Deleting a past or later state in non-linear history leaves the document
unchanged, so a saved document now stays clean (view-only revision bump, as
select_layer and channel eye toggles do). Dragging a later (dimmed) History row
now shows its label at the pointer instead of an empty ghost. Adds a test for
undoing a mixed visible + opacity layer.setProps with visibility changes not
undoable.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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>
A higher Output resolution passed the resample budget (document bytes only)
but then allocated ~40 B/px of f32 composite, gray and output buffers at the
new size. Budget the whole working set before resampling.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With Make Layer Visibility Changes Undoable off, stepping through history kept every layer's
current visibility, so undoing a recorded step that changed visibility (Apply Layer Comp, a
snapshot, a compound) left the new visibility in place. Each history state now remembers the
layers whose visibility its own step changed (History::current_visibility, set by the session
when it records); Undo, Redo, History rows and non-linear Delete carry only the other layers'
visibility, so only the unrecorded eye changes survive. Deleting, purging or branching keeps the
lists consistent by merging them into the following step.
history.deleteState is now enabled in non-linear history while the oldest state is current and
later states exist, so a later state can still be deleted there.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Shrinking a selection's Content-Aware Scale on the Background left the
vacated area transparent. It now goes through Delete's `clear_area`
(background colour on the Background, transparent elsewhere), as
Photoshop does; the transform box's live preview shows the same.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Photoshop gives these painting tools a Mode in the options bar; here they
always painted in Normal.
- Clone Stamp, Pattern Stamp and History Brush offer the layer modes, and
Healing its own list (Normal, Multiply, Screen, Darken, Lighten, Color,
Luminosity). The mode is kept in each tool's own brush, like its size,
and reaches the live stroke and the commit.
- Paint Bucket: paint.bucket takes "mode" (any blend mode, or "clear" to
erase the filled area). algo's bucket fill gains a FillMode for it. The
options bar's Mode has Clear after Dissolve, as the Brush's does.
- A tool's first brush now starts in Normal instead of inheriting the
previous tool's mode, so a Multiply Brush no longer makes the first
Clone Stamp stroke multiply.
- The Shift+Alt letter and Shift+Plus/Minus mode keys reach these tools,
within each tool's menu. Tools with no blend mode (the Erasers, Mixer
Brush, Dodge, Burn, Sponge) now pass the keys to the selected layers.
Not yet: Healing's Replace mode, and a Mode for Spot Healing, Color
Replacement, Blur, Sharpen and Smudge (they ignore the keys meanwhile).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With a printer stand-in installed, file.print.printers and the Print
dialog's driver Printer menu listed its printers on every platform,
while the GDI route answers Unsupported off Windows. Tests on Linux and
macOS then opened the Windows Print dialog instead of the system
printers and paper controls. The stand-in now answers Unsupported off
Windows too, and the tests that read the driver menu expect it on
Windows only.
A frame pulled back out over pixels an earlier crop with Delete Cropped
Pixels off hid treated everything past the old canvas as uncovered, so
Content-Aware Fill painted over the hidden image. The fill now snapshots
the target layer before the crop and counts its opaque pixels past the
canvas as covered (kept and sampled), for unrotated and rotated crops.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Review follow-ups:
- Windows also reserves COM¹–COM³ and LPT¹–LPT³ (the ISO 8859-1
superscript digits), with any extension. `is_windows_device_name`
takes them too; `sanitize` keeps those characters, so `COM¹` became a
device file name before.
- Save for Web's `portable_generated_stem` added `_` at the end, so
`path: nul.v2.png` gave `nul.v2_.html`, still the NUL device, and the
export failed as an unsafe component. Both it and `sanitize` now share
`portable_device_name`, which puts `_` before the first dot.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
File › Package named its folder and document after the document with
`file_cmds::sanitize`, which let Windows device names through. On
Windows 11 a document named `nul` failed to package with "path not
found (os error 3)"; on Windows 10 `con`, `prn`, `aux`, `comN` and
`lptN` are devices too.
`sanitize` now adds `_` before the first dot of a device name
(`nul` -> `nul_`, `com1.v2` -> `com1_.v2`) on every platform, so a
package written anywhere can be copied to Windows. The device check
moves from web_cmds.rs to `file_cmds::is_windows_device_name`, and
web_cmds uses it unchanged.
The new package test fails on Windows without the fix.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
swatches.use converted a CMYK swatch (e.g. a Pantone .ase colour 0/0/100/0) to sRGB and
forgot the inks, so the Color Picker showed 5/1/92/0 and CMYK documents were painted with
the round-tripped inks. It now records them like a colour picked in CMYK (#4196).
The Info panel and Color Sampler readouts used a naive (1-r-k)/(1-k) formula while the
Color Picker converts through the CMYK profile, so the two never agreed (white-to-black
rows differed by tens of percent). device_cmyk now calls convert::rgb_to_cmyk, inside the
document's CMYK space for readouts.
layer.smartObjects.replaceContents takes an ambient path and is refused in
serve/mcp. Add the capability-scoped sibling of doc.place: the file is read
beneath --automation-read-root and handed to the command's own core, as an
RPC method (doc.replaceSmartObject) and an MCP tool (doc_replace_smart_object).