A graph in a legacy .ai, or in a current file's editing data, is written
as `%_Gs`, its data and settings on `%_` lines, its drawn art in a group
with a `%_... Go` tag after each object, and `%_GS`. The reader stopped
on the graph operators as ones it doesn't know, so such a file did not
open through its layers.
The reader now leaves the graph's records out with their operands and
reads the drawn art as usual: the bars, lines and labels come in as the
groups, paths and type they are written as. Paint operators on `%_`
lines between `Gt` and `GT` set nothing. One note per file says the
graphs came in as plain art without their data or settings. A graph
that starts inside another, or doesn't end before its layer does, keeps
the art read so far with a second note. Graph operators outside a graph
stay unknown.
In editing data and legacy .ai files, a lower-case l, c, v or y makes the
anchor the segment ends on smooth and an upper-case L, C, V or Y makes it a
corner. The reader used to guess each anchor's kind from whether its
handles are in line, so a lower-case l always read as a corner and an
upper-case C with handles in line read as smooth. A closed path's first
anchor takes the case of the segment back to it. Points and handles do not
move. Paths, compound paths, clipping paths, guide paths, area type frames
and type-on-a-path paths all keep the kinds.
Area type made from the editing data's text document came in with one
row, one column, the 18 pt gutter and the Ascent first baseline,
whatever its frame kept. Each area frame now keeps its rows and columns,
the gutter between rows and the gutter between columns, and its first
baseline (Fixed, Cap Height, Leading or x Height) with its minimum,
scaled with the art's units. Each frame of a thread keeps its own, and a
key that is missing or can't be read leaves that option as it was. A
first baseline the frame keeps is no longer replaced by the fixed one
that matches the file's first line.
AreaOptions gets row_gutter (rowGutter, left out while unset), which
layout spaces rows by; text.areaOptions takes and reports rowGutter, a
gutter given alone sets both, and the Area Type Options dialog shows
Row Gutter.
Fixes#2763. An object whose editing-data dictionary holds the text wrap
options keeps them (offset in points, 6 pt by default, clamped to
±1000 pt; Invert Wrap), and the objects after the text in a version 7
wraparound group (`*w` … `*W`) are wrap objects at the default options.
Area type below a wrap object in its layer flows around it before the
layers are compared with the page, and threaded type flows on into its
next frames. A Large Canvas file's offset is scaled with its art.
The wrap resolver moves into `vectorcraft_doc::text::resolve_wrap` for
the import and the edit-time refresh, and builds each wrap object's
outline once per layer. A file whose wrap shapes would copy more than
2,000,000 anchors is not read from its editing data.
Docs: docs/file-format-parity.md.
An object whose art style holds Drop Shadow, Inner Glow, Outer Glow,
Feather, Round Corners or Scribble now opens as itself with those
effects live, in their order. Feather is read under the name and key
the files use (`Adobe Fuzzy Mask`, `Radius`).
Each setting is read only when it is written as its kind (`/Int`,
`/Real`, a fill style for a color) with a value in range. An unknown
key or choice, a missing setting, or a Scribble with a color of its own
keeps the drawn look, where the reader used to cast, clamp or default
the value.
Legacy type (`To` ... `TO`) now reads `min opt max TC` and `min opt max TW`
into the paragraph's letter and word spacing, `hangingQuotes Tq` into Roman
hanging punctuation, and `hyphenate limit lead tail max Th` into hyphenation
on or off, After First, Before Last and the Hyphen Limit (0 when the limit
flag is 0). Values are kept to the ranges the Justification and Hyphenation
options take, and a spacing triple out of order goes back to its default.
Operands of the legacy type attributes are read by their place: an attribute
whose operands are missing, are not numbers, or are flags other than 0 and 1
stays as it was. Before, a value that was not a number shifted the later
operands (`5 /x 7 Ti` set indents of 5, 7 and 0), and a bad `TA` or `Ta`
turned kerning on or set the alignment to left. One long `Tx` no longer goes
past the reader's limit on characters.
Version 7 files write each palette swatch as its paint and `Pc`, with no
name. The reader only kept a swatch with a name before `Pc`, so such a
palette gave no swatches and the document got the default set.
