Commit Graph
142 Commits
Author SHA1 Message Date
Tanish Manku 7218285443 Merge branch 'stk-2587' into stack/s2
# Conflicts:
#	crates/doc/src/extents.rs
2026-11-05 19:22:25 -07:00
Zinadin ZidanandClaude Opus 5.5 75175c807d refactor(render): one constant for the dark canvas; the render error limits say they are limits (#803)
render_look's DARK repeated the rasteriser's default background; both now
read RasterOptions::DARK_BACKGROUND. ECHO and LISTED_TABLES are renamed
MAX_ECHO_CHARS and MAX_LISTED_TABLES. No change in behaviour.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-05 19:10:58 +06:00
Tanish Manku ef4b4f7e6c XCLIP: review fixes from CodeRabbit on #2756
- objects are found through a handle index when reading and writing clips, and removed in one pass (no scan per reference)
- the writer's handle seed starts past the clip objects, so a drawing with many clipped references gets no duplicate handles
- a clipped copy that runs out of the block budget marks the list truncated, like an unclipped one
- clipped block contents keep their own layers in plots
2026-11-03 19:34:20 -07:00
Tanish Manku 0d72152142 Merge main into land/2354 2026-11-03 18:21:00 -07:00
Tanish Manku 100d812155 Merge main into land/2354 2026-11-03 18:07:17 -07:00
Tanish Manku e90a197365 XCLIP: keep CADCraft's inverted flag beside group 290 for DWG, whose bridge drops 290; the flag is read only when 290 is absent 2026-11-03 18:07:00 -07:00
Tanish Manku de753897c4 Merge main into land/1944 2026-11-03 18:00:03 -07:00
Tanish Manku 39ee2e5960 SHX fonts: review fixes (search the current folder for a bare drawing name, one lock for the font cache generation, no split_paths on wasm, MTEXT \U+ upper case as the layout, a point budget for repeated linetype shapes, web picks report skipped files and cap bytes read at once) 2026-11-03 17:59:52 -07:00
Tanish Manku 75049718c9 View twist: build with raster images from main (an axis rectangle is turned on a twisted screen, so it cuts an image as an outline) 2026-11-03 17:35:23 -07:00
Tanish Manku 76d09aaddf Merge remote-tracking branch 'origin/main' into land/2354 2026-11-03 17:22:36 -07:00
Tanish Manku c2d62a2842 Merge main into land/1944 2026-11-03 17:11:42 -07:00
Tanish Manku 68e01dd487 Merge main into land/1230 2026-11-03 17:11:25 -07:00
Tanish Manku 945c7f456b Merge remote-tracking branch 'origin/main' into land/1944 2026-11-03 16:52:21 -07:00
Tanish Manku f10e41409d Images: read no more than the limit, keep plot bounds to what a viewport shows, free textures with no images left
Review fixes from CodeRabbit on #2749:
- the image file is read through read_capped, so a file that grows after the size check can't exhaust memory
- an image cut by a layout viewport widens the sheet's bounds only by the part inside it, as other primitives do
- uploaded tiles of closed drawings are dropped also when the next list has no images
2026-11-03 16:52:11 -07:00
Tanish Manku 5bd563caac Merge main into land/2354: charge every block visit in the clipped copy too 2026-11-03 16:42:36 -07:00
Tanish Manku bd22363d5e Merge remote-tracking branch 'origin/main' into land/2505 2026-11-03 16:40:52 -07:00
Tanish Manku a6ef259fec Merge remote-tracking branch 'origin/main' into land/1230 2026-11-03 16:40:36 -07:00
Dylan M. Taylor dad7d1c138 fix(render): charge every block entry, index attribute definitions, and walk dashes and outlines in linear time (#2724)
* fix(render): charge every block entry, index attribute definitions, and walk dashes and outlines in linear time

Array copies skipped ATTDEF entries before charging the expansion budget, each attribute searched its block for a definition, each dash scanned every centreline sample, and outline chaining searched a vertex's segment list from its start on every visit. Each is now charged, indexed or cursored.

