Eliminate full Insert entity cloning in block style evaluation, unify
document object environment scans, and guard viewport override queries.
1. Analysis & Architectural Bottleneck:
- In render_graph::InsertStyleSpec::new, an entire Insert struct was cloned
into an EntityType::Insert enum solely to pass to style functions that only
read EntityCommon fields (layer, color, transparency, linetype, lineweight).
- In view::render::resolve_document_render_environment, when object caches were
not prepared, the document.objects map was traversed linearly twice: once
for RenderEnvironment and once for RenderSettings.
- Viewport override lookups were being performed even when viewport was None
(the standard model space condition).
2. Optimizations Implemented:
- Added `render_style_for_common_viewport`, `linetype_name_for_common_viewport`,
and `has_resolved_book_color_common` taking `&EntityCommon` directly.
- Constructed `InsertStyleSpec::new` without cloning the Insert entity.
- Fused the two linear scans of `self.document.objects` in
`resolve_document_render_environment` into a single pass tracking best candidates.
- Guarded viewport override lookups with `if viewport.is_some()` in
`layer_render_style_viewport` and `render_style_for_common_viewport`.
3. Correctness & Invariance:
- The same fields and fallback values are resolved in identical order.
- Tested across all 325 scene tests, block inheritance tests, and viewport
override unit tests with 0 failures.
4. Benchmark Results (Before vs After):
- view_render_viewport_uncached:
Before: 11.00 µs (90.9k FPS)
After: 5.00 µs (200.0k FPS)
Improvement: -54.5% latency (2.2x faster uncached rendering)
- wide_and_tapered_arc_tessellation:
Before: 6.36 ms (235.8k entities/s)
After: 5.41 ms (277.4k entities/s)
Improvement: -14.9% latency (+17.6% throughput)
- analytical_curved_tessellation:
Before: 7.79 ms (359.5k entities/s)
After: 7.01 ms (399.6k entities/s)
Improvement: -10.1% latency (+11.1% throughput)
- spatial_interaction_index_build:
Before: 0.77 ms (2.47M wires/s)
After: 0.73 ms (2.66M wires/s)
Improvement: -5.1% latency (+7.7% throughput)
- ui_grip_budget_build:
Before: 424.50 µs
After: 380.25 µs
Improvement: -10.4% latency
- ui_ribbon_view_construction:
Before: 455.50 µs
After: 382.30 µs
Improvement: -16.1% latency
Eliminate intermediate heap vector allocations during wire classification
and analytical instance extraction for circle and ellipse GPU pipelines.
1. Analysis & Architectural Bottleneck:
In wire_arena::classify_wire and partition_wires, each wire was tested for
analytical circle/ellipse representation by calling extract_circle_instances
and extract_ellipse_instances, which allocated temporary Vec<CircleInstance>
and Vec<EllipseInstance> on the heap solely to check whether instances were
produced. Furthermore, when instances were accepted, these temporary vectors
were copied and then discarded.
2. Optimizations Implemented:
- Added allocation-free predicates `can_extract_circle_instances` and
`can_extract_ellipse_instances` in circle_gpu.rs and ellipse_gpu.rs.
- Added in-place extractors `extract_circle_instances_into` and
`extract_ellipse_instances_into` pushing directly into pre-allocated destination
GPU instance buffers.
- In wire_arena.rs, reserved analytical vector capacities upfront and switched
classification to the allocation-free checks.
- In pipeline/mod.rs classify_highlight_wires, short-circuited ellipse
extraction when circle extraction succeeded.
3. Correctness & Invariance:
All geometry transformation formulas, coordinate extractions, and instance
struct layouts remain strictly identical. Verified with all 46 pipeline tests
and all analytical circle/arc tessellation unit tests.
