A damaged .ai can give a gradient mesh whose outline is enormous but finite.
draw_mesh_raster flattened that whole outline in device pixels (grown_outline),
and kurbo's flatten then asked for more segments than memory holds: a
"capacity overflow" panic (seen in import_fuzz's ai_written_meshes_never_panic).
An outline wider or taller than 2^24 device pixels now goes to the per-quad
drawing path the function already falls back to.
Running a script or a plug-in now ends the spot color base's link, as a
batch step already did, so a script that deletes the spot color and makes
another of the same name does not make the new one the base. Commands that
only list scripts or plug-ins keep the link.
While the link waits for an Undo, the color it captures before a command is
the one the guide shows, not that of a new spot color of the same name.
A printer name in "Sent “{title}” to {to}" was translated as a message of
its own, so Korean's "{status} ({e})" row turned "Office Laser (Floor 2)"
into "Office Laser(Floor 2)". An unquoted `{to}` value is now translated
only by an exact row, such as "the default printer"; a quoted one stays
as written. The cross-language case uses a printer name with
parentheses, and two quoted `{to}` names that match exact rows.
Overrides::is_set is only called from main under cfg(not(macos)), so it is allowed to be unused there (the repo's cfg_attr pattern). settle_cut_edges walks blended with iter_mut().enumerate() instead of indexing it.
Look::takes let a glyph filled only join a run whose glyphs were also
stroked, so after mode 2 text a mode 0 glyph took the run's stroke:
`2 Tr (B) Tj 0 Tr (C) Tj` came in as "BC" stroked throughout. A glyph
now joins only a run drawn alike (fill, stroke, overprints), so a glyph
filled only starts a run of its own and takes a stroke only when the
matched-stroke path (#8039) strokes it.
flush_stroke no longer strokes a whole run in place: a whole run goes
through the split and join code, so fill and stroke rendering glyph by
glyph still builds one stroked run.
egui places a popup by its size the frame before, so on the frame a late-measured name widens the
menu the filter text can stay where it was while the menu's rect already reaches past the window.
The helper now waits until both the filter and the menu's rect stay put, which fixes the
intermittent failure of a_menu_wider_than_its_field_stays_in_the_window (the menu ending 0.7 pt
past the window's edge).