* address review: drop wall-clock bound from outline test
2026-11-03 18:11:23 -05:00
Tanish Manku 0e19f9d1d1 SHX fonts: build on main (a bottom-justified SHX text sits a third of the height down; explicit render imports after #2239) 2026-11-03 15:41:48 -07:00
Tanish Manku 238aa1789c Images: draw an IMAGE in its own function, under the complexity limit 2026-11-03 15:03:06 -07:00
Tanish Manku 8cdcd6f85c Merge #2354 (XCLIP and XCLIPFRAME) onto main 2026-11-03 15:00:20 -07:00
Tanish Manku 28421dd1d7 Merge #1944 (SHX shape fonts, big fonts and unifonts; linetype shapes) onto main 2026-11-03 14:59:22 -07:00
Tanish Manku 4c00b6732f Merge #2505 (raster images drawn on screen and in PNG output) onto main 2026-11-03 14:52:15 -07:00
SBaksa 457627df33 OLE2FRAME: validate hex strictly, rewrite only the header groups, cheaper corner lookup (review on #2587)
A sign in a hex pair is no longer accepted; only the 310 groups holding the 98-byte header are decoded and rewritten, in place and without cloning the payload; the corner lookup is one allocation-free pass; the io ole_frame test module is not built for wasm like its DWG siblings; Unknown::is_ole_frame replaces five copies of the type check. Tests: a_plus_sign_in_ole_data_is_not_hex, a_large_ole_payload_only_rewrites_its_leading_groups.
2026-11-03 16:44:30 +11:00
mattlafy cbc9db75b1 Merge remote-tracking branch 'upstream/main' into pr/dview
# Conflicts:
#	crates/engine/src/prompt.rs
#	crates/engine/src/session/execute.rs
#	crates/io/src/dxf_read/entities.rs
#	crates/ui-egui/src/canvas.rs
#	crates/ui-egui/src/cmdline.rs
2026-11-02 21:53:26 -05:00
SBaksa a9dbc50ce2 Merge origin/main into ole-frame (drop parity-checklist change) 2026-11-03 13:46:27 +11:00
midasmatterhorn 3f57f8d3d4 render: a display list past the block-expansion budget says what it left out (#1278)
* render: a display list past the block-expansion budget says what it left out

* Report viewport truncation only after the expansion budget is exceeded

* fmt: keep ui-egui control.rs rustfmt-clean
2026-11-02 15:01:46 -07:00
Rahulandrahul05ranjan 03101e8d77 fix(io): a moved DIMENSION draws its block at its insertion point (group 12) (#2634) (#2636)
A DIMENSION moved by an older AutoCAD keeps its *D block contents where
they were and the move in DXF group 12, and is drawn translated by the
insertion point. Group 12 was never read into Dimension, so the block
rendered at its pre-move position and the extents reached there too.

Read group 12 into a new Dimension::insertion_point, draw the block
translated by it, move its bounds with it in the extents, test the
block's text where it draws, and write group 12 back whenever the block
written is the one read from the file. Regenerated geometry (edits drop
the block link) sits at its definition points, so no offset is kept or
written for it.

Co-authored-by: rahul05ranjan <11793455+rahul05ranjan@users.noreply.github.com>
2026-11-02 15:01:26 -07:00
152617f4df Plot: a layout's saved page setup is used by default (part of #2493) (#2586)
* Plot: a layout's saved page setup is used by default (#2493)

A layout whose page setup says plot window, scale to fit or a plot scale was plotted 1:1 from
the sheet origin unless the caller passed fit. plot, exportpdf, the PNG device and preview now
read the layout's saved window, scale to fit, scale and centring when the options don't say
(render::paper::saved_plot); explicit options win, and fit: false without a scale is still the
plain 1:1 sheet. Tests: saved_setup_tests in pdf.rs (four), and
preview_uses_the_layouts_saved_page_setup_by_default.

