Many tools note things on SVG objects in attributes of their own XML
namespace: a plotter's pen, a cutter's passes, a game engine's
collision flag. Import dropped them, so a file that went through
VectorCraft lost what other software had set on its objects. #522
asked for unknown and data-* attributes to survive a round trip; #580
kept the data-* ones, and this keeps the ones in other applications'
namespaces.
- Import keeps every attribute in a namespace that isn't SVG's,
XLink's, XML's, Inkscape's, Sodipodi's or VectorCraft's own, on the
object or layer it belongs to, in order, with its namespace URI and
the prefix the file gave it (ObjectAttributes::foreign, beside the
data-* attributes and the event handlers). The file is untrusted: at
most 64 per object and 64 KiB per value, and what is left out is
warned about. Files that declare no other namespace skip the scan.
- They are the object's own, as its data-* attributes are: a pasted
copy and a duplicate keep them, and deleting the object takes them
with it. The native file keeps them in the object's attributes, a
key written only when an object has some.
- Export writes them back on the same elements and declares each
namespace once, on the root, under the file's prefix. A prefix the
export uses itself, or that another namespace already has, becomes
prefix2, prefix3...; one XML keeps for itself (xml...) becomes ns. A
name or value XML can't hold is left out with a warning.
Nothing in VectorCraft reads them, and nothing in the UI lists them.
svg/tests/foreign_attributes.rs: a layer's and a shape's attributes
survive import, export and import again, Inkscape's and Sodipodi's
own aside and with no warnings, and the native form round-trips;
prefixes are made unique, and bad names and values are left out with
a warning; the caps hold and are warned about once.
engine tests_attributes.rs: a pasted copy and a duplicate keep them, a
deleted object leaves none behind, and the native file and SVG export
keep them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HhvnnnD5fTFupK8WWq35un
Illustrator 30.0's gradient Method, from the public documentation
(https://helpx.adobe.com/illustrator/desktop/paint-and-fill/create-and-edit-gradients/control-perceptual-interpolation-in-gradients.html):
Classic or Perceptual; perceptual interpolation matches colour transitions to
how the eye perceives differences between colours, and perceptual gradients
are rasterised when saved as PDF or SVG.
- Gradient gains `method` (Classic | Perceptual; serde default Classic,
written only when Perceptual, so older files open as Classic and Classic
files are written as before). The colour crate's OKLab helpers open up for it.
- Perceptual mixes neighbouring stop colours in OKLab (sample, and
sample_keep_model for the stroke meshes, Expand and blends); opacity stays
linear. Renderers that mix linearly get stops on the perceptual curve,
placed by halving each half segment where a straight line strays more than
0.4 of a colour level (a nearly straight segment needs none). Classic
gradients expand exactly as before.
- draws_as_pixels includes Perceptual gradients, so SVG and PDF write a
Perceptual fill or stroke as an image at the document's raster effects
resolution, clipped to what it paints. A gradient along or across a stroke
stays slices (PDF no longer draws it as a within-stroke image).
- With Dither: a dithered Perceptual fill is the perceptual mix with dither's
noise (the canvas's dithered grids are keyed by method too); SVG and PDF
write it as an image with the noise, and a dithered Perceptual stroke as an
image without it, as the canvas draws it (GradientPaint::pixels, SVG's
raster_paint and PDF's area are told whether they paint a fill).
Illustrator 30.0's gradient Dither, from the public documentation
(https://helpx.adobe.com/illustrator/desktop/paint-and-fill/create-and-edit-gradients/control-dither-in-gradients.html):
a Dither checkbox in the Gradient panel; dither converts the gradient into
pixels with subtle noise, works with gradient fills only, and follows the
Document Raster Effects resolution.
- Gradient gains `dither` (serde default false, written only when on, so
older files open unchanged and new files keep it).
- The renderer draws a dithered linear or radial fill as an image paint: the
gradient sampled a cell per device pixel or finer (quarter-octave steps,
cached per thread), each channel given seeded triangular noise of up to a
level before rounding. Same pixels every run; strokes and undithered fills
draw exactly as before. PNG and other raster exports go through it.
