Commit Graph
2930 Commits
Author SHA1 Message Date
github-actions[bot] 0889ef9dfc Regenerate generated files and fold pending files [skip ci] 2026-10-29 02:11:11 +00:00
semmelrindeandClaude Opus 5.5 1937dea04b Sketch constraints: refuse one curve picked twice for a two-curve constraint (#8577)
Parallel, collinear, equal or concentric of a curve with itself was added
as a constraint that always holds; perpendicular or tangent to itself was
refused as geometry that can't move. These six now refuse one curve
picked twice, as Coincident already does for a point, and leave the
sketch as it was. Constraints already in a design are untouched.

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 23:52:44 +00:00
Nathan Dane a545f32113 Render stage panels: Surfaces and references, Studios and environments, and the review dialogs (#6281) (#7999)
* Render stage panels: Surfaces and references, Studios and environments, and the review dialogs

* fix: complete stage panel translations and gallery tile fields

* fix: exercise stage catalog sizing in the floating layout

* fix: retain the Appearance library selection query
2026-10-28 23:49:39 +00:00
github-actions[bot] 937d5f0176 Regenerate generated files and fold pending files [skip ci] 2026-10-29 01:43:00 +00:00
semmelrindeandClaude Opus 5.5 05a7a4fef9 Traditional Chinese: translate the 56 messages main gained since batch 1 (batch 3) (#8365)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 23:28:02 +00:00
N3MI-DGandClaude Opus 5 33c27106ff Pick by the body's own size, not the camera's distance (#7919) (#7920)
* claim: #7919

* Pick by the body's own size, not the camera's distance (#7919)

The occlusion test that decides whether an edge or vertex is hidden by the
face in front allowed a slack of `scene_radius(app) * 0.02`, and
`scene_radius` ends with `.max(app.cam.half_height() * 2.0)`: in an
orthographic view that is `0.016 x camera distance`. Zooming out grew the
slack without bound until a whole body fitted inside it and every edge behind
the front face counted as visible, and edges and vertices are offered before
faces, so they won the pick. Cutting the slack from the scene also gave a
small part in a large assembly a tolerance many times its own size.

`pick_slack` now takes a size and returns the mesh's own chord tolerance with
room to spare, `DisplayQuality::Coarse.chord(size) * 4.0`: a facet sits at
most a chord tolerance inside its surface, and that tolerance is already a
fraction of the body's size, so the slack no longer moves when the camera
does. Each body is tested with its own bounding box, which the loop already
holds. `view.pick` does the same, so an agent's pick still matches a click on
the same pixel. Sketch profiles and origin planes belong to no body and keep
a scene-level value, `scene_pick_slack`.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-10-28 23:15:09 +00:00
semmelrindeandClaude Opus 5.5 17807ab6dc Sketch MidPoint: refuse a line's own end as its midpoint (#8562)
A line's own end held at its middle shrinks the line to a point. The
solve accepted that, and a later angle, tangent or parallel on the line
had no direction to hold: its points went millions of millimetres away
while the sketch still reported solved. MidPoint now refuses the pick,
saying why, and leaves the sketch as it was. Constraints already in a
design are untouched (only the command refuses).

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 23:11:05 +00:00
momysnowandClaude 1445b5a4f0 Extend: pick two curves to extend both to where they meet (#4750) (#8392)
* claim: #4750

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

* Extend: pick two curves to extend both to where they meet (#4750)

`sketch.extend` gains a second form: `curves: [a, b]` names two lines or
arcs, with an optional `at: [[x,y],[x,y]]` naming the end of each to grow
(default: the end nearest the meeting point). Both grow along their own
line or circle to where they cross and the two ends are joined with one
Coincident, in one undo step. `moved` is the joined end's id.