* Parity checklist: regenerated after merging main

* Parity checklist: regenerated after merging main

* Plot: saved page setup in preview and PNG matches the PDF (review on #2586)

Preview honours the saved centre when fitting a window and an explicit scale beats a saved scale to fit (PDF too); PNG places a saved window without fit and clips to the printable area.

* Preview: a saved window clips to the printable area (review on #2586)

The preview drew the placed plot over the margins while the PDF and PNG blank them. The blanking helper now lives in the render crate and both the PNG plot and the preview use it. Test: preview_clips_a_saved_window_to_the_printable_area.

* Preview: pin hostile saved page setups and the default setup (review on #2586)

Degenerate or far-away saved windows and saved scales of 0 or not finite already fall back to the plain plot in the PDF, PNG and preview; tests now pin that, and that a layout with no saved setup plots as before.

* Preview: the printable clip follows the view's fit (review on #2586)

The clip assumed the sheet filled the image; when the height cap applies
View::fit scales and centres it. The printable corners now go through
View::to_screen. Test: preview_clip_follows_the_view_when_the_height_cap_applies.

* docs: leave the generated parity checklist to maintainers

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Brandon Thomas <echelon@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-01 14:38:34 -05:00
f0b46e34dc Hit tests don't lay out dashes: moving the mouse over a block stays smooth (#2603)
* Hit tests don't lay out dashes: moving the mouse over a block stays smooth (#1285)

Every mouse move hit-tests the block references under the cursor, and
hit_polylines built them with a full-detail display list, one primitive per
dash. A fine pattern on long lines in a block inserted at 0.001 made millions
of dashes per move.

Options::linetypes off draws every line whole and continuous, and
hit_polylines uses it for blocks, dimensions and multileaders. A dashed line in
a block is now hit across its gaps, as one outside a block already was.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* render: with linetypes off, a lightweight polyline is one continuous stroke

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Steffen <camperking@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
2026-11-01 14:38:27 -05:00
SBaksa ad609fab02 OLE2FRAME: the frame counts in the drawing extents; CLI test uses a self-cleaning dir (review on #2587)
bounds_in now takes the frame from its corner tags (shared ole_frame in the doc crate), so zoom extents shows a frame outside other geometry. The undrawn_entities tests clean up on drop.
2026-11-01 15:38:48 +11:00
SBaksa 4caf26ee60 OLE2FRAME: draw its frame outline; a DWG save no longer lists it as left out (#2422)
The frame is the rectangle from the entity's two corner points (groups 10/20 and
11/21). A DWG write and read-back keeps the OLE data (ole_frame.rs), so the DWG
note leaves the type out of notSaved.
2026-10-31 14:27:48 +11:00
MarekSpodBiedry 189bfcbf08 Draw annotative text with Match orientation to layout upright, ignoring its stored rotation (#2498) (#2504) 2026-10-30 15:46:24 -04:00
9d4b3bb091 Raster images are drawn on screen and in PNG output (#1756)
An IMAGE whose file exists was drawn as a frame with two diagonals. Now
its pixels are loaded when a drawing opens and drawn on screen and in PNG
output, with rotation, clip boundary, fade and object transparency. A file
that can't be read keeps the crossed-out frame and reports a message.

Ported from the raster commit of #1029 onto the split render and canvas
modules. New here: large images are drawn from a pyramid of levels cut
into tiles uploaded as they come into view, instead of one texture with
pixels skipped, so a large scan stays sharp zoomed in and doesn't shimmer
zoomed out.

Limits are listed in docs/gaps.md (FF10): PDF and SVG output write only
the frame, and on screen images lie under all other objects.