- SVG and PDF write a dithered fill as an image at the document's raster
effects resolution, clipped to the shape (the freeform-gradient path);
characters' fills stay vector gradients in SVG, with a warning.
- paint.editGradient and paint.setFill take `dither`; strokes refuse it.
The Gradient panel's Dither box is dimmed behind the Stroke proxy.
- i18n rows in all 24 catalogs; docs: G13, target-app-parity, ui-parity,
mcp.md, plugins.md.
The SVG writer always wrote the document's title in <title>
(crates/svg/src/export.rs), so every file of Export As with Use
Artboards, Export for Screens or an `artboard`/`range` export was titled
after the document ("Icons"), whatever its artboard was called. The
writer's `artboard` option can't tell those apart from Save: Save writes
the first artboard too, standing for the whole document.
ExportOptions has a new `artboard_title` (set by the caller, as
`artboard` is, not deserialized): the file is that one artboard, exported
on its own, and its <title> is the artboard's name (XML-escaped as
before), or the document's title when the artboard has no name. The
engine's SVG encoder (fileio/svg.rs) sets it exactly when the export
names its artboards, which is also when a file holds only that
artboard's art (#550): Export As per artboard, document.export /
serialize with `artboard` or `range`, MCP export, and Export for Screens'
artboards, assets, layers and selections (each written as an artboard
named after it). Save and Save As keep the document's title. Export for
Screens' Full Document draws the art bounds as one artboard, which is now
named after the document, so its SVG keeps the document's title too.
docs/file-format-parity.md's SVG row says so.
Tests: title_control_chars::an_artboard_exported_alone_is_titled_after_it
(crates/svg: two named artboards, the second named with markup;
exporting either alone gives its escaped name, the default and an
artboard without artboard_title give the document's title, an unnamed
artboard falls back to it) and
tests_svgenc::an_svg_of_one_artboard_is_titled_after_it (the engine:
`artboard` 0 and 1, useArtboards range all, Save's SVG, Export for
Screens per artboard, an asset and Full Document). Both fail without the
change (the title was "Icons") and pass with it.
SVG import checks each linked image's path with craft-docpath, and two
of its results were wrong, most visibly on Windows. A missing image's
link kept the path that craft-docpath returned, with its separators
rewritten: an href of /abs/photo.png became \abs\photo.png on Windows,
and SVG export with linked images writes that back as the href. A
folder that is not absolute, such as /tmp on Windows or a relative
folder on any system, made each relative href, and on Windows each
href rooted without a drive, resolve to a path that is not absolute,
so those images were not read. Two vectorcraft-svg tests failed on
Windows because of this.
A link now keeps the href as written (decoded) when the href resolves
to the same path without the SVG's folder, and keeps the path found
when it needs the folder: a relative path, or on Windows a path rooted
without a drive. A file: URL keeps the path it names. The folder is
made absolute against the working directory when it is not, as the
engine's reader does with the document's path. Hrefs that were refused
are still refused: paths on other computers, devices and URLs are not
read.
At 50,000 paths unless noted, against main:
- Pathfinder Outline (300 paths): 145 → 7.5 ms. Each segment was tested against every junction
of the arrangement, and each chain against every chain kept so far. Now a segment tests only
the junctions inside its control box (junctions are sorted by x; a point within reach of a
curve is within reach of that box), with the same nearest-point test. Coincident chains are
found through a grid on their midpoints, taking the first match in the same order a scan does.
- export SVG: 267 → 176 ms. Path data is written straight into one string (`push_num`, the
same digits and trimming as `fmt_num`) instead of a string per number, point and segment.
- open .vectorcraft: 239 → 195 ms. The JSON was parsed once only to find where it ends:
- A file with no NUL has no data after it.
- Otherwise the marker (which holds the first NUL, as JSON can't) is taken when it directly
follows the JSON's last byte, as the writer puts it.
- Anything else takes the old parse.
`perf --deep` prints output hashes of an SVG export and a save → open → save round trip; main
and this branch print the same ones. pathfinder_pinned (Outline included) passes, and so do the
pathops, format and svg suites (345 tests).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VmMmc1J8XeNLxhRihWv9cc
A root such as width="96pt" height="96" took CSS pixels (0.75 pt) along x and points along y,
so a square came in 24 x 32 pt and the artboard 96 x 96 instead of 96 x 72; with a viewBox the
art was stretched too.
