An Art brush set to Scale Proportionately drew its art at the same size
on every stroke weight: the art was scaled across only by how much the
path stretched it along and by the brush's Width. Stretch to Fit Stroke
Length and Stretch Between Guides already multiply by the stroke weight.
Proportional art now does too, so a 3 pt stroke draws it three times as
wide as a 1 pt stroke, still as long as the path. The same applies to
each run with Overlap on and to a Width that follows the pen.
Existing Scale Proportionately art brush strokes at weights other than
1 pt draw scaled by the stroke weight. Strokes at 1 pt draw as before.
The Width menu now offers Tilt, Bearing and Rotation beside Fixed and
Pressure, each with the second width. Tilt goes from Width with the pen
lying flat to the second width upright; Bearing and Rotation go from
Width at -180 degrees to the second width at 180 degrees, as Scatter
Brush Options' modes do. A stroke drawn without that input is half way
between the two widths. brush.options takes width_mode tilt, bearing
or rotation.
A corner tile was turned so that its x axis followed the mean of the
directions the path arrives and leaves in. A tile drawn as a right-angle
corner then came out turned 45° at every corner, so a rectangle drew its
corners as diagonal chevrons.
Outer and inner corner art are now both read as the corner that joins a
side tile on its right to one below it, as in the help's steps for
making corner tiles. Each corner tile is turned so that those two arms
meet the sides: on a rectangle the tile is square to the side tiles and
turns 90° from one corner to the next, and at other angles it is turned
halfway between the two sides. The arms follow Flip Along and Flip
Across, and a corner that takes the other kind's art turns it the same
way. Which art each corner takes is unchanged.
The PatternBrush field docs, the brush.options params text, docs/mcp.md
and the progress doc describe how corner art is drawn.
A menu under each of the five tiles in Pattern Brush Options lists None,
Original and the document's pattern swatches; the corner menus keep the
four automatic corners. A swatch choice copies the swatch's whole art and
its tile frame into the brush, so the brush keeps drawing after the swatch
is deleted, and the menu then shows that art as Original. Side cannot be
None: OK shows "You must choose a side pattern." and keeps the dialog open.
brush.options takes side_tile, outer_corner_tile, inner_corner_tile,
start_tile and end_tile (a pattern name, none, original, the corner auto
ids, or {"pattern": name} for a swatch named like one of those words), and
rejects side_tile "none" and unknown names before editing.
An art, scatter or pattern brush piece whose art carries a linked opacity
mask was bent along the path while the mask's art kept its place in the
brush's own coordinates, so the mask no longer covered the piece. The
mask's art now bends and has its strokes scaled with the object it
masks, on bent runs, on straight runs and where the brush narrows along
the path. A mask whose art cannot bend becomes an empty group. The
anchor budget for a brushed stroke counts the mask's art as well.
Pattern, art and auto-corner brushes used to flatten the art into chords
wherever the path bends, so Expand Brush Strokes left polygons with
hundreds of corner anchors. Each art segment now comes out as cubics
matched to the bent segment's ends and derivatives and halved until they
hold to a quarter of the flattening tolerance (never below the 0.05 pt
the path is flattened to). Anchors stay anchors, lines that stay straight
stay lines, and the halving is capped in depth and in anchors: where the
fit would take more anchors than the chords, or the map gives points that
are not finite, that subpath falls back to the chords, now without the
non-finite points.
A 4-anchor circle tile round a 400 pt circle: 68 anchors a tile -> about 9,
all with handles, 0.04 pt from the bent art against 0.19 for the chords;
stroke_pieces 169 -> 244 us for the 31-tile stroke.
Messhias is right that a sampled check cannot promise a run is straight, and my
own write-up overstated what the grid did: I said the fit was "used only where
it reproduces the real map", which five sample positions cannot establish. A
bend narrower than the sample spacing passes between the samples, and the grid
then approves an affine that is simply wrong.
It is not a theoretical gap. A zigzag of amplitude 4 pt and half-period 20 pt,
turning 11.3 degrees (under CORNER_DEG, so the vertices are not corners and
their tangents blend flat), lands its y = 0 vertices exactly on the five sample
positions. All 25 grid points then agree with the affine to the last digit while
the art between them is thrown 4.000 pt off, against the 0.0613 pt tolerance the
grid believed it was enforcing: 65 times over, measured, not estimated.