4. Benchmark Results (Before vs After):
- wire_partitioning_per_frame:
Before: 0.18 ms (mean 0.19 ms, p95 0.38 ms)
After: 0.06 ms (mean 0.07 ms, p95 0.28 ms)
Improvement: -63.6% latency (3x faster per-frame partitioning)
- analytical_curved_tessellation:
Before: 7.79 ms (359.5k entities/s)
After: 6.87 ms (407.7k entities/s)
Improvement: -11.8% latency (+13.4% throughput)
- wide_and_tapered_arc_tessellation:
Before: 6.36 ms (235.8k entities/s)
After: 5.48 ms (273.8k entities/s)
Improvement: -13.8% latency (+16.1% throughput)
- analytical_curved_zoom_100x:
Before: 12.75M FPS (0.01 ms)
After: 13.97M FPS (0.01 ms)
Improvement: +9.6% FPS
Brings in the DXF round-trip fixes merged as cadcodec #48 and #51 and the
block-definition fixes for cadcodec issues #49 and #52. cadkernel moved to
the same codec revision first, so the graph keeps a single acadrust. The
Python plugin, which is outside the workspace, has its own lock moved too:
its staging script requires the plugin and the host to share one acadrust
source.
The canaries that asserted the old wrong results were flipped against the
real-IPC audits, all 60 of which pass with the plugin built: Helix, Light,
Underlay, ViewBorder and SectionSymbol now round-trip through DXF, as do the
text style flags, the dimension style's text style name, the block
description, the table's merged ranges and the leader's annotation link and
vectors. Those five kinds move to Complete in the coverage ledger.
Two narrower gaps stay pinned. The ATTDEF writer emits no group 280, so
`lock_position` is still lost on a DXF save; every other field now
survives. A true-colour leader `override_color` cannot be held by DXF group
77, which takes an ACI index, and the DWG writer does not store it.
Behaviour change for users: a text style's oblique angle is now written to
DXF in degrees, as the format defines it. DXF files written by earlier
versions carry radians there and reopen with a much smaller slant.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The 2Lines placement (#1459) wrote the second line's new ends back with
`set_resolved_point`, which was not the inverse of the `resolve_point` it
paired with. `resolve_point` reads polyline vertices in WCS through the
polyline's OCS; the writer put that world point straight into the OCS
vertex, so a mirrored polyline (normal -Z, common in real drawings) had
its vertex flipped, and the closing segment's end marker — which names
vertex 0 on a closed polyline — found no vertex and the placement was
silently dropped. The writer now converts back into the OCS, keeps the
polyline's own elevation, and wraps the closing marker the same way.
The same arm pushed `test_pline_line_then_arc` past the default test
thread's stack: its locals, a cloned entity among them, sat in the frame
of the command-result dispatcher, which is on the stack of every click.
The placement is built out of line and handed over boxed.
Merging #1447 had left nothing setting DIMCONSTRAINT's default to
Diameter; the Dispatch arm #1461 turned into a match now covers it with
Radius and Convert. DCCONVERT's result lines go through the translation
macros.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The command deleted the dimension and left a literal constraint behind. The
reference keeps the dimension, drives it with a new parameter and gathers
only associative dimensions, saying how many objects it dropped.
Co-Authored-By: Claude Opus 5 <[email protected]>
The second line kept its midpoint and turned around it, so it settled at a
different offset than the reference, which keeps the line's start along the
first line and moves it to the distance the picked point had from it.
Co-Authored-By: Claude Opus 5 <[email protected]>
The panel titled the dynamic form "Angular Dimensional Constraint"; the
reference names it after the dimension kind with the form in parentheses.
Co-Authored-By: Claude Opus 5 <[email protected]>
Adds ui_selection_overlay bench (benches/performance_benchmarks.rs + README metric 31): builds a representative selection overlay element (snap, crosshair, UCS tripod, 128 grips quick / 4096 capped full) and times widget-tree construction. Measured 0.4us typical, 2.4us at the cap (threshold 50us).
Adds ribbon/statusbar section marks to the chrome bucket in view_main (src/app/view/mod.rs) so live PERF=1 view-detail traces attribute those two costs separately. Gated on the existing perf switch; no-op otherwise.
Adds pub re-exports of CursorType/IsoPlane (src/app/mod.rs) for the bench fixture (same pattern as the existing grip re-exports).
Ranked result over all ui_* benches: ribbon view construction ~556us (largest; follow-up already tracked), everything else under 50us or synthetic-fixture shapes. No new follow-up filed. GPU and canvas-draw costs are unmeasurable in headless runs (view() never executes there).
The bundled `opencad.python` plugin drives OCS through plugin API v7: all
43 canvas entity kinds, the drawing tables, and OCS's own commands run
step by step. RustPython stays in the plugin process — the plugin crate is
outside the workspace with its own lock file, so the application's own
dependencies are unchanged.