When either root side is absolute, both axes now take CSS pixels at 96 per inch (usvg already
parses at that resolution): the unitless or px side is the same CSS length it would be in a
browser. The root scale is one number.
The test opens width="96pt" with height 96, 96px and 72pt, and width 128 with height 72pt: a
96 x 72 artboard and a 24 x 24 square; with viewBox 0 0 32 32 a 72 x 72 square at x 12 (meet),
and with preserveAspectRatio="none" the whole artboard.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
# Conflicts:
# ROADMAP.md
An SVG used as an animation rig lost its hidden alternates and their
rotation pivots: the exporter left hidden objects out (only hidden layers
could be kept), and the importer dropped data-* attributes.
- SVG export: hiddenLayers (on by default for document.save, as before)
now keeps hidden objects too, each written display="none" on its own
element (path, image, text or group). Exports still leave them out
unless asked; the CLI's convert asks, as a conversion keeps the
document.
- An object's own data: ObjectAttributes.data, the data-* attributes in
order (ours, data-name and data-vc-*, aside). The importer keeps them on
the objects and layers they belong to (at most 64 per element), the
exporter writes them back, the native format saves them (missing in
older files), and object.setProps {data: {key: value|null}} sets or
removes them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The svg crate's new CSS writer (css_rules) gives the CSS web pages style objects
with: rectangles and ellipses as background-color / linear-/radial-gradient(),
border and border-radius; type as font, color, spacing and alignment; shadows,
glows and blur as box-shadow, text-shadow and filter. It shares SVG export's
declarations (transparency, fonts, type spacing, colours, gradient stops), which
moved into svg/src/css.rs. Options: units, absolute position, width and height,
unnamed objects, rasterize unsupported art (as a PNG background-image).
Commands css.selection, css.generate and css.export; UI commands css.copy and
css.exportFile; the CSS Properties panel with a highlighted code view.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SVG has no freeform shading: an object's freeform fill or stroke is now
written as an <image> of the colour field over its painted box, sampled
on the grid the canvas uses (twice the box's size in points), clipped to
the shape or to the stroke's outline, with the fill or stroke opacity and
blend mode on the group. Linked image mode writes the PNG next to the
SVG. Character paints keep the linear approximation (with a warning);
gradients along or across strokes stay vector slices. The grid sizing
moved to the colour crate so the canvas and the export share it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SVG export writes UTF-8, UTF-16 (big-endian with a byte order mark) or
ISO 8859-1 (other characters as character references) and SVG/SVGZ
open in any of them. The Tiny 1.2 profile (simplified) writes
presentation attributes only and leaves out filters, masks, blend
modes and symbols with warnings. embedFonts embeds the fonts type uses
as @font-face subsets of the characters used, honouring the fonts'
fsType (whole when subsetting is forbidden, left out when embedding
is). New options: encoding, profile, embedFonts (SVG Options dialog,
document.formats); document.serialize also returns dataBase64 for
non-UTF-8 files.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
document.open {path} (and File > Place) find an SVG's relative <image>
links in its folder: raster files stay linked images (LinkInfo with size,
time and hash), SVG files and data: SVGs import as vector art, and a file
that can't be read becomes a placeholder in its box that keeps the link,
with a warning and a missingLinks row (Update Links / Relink read it
later). The SVG library itself reads no file unless given a reader.
