, and a copied annotative entity other
than a LEADER/MTEXT pair shared the source's context tree.
Fixes#700
Co-Authored-By: Claude Opus 5.5 <[email protected]>
EXPLODE left MTEXT untouched. Each laid-out run becomes a TEXT at its own
baseline origin, height, rotation and width, keeping inline colours.
Fixes#760
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The STEP header timestamp used SystemTime, which panics on wasm, and the
exports wrote through std::fs, which the browser does not have.
Fixes#761
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Middle (72 = 4) carries a Baseline vertical code, so the baseline landed on
the alignment point and the text sat half a height too high.
Fixes#767
Co-Authored-By: Claude Opus 5.5 <[email protected]>
COPY added the entity without its derived image model, so only the frame
appeared at the copy.
Fixes#829
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The synthesised dimension text copied the dimension's ByLayer colour, and
its glyphs take their colour from that entity, so the layer colour showed.
Fixes#898
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The mask is filled with the canvas colour; plotting carried that dark
colour (or the light-to-black rule) onto the sheet.
Fixes#1072
Co-Authored-By: Claude Opus 5.5 <[email protected]>
WBLOCK copied only the entities and their layers, so INSERTs pointed at
missing blocks and the file opened almost empty or needed recovery. It
now goes through the cross-drawing paste kernels, which bring layers,
linetypes, styles, block definitions and baked dimension blocks along.
Fixes#1111
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The width of a wide polyline lives on the wire as a GPU band, which the
custom-arrow extraction dropped, so the arrow drew as a hairline.
Fixes#1130
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Windows reports AltGr as Ctrl+Alt, so '@' and similar layout glyphs were
treated as shortcuts and dropped; polar input like @10<45 failed.
Fixes#1236
Co-Authored-By: Claude Opus 5.5 <[email protected]>
MOVE shifted the *D block contents but left the block cache on the old
epoch, so reopened drawings showed dimensions in place until REGEN.
Fixes#1342
Co-Authored-By: Claude Opus 5.5 <[email protected]>
They went out at depth 0 while lines kept theirs, so some fills
covered lines that DRAWORDER had put on top.
Fixes#1345
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The copy kept the source boundary handles, so editing the source
regenerated the copy onto the source outline and it seemed to vanish.
Fixes#1370
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The reference direction, base point and locked point went through f32,
so the drawn line drifted off true parallel.
Fixes#1387
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The dot-named temp file is stored Hidden by Samba and the attribute
followed it through the replace. Clear it unless the drawing was
already hidden.
Fixes#1414
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The definition took a peeked handle that the image then also received;
the DWG writer dropped the shared-handle definition and reopened images
lost their file.
Fixes#1421
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Resolve the SPATIAL filter through extension-dictionary fallbacks
(missing xdictionary handle, DictionaryWithDefault, case/whitespace
tolerant keys) so clipped INSERTs no longer plot unclipped. The
inverse-block matrix is used as stored; the codec decodes the on-disk
column-major layout (fixed upstream).
#1521 inserted crash_log between web_diag's doc comment and its module, so the comment documented crash_log instead. #1519 wrote the survived marker with a literal line break inside the string.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