`Track::is_straight(s0, s1)` reads the answer off the track instead. A bend in
the span is two segment directions that differ, so no bend can hide in it
whatever its size. It also checks the vertex tangents, because `frame` blends
them across a segment unless the vertex is a corner, so a neighbour outside the
span can bend the tangent inside it: directions alone would call `d, d, e`
straight over the first two segments and be wrong. It fails closed on a closed
track's seam, a non-finite bound and a track with no segments.
`AlongMap::is_linear` covers the other half: Between Guides bends the arc-length
map itself, which no amount of looking at the track would reveal.
The grid stays, demoted to a second opinion that can reject and never admit, and
`affine_fit`'s doc comment now says that rather than claiming to be the gate.
Tests. `a_bend_shorter_than_the_sample_grid_is_still_a_bend` builds the zigzag
above, asserts the sampled check accepts it, measures how far out the approved
affine is, and then asserts the track-based gate refuses it; end to end through
`bend` it asserts the aliased bend is subdivided while a genuinely straight run
of the same span keeps the tile's four anchors. Restoring the sampled gate reds
it with `the aliased bend is subdivided, not affine-mapped: 4 anchors`, so the
old code really did lay a flat tile along a waving track.
`straightness_is_read_off_the_track_not_sampled` covers the predicate directly:
collinear segments and their extrapolation, a 0.2 pt detour, a span around it
and a span before it, the blended-tangent case above, the closed seam and a NaN
bound. Writing it caught a bug in my own guard: `f64::min` discards a NaN rather
than propagating it, so `is_straight(NAN, 10.0)` was reporting straight until
the finiteness check moved ahead of the min/max.
FredApps noted a real divergence: the affine route kept whatever the tile held
(it only required bounds) while warp_node drops every node kind but paths,
groups, layers and compound paths, and it did not clear `live` on paths. For a
tile of paths and groups the result was the same, so nothing was visibly wrong,
but "this changes the curves and nothing else" was a wider claim than the code
made: a symbol instance arrived untouched on a straight run and vanished on a
bend.
affine_node now mirrors warp_node's walk through `prepared` and applies one
transform at the end: the same kinds are kept and the rest dropped, `live` is
cleared on every path, an empty path or a group left with no children is
dropped, and strokes are scaled by `width_scale` rather than by the transform's
own factor. The two routes now differ in the tile's curves and in nothing else,
which is what the change was always meant to be.
Test: a_straight_run_keeps_the_same_node_kinds_as_a_bend puts a symbol instance
beside a path in one tile and asserts both routes return the same kinds. On the
old code it reds with `left: ["group", "path", "symbol"] right: ["group",
"path"]`.
Bending flattens every path it maps: warp_path runs kurbo::flatten and
emits SubPath::polyline, so a circle tile is drawn as a polygon. That
shows in the live stroke, not only after Expand Brush Strokes, and it
happens even on a perfectly straight path where the tile needs nothing
more than an affine.
Bending is only non-linear where the track bends. On a straight run the
arc-length map and the normal offset are both linear, so affine_fit
derives the affine from three corners of the art and then checks it
against a five by five grid, including those corners. Within the
tolerance the flattening would have used, the art is transformed
instead, which keeps its Beziers. Outside it, nothing changes: the old
path runs and the result is what it was.
Strokes are scaled by the same factor the bent path uses rather than by
the affine's own, so curves are the only thing this changes.
Corner tiles already took this route, which is why they kept their
curves while the tiles between them did not.
Two existing tests moved with it, both because they read anchor counts
that only held while straight runs were flattened:
pattern_places_corner_tiles told corner tiles from side tiles by
"four anchors", true only because the side rectangles arrived as
polylines. It now tells them apart by a kept handle, which is what
actually distinguishes an ellipse from a rectangle, and additionally
asserts the side tiles keep their own four anchors.
pattern_output_is_capped_across_subpaths_not_only_per_run counted on
sixteen tracks passing the #3594 anchor budget. A tile is about four
times cheaper in anchors now, so the budget engages later; the test
derives how many tracks it needs from what one costs.
On a corner whose legs are too short for the side tile, the tile is
squeezed along the path but its stroke keeps its width. The mask round
each slice was squeezed with the tile, so it cut off the stroke at the
slice's far end. Its margins along the tile now grow by as much as the
tile is squeezed.
Pattern Brush Options has a menu below the Outer Corner Tile and the Inner
Corner Tile. Each can be the brush's own corner art (Original), None, or one
of four tiles made from the side tile at each corner of the path:
- Auto-Centered stretches one side tile around the corner, centered on it.
- Auto-Between bends a side tile on each side into the corner, the two
meeting on the corner's diagonal.
