- Replace whole-region Vec<f32> read_region/write_region in CutParts::new
with tile-parallel cut_piece + try_fill_surface/try_clear_surface,
sharing interior Arc<Tile> allocations in O(1) and activating the
solid_tile hole-fill fast path on Background and mask targets.
- Cache piece_bounds on CutParts so CutParts::place avoids scanning tile
content bounds on every drag frame.
- Parallelize composite_over across top tiles with Rayon and share fully
opaque top tiles by Arc<Tile> clone in O(1), preserving COW tile
identity for GPU canvas uploads.
- Parallelize Surface::translated across target tiles with Rayon and
prune cleared tiles in-place inside edit_tiles.
- Skip zero-offset floating preview generation on pointer-down in
move_ui::floating_doc.
- Exclude SnapKind::Selection when moving selected pixels with the Move
tool so small initial drags do not snap back to (0, 0).
- Keep navigator_plain active during Move/Brush/Patch drags on a GPU
canvas so navigator_gpu is not evicted into a synchronous full-document
CPU thumbnail composite on the first drag frame.
- Parallelize row normalization in compose::render_reduced_in_bands and
add a direct read_rgba_into fast path for unmasked Normal layers over
an empty backdrop.
A soft shape's transform frame, snap rectangle and Properties X/Y/W/H came from
the path plus the full stroke width on every side, so a centred or inside stroke
grew the frame when Density or Feather was set. The frame is now the pixels the
same shape renders at full Density without Feather (edge strips only, clipped to
the canvas), identical to the hard shape's for every stroke alignment.
feather_matches_photoshops_ramp now enforces the 1-level claim the scorecard
makes (measured max deviation 1 at 8, 16 and 32 bit).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The cfg(test) budget was 4 MB, 2^20 px of 8-bit RGBA only, so the turned
crop test that landed in #5007 (an 862 x 803 px layer at 16 and 32 bits)
was refused. It is now 2^20 px of the widest format (32-bit CMYK with
alpha, 20 MB), so whatever the old 2^20 px test limit let through still
fits; the refusal tests stretch far enough to stay over it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The warp places pixels within +-2^20 px (WARP_REACH) and clamps the rest
away. The byte budget admitted results wider than that (a 2,000,002 px
strip fits 8192 MB), and a plain move past it already emptied the layer
without an error. `warp_output_clipped` says when the result would be cut,
and the budget check refuses it: "would reach past +-1048576 px, the
farthest a transform places pixels". Docstrings for the touched functions.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The fixed 2^30 px cap refused any transform of a stitched layer past about
1 GP, on any machine. A result is now priced in bytes at its pixel format
(with the alpha the warp adds) against Preferences > Performance > Memory
Usage (default 8192 MB), which Session::apply_prefs sets. The report's
37708 x 35259 px 8-bit result (5072 MB) now transforms: 10.7 s, 6.1 GB on
top of the source. Past the allowance the message names it:
"... (10144 MB), more than the 8192 MB Memory Usage allows (Preferences >
Performance)".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The 2^30 px refusal went through allocation::error, so it read "not enough
memory for transforming pixels (...)" although nothing had been allocated.
It now says the result is too large to transform and names the limit.
A real allocation failure still says "not enough memory".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Snap To Layers built each layer's edges and centre from transform_bounds, which clips the
layer to the selection. With a selection, every layer's targets were the selection's own
bounds, so dragging the outline with a marquee snapped it back to where it started (shown
as layer alignments) and beat nearby guides and document edges. Layer targets now use the
layer's whole content bounds, as in Photoshop; the moving box (layer_rect) is unchanged.
