Commit Graph
385 Commits
Author SHA1 Message Date
David MannandClaude Opus 5.5 c0007ed02d Address CodeRabbit review: frame stroke bounds from the compiled stroke
A mitered corner reaches up to miter_limit widths past the path, past the
path bounds grown by the width; take the bounds from the compiled stroke.
An empty frame's effects now count its stroke too, as layer_bounds and
composite_bounds already did.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-21 12:24:36 +02:00
David Mann 0f8a73b3a4 Merge remote-tracking branch 'origin/main' into fix/5692-tsl-off-blend
# Conflicts:
#	crates/compose/src/lib.rs
2026-10-21 12:21:07 +02:00
David MannandClaude Opus 5.5 e5f4f8e260 Address CodeRabbit review: soft shape frame from the hard render, feather claim
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>
2026-10-21 12:14:46 +02:00
David Mann 86811ddeee Merge PR #6002: Stroke: keep Overprint (overprint)
# Conflicts:
#	crates/engine/src/layer_style.rs
#	crates/ui-egui/src/layer_style.rs
#	docs/file-format-parity.md
2026-10-21 11:36:58 +02:00
David Mann 5bf2898f4e Merge remote-tracking branch 'origin/main' into feat/tool-214-12-shape-feather
# Conflicts:
#	docs/scorecard.md
2026-10-21 11:36:04 +02:00
David Mann a492716367 Merge PR #5990: Bevel & Emboss: keep the Gloss Contour's Anti-aliased (antialiasGloss) 2026-10-21 11:29:05 +02:00
David Mann 14b928e8a8 Merge remote-tracking branch 'origin/main' into HEAD 2026-10-21 10:56:06 +02:00
David Mann d25387a459 Merge remote-tracking branch 'origin/main' into HEAD 2026-10-21 09:35:08 +02:00
aiongg fc371711e6 Merge PR #5847: Shapes: Density and Feather on shape layers (#5846)
# Conflicts:
#	docs/scorecard.md
#	docs/target-app-parity.md
#	docs/ui-parity.md
2026-10-21 05:04:03 +00:00
xsvm f2dce80718 Compose: quantize floors without a libm call
Baseline x86-64 has no SSE4.1 roundss, so f32::floor is a software routine there, called once per sample of the backdrop under every adjustment layer of an 8- or 16-bit document. branchless_floor rounds through 2^23 and steps down where that went up: bit-identical to floor for every f32 (a new test sweeps every 4099th bit pattern and the edges around whole numbers and 2^23), and with no branch the loop vectorizes.
2026-10-21 12:26:49 +08:00
xsvm 420353c5ae Compose: adjustment layers blend back in Normal mode without a per-pixel mode dispatch
An adjustment layer's result was blended back into the backdrop pixel by pixel through blend_rgb, which looks up the HDR flag and dispatches on the mode, three channel calls deep, for every pixel. Normal (nearly every adjustment layer) gives the source exactly, so blend_back skips the dispatch for it, and walks the backdrop, the result and the coverage in step instead of indexing by coordinates. Bit-identical.
2026-10-21 12:26:47 +08:00
xsvm d278a85696 Compose: tone and hue-range lookups truncate instead of calling floor
Baseline x86-64 has no SSE4.1 roundss, so f32::floor is an out-of-line libm call there. lut, which every Levels, Curves, Hue/Saturation range and modern Brightness/Contrast sample goes through, floored its non-negative table position. Truncating is the same there (NaN casts to 0 either way) and inlines. Bit-identical.
2026-10-21 11:46:27 +08:00
xsvm aab9cdc045 Compose: Color Balance works out each channel's curve once, not per pixel
color_balance_channel recomputed the folded black point, white point and midtone gamma, and four of the five powf calls of the gamma > 1 Hermite join, for every sample. Those depend only on the sliders, so BalanceCurve works them out once per channel and the per-sample step is the clamp and one powf (or the Hermite polynomial). The CPU compositor builds the three curves once per buffer and the GPU plan once per LUT (it sampled color_balance_channel 3 x 4096 times). Results are bit-identical: a new test compares the curve against the per-sample formula it replaced, to the bit, over 123 slider settings, both Preserve Luminosity modes and a 1/1024 ramp.
2026-10-21 10:10:57 +08:00
aiongg 9b889e1a58 Merge PR #6640: Layers: toggling a clipped adjustment damages only its base (#6535) 2026-10-21 01:06:16 +00:00
aiongg 68f8be3967 Merge PR #5791: PSD Multichannel: alpha channels are stored as-is, not as ink (#5790) 2026-10-21 01:06:14 +00:00
aiongg b097800e78 Merge pull request #8095 from aiongg/integration-7
Integration batch: 9 PRs on main
2026-10-20 20:44:51 -04:00
David MannandClaude Opus 5.5 627bb8f1bf Integration: merge main, resolve conflicts (#6002)
docs/file-format-parity.md: both sides added revision-history rows; keep both.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-21 00:11:01 +02:00
0073b395d4 Integration: merge main, resolve conflicts (#5990)
docs/file-format-parity.md: main added three 2026-10-15 revision rows
(DDS save, PNG sTER, PSD guides) where the PR added its own; keep all four.