Nested clip paths become clip groups in clip groups; undisplayed objects
(display:none) come back hidden; spreadMethod reflect and repeat are
expanded into stops over the painted area (matching resvg). Midpoint stops
and focal points keep folding back as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
<symbol> with <use> imports as a symbol (named after data-name or its id)
with one instance per <use>; a <use> that shows the symbol differently
(inherited paint, a viewport that cuts it) or that the canvas would paint
differently under its transform stays plain art. feGaussianBlur,
feDropShadow and the usual shadow chains (offset, blur, flood or colour
matrix, composite, merge) become Gaussian Blur, Drop Shadow, Outer/Inner
Glow and Feather effects; other filters are still reported. Export records
the symbol name and the shadow/glow blend mode so export -> import keeps
them. <pattern> keeps importing as a pattern swatch (now tested).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
file.info takes and reports the title, author, author title, description,
keywords, rating, copyright status, notice and URL in one undo step, plus
the read-only created (file.new) and modified (every save to a file) dates;
junk values are refused. Document.metadata (DocMetadata) round-trips
natively. PDF export writes author, subject and keywords to the info
dictionary and XMP, PNG export writes tEXt/iTXt chunks, SVG export writes
<desc> and, with the metadata option, every field as Dublin Core. The File
Info dialog (fileInfo, ui.fileInfoDialog) edits it with keyword chips and
star ratings.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The editing data now sits in CDATA inside <metadata> with a hash of the
markup around that element (whitespace between tags and line endings
don't count). document.open restores the native document when the hash
matches (data written without a hash is trusted as before); when another
app changed the SVG, or the data can't be read, it imports the SVG and
says why in its warnings. vectorcraft_svg::editing reports the data and
whether it is intact; editing_data stays. Live effects, symbols and
patterns come back exactly; the import fuzz test now mutates saved SVGs
with editing data and hidden layers.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Symbol instances export as one <symbol> def per symbol and a <use> per
instance. An instance the def can't stand for exactly (stained by its fill,
or a transform under which the canvas paints differently: scaling with
strokes, effects or live objects, turning with effects or brushes, any move
with pattern paints or unlinked masks) is written as its own art, now
stained like the canvas and with its patterns and unlinked masks left on
the page. Symbols nested in themselves stop at a fixed depth.
New SVG option hiddenLayers keeps hidden layers as display:none groups;
document.save (and Save As / Save a Copy) turns it on unless told
otherwise, exports leave hidden layers out. Import shows display:none
layer groups to usvg and marks them hidden, so they reopen hidden
(groups a <use> refers to stay templates). File Info metadata goes through
one Dublin Core list. The symbol stain helper moves to vectorcraft-brush
so the renderer and the SVG writer share it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
svgz is a writable format with the SVG options (document.export,
serialize, save, Export for Screens): exactly the SVG export, gzipped.
SVGZ is recognised by its extension or the gzip magic bytes and opens,
places and pastes like SVG (clipboard.importSvg takes dataBase64 of SVG
or SVGZ); Export As SVG or Save As to a .svgz name writes it
compressed. Gzip lives in the svg crate (pure-Rust flate2, builds for
wasm) and unpacking stops at 512 MB.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every laid-out line of type is written at its own position (wraps,
indents, tabs, hyphenation and all caps as on the canvas): a positioned
tspan per style run and, on justified lines, per word. Centred and
right-aligned lines are anchored at their centre or right end, so they
stay aligned in a viewer whose font differs. Leading comes from the
layout's baselines, baseline shift is written as dy, horizontal scale as
textLength, vertical scale as the size, tracking and manual kerning as
letter-spacing. The fewerTspans option (engine and dialog) writes one
positioned tspan per line.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
document.export, serialize and save, and the Save As / Save a Copy UI
commands, take the SVG options flat or as svg: {…}: styling
(presentation, style, entities, css), outlineText, images (embed or
link; linked files are written next to the SVG), objectIds (layerNames,
minimal, unique), decimals 1–7, minify, responsive, useArtboards and
range "all" (one SVG per artboard), preserveEditing (the SVG reopens as
the full document) and metadata. Unknown keys are rejected and
document.formats lists every option with its default. Results carry the
SVG writer's warnings.
Save writes SVG to an .svg path and Save again reuses its options.
Export As SVG and Save As / Save a Copy to .svg open SVG Options (first
use: Internal CSS and Responsive, then the last-used choices) with a
read-only code view. Save a Copy no longer retargets the document.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each <text> is swapped for a placeholder group before usvg runs, so the text keeps its z-order,
parent group, clip, mask, opacity, link and <use> instances, and usvg resolves its gradient and
pattern paints (in text space). <textPath> becomes type on a path (startOffset, side, anchor),
vertical writing-mode type on a vertical path. Per-character x/y/dx/dy/rotate lists become line
breaks (bounded), kerning (absolute x measured against the layout), baseline shift and rotation;
baseline-shift, numeric and relative weights (style names), letter/word spacing, kerning and the
character stroke's cap, join, miter and dashes are kept. The style cascade supports type, #id,
multiple-class, attribute, descendant and child selectors with !important and specificity.