Co-Authored-By: Chanthawat Kiriyadee <enwuft@icloud.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 17:05:02 +01:00
Dylan M. Taylor d6de2db6db Update prompt shows the app icon and accent color instead of a letter (#2479)
UPDATE_APP (native builds; craft-update is a native-only dependency) carries our 128 px icon and accent (#3ea6e8, its colour in ArtCraft Launcher). The updater, the self-update
helper and craft-update's startup prompt all get them through crate::UPDATE_APP; the
prompt draws the icon, the accent and "by ArtCraft" (craft-update's default developer).

craft-update is pinned to storytold/craft-libs main at 1065653e7 (#434).
Clippy fixes for Rust 1.99 in untouched code (same logic):
- crates/constraints/src/auto.rs
- crates/engine/src/audit.rs
- crates/engine/src/cmd/inquiry.rs
- crates/engine/src/mtext_edit.rs
- crates/geom/src/nurbs.rs
- crates/io/src/dxf_raw.rs
- crates/io/src/dxf_read/document.rs
- crates/render/src/text.rs
- crates/ui-egui/src/cmdline.rs
- xtask/src/layers.rs
2026-10-30 01:28:46 -04:00
Brandon ThomasandClaude Opus 5.5 7546277128 fmt: rustfmt the DIMTMOVE 2 clearance port (#2302)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-29 21:39:34 -04:00
BooksandBrandon Thomas d15de75406 Dimensions: report the measurement as soon as one is created (#2421) (#2438)
* Dimensions: fill in the measurement when one is created (#2421)

A dimension made in the session kept measurement 0.0 until it was saved and reopened.
add_dim now sets it with the geometry call the DXF/DWG writer uses for group 42.
Block names are still assigned at save time. Test: a_new_dimension_reports_its_measurement.

* Parity checklist: regenerated after merging main

* Dimensions: a jogged radius dimension stores its radius as the measurement (#2421)

draw_jogged_radius set the text but not DimGeometry.measurement, so DIMJOGGED dimensions
kept measurement 0 (and the DXF/DWG group 42 was 0). It now sets the radius.
Test: jogged_radius asserts the stored measurement.

* Parity checklist: regenerated after merging main

* Dimensions: fill in a new dimension's text position when it is created (#2421)

A dimension made in the session kept text_mid (0,0) until it was saved and reopened.
add_dim now sets it from the same dimension_geometry result as the measurement, the
position the writer puts in group 11, unless the user placed the text.
Test: a_new_dimension_reports_its_text_position.

---------

Co-authored-by: Brandon Thomas <echelon@gmail.com>
2026-10-29 21:31:22 -04:00
dcbcb6ab32 fix: keep WIPEOUT closure coherent across grips and outlines (#2284)
* fix: keep WIPEOUT closure coherent across grips and outlines

Treat WIPEOUT as a closed polygon with an optional exact terminal closure marker, not as an open polyline. Centralize logical vertices, explicit outlines and closure-preserving vertex edits in the document type. Reuse this contract in grips, geometric primitives, rendering and DXF export without changing the stored schema or writer closure convention.

Add six-version DXF ASCII/binary and DWG regressions for implicit/explicit closure, both windings, all corners, API/interactive edits, undo/redo and repeated saves. Keep open/closed polylines and their coincident endpoints distinct. Cover multiple hot grips, hostile inputs and equivalent mask/frame output.

Validation: cargo xtask ci, all seven stages passed (2571 tests passed, one pre-existing skip). Two unrelated network tests failed transiently in earlier runs; each passed in isolation and the complete final run passed without exclusions. Direct and reopened grip edits produce byte-identical CLI PNG output.

Fixes #2256; related test-only matrix PR #2237 and tracker #2238 B3.

* fix(doc): keep coincident WIPEOUT endpoints separately editable

When an edit makes the endpoints of an implicit boundary coincide, add a separate closing marker rather than reclassifying the edited corner as that marker. Retain the logical vertex count and allow either coincident corner to move independently afterwards.