A swatch without a name is now named by its spot color, gradient or
pattern, or by its values in the format the swatch import already uses
for unnamed colors ("Gray K=25", "C=25 M=0 Y=0 K=0"). `Pn` and a `Pc`
with no paint before it add nothing. Named swatches keep their names.
An operator the reader does not know, or a `Pg`, now ends the paint
before it, so a later `(name) Pc` no longer takes that paint. The
swatch limit is kept with a running count instead of summing every
group at each `Pc`.
An envelope in a .ai's editing data opened as its distorted art, with
its mesh and contents dropped. It now opens as a live envelope of its
contents: a warp with its style, bend, distortions and orientation, a
mesh with its points and handles, or a top object with its outline,
with Envelope Options, Fidelity and Edit Contents.
The envelope replaces its drawn art only when VectorCraft draws it like
that art, with a budget so that live envelopes don't make the file open
from its PDF part. When the layers still differ too much from the page,
the drawn art is put back. An envelope that can't be read, or that
draws differently, keeps its drawn art with a note. Envelopes of type
keep their drawn art and hidden type as before.
The editing data's layer options set a template layer from the flag four
places after the colour index. The colour index of nine-operand options,
and the colour index and dimming percentage of eleven and twelve operand
options, are read from their own places (they were read one place later).
A straight (N) * guide path directly on a layer that runs across the
whole canvas, or from edge to edge of one artboard, opens as a ruler
guide on that layer (an artboard guide of that artboard); other guide
paths stay guide paths. Guides scale with the document when its editing
data is in the units of a UserUnit page, and an EPS that keeps only its
first artboard keeps only that artboard's guides.
Fixes#779
File › Open and File › Place read JP2 files and raw JPEG 2000
codestreams (.jpf, .jpx, .jp2, .j2k, .j2c, .jpc) through hayro-jpeg2000,
and store the pixels as PNG. Gray, RGB, CMYK and ICC-tagged images, with
or without alpha, are converted to RGBA. Several codestreams, JPX
composition or layers, other color spaces, inconsistent or oversize
palettes, inconsistent channel definitions, oversize tile grids and
images past the pixel allowance are refused with a message before the
pixels are decoded.
The capture and display resolution boxes size the image: the box that
comes last sets each axis, and an axis without a usable value uses
72 ppi.
IccProfile::to_srgb_into converts a pixel with the cached transform and
no heap allocation.
Art dragged off the canvas and colours dragged in from elsewhere (swatches, another library's)
used to be added at the end of the ungrouped items, wherever they were dropped. They now land
as items dragged within the list do: on another item of their kind, just before or after it by
the pointer's half and in its group; on a group, into it; away from the items, ungrouped at the
end as before.
library.add takes group and before/after, placed as library.moveItem places an item (the
placement is shared). A library change that fails partway now leaves the library as it was.
A new reader, written from the ZSoft PCX Technical Reference Manual, decodes
the run-length encoded scan lines of 1-bit images in the header palette's
first two colors (black and white when the version has no palette or the two
colors are the same), 1-bit images in two to four planes and 4-bit images
with the header's palette (the EGA colors for versions 0 and 3), 2-bit CGA
images in the colors that the header's CGA background and palette settings
pick, 8-bit images with the 256-color palette after the image data (gray
levels without one) and 24-bit images in three planes. The image is stored as PNG at the resolution of the header's HDpi and
VDpi fields.
Open and Place list PCX, and pasting a copied .pcx file and Glass's Load
Texture… read it by its header. A file of another depth or plane count, one
cut short and one whose pixels as RGBA pass the pixel allowance are refused
with a message before its pixels are decoded. The desktop packaging does not
associate .pcx files, and no export writes them.
Docs: docs/file-format-parity.md, docs/target-app-parity.md, docs/mcp.md,
docs/releasing.md.
Pages that share one Contents array made the name of a refused stream read that array once per page, quadratic in a crafted file. Each object is now read once, and a page's own array as before. From the review of #4612.
A page's Contents can be an array written as an object of its own (/Contents 6 0 R). Its streams were named "a stream in it": the reference was compared with the stream, not the array's items. From the review of #4603.