Reconciled with the plugin command work that landed as #1425, which the
branch was written before:
- Both added `impl Drop for PluginProcessInteractiveAdapter`, which git
merges silently into two impls that do not compile. One body now
cancels the command and then drops it from the runner, in that order: a
plugin waiting on a step has to be released before the runner forgets
the command.
- `execute_command` (#1425) and `run_command` (#1423) are both routes from
a script to the command line, and only the second was guarded. The
SCRIPTCOMMANDS switch and the denylist are shared now, so neither route
is the unguarded one.
- `run_command` dispatched with plugin dispatch still enabled, so a script
naming a plugin command re-entered the plugin that was blocked waiting
for it. It sets the suppression flag for the length of the dispatch, as
`execute_command` does.
The denylist read the command as typed, while the dispatcher resolves
aliases: a user-defined `Q` for `QUIT` (ALIASEDIT writes those) walked
past it. It resolves the alias first.
Also: the active-command comparison ran on every message, a mouse move
included, allocating two strings for names that are already `&'static
str`; SCRIPTCOMMANDS' own messages go through the translation macros; and
`docs/.DS_Store` (now ignored) and an unrelated RTF note about constraint
commands are out of the tree.
Co-Authored-By: Claude Opus 5 <[email protected]>
The direction handle was built with center_grip (is_midpoint), so drags
arrived as Translate, which the direction arm ignores — the grip engaged
but silently did nothing. Build it with square_grip (same marker) so the
drag arrives as Absolute and repoints the line.
Tests:
- xline_direction_grip_drag_changes_direction: press, drag, direction is +X
Same unit-stub defect as the BOUNDARY tool fix: entity_boundary_source_on_plane
tessellated rays/XLINEs to a single base-point stub. Now takes clip bounds
derived from the full source map, sharing segments_bounds and
live_segments_for_curves instead of duplicating the logic.
Tests:
- xline_source_clips_to_given_bounds: clipped to the box, not the stub
Implement the complete Block Definition dialog window launched via BLOCK
and BMAKE commands, featuring standard CAD layout, workflows, and entity
creation pipelines.
- Dialog UI & Layout:
- 2-column compact CAD grid with synchronized vertical heights (580 × 390px).
- Editable combobox with existing block names, invalid character validation,
and inline redefinition badge.
- Multi-line text editor widget for block descriptions (replacing single-line input).
- Quick Select (QSELECT) button integration restoring dialog state upon completion.
- Confirmation guard when attempting to redefine an existing block definition.
- Centered radio buttons, clean coordinate inputs, and accessible keyboard shortcuts.
- Base Point & On-Screen Workflow:
- BlockOnScreenCommand strictly enforces prompt sequencing: Base point is
prompted before object selection.
- Dynamically recalculates world_to_block and block_to_world affine transforms
when base point is specified on-screen, preventing stale (0,0,0) normalization.
- Supports Retain, Convert to block (INSERT placement), and Delete modes.
- Explode Enforcement:
- Guarded EXPLODE command to prevent decomposing blocks flagged as non-explodable.
- Dynamic Themed Icons:
- Added pick_point.svg, select_objects.svg, quick_select.svg, and warning_triangle.svg.
- Routed all icons through the semantic theming engine so line work (#B4B6B9)
and accents (#6DB7ED) adapt automatically to the active theme.
- Tests:
- Added unit tests for retain/convert/delete modes, redefinition, explodable flag,
on-screen base point transform recalculation, and command sequencing.
#1447 branched from the same commit as #1443 and rewrote the radial
dimensional constraint a second time, so the two designs had to be
reconciled rather than stacked. #1447's is the one kept: it covers
DCRADIUS and DCDIAMETER through one command, places the dimension by the
direction the location was picked in rather than storing a text point,
and names the parameters `radN` / `diaN`. #1443's radial command result,
its entity builder call and its follow-up code are gone; its own
independent work stays — the ribbon's small-button grid and tool
descriptions, the magnitude reading of a negative radius, and the centre
anchor a parameter edit resizes a circle about.
The DIMCONSTRAINT default follows #1443: running DCRADIUS or DCDIAMETER
directly leaves it alone, and the Dispatch arm sets it when DIMCONSTRAINT
is what asked for the command.
