A right-click while the left button still dragged a crop frame or a
Straighten line opened the crop menu over the live gesture (or, outside
the image, the pasteboard menu). The canvas now opens no menu with the
Crop tool while a crop drag or Straighten line is under way and the
primary button is still down. It needs both: egui ends the drag on the
right button's release without an Up, so the drag flag can stay stale,
and a lost release (Alt+Tab, a cancelled touch) leaves the button
"down". A finger held still (egui's press-and-hold right-click) is
excepted. The agent path (ui.pointer) opens no menu while its own crop
or Straighten drag is down, also the first one on the default frame.
Tests: no_crop_menu_while_the_left_button_drags_the_frame (real pointer
events), a_left_button_down_without_a_crop_drag_keeps_the_crop_menu,
no_crop_menu_while_the_left_button_draws_a_straighten_line,
a_finger_held_on_a_pending_crop_opens_the_crop_menu,
agent_right_click_during_its_crop_drag_opens_no_menu, and
enter_under_the_crop_menu_does_not_crop (only Esc was tested). The
parity row notes Photoshop's Reset Crop and Rotate Crop Box, both still
missing.
Co-authored-by: Marco Arnulfo <103679419+marcoarnulfo@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Folding #5236's performance notes into docs/seam-carving-performance.md
dropped main's "P28's recorded number is stale" section (the 2026-10-05
baseline commit at 266 s, the loaded machine behind the 954 s). None of
it depends on #8526's code, so it is back, dated and attributed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Merged with our fix-up d5139d177 (pr-1199-fixup), which merged main as of 8f09d122d.
Conflicts with main since then and with batch 8:
- photocraft-cli: main split the CLI into clap modules (#7071); the model wiring
(local_headless, configure_cli_models, batch's shared backend, droplet) is re-applied there.
- Cargo.lock regenerated: crates/ml pins ureq =3.4.2, so the app's ureq goes 3.5.0 -> 3.4.2.
- cutout_cmds: main's subject area (#8235) kept for the classical path beside the model path.
- file_cmds batch: main's per-input closure takes `s` later, so the backend and
integrations prefs are read before the closure instead of inside it.
- prefs_ui: model-choice filter with #4573's pref_choice_label; Integrations stays visible
(#4097) with the model choices only where models run; the translation-coverage test main
deleted (c3aa3f9e5) stays deleted; old section_fields test calls take PreferenceFieldOptions.
- Semantic clashes git did not flag: #4097's Integrations test literals in
apps/photocraft take ..Default::default() for the new model fields; main's and #4573's
section_fields test calls pass PreferenceFieldOptions.
- engine lib.rs, prefs.rs, smartselect tests, snapshot example, jobs_ui_tests, people.toml,
docs and i18n catalogs merged as unions.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
panels.rs: the PR's snapshot rows next to main's History Brush source
and scroll-to-current; ui-parity.md: main's status line.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Main moved the shape options bar and paint swatch out of vector_ui.rs into
shape_options.rs and shape_paint_ui.rs (#5452). Carry the PR's gradient paint
over: PaintChange::Gradient, paint_param and the swatch's Gradient tab live in
shape_paint_ui.rs, the tool gradients and set_tool_paint in shape_options.rs,
and shape_gradient_ui.rs is its own module. Both sides' new tests are kept.
Co-authored-by: Denis <179751626+deniskhud@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
#6923 develops the Camera Raw dialog's preview on the GPU whenever no spatial stage is set,
from `per_pixel`'s stages. #5128 adds Calibration (primaries and Shadows tint) to the CPU
develop between white balance and Dehaze, which those stages don't run, so with both merged
a Calibration-only change previewed unchanged on the GPU while Open applied it. `per_pixel`
now returns None when Calibration is set, so the dialog develops on the CPU then. The
white and black points stay as they were, so the dialog's points cache is still right once
Calibration goes back to neutral. The hardware parity doc and BUG-204-3's note now say so.
