Review follow-up (#3785, CodeRabbit): commands in SELECTS_GIVEN_IDS run select.set with the given ids before their body; when the body then returned an ordinary error, the ids stayed selected (and their layer current), while a panic restored the selection and refused ids never changed it. The dispatcher now keeps the active document's selection and current layer from before select.set and restores them when the body errors. This affects every command in SELECTS_GIVEN_IDS (Duplicate, Arrange, Clipping Mask, Compound Path, Group/Ungroup, Hide, Lock, Move, the Pathfinder panel commands including Trap, and the transforms). Tests: Trap with ids whose edges don't meet, and Intersect with ids of shapes that don't overlap, leave the selection as it was.
Review follow-up (#3785, CodeRabbit): traps() tried every pair of visible solid paths on each redraw of the live effect and in the panel command, so many separated shapes cost quadratic work that neither the segment cap nor the overlap budget bounds. Each visible part's bounds are now computed once and a sort-and-sweep on x finds the pairs whose bounds come within the trap's reach (the clamped thickness across x, thickness × height/width across y, plus a hair for rounding); only those reach same_color, lighter and trap_band, which rejected the rest by the same bounds test, so the traps and their order are unchanged. Tests: 2,000 separated squares of two colours and one abutting pair give one candidate pair and one trap; the traps of a mixed scene match those of each pair alone, in order.
The Pathfinder panel menu's Trap… (object.pathfinder.trap) and Effect › Pathfinder › Trap… (pathfinder.trap, live on groups) share one geometry (vectorcraft_pathops::trap_band, vectorcraft_effects::trap): Thickness, Height/Width, Tint Reduction, Traps with Process Color and Reverse Traps, plus the effect's Precision and Remove Redundant Points. The panel's Trap is one undo step and groups the traps above the topmost selected object. Behaviour from https://helpx.adobe.com/illustrator/using/trapping.html and https://helpx.adobe.com/illustrator/using/pathfinder-panel-overview.html.
Review follow-up (CodeRabbit): -80 pt at 0° drew the line leftwards but
kept 80 pt at 0°, so the next click drew it rightwards. Both now use
80 pt at 180°.
Step 3 of the Rectangular Grid tool and Arc tool help pages
(https://helpx.adobe.com/illustrator/using/tool-techniques/rectangular-grid-tool.html,
https://helpx.adobe.com/illustrator/using/tool-techniques/arc-tool.html)
and the Tools page (https://helpx.adobe.com/illustrator/using/tools.html)
say a double-click on the tool opens its Tool Options. For these four tools
the double-click (tool.options, and Enter once #2891 lands) said the tool
had no options (#2691).
The Tool Options are the tool's click dialog in dialogs/shapes.rs, opened
without a place to draw (shapes::open_options; a `__toolOptions` field
marks it), as the review of #2979 asked: no copied form code. Its OK runs
tool.setOption and draws nothing; Cancel changes nothing. The grids' dialogs
are titled Rectangular Grid / Polar Grid Tool Options there. The options
are what the next drag draws with and what a click's dialog opens at:
- Line Segment: Length and Angle (now kept by the tool and used by the
click dialog) and Fill Line (new: shape.line {fill: true} gives the line
the current fill; the click dialog has it too);
- Arc: lengths, Reference Point, Type, Base Along, Slope, Fill Arc;
- the grids: dividers, skews and switches, plus the Default Size and
Reference Point, now kept by the tool (width, height, reference) for the
click dialog; a drag leaves them out of its command.
The grids' Reference Point is drawn only as the 9-point locator (it also
showed as a plain number field).
New strings (Rectangular Grid Tool Options, Polar Grid Tool Options, Fill
Line) have rows in all 24 catalogs.
Verification:
toolbar::tests::double_clicking_the_line_arc_and_grid_tools_opens_their_tool_options
double-clicks each tool's toolbar button: the dialog and its title, Cancel
leaves the next drag as it was, OK draws nothing, the next drag uses the
option and a click's dialog opens at it (a grid at its Default Size);
line_segment_options_and_fill_line, grid_default_size_is_an_option_the_drag_leaves_out,
fill_line_gives_a_new_line_the_current_fill. Grid dialog tests now expect
the stored Default Size. Live: double-clicking the Rectangular Grid tool in
the release build opens Rectangular Grid Tool Options; OK with 2 and 3
dividers, then a drag, draws a 3 × 4 grid.
On Linux under X11, and under XWayland, the X server lists a pen's
eraser end as a pointer device of its own, and winit 0.30 hands it to
the window as a mouse. Turning the pen over and dragging the eraser
end drew with the active tool, for example a Paintbrush stroke.
The tablet reader now marks a pen device whose name or XInput type
contains "eraser", and Services::stylus carries the pen's end with
its pressure. Once a frame while no press is held and the pointer is
in the window, the canvas selects the Eraser when the eraser end
moves, and it selects the tool used before when the tip moves again.
The return goes through Session::switch_tool, so a perspective grid
hidden meanwhile stays hidden. A tool chosen while the eraser end
stands in stays when the tip comes back. The feed starts afresh when
the pen turns over, so a press of one end does not take the pressure
the other end last touched with. Native Wayland is unchanged.
The switch follows PhotoCraft's sync_eraser_tool
(storytold/photocraft#255, and storytold/photocraft#5043 for the
cleared tool).