Co-authored-by: varanus-salvator <262257003+varanus-salvator@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-20 23:50:13 +02:00
David Mann d089eaa81b Merge PR #7693: perf(compose): Reuse bounded scratch and tone tables 2026-10-20 22:39:29 +02:00
David Mann 651c32b9eb Merge PR #7566: perf(compose, gpu): many-layer composites, P60/P61 on the GPU 111 -> 28 ms 2026-10-20 22:39:26 +02:00
Sir Starch 5579cf5439 Compose: Extend Softer outer glows past Size (#7899) 2026-10-20 22:09:33 +02:00
aiongg ea7553f031 Merge PR #5111: Large documents: a gigapixel New Document draws without freezing (#4017) 2026-10-20 18:39:28 +00:00
David Mann 0b39361076 Merge main into feat/tool-214-12-shape-feather 2026-10-20 17:12:29 +02:00
Rene Leonhardt a686a932c0 perf(compose): reuse bounded scratch and tone tables
Reuse tile scratch and tone tables across rendering work. Skip fully masked
adjustments and identity float curves. Previous-path tests preserve output
across sample depths, masks and clipping.
2026-10-20 15:52:44 +02:00
Nigel Ihza Firdausy f0aae59938 Merge upstream main and retain CPU/GPU Dissolve parity coverage 2026-10-20 19:44:22 +07:00
Hector Villarreal b517b6af02 Merge remote-tracking branch 'upstream/main' into fix/7360-type-bevel-fill 2026-10-19 16:23:08 -06:00
NetterSilverandClaude Haiku 5.5 8591ca1904 Scale layer-effect noise by coverage so it follows the glow
Fixes #7833

Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19 19:44:44 +00:00
aiongg acc853b517 Merge upstream main into integration-5
docs/file-format-parity.md: keeps both 2026-10-15 rows, main's #6063 PNG sTER row first, then #5976's PSD guides row
2026-10-19 17:20:29 +00:00
David MannandClaude Opus 5.5 5c1b9425d1 Gradient Overlay: repaint the whole layer when a stroke moves its frame (#4053) (#5340)
A Gradient Overlay (and gradient strokes, linked patterns, linked bevel
textures) is laid out over the layer's bounds. Painting grows those
bounds, but the stroke commands reported only the stroke as damage, and
the canvas's committed-stroke path repainted only what the live stroke
had not shown. The canvas therefore kept the old gradient around earlier
strokes: one gradient per stroke instead of one over the whole layer.

Session dispatch now grows a command's reported damage by
photocraft_compose::relayout_damage(before, after): the whole composite
of any visible layer whose effects span its frame, when that frame moved.
The canvas adds the same area to a committed live stroke. Strokes inside
the existing bounds and frame-independent effects keep tight damage.

Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-19 13:04:52 -04:00
David MannandClaude Opus 5.5 bc4f2a783d Scripts: Flatten All Layer Effects bakes with the document's Global Light and patterns (#5719) (#5720)
`bake` rendered each layer through `render_layer`, which has no document context (default Global
Light, no patterns, the render rect as canvas). It now uses the new doc-aware `render_layer_in`.

Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-19 13:04:46 -04:00
Akos Arinton dfeba9e974 test(compose): compare NaN as NaN in the vector span test, not by its bits 2026-10-19 18:11:13 +03:00
David Mann c2f3e23a40 Merge PR #5477: perf: chisel bevel size changes, CPU passes on threads and GPU fields share their sharp edges (results unchanged) 2026-10-19 16:51:22 +02:00
David MannandClaude Opus 5.5 de2cb1f0c0 Compose: a clipped adjustment inside a pass-through clip base damages the whole group (#6535)
A pass-through group that is the base of a clipping group renders isolated while it holds a
visible adjustment (`clip_base_adjusts`), so showing, hiding, adding or removing a clipped
adjustment inside it can switch the group's other children (a Multiply layer beside the base)
between isolated and pass-through rendering. `change_bounds_in` now takes the list holding the
parent group and reports that group's change bounds in this case; `layers_damage`, `step_damage`
and `one_sided_bounds` pass it.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19 14:14:06 +02:00
Akos Arinton 80f896210a perf(compose): write band tiles as they finish, not after the band
Parallel bands each collected their finished tiles until the band's last
one was done, so with work stealing many bands could hold most of their
tiles at once, close to a second copy of the output. Each tile is now put
into its band behind a lock as soon as it is rendered; only the tiles in
flight are held, as before the bands ran in parallel.
2026-10-19 13:34:32 +03:00
David MannandClaude Opus 5.5 8534f0c935 Compose: a clipped adjustment with no base inside a pass-through group reaches everything (#6535)
The compositor draws a clipped adjustment that has no unclipped layer
beneath it as an ordinary adjustment, but change_bounds counted it as
contained in a pass-through group. Hiding the layer clipped to such a
group (which ends its isolation) or the group itself then recomposited
only the group's pixels and left the rest of the canvas stale.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-19 10:03:54 +02:00
Hector VillarrealandClaude Opus 5.5 a42e831698 Map the bevel's shading through its gloss contour, not its light (#7360)
The gloss contour was applied to the highlight and shadow amounts after
the lighting split, so a contour that starts high (Rolling Slope -
Descending) turned every unlit pixel into a full highlight: a bevelled
glyph's flat face painted (255, 40, 42) over its (151, 22, 23) fill. It
now maps the shading before the split, with the flat face's mapped
shading as the neutral, on the CPU and the GPU. Linear is unchanged.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-18 23:01:11 -06:00
Ni-zavandVikram Bhaskaran a221622ad4 fix(compose): cover clipped Dissolve adjustments
Port the clipped-adjustment coverage path from #4711 and verify depth, alpha, masks, and document-coordinate GPU parity.