A page drawing one form over and over made its art again for every drawing
with no limit (a 1 KB file took 877 MB), and a form decoding to 64 MiB drawn
a thousand times hung. The import now makes at most half the objects a saved
document may hold (2M, 256K on the web), leaving the rest out with a warning,
and the scan reads at most 2 GiB of form content per page before refusing it
as past the reader's limits.
Review follow-ups to the live blend import (#4127):
- The check's grid saturates its neighbours' keys: a step far out (a
coordinate like 1e300) saturates its cell, and the cells round it
overflowed (a panic in debug builds, a wrong cell in release).
- Each laid object matches one drawn step: two drawn steps near one laid
object no longer both take it, so a drawing the blend doesn't lay out
isn't taken for it (the drawn steps stay).
- A blend's record is read to the length its XP gives: hex digits after
it on the last line no longer spoil it.
Tests: a laid object matches one drawn step, 1e300 coordinates in the
check and in the import fuzz property, and a record line longer than
the record.
Review follow-ups for the live blend import:
- The check of a blend against its drawn steps matched every drawn step
against every laid one and measured centres inside the loop, so a
crafted blend (64 keys, 1,000 steps between each) could stall the
import. Centres and fills are now measured once, and each drawn step
finds its neighbours through a grid of cells the tolerance wide.
- A blend lays out at most 5,000 objects to be made live, and all the
blends of one import at most 50,000 (each laid out twice, once per
orientation); past either, the drawn steps stay, with the note. The
live steps take the place of drawn steps already counted against the
reader's object limit.
- A fill the check can't compare (a gradient, a pattern) or a missing
one no longer passes: the fills must both be none or solid colours
within the tolerance.
- When what follows the blend's own art isn't a group of drawn steps,
that art (its keys and spine) goes too, with the note, instead of
staying drawn.
- The spine test can no longer pass without asserting: a spine written
backwards keeps the keys' order (live), and a curved spine is read
exactly while its straight-drawn steps stay drawn. New tests for steps
with a gradient, for the blend's art when its steps aren't a group,
and for the budget; import_fuzz gains a mutated-blend property and a
large-blend test (998 steps live, 1,500 kept drawn, both quickly).
A blend in a .ai file's editing data came in as plain art twice over:
the blend's own art (its key objects and spine, on `%_` lines) and its
drawn steps were both kept visible, so the keys lay doubled under the
drawn ones and the spine showed as a stray path, and nothing was live.
The spine itself was read exactly (its line and curve segments match).
The `(Adobe Path Blends) 1 0 n XP` record is now read (n bytes of hex on
the `%` lines after it, capped, checked: `DCBA`, version 1, the key
count and a record per key starting with its id). Each object's
dictionary names its part (`ShapeBlendPathType` K or S) and a key's id
(`ShapeBlendPathID`); the keys the record lists and the spine become one
live blend, the keys sitting on the spine's anchors. The record's other
fields aren't known, so the spacing is the drawn steps between two keys
(Specified Steps) and the orientation whichever lays the blend out as
drawn; the blend replaces the drawn steps only when it comes out as
drawn (each step within 0.5 pt and 0.02 of its colour). Otherwise the
drawn steps stay, without the blend's own art over them, with a note.
Behaviour as documented: https://helpx.adobe.com/illustrator/using/blending-objects.html
Save As .ai writes the PDF-compatible part only (no editing data), so a
blend is written back as its steps drawn, not live.
Hand-written tests: two keys and a straight spine as one live blend with
three steps and interpolated colours; a curved spine (l, c, L) in both
directions keeping its points and the keys' order; a garbled or short
record, a key without an id and ids the record doesn't list keeping the
drawn steps with a note; moving a key moving the steps; record parsing
on hostile input. The import fuzz target gains the record and the new
mesh keys.
Review follow-up. A function-based shading is read once per import (kept
per shading pattern and inks), and all of them draw on one budget of
200,000 function evaluations; past it a shading comes in as its average
colour, then flat grey, each with a warning that names the cause. A
gradient read leaves enough of the budget to average another shading.
A colour that changes in step with the sampling grid is checked across
the middle of the domain before an axis is chosen, so it isn't read
along the wrong one. The grey fallback warning for other shadings names
only them.