- Where two crossings are possible (line and arc, two arcs) the one with
  the least total growth is used; a tie is refused.
- A curve that already reaches the point does not grow. An end held by
  other geometry or a constraint stays and gets a new line or arc.
- Parallel or collinear lines, curves that never meet, a point behind the
  chosen end, the same curve twice, splines, ellipses, infinite lines,
  offset geometry and malformed `curves`/`at` are refused with no change.
- The single-curve form is unchanged; its code is shared through
  `grow_line` / `grow_arc`.

Canvas: in the Extend tool, Shift-click two curves. Shift goes through
the tool modifier registry as the new `extend_pair` action, and Extend
turns off Shift's midpoint snap. Escape drops a pending first pick. The
sketching guide describes the gesture.

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

---------

Co-authored-by: Claude <noreply@anthropic.com>
2026-10-28 23:10:24 +00:00
github-actions[bot] 066de1b005 Regenerate generated files and fold pending files [skip ci] 2026-10-29 01:07:21 +00:00
github-actions[bot] d6876d6e67 Regenerate generated files and fold pending files [skip ci] 2026-10-29 00:58:24 +00:00
raykholo a1e7337d3c Toasts: one stack in the view's bottom-right corner, one width, buttons as in dialogs (#7893) (#7959)
* claim: #7893

* Toasts: one stack in the view's bottom-right corner, one width, buttons as in dialogs (#7893)

* Control protocol: toast_place in the sorted ui.prefs list (#7893)

* Toasts: a sketch test clears the status notice before clicking the view's corner (#7893)

The test's line end lands in the bottom-right corner, where notices now stack
and take clicks as they did at the bottom centre; the test is about the circle.

* Toasts: the pattern scenario's circle sits clear of the view's corner (#7893)

After the refit, a circle at y 0 landed in the bottom-right corner, where the
tool's prompt now stacks and takes clicks. The circle moves 40 mm up; the
pattern it checks is the same.
2026-10-28 22:35:24 +00:00
semmelrindeandClaude Opus 5.5 951b5eb1d3 Indonesian and Traditional Chinese: the Appearances panel's 27 Render messages (#8336)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 22:27:28 +00:00
Colin Faulkingham 5221103369 Components: a component named like the design is reachable by name (#8358) (#8423)
* claim: #8358

* Components: a component named like the design is reachable by name (#8358)

Document::find_component resolved the design's own name to the root component (id 0) before
searching the component list, so a component that shares the design's name was unreachable by
name: appearance.assign reported `no component `CurtainRailPlug`` (the same lookup serves every
command that takes a component by name).

An exact component name (or component id) match now wins over the design-name fallback; "0" and
"root" still always mean the design.

Regression test: a_component_named_like_the_design_takes_its_appearance_by_name (fails on main
with the reported error, passes with the fix).
2026-10-28 22:07:58 +00:00
JK3D 9b7a6ea04d History: feature folders get their own symbol (#8521) (#8522)
* claim: #8521

* feat: give History feature folders their own symbol

A folder of features in the History on the left drew the same folder icon
as the Browser's folders. It now draws feature_group, a pile of history
chips with the front one in blue, and so do Add to New Folder and Remove
from Folder in the History's menu. Other uses of folder keep it.
2026-10-28 22:04:12 +00:00
Lars Jansen fe265d74b1 Sketch: draw conflicting curves and points in the conflict colour (#7244) (#7249)
* claim: #7244

* feat(sketch): draw curves, points and constraints that conflict in their own colour (#7244)
2026-10-28 22:03:54 +00:00
NOWN-DEVandClaude Opus 5.5 eb29c4b2d8 Sheet metal feature commands: typed parameter contracts (#6581) (#6582)
* claim: sheet.* typed parameter schemas

* Sheet metal feature commands: typed parameter contracts

sheet.flange, sheet.perforate, sheet.hem, sheet.fold, sheet.unfold, sheet.refold and
sheet.convert read their parameters by hand, so a misspelt key or a value of the wrong
shape was ignored or failed late, and describe_command had only prose for them.

