group3_tiff_rgba decodes each strip or tile with
craft_stream_codecs::decode_ccitt_raw instead of hayro-ccitt; the error
messages stay as they were. vectorcraft-engine's group3_reference test
keeps the hayro-ccitt reader as the reference: over 190 Group 3 TIFFs and
5,500 cut or edited copies the pixels and refusals match, except that a
strip or tile wider than MAX_CCITT_COLUMNS (2^20) is refused as an
overflow where hayro-ccitt ran out of data. Needs craft-libs#1772.
The CMYK reader's JPEG side (inks, component count, size, ICC profile)
uses craft_image_codecs::jpeg_inks instead of zune-jpeg, which stays only
as the dev-dependency the comparison test reads with: inks, sizes,
component counts and profiles match it over CMYK, YCCK, RGB and gray
JPEGs, whole and cut off. Frames zune-jpeg panicked on read as no inks.
Needs craft-libs#1779.
header_size reads every format's header through craft-image-codecs
(probe_allocation_size: stored size with no limit; a GIF's logical
screen through probe_size), no longer through the image crate's
readers. Metafile DIBs are written as PNG by craft-image-codecs.
Area type with wrap shapes was always written as outlines in the editing
data for other apps. It is now written as type when the reader works the
same wrap out again from the objects above it: each wrap object is a path
of one subpath, a compound path of visible members, or an unclipped group
of such objects, with no paint, or a solid fill, a plain solid stroke or
both, without effects, art style, opacity mask or clipping, with an
offset within -1000 to 1000 pt; the type's stored shapes match the ones
those objects give it; and the shapes of all area type stay within the
2,000,000 anchors the reader copies. Each frame of a thread is checked on
its own. When art is left out at the object limit, such type is written
as outlines again. Other wrapped area type stays outlines.
The reader recomputes the wrap with resolve_wrap after reading
(resolve_text_wrap in crates/eps/src/import/native/build.rs) and reads the
options with text_wrap in crates/eps/src/import/ai/read/document.rs; the
round-trip tests check the reopened characters, frame, wrap shapes and
wrap objects. resolve_wrap's layer walk is shared through a new
wrap_sources so the check follows the same traversal.
symbol.new and symbol.options take alignToPixelGrid (off by default),
symbol.list rows report it, and the New Symbol and Symbol Options dialogs
show the checkbox. When it is on, the horizontal and vertical straight
segments of the paths in the symbol's art, opacity mask art included,
snap to whole points (half points for an odd centered stroke) at the
symbol's natural size, and its stroke weights become whole points. Each
instance placed at that size, reflected across one axis or not, moves to
whole points and goes back to them after every move. A scaled, rotated or
sheared instance, and an instance in another symbol's art, stays where it
is. In symbol editing mode the stored art and the editing layer snap
after each edit.
Duplicate Symbol, rename, copy and paste and symbol libraries keep the
setting, and Delete Symbol removes it.
Docs: docs/mcp.md, docs/ui-parity.md.
Tsume no longer stops at horizontal type. In vertical type an upright
character loses its share of the blank space above and below its ink,
inside its cell down the column (the font's vertical metrics, else at
least one em, with the Proportional Metrics change), in point type and
on a path. A character lying on its side is measured along its own
width, and the characters of a tate-chu-yoko block close up across the
column while the block keeps its one em.
Tsume now measures what a glyph paints: its COLR or SVG art, else its
color bitmap, else its outline, so a color glyph is no longer measured
by the outline its art replaces.
Docs: docs/mcp.md, docs/target-app-parity.md, text.setRangeStyle and
text.setFormat params.
craft_image_codecs::probe_size only reads WebP, EXR and the GIF/BMP/TGA family;
for PNG, JPEG and TIFF it returns Unsupported. The cutover used it for header
sizes, so DXF export left every PNG out (cad tests), plugin objects lost PNG
pixels, and EPS image-mask and SVG import fell back to decoding whole images to
learn their size. vectorcraft_doc::header_size reads PNG's IHDR and a JPEG's
frame through craft-image-codecs and other formats through the image crate's
header readers, as main does, without decoding pixels. The SVG filter's JPEG
size comes from jpeg_info instead of a full decode.
Select Behind gave the general traversal a fresh limit of symbol art when
it fell back to it, for an envelope in Edit Contents or a root that is
not a layer. One Select Behind could then walk twice MAX_SYMBOL_NODES
objects of symbol art.
A root that is not a layer now sends Select Behind to the general
traversal before the first pass walks any art. For an envelope in Edit
Contents, the objects the first pass found stay first, and the general
traversal finds the rest with what is left of the limit.
When Profile inclusion is set to Don't Include Profiles (`profiles:
"none"`), gray, RGB and CMYK images are written in device color spaces
without their ICC profiles, and a JPEG that is kept as is loses its
APP2 ICC_PROFILE segments. PDF/A-2b, PDF/X-3 and PDF/X-4 keep tagging
their images.