The document regression fails before the fix. Add API/interactive coverage for both endpoint directions and both closure forms, undo/redo, and subsequent separate edits across all six DXF/DWG versions including binary DXF and save-after-edit. Update the documented editing contract.

Validation: cargo xtask ci passed all seven stages; 2573 tests passed, one pre-existing skip.

Addresses the coincident-endpoint review on PR #2284; related #2256.

---------

Co-authored-by: cloudj <jonbahnes@gmail.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
2026-10-29 21:31:17 -04:00
0d9e62bc9a Stop DIMTMOVE 2 text taking the DIMTAD 1 clearance twice (#2302)
Fixes #2222

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
2026-10-29 21:31:13 -04:00
heejunB f817ea9a3a fix(render): fill font glyphs with the nonzero winding rule (#2189)
TrueType and OpenType outlines are defined with the nonzero rule, but glyph
meshes were tessellated even-odd, so any area covered by two contours came
out empty. Arial Unicode (the macOS fallback CADCraft picks for Hangul)
builds 진 (U+C9C4) from a component that already contains the final ㄴ plus
a second copy of that ㄴ; even-odd cancelled the pair and 진 rendered as 지.
Overlapping contours in other fonts lost their overlaps the same way.

Tessellate glyph outlines with FillOptions::non_zero(). Outlines placed in
drawing coordinates without a TextGlyph (complex-script runs) go through the
same tessellator, offset to their first point to keep f32 precision, instead
of the even-odd slab filler.
2026-10-29 21:21:38 -04:00
Francesco BVandClaude Opus 5.5 7917afce5d render: explicit imports for XCLIP clipping after #2239
#2239 replaced the glob imports in blocks.rs and viewports.rs with explicit
lists; the XCLIP code relied on the globs for `Clip`, `append_clipped_keep`,
`clip` and `Pen`. Import them by name.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-29 11:42:25 +01:00
Francesco BVandClaude Opus 5.5 14296dc4f5 blocks: XCLIP and XCLIPFRAME, clipped references kept through DXF and DWG
XCLIP clips block and xref references to a rectangular, polygonal or
closed-polyline boundary (Invert clip, ON/OFF, Delete, generate Polyline;
Clipdepth is ignored in 2D). The boundary is Insert::clip, in block
coordinates, so it follows its reference; other references stay whole.
Clipped references are drawn, plotted and framed (XCLIPFRAME 0/1/2) cut
to the boundary, attributes included, and their extents are cut to it.
DXF and DWG read and write it as the reference's ACAD_FILTER/SPATIAL
SPATIAL_FILTER; filters the model can't hold stay foreign data. Gap F2.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-29 11:06:23 +01:00
mattlafy e1d0a4747e Fix imports after merging upstream/main 2026-10-28 15:34:34 -04:00
mattlafy b7c62eb93d Merge remote-tracking branch 'upstream/main' into pr/dview
# Conflicts:
#	crates/doc/src/entity.rs
#	crates/engine/src/cmd/layout.rs
#	crates/engine/src/cmd/modify2.rs
#	crates/engine/src/lib.rs
#	crates/io/src/dxf_read.rs
#	crates/io/src/dxf_write.rs
#	crates/render/src/lib.rs
#	crates/ui-egui/src/canvas.rs
#	docs/gaps.md
2026-10-28 15:30:03 -04:00
Dylan M. Taylor 2721aa8982 Enforce Rust coding standards via Clippy (#2239)
* chore(clippy): enforce workspace Rust quality gates