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Free Transform and Warp moved a fill layer's linked masks (and so the frame its
gradient is laid out in) but kept the cached render, which the compositor draws
as-is while the Fill is unchanged. Drop the cache in apply_transform and
warp_layer, as transform_layer_geometry and Perspective Crop already do.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A shape below 100 % Density renders over the whole canvas, and a Feather
grows its pixels, so Free Transform, the Move tool's transform controls,
snapping and Properties > Transform framed the canvas or the blur. As in
Photoshop the frame now follows the path and stroke whatever the mask
Density and Feather: photocraft_compose::hard_shape_bounds (shape_bounds
without the soft reach) feeds whole_layer_bounds and the Properties
Transform fields. Rendering keeps the soft bounds; other layers and hard
shapes are unchanged.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
With a selection, ⌘T and the Move tool's Show Transform Controls framed the
layer's content clipped to the selection, so the box was smaller than the
marching ants wherever the layer did not fill the selection. The ants then
stayed where the selection started until the transform was applied.
- transform_frame / target_frame (engine): the box is the selection's own
bounds when a selection limits the transform. transform_bounds is unchanged,
so snapping and the command defaults are as before.
- The selection outline is drawn through the open box (homography, or the warp
in Warp mode), so it follows move, scale, rotate, skew, distort, perspective
and warp before the transform is applied. The whole outline is traced while
such a box is open, not only the part in view where it started.
- marching_ants_lines: ants along screen-space lines; non-finite lines are
skipped and very long ones are clipped to the view.
Part of #3402.
Claude-Session: https://claude.ai/code/session_015vKPoMXFhSvWmX5D9XM646
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
Stretching a 1 px row or column faded out along the stretch: bicubic and
bilinear sampling read transparent pixels past the source edge, and with a
selection the edge was the whole layer's bounds (lifted pixels kept their
colour at zero alpha). Samples past the source rect now repeat the edge and
the outline is antialiased by coverage, and the lifted pixels' bounds are the
selected pixels'.
Fixes#3773
Co-authored-by: Fabio Lussi <198260595+itrycodinganditsbad@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Committing Free Transform on a 300-550 MP layer ran the bicubic warp on the UI
thread: several seconds of a frozen window with no feedback. On a 20000x16000
layer the warp alone took about 6.6 s.
edit.transform is now a background job (edit_job): the warp reports progress per
tile and stops when cancelled. The transform session stays open while it runs,
showing the moving pixels without the box, so the layer doesn't jump back to its
old place until the result lands. The modal progress dialog shows after the
usual delay, and Esc or Cancel cancels the job. A copy or placed layer the
session made is folded into the step, or taken back, in its own document.
Undo and Redo are refused while the commit runs.
Tests and scripts run inline as before.
Co-authored-by: Brandon Thomas <echelon@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A large scale (e.g. 200x on a thin column) made the warp fill every tile of
a result up to 2^21 px on a side, enough to exhaust memory. Free Transform
now prices each warp before building anything and refuses with an
allocation-style error ("...; the document was not changed") once a result
would exceed MAX_TRANSFORM_PIXELS (2^30, the same bound as Puppet Warp's
MAX_WARP_PIXELS). Only the result's size counts, so moving content far off
the canvas still works, and nothing is clipped.
The check covers the layer, a smart object's cache and filter mask, linked
masks, a lone mask/channel target and the selection outline, and runs over
the whole layer tree before any layer is changed. Transform Selection gets
the same check. The transform box now stays open when a commit is refused.
Part of #3773
Co-authored-by: Fabio Lussi <198260595+itrycodinganditsbad@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
- edit.transform gains the relative form (scaleX/scaleY %, rotate, dx/dy about the frame
centre), as a Photoshop action records a Free Transform; image.applyImage takes a source
layer by name.
- .atn events: transform (layer → translate or relative transform; selection →
transformSelection), applyImageEvent, export (Save for Web, with the txtu path record
decoded and written back; kept verbatim without a destination), newPlacedLayer /
placedLayerEditContents / updatePlacedLayer, close with the Save answer → save + close,
delete of the target channel → layer mask delete, Quick Mask on/off, polygon selections →
select.lasso, Gradient Map (direct and as an adjustment layer). 92 → 105 events.