- Auto-Sliced lays a side tile on each side past the corner and masks it
at the diagonal, so the two meet like a mitered frame.
- Auto-Overlap lays a side tile on each side past the corner, overlapping.
The tiles are straight at the corner and follow the path towards the side
tiles, so they meet a curve that runs into the corner. A side tile taller
than the corner can reach is stretched along until the two sides meet. On a
short path, a corner's tile is squeezed along so that it covers at most the
path up to an open path's end and half way to the next corner, and its two sides still meet on the
corner's diagonal.
The definition stores the choice as `outer_source` and `inner_source`
(`none`, `autoCentered`, `autoBetween`, `autoSliced`, `autoOverlap`), which
`brush.options` sets. Corner art wins over a source. A brush saved without
a source draws as before: a kind of corner without art takes the other
kind's art. None leaves that kind of corner empty.
Docs: docs/mcp.md, docs/target-app-parity.md.
Art Brush Options' Width had no pen tablet option, so an art brush
drew at one width whatever the pressure. The Width row now has a
Fixed or Pressure menu and a second width. With Pressure the art is
Width wide at the lightest pen pressure and the second width at the
heaviest, following the pressure the Paintbrush recorded along the
stroke, and half way between the two on a stroke drawn without
pressure. brush.options takes width_mode and width_max; a brush
without them is saved as before.
Pattern tiling and scatter cap each factor on its own: 20,000 tiles a
run, MAX_SUB subdivisions a segment, 20,000 copies a track. Those caps
are per track, and a path has as many tracks as it has subpaths, so
they multiply rather than bound.
The audit called the end-to-end impact unverified, so it was measured.
A pattern brush on parallel lines, each long enough to want more tiles
than a run allows:
1 subpath 20,000 nodes 320,000 anchors 36 ms
4 subpaths 80,000 nodes 1,280,000 anchors 142 ms
16 subpaths 320,000 nodes 5,120,000 anchors 571 ms
Exactly 20,000 nodes a subpath, linear with no ceiling. Scatter is the
same shape.
MAX_ANCHORS bounds the total for one brushed stroke, threaded through
both. It is 2,000,000, the figure the stacked-effects budget already
uses (effects::distort::MAX_COPY_ANCHORS, #3578), and bounded the same
way: per stroke rather than per document. One track at its own cap is
well inside it, so the budget bites only where the caps were
multiplying; sixteen subpaths now stop at the budget in 113 ms.
A piece is taken whole or not at all, so art is never half-drawn. The
walk that counts an art tile's anchors is iterative, and a tree it
cannot finish within MAX_WALK nodes is charged the whole budget rather
than undercounted, so nesting buys no room.
The per-factor caps are untouched: the issue asks for a total, and
changing both at once would hide which one did the bounding.
Double-clicking a Scatter, Art or Pattern brush in the Brushes panel (or the
panel menu's Brush Options...) now opens its options dialog, as Calligraphic
and Bristle brushes already did. OK runs brush.options and every stroke
painted with the brush redraws.
- Scatter Brush Options: Name; Size (10-1000 %), Spacing (10-10000 %),
Scatter (-1000-1000 %) and Rotation (-180-180 deg), each Fixed (the first
value), Random (between the two values) or Pressure (the first value at
the lightest pen pressure, the second at the heaviest); Rotation relative
to Page or Path. Scatter brushes now keep these modes (`modes`); brushes
saved before keep their look (Random where the two values differ).
- Art Brush Options: Name; Width (1-1000 %); Scale Proportionately, Stretch
to Fit Stroke Length or Stretch Between Guides with the guides' Start and
End (% of the art length); the four Directions; Flip Along and Across.
- Pattern Brush Options: Name; Scale (1-10000 %), Spacing (0-10000 %); the
five tiles (Outer Corner, Side, Inner Corner, Start, End) drawn with their
art or None; Flip Along and Across; Stretch to fit, Add space to fit or
Approximate path.
- All three: Colorization (None, Tints, Tints and Shades, Hue Shift with its
Key Color picked from the art's colours, the most used one by default) and
a live preview of a stroke painted with the edited brush.
Brush definitions from files and commands are now kept within these ranges
(Brush::sanitize, also for Calligraphic and Bristle values); unknown modes,
directions, fits or colorization methods are refused. The brush.options and
ui.brushOptions docs list the new params and fields. Tests: the brush model
(modes, pressure, sanitizing, art colours), engine params reaching the
brush and strokes redrawing, junk params in command_sweep, and headless
dialog tests for each dialog.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>