Docs: docs/development.md, docs/gaps.md, docs/hardware-parity.md,
docs/target-app-parity.md.
The list view gives colours and graphics a row each with the item's name (a colour's hex value too, where it fits); items drag and drop by row there. Graphic thumbnails are shrunk on the CPU to the size they are drawn at, as egui-wgpu doesn't mipmap textures and thin strokes broke up in small tiles.
Two Polish rows added this week:
- "Plug-ins & Scratch Disks" read "Wtyczki & Dyski robocze": every other "&"
in the catalog is "i", and Polish doesn't capitalize the second noun.
- "Geometric Guides" read "Prowadnice geometryczne"; Guides are "linie
pomocnicze" in the rest of the catalog (Guides, Smart Guides …), as in #2870.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DukHNvUY45rjRnvtz91x33
- Packaged … in {folder} {note}: keep the space before {note}, as the
English does, and use "Paketoitiin" in the status line like the @msg row
- Colored Pencil: Värikynä (Väriliitu is a crayon)
- Underpainting: Pohjamaalaus
- Select artwork to apply effects: use "kuvitus" like the rest of the catalog
- Arc Segment Tool Options: Kanta akselilla, X-/Y-akselin pituus
- Plug-ins & Scratch Disks: Laajennukset ja työlevyt
- Add the Finnish row for "PDF security settings are not an object" (#3836)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* i18n tests: report missing translations instead of failing
The scans for untranslated menu strings, tl! literals, undo labels, menu labels and status
messages print what a complete language lacks. A missing translation shows the English
text until a translator adds the row, so it no longer fails the gate.
* i18n: one untranslated report in xtask ci; drop one-language wording tests
`i18n::tests::untranslated_report` lists, per language, the menu strings, tl! literals, undo
step names and templates, menu labels, render-site labels, status and error messages,
PostScript error meanings and default swatch names its catalog lacks; strings every language
lacks are listed once. It never fails on a gap, only on a scan that finds too few strings.
`cargo xtask ci` runs it in an `i18n-report` step, and `cargo xtask i18n-report` runs it alone.
The five scan tests, the render-site and swatch coverage tests and untranslated_messages.txt
fold into the report. Tests that pinned one language's wording are gone; where they also
checked code (lookups, plural rules, locale tags, menu and History step names, engine
parameter errors), that check stays and compares with the catalog lookup.
As Illustrator's help describes (Filter fonts by classification, Font Technology › Variable), the font menu gets a Font Technology row whose Variable choice lists only the families with variation axes, together with the script, kind, favourites and search filters, and stays set from one opening to the next. A family is variable when any of its faces, loaded or scanned, has an `fvar` table, so a family installed as static files beside a variable one counts.
Spec: https://helpx.adobe.com/illustrator/desktop/design-with-text/fonts-and-scripts/find-apply-and-adjust-variable-fonts.html
COLR and SVG glyphs, which have no outlines, were left out wherever type was taken as its outlines: geometry effects, Expand Appearance, EMF/WMF export.
A root size without a viewBox sets up the em, a viewBox that isn't square is centred in it, and a glyph element nested in others is drawn on its own, as if used.
A PDF names fonts by PostScript name. When the font isn't installed at import, the importer
splits the name (`Honoka-Mincho` → family Honoka, style Mincho) and the document keeps that
pair. Installing the font later didn't help when its family name shares nothing with its
PostScript name (ほのか明朝): the pair stayed missing. FontDb::canonical now also looks the
pair up, joined, among the PostScript names, and answers with that face's own family and style.
The renderer placed symbol art that holds a blend, an envelope or a
repeat again for every instance in every frame. The live object in the
placed copy was a new node each time, and the renderer evaluated it
again. Symbol art that is itself a live object, an opacity mask whose
art is one, and a live object in a clip group in Outline view were
drawn from an uncached copy in every frame. At 1000 x 1000 pixels,
2,000 instances of a symbol holding an envelope took 126 ms a frame on
main, and the same 2,000 envelopes on the page took 30 ms.
The renderer now keeps the placed art of instances and inline graphics
when it holds blends, envelopes or repeats, and when the art is itself
a live object. It draws kept live art, opacity-mask art and Outline
view clip members that are live objects from the live cache, which
evaluates them again when the symbols, patterns, fonts or color
settings they show change. The 2,000 instances take 28 ms a frame,
with the same pixels.
Docs: docs/target-app-parity.md
Review follow-up (CodeRabbit, clean-room): the test header, gaps.md and
target-app-parity.md quoted a phrase from the cited help page. Comments
and docs only.
A file opened or placed off the UI thread waited while the active tool
reported busy. The Dimension tool also reports busy between its clicks,
so after one click a ready file stayed queued with its spinner until
the dimension was finished or canceled. Files also waited between the
clicks of the Blend tool.
The session now records whether the pointer button is held. A file
waits for a busy tool only while the button is held. A file that
opens between a tool's clicks resets the tool, as a switch of
documents does; a file placed in the active document leaves the
tool's clicks in place. Files still wait while a dialog is open, a
drag is under way or a preview is open, and a tool's preview between
its clicks counts as one (the Flare tool after its center drag, the
Dimension tool's line once the pointer moves after its second click).
Taha-Mahmoodi