- Export tries every event listing a command id (Quick Mask on → set, off → clearEvent).
- Docs: event counts in the parity docs.
* Load and save Photoshop Actions (.atn) (#219)
- photocraft-psd: `atn` parser and writer for version-16 action sets, exact round trip,
truncation/garbage safe, fuzz target and `atn_dump` example. Descriptor fix: `indx` and
`Idnt` reference items carry a class before the number (real action files failed to parse).
- photocraft-engine: `atn_cmds` maps 92 Photoshop events to commands and back (filters,
adjustments per channel, image/canvas/crop/trim/rotate/flip/mode, make/set/select/delete/
hide/show/move/duplicate, fill/stroke, selection, clipboard, purge, profiles, play, stop).
Other steps load as `actions.photoshopStep` (descriptor kept, written back by save, fails at
playback by name); tool selections load switched off. Commands `actions.load`,
`actions.save`, `actions.stopMessage`, `actions.photoshopStep`. `Action` gains `dialogs`
and `set`. `layer` params accept a layer name, so loaded steps replay in any document.
- UI: Actions panel menu › Load Actions… / Save Actions…; Photoshop steps labelled.
- CLI: `batch --actions set.atn`. Automation policy covers the new path params.
- Docs: file-format parity (.atn R/W), gaps G11, target-app parity, scorecard AUTO-219-2
partial, development guide.
* Translate the Actions panel's Load/Save strings in every catalog
The eight new panel strings (Load/Save Actions…, the loaded/saved counts, the kept-step label)
were missing from all 19 catalogs; the languages that were complete are back at 100%.
* Show the Actions entries in the default dock's panel menu
The legacy dock renders its own group menu and only knew the Layers and Swatches entries, so
Load Actions… and Save Actions… were missing there.
Photoshop's Background is the bottom layer of a file without transparency: a pixel layer that is
transparency- and position-locked, whatever the file's language calls it ("Hintergrund", "背景",
"Arrière-plan"). We matched the name "Background" in compose (deep / shallow knockout), the
engine (cut/clear, merge visible, transforms, warps, distort, erasers, styles, masks, crops,
straighten, smart-object unpack) and the UI (Properties, layer row, tab title), so a localized
file's deep knockout revealed transparency instead of the Background.
- `Document::background` / `background_mut` / `is_background` and `Layer::has_background_traits`
in the doc model replace every name check; `extra_cmds::background_to_layer` is the one
"Background → Layer 0" conversion.
- PSD import marks the bottom record position-locked when it is transparency-protected and has
no transparency channel, with or without an `lspf` block. Affinity import and HDR Toning now
lock the Background's position like every other place that creates one.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Edit › Transform › Distort / Perspective and Warp on a shape layer let the
user edit the box or mesh and then refused at commit ("needs a rasterized
shape"). Photoshop maps the shape's path and keeps it a shape.
- photocraft-vector: new `map` module. `map_points` sends every anchor and
control point through a map (Distort / Perspective: exact for straight
segments, exact end tangents for curves). `map_path` follows a curved map
(Warp), halving each segment until the mapped control points follow the
mapped curve within a tolerance (at most 64 pieces a segment, bounded in
total); straight segments whose image is straight stay straight.
- engine: `transform_layer` and `warp_layer` map a shape's path instead of
refusing; a live rectangle/ellipse/polygon/line becomes a plain path. A
linked vector mask now follows Distort / Perspective / Warp too. A map that
sends a point to infinity is an error, not a corrupt path.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
edit.transform without a "target" (or "layer") param now follows the
Channels panel: a targeted alpha channel or the Quick Mask transforms
instead of the layer. Edit › Transform › Rotate 180°/90° and Flip
Horizontal/Vertical are routed to the panel target like the paint and
filter commands, so they turn a targeted alpha channel, the Quick Mask or
an unlinked layer mask within its frame (one undo step). The Background
no longer greys the transforms when a channel is targeted.