Tests: the same shading on 2,000 paths is read once; 40 distinct
shadings stay within the budget; a hostile file (5,000 fills, 20
shadings with 500-step functions) opens in 0.8 s, 253 s before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The editing data's palette (%AI5_BeginPalette) was skipped, so a .ai file
opened with the default swatches plus only the spot colours its art used:
inks the art doesn't use (a Pantone set kept in the file) and colour groups
were lost. The palette is now read in order (process, spot in CMYK, RGB or
Lab, and global colours, gradients and patterns, colour groups; versioned
content read once) and replaces the default swatches.
Xs/Xz (Registration at a tint, as printer's marks use it) are read as the
Registration swatch: before, art painted with them stopped the editing data
from being read at all.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- One budget for the whole read (layers, pixels, characters, table
cells), shared with embedded documents and charged also when their
reading fails; the first reading is dropped before the second.
- Tables: at most 65 536 cells; lines and cell formats that don't
match the grid are left out instead of shifted.
- Paragraph formats matched to runs by binary search; font names
longer than 256 bytes are dropped.
- Overlays: one work budget per import; encoded originals decoded
within their layer's size; table cell outlines recoloured too.
- Engine: master page copies capped, each page's master found in
linear time, line gaps looked up once per font, the working CMYK
table built once, objects past MAX_COORD left out.
- Embedded content stays with its node; pixel masks under artistic
text are read whole and include stroke reach; gradient overlays on
hidden layers are kept.
- Render: contexts and raster effect images stay within
MAX_RASTER_SIDE per side (vello overflowed near u16::MAX; found by
the new fuzz target).
- Tests: affinity tests/import_limits.rs, feature samples in
import_fuzz.rs, shared builders in engine/tests/support/.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The list fills the panel down to its bottom bar and scrolls, also while a drag nears its top or
bottom edge (in the dock and in a flyout, as the Layers panel does).
Dragging in the list:
- an item (colour, text style or graphic) dropped on another of its kind lands just before or
after it, by the pointer's half, an accent line showing where, and joins that item's group;
- dropped elsewhere on a group (its header too, even collapsed) it goes into that group, and on
the ungrouped items out of its group, the section outlined; with every item grouped, a
"Drop here to take it out of its group" strip shows while one is dragged;
- a group dragged by its header goes before or after another group.
Colours and graphics still drag onto art and the canvas as before.
Engine: library.moveItem takes `before` / `after` (an item of the same kind) to place the item
in its kind's order, which the panel shows; the new library.moveGroup {group, to} moves a group.
Text styles drag as PanelDrag::LibraryItem, groups as PanelDrag::LibraryGroup, and a library
colour's Paint drag names its library item, so the panel tells it from a swatch dragged in.
Tests: reordering and moving in the engine, the commands in the library fuzz test, and panel
tests that drop styles and group headers and check the list scrolls and scrolls while dragging.
Strings translated in the 23 catalogs; docs/mcp.md and the parity row updated.
The PostScript interpreter that reads EPS and PostScript .ai files
accepts a FlateDecode or LZWDecode filter with Predictor 2 at
BitsPerComponent 1, 2, 4, 8 and 16, but undid the horizontal
differencing only at 8 bits. At the other depths the filter gave back
the stored differences, and the image drew them instead of its
samples.
The predictor is now undone at every depth it accepts, as ISO 32000-1
§7.4.4.4 describes: each component adds the same component of the
pixel to its left, within its bit width, and each row starts again.
16-bit components add as big-endian 16-bit numbers. The padding bits
at the end of a row stay as they are, and 8-bit data takes the same
path as before. The PNG predictors are unchanged.
Docs: docs/target-app-parity.md.
A .ai can keep its variables and data sets in a variable library in
its editing data. VectorCraft doesn't read that library, and such a
file opened with an empty Variables panel and no warning about it.
The reader now counts the library's variables and data sets in the
editing data's /Document dictionary, and opening the file adds a
warning: "the file's variables and data sets weren't read (2
variables, 3 data sets): the Variables panel doesn't list them". The
same warning appears when the layers can't be read and the file opens
from its PDF part. A PDF that opens from the editing data it carries
gets the warning too. A library with no variables and no data sets
gives no warning.