<pattern> paints become pattern swatches. The import fuzz covers the new attributes and CSS.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A pixel (and a unitless user unit) is a point and absolute lengths convert at 72 pt per inch
(`font-size="12pt"` is 12 pt, `1in` is 72 pt); a root size in absolute units keeps the physical
size main already gave it (`210mm` opens as a 595.28 pt artboard, user units being CSS pixels of
it). The document's units follow the unit of the root width (pixels when it has none). Text font
sizes honour their units (em/ex/% relative to the parent size, CSS size keywords), as do text
positions, letter spacing, stroke widths and dashes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Gradient panel's Stroke buttons now work: with the Stroke proxy on a
linear or radial gradient they lay it within the stroke (placed on the
page, as before), along the path (start to end of each subpath) or across
it (left edge to right edge; inside/outside strokes span their visible
side). `paint.editGradient {stroke: true, strokeMode}` sets it (new art
too with nothing selected); `document.inspect` reports
strokeOptions.gradientMode. Model: StrokeLayer.gradient_mode (default
Within, not written).
Shared geometry in effects::stroke::gradient: ribs on the bisectors of
the flattened path (fans round sharp corners, widened under arrowheads,
following width profiles) cut slices each painted with a linear piece of
the gradient (Gradient::span). The canvas and raster export clip them to
the stroke's outline and arrowheads, grown a pixel and composited with
copy so there are no seams and translucent stops paint once. SVG and PDF
write the same slices clipped to the outline, with a warning (new
vectorcraft_svg::export_with_report). Outline Stroke and Expand make
gradient meshes clipped to the outline. Mesh quads now grow along their
edges so thin quads show no seams. Dashes, profiles and arrowheads keep
working in both modes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Exported masks carry data-vectorcraft-mask="noclip invert" (the options that
differ from clipping, not inverted). Import restores the flags and the original
mask art, dropping the backdrop rectangle and inverting filter group that draw
the options in SVG; a single mask object comes back as itself, not wrapped in a
group. Masks from other apps import as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Both writers now ask the shared stroke geometry how to write each stroke
(effects::stroke::for_writer): a plain stroke where that is exact, else the
canvas's filled outlines (arrowheads, width profiles, dashes fitted to corners,
dots with round or projecting caps), grouped under the stroke opacity when the
line and its heads overlap. Brushed strokes are written as their brush art.
Inside/outside alignment only applies to closed paths, so open paths export as
centred strokes. The PDF outside-stroke clip frame and the SVG outside mask now
cover the whole stroke reach (miter spikes, projecting caps). SVG stroke
gradients use the same paint box as the canvas. The svg crate docs no longer
claim arrowheads, profiles, brushes and effects are dropped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
VectorCraft writes and reads SVG at 1 user unit (CSS px) = 1 pt, but a root width/height in
absolute units was taken in CSS pixels (96 per inch). An A4 document from Inkscape
(width="210mm" height="297mm" viewBox="0 0 210 297") opened on a 793.7 × 1122.5 pt artboard,
i.e. 280 × 396 mm, with all art 4/3 too large, and its PDF export came out on that page size;
width="8.5in" gave an 11.3 inch artboard and width="300pt" a 400 pt one.
When the root width or height has an absolute unit (in, cm, mm, pt, pc) the import now scales by
72/96 along that axis, so the artboard and the art keep their physical size (595.3 × 841.9 pt for
A4); geometry, stroke widths and text follow through the same transform. Unitless, px, % and
viewBox-only files are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
VectorCraft exports type on a path as <text><textPath href="#text-path-N"
startOffset="…%"> with the path in <defs>, but the XML text fallback read
every <text> as point type at the text element's x/y. Text on a path
therefore re-imported as a straight line at (0, 0), above the artboard
and mostly out of view, and lost its path.
Resolve the textPath's href (SVG 2 href or xlink:href) to the <path>,
bake the text's user-space transform (and the path's own transform) into
it, and read startOffset (a percentage, or a length along the path) into
TextKind::OnPath.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
el-j