Each now has a contract in schema/sheet.rs, checked in front of the unchanged handler,
and returns FeatureResult. The values no shared type fits (sheet-edge points, an edge
flange's edges, a fold's line and fixed point, the flange type, the positions, the
relief shape and the perforation pattern) get types that call the handlers' own
readers, split out of cmd/sheet.rs, and give their messages. The rule's values
(thickness, K-factor, bend radius, reliefs) stay out of the contracts: the handlers
refuse them, and strict refuses them as unknown keys.

A test passed a nested `base` object whose `sketch` the flange never reads; it now
names the sketch at the top level.

Part of #2829 (#6581).

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

* Sheet metal contracts: a further bend's angle stays as lenient as before

The flange handler reads a `then` bend's `angle` that is no expression
(`true`, a list) as left out and bends it 90 degrees. The rebased
contract had moved that check into the shared reader, so the handler
refused such a call even with the check off.

The reader is back to the handler's behaviour, and `ThenSteps` is the
stricter one, as the schema guide asks: warn reports such an angle and
strict refuses it. Tests cover the reader's leniency, the type's
refusal and the three check levels on one flange.

Part of #2829 (#6581).

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

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 22:01:35 +00:00
github-actions[bot] f1ab19422b Regenerate generated files and fold pending files [skip ci] 2026-10-29 00:30:57 +00:00
github-actions[bot] 0a187f5cb1 Regenerate generated files and fold pending files [skip ci] 2026-10-29 00:21:59 +00:00
tr1pp4andClaude Opus 5.5 7adbe00377 History in the browser: a History layout that lists each component's history in the browser (#7709) (#7710)
* Browser: the history in the browser, as a History layout (compact mode)

Settings > Interface > History gains "In browser" (view.history_layout dock: browser):
neither timeline is shown, and each component's History folder lists its sketches and
features in history order with the marker. Dragging the marker rolls the history back
through timeline.roll_to; Find in Timeline reveals the feature in the browser.

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

* Browser history: sketches listed under their feature as in the vertical History

With Sketches: Under their feature, the history in the browser (compact mode) lists each
sketch under the first feature built from it, through the shared history_rows (#6494)
without reference rows: same rule, same fold key, so switching the History layout keeps
what is open. A drop above a row rolls back its listed sketches with it. Selecting a feature
lights the rows of its sketches, and selecting a sketch lights the features built from it.

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

* Browser: with the history in the browser, the Sketches folder starts closed

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

* Browser history: sketch rows have the eye of their Sketches folder row

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

* History in the browser: Arabic, Hebrew and Vietnamese for the layout's name

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

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:49:21 +00:00
semmelrindeandClaude Opus 5.5 0243afdc0b Dialogs: scroll a long refusal into view with its Create Issue (#8454) (#8506)
* Dialogs: scroll a long refusal into view with its Create Issue (#8454)

A dialog's refusal comes last in its fields. When the fields fill the
dialog, the refusal sits under the fold: an Extrude Cut that the kernel
refused showed only its first line, "boolean Cut: the operation failed: no",
and no Create Issue. The dialog now scrolls a new refusal into view once:
a refusal on OK at once, a preview's once no field is being typed in, so a
field never moves away while it is typed in.

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

* Dialogs: decide the reveal in one function and test it

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

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:48:09 +00:00
raykholo 8e6a107f6e Lists many PRs add to: no written lengths, one file per theme (#7747) (#7898)
* Lists many PRs add to: no written lengths, one file per theme (#7747)

* xtask lists: the fixed-table allowlist in order, each with its own reason

* Lists: the wide-character blocks and the drawing scale presets as slices too (#7747)

* Lists: the drawing units test reads UNITS as a slice (#7747)
2026-10-28 21:33:34 +00:00
semmelrindeandClaude Opus 5.5 4ad18407dc Paste: leave copied body transfers out, so the design still opens (#8478)
A copied body transfer was pasted like any feature and given the target
component as its owner. A transfer belongs to the component it moves bodies
to, so the saved design was refused on open ("body transfer owner must match
its destination"), and a pasted transfer had nothing to move anyway: the
copies go into the component they are pasted into. Paste now leaves body
transfers out and says so in its notices; a clipboard of only transfers is
refused.