Co-authored-by: Fabio Lissi <3223836+FabioLissi@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Open wrote PhotoCraft's crs: subset over any existing sidecar, on by
default: a Lightroom or Bridge IMG_0001.xmp lost its crop, masks, rating
and keywords, even when the raw was only opened with Open in Camera Raw
off (FredApps' review). Now a sidecar is only written when the settings
changed, never over one whose x:xmptk isn't PhotoCraft's (a notice says
it was kept, camera_raw_xmp::is_photocraft_xmp), and never for a control
or MCP request, which has no authority to write beside the user's files.
A write error goes to a notice instead of a status line the develop
overwrote at once.
The Camera Raw Defaults' colour space, 8-bit and smart-object steps ran
before the Camera Raw step, so with the default Adobe RGB every raw opened
with the dialog's settings applied to other pixels than it previewed. They
now follow the step (also when a background re-develop lands), as
Camera Raw converts to the workflow space last. Main already sets the raw
resolution at import (#7811), so the PR's Image Size step is dropped.
Two raws opened back to back with raw-stage edits: the second one's
background re-develop replaced the first one's pending entry, so the first
never got its Camera Raw step, conversions or unmodified mark. Pending
re-develops are now a list keyed by job. Sidecars are read only from
regular files of at most MAX_XMP_BYTES (a pipe or a huge file named
IMG.xmp no longer blocks or fills memory), and a conversion that fails
after the Camera Raw step is reported instead of leaving the dialog open
for a second Open that would apply the step twice.
Tests: a_foreign_sidecar_is_never_overwritten,
a_rating_only_sidecar_is_left_alone,
opening_without_the_dialog_does_not_rewrite_the_sidecar,
a_control_request_never_writes_the_sidecar,
a_sidecar_write_error_stays_reported,
the_camera_raw_step_runs_before_the_workflow_conversion (inline and
background), two_raws_opened_back_to_back_both_finish_their_open,
sidecars_are_read_from_small_regular_files_only,
is_photocraft_xmp_reads_the_toolkit_marker; the existing
sidecar round trip now uses a PhotoCraft-written fixture.
Co-authored-by: Fabio Lissi <3223836+FabioLissi@users.noreply.github.com>
Co-authored-by: FredApps <153173854+FredApps@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
docs/file-format-parity.md: main's PSD row with the round-trip count
258/258. docs/scorecard.md: regenerated with `cargo xtask scorecard`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The PR fixes the stale Photoshop corpus count (256 -> 258, the pinned
manifest xtask/photoshop-corpus.sha256 lists 258 PSDs) in
docs/development.md and docs/scorecard.md. The same stale number was
left in the io corpus test comment, docs/file-format-parity.md,
docs/roadmap.md ("258/256" round-trip floor) and ATTRIBUTION.md.
Co-authored-by: Simone <185146821+Lucenx9@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Resolves the conflicts with main's area type (TextShape::Area): EngineData keeps main's area arm beside the path comment, and a Txt2 frame of an area layer becomes a text box over the path's bounds, as its EngineData describes it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
prefs.rs keeps main's demosaic default and the PR's write_xmp_sidecars, and
unhides the Camera Raw Defaults the PR makes live; the raw-open RawOpen built
without the dialog gains main's developed_with/demosaic; i18n and docs as
unions; docs/scorecard.md is regenerated in the fix-up.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Tolerance defaulted to 50 and averaged alpha into the difference, so an
ordinary edit (e.g. #c03030 -> #d04848) was never painted back, and a stroke
that wrote nothing still left an empty History step. Photoshop's default is
0; the difference is now the larger of the mean colour change and the
alpha change (the Eraser keeps colour and lowers alpha), and a stroke that
writes nothing is an error instead of an empty step.
- Each tile copied a source window of tile +- halo per dab: a leftward stroke
or a curl across x = 64 read empty pixels there (seam), and a size-1000 click
copied ~5e9 floats (11.9 s). The rotated read stays inside the dab, so it now
samples the source directly; bilinear reads at pixel centres, so an
unrotated dab copies the history pixel instead of a 2x2 blend.
- The brush used its own history lookup (state 0 by default): it now takes
the History Brush's source (history.setSource, snapshots, refusal after a
canvas size change).