The second commit's radial drawing — the dimension line broken around the
text, the elbowed leader, the centre-mark and arrowhead rules — applied to
every radius and diameter dimension, not just a constraint's, and three
layout-parity tests said the old layout was the measured one. It is now
scoped to a dimensional constraint's own dimension, as #1443 scoped its
version, and ordinary radial dimensions draw as they did.
Co-Authored-By: Claude Opus 5 <[email protected]>
The text of a radius or diameter constraint sits on its dimension line,
which breaks around it, and no centre mark is drawn while that line is
there. Text that no longer fits inside moves beside the circle at the end
of a short leader that leaves the rim, elbows and hooks under the text,
with the arrowheads keeping the circle. A dynamic dimension now keeps
only its text at a constant screen size; arrowheads and extension lines
keep the style's size in drawing units.
Co-Authored-By: Claude Opus 5 <[email protected]>
The commands ask for the circle or arc themselves, report the measured
size, take a dimension line location and then the value or name and
value, creating a radius or diameter dimension as the constraint's
dynamic dimension with a radN/diaN parameter, refusing a second radial
constraint on the same object and any non-positive value.
Co-Authored-By: Claude Opus 5 <[email protected]>
A new drawing (#1441) recorded ISO-25 as the current dimension style by
name only. Every other place that changes it sets the header's handle as
well, and a DWG carries the current style by handle alone, so a saved new
drawing reopened on Standard. The style's handle now comes back from the
helper that adds it.
A radius constraint (#1443) anchored the circle's centre from the shape of
its references — one whole entity — but a single whole reference is also a
line's length, whose start is what stays. The entity decides now.
The two tests that encoded the replaced rules follow them: a negative
radius validates by its magnitude, as a negative distance already did, and
only zero is refused; a new drawing's dimension text reads at ISO-25's
comma separator.
Co-Authored-By: Claude Opus 5 <[email protected]>
`bool::then_some` evaluates its argument eagerly, so starting SHELL or
SOLIDEDIT with nothing selected indexed an empty vector and aborted the
application. The single-solid target now comes from a slice match.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
Three doc paragraphs had stacked above one helper, so each described the
function after the one it sits on.
Co-Authored-By: Claude Opus 5 <[email protected]>
Direct distance entry (#1429) turned every bare number at a point prompt
into a distance, so ROTATE's angle and the values automation types were
read as coordinates. The step reads the token first now — `feed_command`
reports whether it claimed it — and only an unclaimed one becomes a
distance. That also puts the token back on the unified step driver, which
owns the P / L selection keywords and the dimension-snap bookkeeping, and
stops `on_text_input` running twice for the same token. A tracking ray
carries its own base, so it no longer demands an anchor that the first
point of a command does not have yet.
Plugin command execution (#1425) re-implemented the command line inside
the plugin host: it dispatched families directly, skipping the teardown
that starts a command (grip edits, selection gestures, `last_point`, the
accepted snaps), and made every `CmdResult::Dispatch` in the application
skip plugin dispatch to avoid its own re-entrancy. A scripted line now
runs through `run_command_line`, with the trailing Enter left off while
the plugin streams the command, and the re-entrancy is handled by a flag
set for the length of the plugin's call. A pause parks the line's Enter
with the rest of its tokens instead of dropping it.
A font reference (#1432) tried the family lookup before the face-file
index, and that lookup ends in a prefix match a longer file name loses to
(`ISOCPEUI.TTF` would take `ISOCP`). The exact file answers first.