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:29:05 +00:00
semmelrindeandClaude Opus 5.5 5695785e3e Dutch: translate the 117 messages the catalog still lacks (batch 4) (#8486)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:26:12 +00:00
semmelrindeandClaude Opus 5.5 c2e1dbe6e4 Italian: translate the 125 messages the catalog still lacks (batch 4) (#8491)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:26:00 +00:00
Logan NightandClaude Opus 5.5 bda5f69c5c Preferences: reverse trackpad orbit direction (#8410) (#8484)
* claim: #8410

* Preferences: reverse trackpad orbit direction (#8410)

Add "Reverse trackpad orbit direction" under the pan settings in
Preferences > Navigation. orbit_reverse defaults off, so older settings
files load unchanged, and turns the Shift + two-finger orbit the other
way on both axes, as natural scrolling flips both. Pan, zoom and mouse
orbiting are untouched. Labels in en, ar, de, he, ja, ko, vi and zh as
a pending locale file; orbit_reverse in the control protocol's sorted
settings list.

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

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:25:24 +00:00
semmelrindeandClaude Opus 5.5 78d05e35bb Malay: translate the 125 messages the catalog still lacks (batch 4) (#8492)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:25:18 +00:00
semmelrindeandClaude Opus 5.5 70fdc27703 Czech: translate the 121 messages the catalog still lacks (batch 4) (#8485)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:24:15 +00:00
semmelrindeandClaude Opus 5.5 b650d7b49b French: translate the 121 messages the catalog still lacks (batch 4) (#8487)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:24:12 +00:00
semmelrindeandClaude Opus 5.5 883aff8816 Traditional Chinese: translate the 125 messages the catalog still lacks (batch 4) (#8498)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:23:58 +00:00
semmelrindeandClaude Opus 5.5 43cf7774aa Galician: translate the 125 messages the catalog still lacks (batch 4) (#8488)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:23:54 +00:00
raykholo 29749c73f3 Report missing translations for complete catalogs before review (#8542) (#8543)
* claim: #8542

* CI: report missing translations for complete catalogs before review (#8542)

`cargo xtask i18n --check` already reads the pending locale files and fails
when a complete catalog lacks an English message, but no workflow ran it, so
the gap surfaced only in the full test run after approved-for-ci. It now runs
in Format and repository checks, and CONTRIBUTING names the complete catalogs.
2026-10-28 19:51:47 -04:00
github-actions[bot] 05081e8428 Regenerate generated files and fold pending files [skip ci] 2026-10-28 23:50:48 +00:00
dependabot[bot] 00df48dfe4 Bump zip from 9.0.4 to 9.0.5 (#8572)
Bumps [zip](https://github.com/zip-rs/zip2) from 9.0.4 to 9.0.5.
- [Release notes](https://github.com/zip-rs/zip2/releases)
- [Changelog](https://github.com/zip-rs/zip2/blob/master/CHANGELOG.md)
- [Commits](https://github.com/zip-rs/zip2/compare/v9.0.4...v9.0.5)

---
updated-dependencies:
- dependency-name: zip
  dependency-version: 9.0.5
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-10-28 19:41:29 -04:00
bherbruckandClaude Opus 5.5 0656819123 Release 0.26.0: fold point-clouds into format 43 (#8579)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 19:32:15 -04:00
github-actions[bot] 926d753a40 Regenerate generated files and fold pending files [skip ci] 2026-10-28 23:29:43 +00:00
github-actions[bot] c03dccabb6 Regenerate generated files and fold pending files [skip ci] 2026-10-28 23:20:35 +00:00
bherbruckandClaude Opus 5.5 776653dd7c Release 0.26.0 (#8574)
Fold the pending design format change (linked-body-properties) into format 42.