- TOOL-213-20 is partial (8 of 10 styles, Area scales spacing); docs follow.
Co-authored-by: Matt Van Horn <455140+mvanhorn@users.noreply.github.com>
Co-authored-by: FredApps <153173854+FredApps@users.noreply.github.com>
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Conflicts with #5547 (Animated Zoom) resolved in wheel_nav.rs, canvas.rs and
panels.rs: the Navigator and canvas pass both Animated Zoom and
!animations_off; with Animation Effects off, the sideways part of an
animated wheel zoom pans by the frame's own events folded by their
modifiers (#4573's unsmoothed scroll) instead of the plain-event raw.x,
so Ctrl + wheel still pans sideways there. #4573's modifier test runs
with Animated Zoom off and on.
Translation catalogs keep main's strings plus the Art History Brush ones
(fr: main already has Area; hi: added the brush strings). docs/scorecard.md
regenerated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The review round of the #7071 stack, which #9167 merged without
(it took the pre-review heads):
- A bare --help or -h before `--` is help after any parse error, not
only after a missing value: `mcp stray.psd --help` and
`serve --port abc --port 1 -h` print the subcommand's help.
- mcp and serve --help carry #8035's rules: what the automation roots
grant, that bridge roots are refused, and that the token flags need
--bridge or --port (or PHOTOCRAFT_CONTROL_PORT).
- check_server takes the subcommand and whether a transport is set;
the bridge-roots check moves next to the other mcp checks.
- The server tests match clap's "a value is required for" text, and
mcp.md in the book and docs names PHOTOCRAFT_CONTROL_PORT.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Say which packages install them and how to generate and install them
by hand from a checkout. The desktop app has neither.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The FreeBSD tarball carries them in the ports layout: man/man1,
share/bash-completion/completions, share/zsh/site-functions and
share/fish/vendor_completions.d.
The man-and-completions example is built in the same cargo build as
the binaries (--bins --example), and package.sh runs it from the
target dir. With --skip-build it must already be built, or the script
says so and fails. A dry run builds nothing, so it leaves them out.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The deb, rpm and tarball install the man pages in /usr/share/man/man1
and the bash, zsh and fish completions in bash-completion's
completions, /usr/share/zsh/site-functions and fish's
vendor_completions.d. Fedora's and openSUSE's zsh read site-functions;
Debian's reads /usr/share/zsh/vendor-completions instead (its zsh is
configured with --enable-additional-fpath), so the deb also carries a
symlink there.
The man-and-completions example is built in the same cargo build as
the binaries (--bins --example, same profile, features and target),
and package.sh runs it from the target dir. A cross build runs it
through $EMULATOR, as the --version smoke test does, and without one
packages no pages or completions, with a warning. With --skip-build the
example must already be built, or the script says so and fails. The
pages aren't gzipped, like the other packaged docs.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Say that `photocraft --help` lists the flags and their variables, that
a flag wins and an empty variable is unset, and what is a usage error;
that a bad --present-mode or --frame-latency is one while a bad
variable is still ignored; list PHOTOCRAFT_IN_WINDOW_MENUS; and that a
token with a token file is a usage error for the control server too.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
With the environment fallbacks, clap's conflicts_with on the token pair
made a stale exported PHOTOCRAFT_CONTROL_TOKEN and
PHOTOCRAFT_CONTROL_TOKEN_FILE a usage error for every mcp and serve, though
headless mcp and stdio serve read no token. As on main, the pair is now
refused only by serve on TCP and mcp bridging, still as clap's either/or
usage error naming each half as it was given.
An automation root on the command line with PHOTOCRAFT_CONTROL_PORT set
is a usage error too: the variable would switch mcp to bridging, which
drops the roots, with only a note on stderr.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Say that PHOTOCRAFT_CONTROL_PORT is the fallback for serve --port and
mcp --bridge (with the stderr notice and the empty-is-unset escape),
and list the CLI's variables on the book's CLI page, which also says
the automation root variables stay the desktop app's.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Step 4 of the issue: the control token flags and the automation roots
move to photocraft-automation's new args module as two clap structs,
which mcp and serve flatten and the desktop app can flatten in step 5
without moving them again. clap is a plain dependency of the crate.