Co-authored-by: Vikram Bhaskaran <vikram.bhaskaran@outlook.com>
2026-10-19 07:18:54 +07:00
Nigel Ihza Firdausy 9050e86dac test(compose): preserve clipped Dissolve coverage across fill, boundaries, and regions 2026-10-19 07:18:54 +07:00
Nigel Ihza Firdausy 157711d1be style(compose): apply rustfmt layout from CI for Dissolve regression 2026-10-19 07:18:54 +07:00
Nigel Ihza Firdausy e0d3d940c6 fix(compose): resolve Dissolve coverage across adjustment, effects, and clipping paths (#3050-#3052) 2026-10-19 07:18:53 +07:00
Akos Arinton 5418399dda perf(compose): blend pixel layers row by row, Normal in 8-pixel vector blocks
A pixel layer with nothing clipped to it is read from its tiles a row at a
time and blended straight onto the backdrop, instead of being copied into a
tile-sized buffer first (a megabyte per layer per 256² render tile). Normal
RGB rows go through a branch-free kernel over blocks of 8 pixels that the
compiler vectorizes; few-band renders also run their bands in parallel.

Bit for bit the same: the row kernel is tested against the scalar loop (NaN,
infinities, subnormals, signed zeros), the row path against the tile copy at
8/16/32-bit with every kind of mode, fill and mask, and the whole GPU parity
suite's CPU references (1620 renders) are unchanged.

P60 on the CPU compositor: 627 -> 530 ms (the render without effects
300 -> 199 ms).
2026-10-18 22:26:05 +03:00
SBaksa 1e0673432b CMYK split route: convert stand-ins a band at a time, skip hidden layers
A layer's stand-in surface converted every stored tile to floats before
writing any, a transient several times the layer's stored size (twice,
once per stand-in). Convert about BAND_PIXELS of tiles at a time instead,
and leave hidden layers unconverted, since the compositor never draws them.
2026-10-19 01:25:39 +11:00
SBaksa df17ad7520 CMYK ink compositing: isolated groups keep the special Fill rule
An isolated group in one of the special eight blend modes below 100% Fill
blended with Fill as plain coverage on the exact ink path, while the
compositor (and so the split route) keeps Fill apart from opacity, so its
inks differed from the canvas. Blend groups through blend_into_fill like
other layers, and drop the group exclusion from the test reference.
2026-10-19 01:25:39 +11:00
oRE-o 8fb5d94d2b Gradient Overlay: expose interpolation methods and retain source stops 2026-10-18 18:36:30 +09:00
SBaksa 47ff3a99bf CMYK ink compositing: keep layers without an alpha channel
The ink compositor skipped a layer with no tiles in the rect whose default
pixel was all zeros, reading the last value as alpha. A CMYK layer without
an alpha channel defaults to no ink, which is opaque blank paper, so under
an adjustment (Threshold, Invert, ...) its untouched pixels composited as
transparent. Skip only surfaces with an alpha channel and a transparent
default, as `empty_in` does.
2026-10-18 17:11:35 +11:00
SBaksa a4cb23a33f Merge remote-tracking branch 'origin/main' into cmyk-native-composite
# Conflicts:
#	crates/compose/src/lib.rs
#	crates/io/tests/cmyk_composite_export.rs
2026-10-18 17:11:34 +11:00
Korg1ch 00e4423fcc test(compose): Dissolve effects are speckled, tile-independent and match on the GPU (#6995) 2026-10-18 02:57:55 +03:00
Korg1ch e0d90b2896 fix(compose): layer effects in Dissolve mode speckle instead of painting as Normal (#6995) 2026-10-18 02:57:44 +03:00
seotarek ac1bc012ef fix(compose): prevent outside stroke from bleeding into opaque character interior (#5861) 2026-10-18 02:21:02 +03:00
David Mann 0739ff48c5 Merge remote-tracking branch 'origin/main' into fix/4017-huge-new-document
# Conflicts:
#	crates/ui-egui/src/gpu_canvas.rs
#	docs/scorecard.md
#	perf/budgets.toml
2026-10-17 01:07:37 +02:00