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 19:17:37 -04:00
semmelrindeandClaude Opus 5.5 9b8be3b8a5 Romanian: translate the 121 messages the catalog still lacks (batch 4) (#8494)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:23:47 +00:00
semmelrindeandClaude Opus 5.5 d755b70e25 Swedish: translate the 121 messages the catalog still lacks (batch 4) (#8497)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:23:33 +00:00
semmelrindeandClaude Opus 5.5 0c4673a89d Russian: translate the 125 messages the catalog still lacks (batch 4) (#8495)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:23:29 +00:00
semmelrindeandClaude Opus 5.5 1b000c92f3 Hungarian: translate the 121 messages the catalog still lacks (batch 4) (#8489)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:23:28 +00:00
semmelrindeandClaude Opus 5.5 eb8cfdf8cc Indonesian: translate the 125 messages the catalog still lacks (batch 4) (#8490)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:23:25 +00:00
semmelrindeandClaude Opus 5.5 2da2c85f27 Norwegian Bokmål: translate the 121 messages the catalog still lacks (batch 4) (#8493)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:23:02 +00:00
semmelrindeandClaude Opus 5.5 359a7253e9 Ukrainian: translate the 121 messages the catalog still lacks (batch 4) (#8500)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:22:18 +00:00
semmelrindeandClaude Opus 5.5 64bbfe9ed3 Turkish: translate the 121 messages the catalog still lacks (batch 4) (#8499)
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 21:22:10 +00:00
e6dc105134 Point clouds: insert PLY, XYZ and PTS scans as a modelling reference (#5151) (#5381)
* Point clouds: insert PLY, XYZ and PTS scans as a modelling reference (#5151)

A point cloud is kept in the design as a reference, drawn as depth-tested
dots and never a body.

- io: a PLY reader (ASCII, binary little/big endian; x, y, z of any numeric
  type, optional colours, other elements skipped; counts checked against the
  file) and XYZ/PTS text, thinned evenly past max_points. PLY meshes with
  faces can also come in through file.insert_mesh.
- doc: Document.point_clouds (shared f32 points relative to an f64 origin,
  colours, units, rigid placement, size and colour), stored as a binary
  container part; format change in pending/point-clouds.rs.
- engine: pointcloud.insert / edit / delete / list; doc.open opens a cloud as
  a new design; selection.delete takes point_clouds; view bounds include them.
- gpu: an instanced point-sprite pipeline (16 bytes a point).
- ui: Browser Point Clouds folder with eyes, menu and an Edit Point Cloud
  card; drop, Open, Insert and file dialogs take .ply/.xyz/.pts; a CPU
  viewport fallback; messages in en, en-GB, ja, zh, de and ko.

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

* Point clouds: colour by the file, one colour or height, from the UI (#5151)

The Edit Point Cloud card gets Color By (File Colors, Single Color,
Height). Height is a blue-to-red ramp over design Z, between the cloud's
own two heights or, by default, a range worked out from the points (all
but the lowest and highest 2 %, from an even sample). pointcloud.insert
and pointcloud.edit take color_by and height_range; pointcloud.list
reports them. The colour of each point is worked out in one place
(PointCloud::shade) for the GPU and CPU viewports.

Also: the card's dropdowns compare the picked index (a dropdown's
response doesn't report the pick, so File Units did not change either).

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

* Point clouds: integrate format registration and viewport caches (#5151)

* Point clouds: satisfy current UI audits and document comparisons (#5151)

Route model RGB data through the theme conversion boundary, provide a compact ribbon label, and include point clouds in the new document comparison. Gate the Unix-only CLI-link helper and test so Windows all-target checks remain buildable after the rebase.

* Point clouds: isolate translations for current main (#5151)

Move the existing 74 messages unchanged into one pending locale file and attribute it through a pending row, following the current repository workflow. Repository checks, 50 tooling tests, final workspace Clippy, wasm and generated-catalog equality checks passed after the rebase.