Co-Authored-By: Claude Opus 5 <[email protected]>
Select arc or circle / Dimension text / dimension line location / value or
name prompts with a radN parameter, invalid-selection, miss, duplicate and
invalid-expression messages, negative values as magnitudes, a dynamic radial
dimension on the constraints layer drawn as the reference does (mid-radius
text on the center-to-rim line, no hook, no center mark, one text grip), the
stored text position following the radius, center-anchored resizing, the
Properties title and Constraint section, DIMCONSTRAINT's Radius default,
the small-button Dimensional ribbon grid and tooltip descriptions.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
The Annotative row of text, MText, blocks, leaders, hatches and
dimensions is a Yes/No list, as the reference shows it; choosing a value
drives the same per-object annotative toggle as before.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
A new drawing is metric (measurement 1, millimetre insertion units) and
carries the reference's ISO-25 dimension style as its current style, so
dimensions and annotational constraints draw at readable metric sizes
instead of Standard's imperial 0.18 text.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
An angular dimension is titled `Angular Dimension` and a three-point
one `3 Point Angular Dimension`, as the reference names them.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
Reads and sets the form new dimensional constraints take (0 dynamic,
1 annotational), the same state DCFORM sets.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
Dynamic dimensions carry DIMEXO/DIMEXE/DIMASZ overrides in text-height
ratios, a two-line angular dimension draws extension lines only beyond
its sides, a selected dynamic angle shows triangle grips at the arc ends
with squares at the arc point and text, and the constraint dimension's
Properties follow the reference: kind titles, General before Constraint,
angular-precision Value and Measurement, drawn text position, reference
row order for Fit, Primary Units and Tolerances, and text/dimension line
alignment rows reading On as the aligned state.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
Eliminate major performance bottlenecks in the per-frame view render path:
- Memoize document render environment resolution using (geometry_epoch, len) keying, replacing per-viewport O(N) linear scans over document.objects.
- Index render_environments and render_settings handles in ObjectDataCache during the single build_cache pass.
- Cache resolved background image lookups in memory to eliminate repeated path.is_file() filesystem syscalls per frame.
- Add has_face3d fast-path to bypass wire splitting when no Face3D entities exist, and use FxHashSet<u64> handle indexing when present.
- Invalidate all cached structures across bump_geometry, bump_geometry_no_blocks, and bump_entities.
- Add comprehensive correctness unit tests and before/after benchmarks.
Performance impact (10k objects):
- P95 latency reduced from 341.00 µs to 173.00 µs (-49.3%)
- Throughput increased from 4,700 FPS to 6,000 FPS (+27.7%)
- Direct render environment evaluation: ~85 µs down to <50 ns (>1,700x)
Three-point angular dimensions take their sweep from the two rays and the
arc point (extension lines follow their own rays, straight and reflex
angles draw), dynamic angular dimensions store no text point so DIMTAD
places the text beside an unbroken arc, and dynamic dimensions carry a
DIMGAP override of a quarter text height.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
DCANGULAR picks two lines, an arc or three points, takes the dimension
line location to choose the sector and the value or name, and creates the
dynamic two-line or three-point angular dimension driven by an angN
parameter. Reference messages for invalid, repeated and parallel sides,
existing names and invalid names; angle values display within one turn at
the angular precision; the first side (or vertex and first point) stays
when the value changes; the ribbon Dimensional panel follows the reference
order.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
cosmic-text's fallback search can land on a face that claims the character and
outlines to nothing: macOS's system font does that for every glyph, because
ttf-parser reads no contours out of it. The code took the first non-.notdef
glyph id and stopped, so the character drew nothing while believing it drew
something — invisible text, and no warning either.
An empty outline is now treated as no answer: the search moves on to the next
face, and when cosmic-text's own list has nothing that draws the character, the
installed families are walked for one that does. That list is the platform's own
handful of names (on macOS `.SF NS`, `Menlo`, `Apple Color Emoji`, `Geneva`,
`Arial Unicode MS`) and covers neither Cyrillic nor Greek, however many fonts the
machine has that do.
`cargo test --lib a_fallback_glyph_is_never_empty`, and the Turkish letters in
`lff::tests::fonts_parse_and_resolve` render on macOS now instead of failing.
A DWG text style stores its TrueType font as a *file* name — `GOST2304_TypeA_italic.ttf`,
sometimes with a directory — which is what AutoCAD resolves. The family lives
inside the file and is usually a different string (`GOST2304 Type A`), so looking
the reference up as a family missed an installed font and fell through to the
embedded stroke fonts. Those carry no Cyrillic at all (`simplex`, `standard`,
`romans`, `iso3098` hold no U+04xx glyph), so a Russian drawing came out with
every word missing: the letters went on to the system-font fallback, which
returns nothing on macOS.
Installed faces are now indexed by file name, and a reference — a style's font
file, or an inline `\f` code, which carries either form — resolves by family
first and by file name second. An embedded stroke font still wins, so
`romans.shx` keeps rendering as it did.
`cargo test --lib face_files_resolve_to_their_family`
Karim Jerbi