- Check Spelling: with #8726's measured form a `change` made in the dialog records no step
(the reference application records one `spellCheck` step), so #4514's two tests that
expected recorded `change` steps are removed; its portable recording stays for all else.
- Clippy `chunks_exact_to_as_chunks` in #10765's tests; a single-element loop in #10755's.
- docs/parity-checklist.md and docs/scorecard.md regenerated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Link Layers keeps main's measured toggle mapping (batch 12, #10628): played twice in the
reference application it links, then unlinks. #10755's second, explicit `linked: true`
entry for the same event was unreachable on import and contradicted that measurement, so
it is dropped; Unlink stays explicit. Its tests now cover Unlink only, and `{}` exports.
- #10765: an ordinary Gradient Map now saves as `gradientMapEvent` (#10702 landed in batch 12).
- #10755: flattened pixels read with `Buffer::into_px()` since `Buffer` implements `Drop`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Reveal All runs its crop through crop_keeping_background instead of
Session::execute, so the session's authorize gate no longer saw the
image.crop it runs on its behalf. It now calls the gate with the crop's
params before the crop, as dispatch would.
The image.crop params doc now says the Background becomes Layer 0 with
Delete Cropped Pixels off only when contentAware is false.
Tests: the action-step case now goes through a binary .atn (the Crop
descriptor carries Dlt false, import decodes it, actions.play runs it);
new tests for the Reveal All gate and for Content-Aware with Delete off
keeping the Background.
Main's new profile dither (8a0f959d2) matched on the tile bytes by value inside the
per-pixel loop. On this branch Tile::bytes returns a TileBytes guard rather than a
slice, so borrow it for the loop instead, and compare the test's tiles as owned
bytes since the guard has no Debug or PartialEq.
The GIF branch checked one frame against MAX_GIF_FRAME_PIXELS but kept every
encoded frame until the file was written, so frames x pixels was unbounded: a
4000 x 4000 canvas on a long timeline staged terabytes. Check the total
(one index byte per pixel plus the palette, 2 GiB like imported footage) and
the 16-bit GIF side before the first frame is rendered, with checked u64
arithmetic and an error that names the limit.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
- A loaded spellCheck step reads null and Usng by name, in either order.
- A correction step (action set) with its dialog on runs without opening
the Check Spelling dialog.
- Done continues a waiting action only on the document it was playing on;
on another document the action ends there and the status says why.
- When recording the step fails at Done, the status keeps that error next
to the playback's outcome.
Main moved the Save As, Save for Web and .atn code (cmd/atn_cmds,
cmd/web_cmds, app/files.rs, dialogs/) and added DDS and JPEG Options
prompts, so the fix is applied again there:
- Save As with the suggested name records {"folder", "extension"}, also
when answered in the Targa, DDS or JPEG Options prompt, and a Save of
the document's own file is still journaled as File > Save.
- A loaded "save in directory" (DIDr) Save As imports as that folder step
and exports back with DIDr, instead of staying a kept step; a copy into
a folder stays kept.
- Playing it on two documents writes one file for each, under its name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A scratch write or create failure, or a volume about to drop below max(1 GiB, 5%)
free space, now pauses spilling with exponential backoff (1 s doubling to 60 s)
instead of disabling it for the session. Tiles stay resident, set_budget keeps
the latest budget for the resume, preference changes retry at once, and
spill::epoch lets the engine re-apply History's byte limit while paused.
spill::set_size_cap adds an explicit size limit; free space comes from sysinfo's
disk feature.
One unreadable tile used to block editing, saving and exporting every open document
(spill::check_integrity was process-global). Add Tile/Surface/Document::scratch_health
(free while no tile has failed) and Document::for_each_surface (ops' history walk now
delegates to it). edit_with checks the document before and after, jobs the document they
captured, import the result, save/export the document written; check_since is now
generation-only so a failure or recovery during an operation refuses it once and a retry
works. Caches check their own sources and purge once per new failure instead of every
frame. Fault-injection tests use a duplicated handle (the Unix file has no name).
The spill manager is process-global and never dropped, so the Unix scratch file was only
removed by the next launch's reap. Unlink it immediately after opening and read/write
through the one handle with pread/pwrite (readers can no longer reopen it by path).
Windows keeps delete-on-close and per-thread readers. The abrupt-exit test now asserts
that no file remains on every platform; docs/scratch-disks.md updated.
Resolve conflicts with main's prefs UI split, session/execute.rs move, GPU memory
budget (gpuMemoryMb, 0 = auto, kept from main) and effective_memory_usage_mb history
budget. Port the Memory Usage % and Scratch Disks preferences into prefs_ui, the edit
integrity guard into session/execute.rs, and adapt main's new pixel readers to the
guarded TileBytes API. Cargo.lock regenerated; scorecard regenerated.
Delete Cropped Pixels clips pixel layers, fill caches and video frames only. When a layer mask,
channel, the quick mask, the selection, type, a shape or a smart object holds the hidden pixels
over the budget, the refusal now says so instead of pointing to an option that can't clear it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>