* Point clouds: fit the rebased code to main's GpuTarget and Frame

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

* Compare point clouds when reusing placement-independent results

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

* Point clouds: complete command help and required translations (#5151)

* Point clouds: size the save cap from the raised limit after the rebase

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

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
2026-10-28 21:17:02 +00:00
866029aaa2 Engine: one sketch store for the sketch commands (W38.0, #5425) (#6336)
* Engine: one sketch store for the sketch commands (W38.0, #5425, no behaviour change)

The sketch commands found and kept the sketch they work on in about forty places, each its own
way. They now go through one module, cmd::sketch_store, so a drawing session (W38a/b) can put
sheet and view sketches behind the same calls without the commands changing again.

- target: the command's `sketch` param or the sketch being edited (target_sketch, moved).
- active: the sketch being edited, if any. The thirteen commands that need one keep their own
  refusal text; unifying the three texts is a separate, visible change.
- read and write: the only calls from the sketch commands to Document::sketch and sketch_mut.
- update: change a sketch on a copy of the document and keep both, the edit the text and link
  commands made by hand (sketch.text, sketch.edit_text, the projection, intersection and
  include links, sketch.fit_curves_to_section, sketch.break_link). As before, those paths
  neither solve the sketch nor run its checks.
- checked: the checks an edited sketch passes (the periodic splines), shared by edit_solved and
  sketch.offset.break.

edit and edit_solved stay in sketch.rs and load and keep their sketch through read and write.
Imported text, snap points and the face projections keep their own shape (a text that can't be
linked is skipped, snap points copy the document only when one is needed) through read and
write. The reads and writes that are about the design's own sketches stay as they are, each on
a line marked `// design-only:` (Redefine Sketch Plane, a joint's curve, a motion trace's wire,
and the session's live drag, parameter edits and `created` diff).

A source scan, each_sketch_command_goes_through_the_store, fails when a command reads or writes
a sketch, or asks for the active one, anywhere but the store or a marked design-only line in the
files that work on the design's sketches. Every existing sketch test is unchanged.

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

* Engine: the sketch store's scan also catches the active sketch taken as a value (#5425, review of #6336)

The scan only caught `active_sketch.ok_or`, so main's `let Some(id) = s.active_sketch else …`
could have come back unseen (0xlillith's review). It now also flags the value-read forms:
`) = s.active_sketch` (let-else and if let), `=> s.active_sketch` (a match arm),
`s.active_sketch.and_then(` and `s.active_sketch.map(`. State checks (`is_some`, `is_none`,
`== Some(`), assignments and snapshot copies stay allowed.

The wider scan found five commands that hand-rolled the store's `target` rule (a `sketch` param or
the sketch being edited). They now call `sketch_store::target`: sketch.inspect, sketch.glyphs,
sketch.export_dxf, inspect.curvature_comb and inspect.minimum_radius (their `curves` form).
Their refusals become target's:
- no `sketch` and none being edited: "no sketch is being edited" (was "no such sketch (and none
  is active)", or "no such sketch" for sketch.glyphs);
- an unknown name or id: "no sketch `X`" (an unknown id used to reach "that sketch is not
  evaluated");
- a feature that is not a sketch: "`X` is not a sketch";
- a `sketch` that is neither an id nor a name (true, a list, an object): refused,
  where it used to fall back to the sketch being edited. A null `sketch` still means none given
  (the scenario harness sends one): target now reads it that way for every sketch command
  (target_reads_a_null_sketch_as_none_given).
No test, doc or schema text named the old wording; the params docs ("sketch?: id|name (default
active)") still hold.

The two remaining value reads, the default sketch for a feature (features.rs) and the sketch a
measurement prefers (measure_sel.rs), go through `sketch_store::active`, with no change.