- ControlArgs holds --control-token and --control-token-file;
AutomationArgs holds the roots, and AutomationArgs::workspace opens
them (replacing the CLI's automation_workspace).
- The port value parser is photocraft_automation::args::control_port,
with its unit tests, for serve --port and --bridge now and the
desktop app's --control later. It ignores surrounding spaces and
refuses anything else with a typed NotAPort.
- --bridge parses to Bridge::Port or Bridge::Addr: it takes a bare
port, meaning 127.0.0.1:<port>, or any value containing a ':', which
the bridge client then validates. Anything else is a usage error
(exit 2) instead of a bridge error.
- 127.0.0.1 addresses are built from Ipv4Addr::LOCALHOST.
mcp_bridge_takes_a_port_or_an_address is new: `--bridge nope` is a
usage error, and `--bridge 1` bridges to 127.0.0.1:1, which a tool
call names when nothing listens there.
No new environment behaviour: the token variables are still read at
run time through security::token_inputs, and the roots stay flag-only.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A layered TIFF carried an empty `LMsk` block in its ImageSourceData
(tag 37724). Photoshop rejects that: it reports a problem reading the
layer data and opens only the composite. Photoshop writes 14 bytes there
itself (overlay colour space 0, colour 65535/0/0/0, opacity 50, kind 128).
The TIFF writer now uses those bytes when the model's global layer mask is
empty. PSD output is unchanged; an empty block is fine there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* perf(seam): drop the original-column map from the shrink proxy, same pixels (-15 % at 24 MP)
Content-Aware Scale's proxy seam search tracked each panel column's original
column in an idx map, one usize per proxy pixel, shifted by every seam. The
original column is x plus the number of columns already taken at or left of it,
so a per-row list of the taken columns answers the same question without a
buffer the size of the energy map.
The recorded P28 budget number predates the proxies: the 2026-10-05 baseline
commit carves at full height and recomputes energy per seam, and measures
266 s for the 24 MP case on the machine this was timed on, against 2.6 s now.
Documented in docs/seam-carving-performance.md; enforce stays off until a
budgeted-class run can vouch for the whole row.
* review: generate exactly ch samples per pixel in bench_seam, reject ch > 4
The fixture flattened four samples per pixel and then truncated to w*h*ch, so a
three-channel run grouped samples across pixel boundaries and any ch above four
built a buffer shorter than the rows carve_width indexes.
Generate ch samples inside the per-pixel closure and refuse ch outside 1..=4.
The three-channel row is dropped from the perf note rather than re-measured
here: it moved inside the timer noise either way, and the other agents' builds
on this machine make a fresh number worthless. The four-channel hashes are
unchanged, which is the check that this edit touched only the fixture.
* perf(seam): find the removed column by binary search, not a linear scan
Recording a seam recovered each removed original column by scanning the row's
list of already-taken columns, so a deep shrink paid a comparison per removed
column on every row and every seam: quadratic in the seam count, and worse than
the index map this PR replaced once a shrink got deep enough.
`o - i` never decreases along the row's sorted list, so that count is a binary
search, and the original column belongs where the search stopped; the rows are
independent and update in parallel, like the pixel shift. What is left per
removed column is one insert.
Output is unchanged: same columns, same order, same hashes (10 % and 50 % width
cases below), and `counted_removed_columns_match_the_tracked_map` now runs a
tracking and a counting carver side by side to one column and compares every
row's list after every seam.
docs/seam-carving-performance.md gains those cases, states that the ~104x
figure is the two example runs rather than the baseline row, and names
perf/baseline.json as the source of the 954 s.
---------
Co-authored-by: ebosido444-ai <ebosido444-ai@users.noreply.github.com>
* Speed up Crystallize and Pointillize with cell-space rasterization
* Measure pixelate commands on unlocked layers and Background separately
* Record pixel-exact 24 MP filter and command benchmarks