* refactor(doc): use explicit entity imports

* refactor(constraints): split residual evaluation

* refactor(fonts): split mtext layout

* refactor(render): use explicit entity imports

* fix(fonts): retain height for empty paragraphs

* fix(doc): repair dimension and dynamic block links

* refactor(render): split entity drawing helpers

* docs: document workspace lint inheritance

* refactor(render): isolate dimension geometry

* refactor(engine): use explicit imports

* refactor(engine): split complex edit paths into helpers

* refactor(io): split DXF read and write helpers

* refactor(ui): split canvas interaction and drawing

* refactor(ui): split command and drawing helpers

* test(i18n): split locale preference checks

* fix(ui): use current egui scope id API

* fix(dim): preserve text offset and ellipse grip behavior
2026-10-27 15:55:09 -04:00
Francesco BVandClaude Opus 5.5 52c1e6de98 fonts: read SHX shape fonts, big fonts and unifonts; draw linetype shapes (FF8, FF9)
SHX files referenced by text styles are found by file name through a
font map, registered bytes, an app provider, the drawing's folder and a
search path (CADCRAFT_SHXPATH); never in a CAD program's install folder.
TEXT, MTEXT, attributes, dimension text and table cells draw with the
style's font, big font and vertical flag; linetype shape elements are
drawn when their file is found. The web build registers .shx files
picked or dropped with a drawing. All test fonts are compiled in code
from shape source.

Plain MTEXT text decodes \U+XXXX escapes as the layout does, and an SHX
file whose name differs in case from the one asked for is found on
case-sensitive file systems too (the folder is looked through).

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-27 18:10:33 +01:00
NetterSilverandClaude Opus 5.5 850dd94f3c Stroke runs of hairline polylines as one path in the rasteriser
PNG export (and everything else drawn by raster::render) stroked every
polyline primitive with its own tiny-skia stroke_path call: a new path
and a newly built paint pipeline per primitive. Hatch patterns, linetype
dashes and text strokes produce long runs of short polylines in one
colour, so that per-call setup dominated the render.

Consecutive hairline polylines (zero width, or at most one pixel when
anti-aliased; the default for PNG export) with the same colour, alpha
and width are now collected into one path and stroked once. tiny-skia
draws hairlines segment by segment straight onto the pixmap with the
same per-contour round caps, so the pixels and their blending order are
unchanged. Thicker strokes (filled outlines), fills, points, infinite
lines and polylines with non-finite or far-out screen coordinates break
the run and are drawn exactly as before.

Adds crates/io/examples/export_bench.rs, which exports examples/
apartment.dxf repeated on a grid (10x10: 5,800 entities) and reports
median/min wall time per format.

PNG export, 10x10 grid, 11 interleaved runs (Ryzen, 16 threads, Linux):
  before: median 1790.7 ms, min 1703.3 ms
  after:  median 1037.6 ms, min  926.0 ms   (-42.1 %)
A/A noise on the same setup: 1.1 %. Output PNGs are byte-identical
(bench at 10x10 and 20x20, CLI export of apartment.dxf and a 23,200-
entity drawing). New tests compare render() against stroking each
primitive on its own.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 15:47:42 +01:00
mattlafy 1823eeab19 fix(view): twist follows named views in viewports, undo, screen-direction crossing, clipped DVIEW preview, DCS target
- VIEW Save inside an MSPACE viewport stores the viewport's twist; VIEW Restore into an
  unlocked viewport turns the viewport to the saved twist (one undo step).
- Undo and redo keep the untiled Model view's twist record in step with the live view.
- A selection window is a crossing window when dragged leftwards on screen, also in a
  twisted view (preview and selection alike).
- DVIEW's preview inside a clipped viewport is clipped to the viewport's outline.
- DXF view centres are read relative to the view target (VIEWPORT/VPORT group 17, VIEW
  group 12), the origin of display coordinates.
2026-10-26 16:19:28 -04:00
aiongg c042e98abb refactor(render): move dim.rs items to dim/arrowheads.rs and dim/text.rs 2026-10-26 19:44:13 +00:00
aiongg 994af5cac9 refactor(render): move lib.rs items to display, pipeline, viewports, props, draw, text_entity, dim_draw, blocks 2026-10-26 19:44:13 +00:00