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

* Engine: a number in `sketch` is an exact sketch id, never a name (#5425, review of #6336)

`sketch_store::target` turned a JSON number into text and looked it up with
`Document::find_feature`, which falls back to a name when no feature has that id. So
`{"sketch": 999999}` could pick a sketch named "999999" whose id is something else, and export
or inspect it. sketch.export_dxf, sketch.glyphs and the two curvature inspects read an exact id
before they moved onto the store; the other sketch commands (the old `target_sketch`) and
sketch.inspect read a number as text.

One rule now, for every command that takes `sketch`: a number is an exact id (`as_u64`, then
the feature with that id); text is an id written out or a name, as before. A number that isn't
a whole number of 0 or more (negative, fractional) is refused with "`sketch` must be an id or a
name"; an id no feature has gives "no sketch `N`", as before. Nothing in the tests, docs,
scenarios or examples names a sketch with digits, so nothing relied on the old fallback.

Test (sketch_store): a sketch named "999999" with another id. sketch.export_dxf, sketch.glyphs
(with another sketch being edited, for its enabled check), inspect.curvature_comb and
inspect.minimum_radius (with a real curve id) refuse `{"sketch": 999999}`, find the sketch as
`"999999"` and by its own id, and refuse -1 and 1.5.

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

* Engine: MidPoint finds its sketch through the store too (#5425, after #5011)

#5011 added a caller of the old `target_sketch` (MidPoint's `line_stays`, which swaps a point
picked after its line) after this branch's base. It now reads the sketch through
`sketch_store::target` and `sketch_store::read`, as the other sketch commands do, so a number
there is an exact id too.

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

* Engine: standard parts fix their sketch's entities through the store (#5425, after #6376)

#6376 (a standard part's sketches are fully defined) reads the sketch being drawn straight from
the document's features, after this branch's base. `fix_all` now takes it from
`sketch_store::active` and `sketch_store::read`, as the store's scan requires, with the same
result.

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

* Engine: Toggle Diameter and Delete on a profile find their sketch through the store (#5425)

Two sketch commands landed on main after the store: sketch.toggle_diameter (#7729) still
called the removed sketch::target_sketch, and Delete on a picked profile (#7107) read the
sketch straight from the document. Both now go through sketch_store (target and read), and
the store's source scan passes again.

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

* Sketch dimensioned: use the sketch store's target after the rebase

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

* Sketch dimensioned: read the sketch through the store

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

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
2026-10-28 20:56:05 +00:00
Nathan Dane 85a60c1af6 Render output: Output editor, size presets, Check list, Copy, progress pill and result viewer (#6281) (#6433)
* Render output: size presets, Check list, Copy, progress pill, R and Esc, and the result viewer

* Complete Output GPU help translations and format choices

* Remove Output size labels already present in catalogs

* fix: compile the legacy finishing controls only for tests
2026-10-28 20:55:55 +00:00
semmelrindeandClaude Opus 5.5 683535a18f Evaluation: a pattern of another component's body follows its own component when that moves (#8467)
A pattern of bodies copies them about an axis in its component's frame,
so its copies of a root body depend on where that component is placed.
placement_digest never counted a pattern's bodies, so moving the
component kept the cached copies. Read each body through placed_body,
as Combine does.

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-28 20:49:55 +00:00
Logan NightandClaude Opus 5.5 d67c8e6231 Sketch text: Arabic and Hebrew are ordered right to left and joined (#8237) (#8324)
* claim: #8237

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

* Sketch text: right-to-left lines are ordered and shaped (#8237)

A line with Arabic or Hebrew letters is split into display-order runs by the
Unicode bidirectional algorithm (unicode-bidi) and each run is shaped with the
font's own rules (harfrust), so letters join and marks sit where the font puts
them. Lines without right-to-left characters keep the glyph-by-glyph layout.

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

* Sketch text: a line break inside a right-to-left line keeps the text after it (#8237)

The bidirectional algorithm ends a paragraph at a line break; every
paragraph is shaped now, not only the first, so measuring such a line
counts all of it.

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

---------

Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-10-28 20:31:02 +00:00