Docs: docs/target-app-parity.md.
When a PDF, PDF-compatible .ai or .ait opens, crates/pdf/src/file_info.rs
reads its File Info from the XMP packet with roxmltree, which parses
nested elements recursively and has no limit on their depth. A packet
with elements nested 10,000 deep (70 KB) overflowed a 1 MB stack, and
one nested 100,000 deep (700 KB) overflowed an 8 MB stack. A stack
overflow aborts the app.
File Info now counts how deep the packet's elements nest before roxmltree
parses it, and leaves out a packet nested more than 256 deep, as it
leaves out a packet that is not XML. The document information
dictionary then gives the File Info.
dataset.import refused every header cell that wasn't the name of a
variable the document defines. A data source file could only fill
variables made beforehand, and `#` and `@` in a header were read as part
of a variable's name.
A header cell that names no variable now defines a new, unbound
variable in the import's undo step. `#Shown` defines a visibility
variable `Shown`, `@Photo` defines an image variable `Photo`, and a
cell without a prefix defines a text variable. A prefixed cell whose
name is already a variable of the prefix's kind names it. A cell that
is a variable's exact name keeps naming it, even when the name starts
with `#`, `@` or `%`, so files that dataset.export writes import back
as before. The import fails, naming the column, on a prefixed cell
whose name is a variable of another kind and on a `%` cell that names
no variable (graph data files). While the variables are locked, a
header that would define a variable fails the import with "variables
are locked", as New Variable does. The reply's `defined` lists the new
variables.
Docs: docs/mcp.md, docs/target-app-parity.md.
1-bit TIFFs compressed with CCITT Group 3 did not open, place or
relink: the image decoder reads Group 4 only and refuses Compression 2
(modified Huffman) and Compression 3 (ITU-T T.4).
They are now decoded with hayro-ccitt, the CCITT decoder that EPS
import uses. Compression 2 rows start at a byte and have no EOL codes.
Compression 3 rows follow an EOL code and are one-dimensional, or, with
T4Options bit 0, one- or two-dimensional as each row's tag bit says.
Fill bits, FillOrder 2, either PhotometricInterpretation, strips and
tiles are read.
A file whose pixels would take more than the decoding allowance as
RGBA is refused as too large to draw. A strip or tile whose data ends
before its last row is refused as damaged, and so is a tile larger
than the allowance. The strips and tiles together are read from at
most as many bytes as the file holds.
Part of #1466.
Docs: docs/file-format-parity.md, docs/target-app-parity.md.
VectorCraft listed a variable font's named instances as styles, but
type could not be set between or past them. The reference app's
Character panel has a Variable Font control that adjusts a variable
font's axes, such as Weight, Width and Slant.
Type now keeps Variable Font settings by axis tag
(CharStyle::font_axes, saved as fontAxes only when set). text.setFormat
and text.setRangeStyle take fontAxes: each value is clamped to its
axis's range, a tag the font has no axis for is left out, and null
goes back to the style's own setting. Another font or style clears
them, including one from a character style that does not set them. A
character style captured from type sets them, with {} for none.
text.fontList {family, style} lists a font's axes.
Layout, outlines, the substituted-glyph highlight and PDF export use
the font instanced at the settings, and neighboring runs at different
settings are shaped apart, as runs in different fonts are. SVG export
writes them as font-variation-settings, the wght setting as the weight
and the wdth setting as font-stretch, and embeds the font instanced
there; SVG import reads font-variation-settings back. Legacy .ai export
warns that the settings are left out.
The Character panel shows a Variable Font button for a variable font.
It opens a slider and a field for each axis the font does not hide,
labeled with the font's name for the axis, and screen readers get each
slider under that name. A drag is one undo step. Reset Panel leaves
the settings, as it leaves the font and style.
Missing Glyph Protection now keeps the settings and the font version
on the characters that keep their font; before, it kept only the
family and style. A character style applied or redefined with another
font now clears the old font's version, as text.setStyle does, unless
the style sets one. SVG export with embedded fonts warns when faces of
one weight, style and width share characters, which browsers then draw
in one of the faces.
Docs: docs/mcp.md, docs/target-app-parity.md, docs/type-parity.md.
dotCooCoo