Leaving the setting out keeps the earlier output. `profiles` now
serializes as null when not set, `"default"` clears a preset's value,
and Save PDF > Output lists Default first. Presets and stored settings
saved earlier hold `"profiles": "none"`, so exports from them now drop
image profiles.
A click walks a symbol instance's art within one limit of objects, but
the members of an Intertwine group in that art, and the objects of groups
nested deeper than the part path depth, were tested without taking from
it. Each object hit_any visits in placed symbol art now takes one object
from the limit, and testing stops once the limit is used up.
A click or marquee outside a symbol instance's box walks the instance's
art. An instance nested in that art was hit or touched anywhere in its
own box, so empty space inside the nested box, where a moved clipping
path reaches past the outer box, selected the outer instance. In that
walk a nested instance now counts only where its own art is, by the
same rule at every level. Inside the outer box nothing changes.
A click or marquee outside a symbol instance's box walks the instance's
placed art, and an instance nested in that art walked its own art with
no shared limit, so deeply nested or widely repeated instances could
make one click walk far more objects than the canvas draws.
Clicks, Select Behind, the locked-object and hidden-selection lookups,
Direct Selection's part lookup and marquees now share one limit of
MAX_SYMBOL_NODES objects across every level of nested art. Once it is
used up, an instance is no longer hit or touched outside its box;
inside its box it still is. The art walk also checks the instance's
reach before starting.
An image in a four-component ICCBased color space opened with its ink
amounts but without the profile its color space holds, so PDF export
could not write it in that profile again.
- The profile is read from the color space stream (at most 16 MiB) and
kept when it is a well-formed CMYK profile. A CMYK TIFF carries it as
tag 34675; a CMYK JPEG kept as it is carries it in ICC_PROFILE APP2
segments in place of its own, with its compressed data unchanged, and
one whose segments cannot hold it is stored as a CMYK TIFF.
- jpeg_with_icc passes over fill bytes (0xFF before a marker, ITU T.81
B.1.1.2) and leaves them out of the output.
- A malformed, non-CMYK or larger profile leaves a decoded image's TIFF
untagged and a kept JPEG as it is. DeviceCMYK and DeviceN images stay
untagged.
- A CMYK TIFF's tag 34675 is read up to 16 MiB; before, a 4 MiB profile
was read back and a 16 MiB one was not. The comments of
tiff_icc_profile and source_icc give that limit.
The Character panel shows Adjust Horizontal and Adjust Vertical Diacritic
Position with the Middle Eastern options. text.setFormat and
text.setRangeStyle take diacriticHorizontal and diacriticVertical, in
thousandths of an em from -1000 to 1000. Positive values move the
characters' mark glyphs (GDEF class 3) right and up, and the letters keep
their places. An Arabic harakat or a Hebrew point styled apart from its
letter takes its own values without splitting the shaping cluster.
Character styles and Character Style Options carry both values. They are
saved only when set, and SVG export writes such type as outlines.
Docs: docs/mcp.md, docs/type-parity.md, docs/target-app-parity.md.
Type > Composite Fonts... has Custom: named character sets, each setting
the characters it lists in its own font, style, size, baseline and
scales ahead of their class. The later of two sets that list a character
sets it. Characters are added by Direct Entry or by Unicode code and
deleted one at a time, and the parts table scrolls when it grows.
- type.compositeFonts.set takes `custom`; part keys left out take the
Kanji part's values. Sets are saved with the document and in
.vccomposite files, and Export refuses a file over 4 MB.
- VectorCraft's own limits: 64 sets per font, 8,192 characters per set,
each character once per set, up to U+FFFF and no control characters.
- The font database caches each font's character lookup.
Docs: docs/mcp.md, docs/type-parity.md, docs/target-app-parity.md.
With Direct Selection, the point a Shift-click adds to two or more
selected anchor points is the key anchor. It is drawn as a larger filled
square in a frame, and the Align panel and the Control bar show the key
choice chosen as Align to Key Anchor. The Align buttons move the other
points to its x or y, and Distribute with a spacing places them from it.
A plain click still selects one point alone.
select.key takes {id, anchor} to set it, select.key {} cancels it, and
document.inspect reports it as keyAnchor. It goes when it is deselected,
deleted or left as the only selected point.
A clipping path that paints a fill or stroke kept the geometry effects of
those fills and strokes live, so a geometry effect on the clip group or on
a group around it reshaped the path first and the fill's or stroke's
effect applied after it, on the canvas and in export alike. Unpainted
clipping paths already applied them first.
The clipping path now becomes one path per fill and stroke with those
effects applied, as an unpainted one does, in a group marked as such
pieces (`NodeExtras::clip_pieces`, saved with the document). `clip_paint`
paints that group's fill pieces behind the clipped art and its stroke
pieces over it, as the path paints its fills and strokes, also after an
effect around it turns a piece into a compound path or puts it in a group.
The clip group's visual bounds take in the stroke pieces' width, and HP-GL
export draws the clipping path once, in its stroke pieces.
An export of a painted clipping path with such effects in a clip group
without effects of its own also keeps its fill and stroke now, where the
baked pieces painted nothing.