Flips scorecard row BUG-207-4.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Free Transform on type: whole-layer frame, grey Distort/Perspective (#2630)
With a selection active, Free Transform on a type layer framed only the
selected part of the text and the preview lifted half of the rendered
glyphs, but the engine then transformed the whole type layer, so the box
and the result disagreed. Type, shapes and smart objects always move
whole, so transform_bounds no longer intersects their frame with the
selection and the preview no longer lifts pixels out of them.
Distort and Perspective stayed enabled for editable type and then failed
on commit. They are now greyed for type layers (menu bar, marquee and
transform-box context menus), as in Photoshop, unless an unlinked mask or
channel is targeted.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* transform_type_tests: set CanvasToolMenu::pasteboard (#2630)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Fabio Lussi <198260595+itrycodinganditsbad@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Cmd+T (and Edit > Transform, the Move tool's transform controls) on a
layer locked in place opened the box and previewed every drag; only the
commit failed. Photoshop refuses before the box opens with "Could not
complete the Free Transform command because the layer is locked."
transform_cmds::locked_for_transform applies the commit's own rule (its
own position/all lock or an enclosing group's, a group's locked layers,
never the Background, which becomes Layer 0) and transform_tool::begin
asks it first, opening the alert in the Error dialog as the Move tool's
lock message does. Translated in every catalogue.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
* Make large pixel operations allocation-fallible
* Warp: drop the fixed 400 MP cap; fallible reservations already cover large transforms
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Adapt #2530 to main: fallible steep-Distort warp, Result from mask paste in float moves; regenerate scorecard
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Brandon Thomas <echelon@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Smart objects keep their four-corner (Distort / Perspective) placement
A smart object's placement was an affine transform, so a perspective
placement lost its fourth corner: a mockup PSD's distorted screen looked
right on open (Photoshop's cached pixels) but re-rendered as a
parallelogram after Edit Contents → Save.
- SmartObject gains `perspective`, the full projective map (source px →
document px); `transform` keeps its affine approximation
- Re-renders go through it (place_source_projective), as do moves,
canvas rotations and edit.transform, so Distort / Perspective now work
on smart objects instead of being refused
- PSD import reads the quad (nonAffineTransform / Trnf) and export writes
it back; .pcraft stores the field (absent for affine placements)
- Test: a distorted smart object survives Edit Contents → Save
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Smart objects: warp maps through a distorted placement; perspective round-trip tests
Warp handles and stored warps on a Distort / Perspective smart object go through its full
projective placement (smart_cmds::placement) instead of the affine approximation, so they sit
on the distorted corners. Tests: SoLd perspective round trip and quad_moved (io), a perspective
smart layer in the .pcraft rich_doc, warp on a distorted smart object (engine). Comments no
longer name another product.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Jason Short <3344856+jason4short@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: Brandon Thomas <echelon@gmail.com>
Rust 1.99's clippy adds chunks_exact_to_as_chunks and some_filter, which
broke the Clippy step on every CI runner (it uses the latest stable). All
sites are mechanical: as_chunks drops the remainder exactly like
chunks_exact.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
CI's `cargo fmt --all -- --check` has failed on every run because the
repo had no rustfmt config and was never formatted. Use the same settings
as EffectCraft/FilmCraft (max_width 160, use_small_heuristics Max), which
also keeps the churn small. Formatting only; no code changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
warp_gray cleared the old + warped region as one dense buffer spanning their
union, so moving a selection/mask/channel far away (e.g. dx=1e5) allocated a
~1e5x1e5 buffer and hung the app. Write the two regions separately instead —
Surface is tile-sparse, so a distant region costs only its own tiles. Found by
the panic_hunt fuzz (now also catches hangs via a per-command thread timeout).
Regression test covers a far transform.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
aiongg