Commit Graph
97 Commits
Author SHA1 Message Date
BFlat 499903878e Merge pull request #7055 from dotCooCoo/feat/mojikumi-presets
Type: the Mojikumi Set menu offers the predefined spacing sets
2026-11-13 13:17:36 +07:00
dotCooCoo 7b6b6af7b1 Type: the Mojikumi Set menu offers the predefined spacing sets
The Paragraph panel's Mojikumi Set lists None, YakumonoHankaku,
GyomatsuYakumonoHankaku, GyomatsuYakumonoZenkaku and YakumonoZenkaku,
then Line-end Punctuation Half Width, and text.setFormat {mojikumi}
takes yakumonoHankaku, gyomatsuYakumonoHankaku, gyomatsuYakumonoZenkaku
and yakumonoZenkaku besides none and lineEndHalf.

The spacing of each set is VectorCraft's own rule, built on the help
page's one-line description of each set:
- YakumonoHankaku sets every bracket, comma and full stop half width,
  takes a quarter em off each side of a middle dot, and puts a quarter
  em between Japanese and Latin letters or digits.
- GyomatsuYakumonoHankaku keeps punctuation full width inside a line,
  sets an opening bracket starting a line flush and a closing mark
  ending it half width, and takes the quarter em after a middle dot
  ending a line off.
- GyomatsuYakumonoZenkaku keeps punctuation full width and adds the
  quarter em between Japanese and Latin text.
- YakumonoZenkaku sets characters as None does.

Both composers and vertical type measure each line as it is set. A
right-to-left paragraph, whose line ends are measured untrimmed, also
draws them untrimmed, and a right-to-left line that ends between a
Japanese and a Latin character is measured without the quarter em the
set adds there. With a Desired glyph scaling other than 100%, the
space a set takes off before an opening bracket or middle dot, and the
Japanese-Latin space recorded at a line's end, scale with the glyph.

The sets save with the document and paragraph styles; new type and
Reset Panel keep Line-end Punctuation Half Width. SVG export lays type
with middle dots out again for its positioned spans. A .ai save
reports a paragraph with one of the new sets as left out.

Docs: docs/mcp.md, docs/target-app-parity.md, docs/type-parity.md,
docs/gaps.md.
2026-11-12 21:07:44 -08:00
jmverges f94eecd439 Merge remote-tracking branch 'upstream/main' into fix/ai-import-font-match
# Conflicts:
#	docs/file-format-parity.md
2026-11-13 01:55:35 +00:00
jmvergesandClaude Opus 5.5 bb50d438c8 .ai import: type from the text document takes the face of its exact PostScript name and recorded version (#6983)
Type read from a text document's font record was set in whichever installed
face first answered to the record's PostScript name (ignoring case and
punctuation), and its family was then swapped for any available family
spelled alike but for case and punctuation (`FontDb::find_family`): a record
naming `SampleInk`, the PostScript name of `Sample/Ink`, set its type in
`SAMPLE INK`. Copies of one family and style under other PostScript names or
versions were not told apart either, so the type could set at other widths
than the file's own page.

- `named_font` / `document_font` (crates/eps/src/import/text.rs) give the
  family and style of the installed face of exactly that PostScript name,
  else one that differs only in case or punctuation, and name its version
  when the family and style alone resolve to another installed copy.
- The font record's key 5, its version string (name ID 5), is read and picks
  that version among the faces of the name. Each record is resolved once.
- `FontDb::postscript_faces` lists every installed face a PostScript name
  answers to, loading their families; the import prefers among them one of
  exactly that name.
- Type a PostScript program sets (EPS import) and legacy .ai type take the
  face of their font's name the same way.
- The .ai save writes a static font's version (`/5`) when its PostScript name
  alone would give the reader another installed copy (another version, or the
  same name in another family), keeps the version in the style it compares,
  and checks that the type comes back in the same family, style and face,
  taking the record's version as the reader does (up to 256 bytes), so such
  type round-trips without a note.

Measured on 224 users' files (used locally, never committed): 16 files set
some type in another face; 12 render differently against their own page,
10 closer and 2 further by 0.03% of the page; envelopes of type that open
live go from 154 to 159, none fewer in any file; no file opens another way.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 00:33:13 +00:00
dotCooCoo 705b688f2b Type: Insert Aki keeps a widened glyph's own width and an optical pair's letter
A glyph that justification widens to fill the line keeps its Insert Aki
before it at its width, and its own width, which ligature carets stand
in, now starts where the glyph is drawn. Optical kerning measures a pair
from the letter of a cluster rather than from a combining mark whose
advance holds the cluster's Aki.

The SVG test also checks centered and right-aligned lines with Insert Aki.
2026-11-12 15:48:12 -08:00
dotCooCoo 89748b0360 Type: Insert Aki Left and Right add space before and after a character
Characters take two new attributes, `akiLeft` and `akiRight`: space
before and after each character in reading order, in thousandths of an
em from 0 to 1000, measured as tracking is. `text.setFormat`,
`text.setRangeStyle` and character styles set them; values that are not
numbers or are outside the range are refused, and a file holding one
outside the range opens with the nearer end. Both are saved only when
not 0.

- Layout: Left is the space above a character in vertical type and on
  its right in right-to-left type. Wrapping counts the space, a line's
  end keeps it when aligned, and justification and Optical kerning
  leave it at its set width. A ligature, a letter with its marks and a
  tate-chu-yoko block take the values of their first character.
- With Fractional Widths off or System Layout on, the space before and
  the space after a character are each set in whole points and added
  to its rounded width, so justification keeps them at that width. A
  tate-chu-yoko block rounds the space above and below it the same way.
- Character panel: with Show East Asian Options on, Insert Aki Left and
  Insert Aki Right fields with 0, 250, 500 and 1000 in their menus;
  Reset Panel sets both to 0.
- Character Style Options › Advanced Character Formats shows Insert
  Aki Left and Insert Aki Right, with the Character panel's 0, 250, 500
  and 1000 in their menus; a typed value below 0 or past 1000 sets 0 or
  1000, and Reset Panel unsets both.
- SVG export: live text places each character the space moves, and
  type on a path with Insert Aki is written as outlines.

Docs: docs/mcp.md, docs/target-app-parity.md.
2026-11-12 10:24:38 -08:00
BFlat e618783377 Merge pull request #6084 from Taha-Mahmoodi/feat/glyph-alternates
Glyphs: alternates keep the word joined (feature values, one-piece shaping, unencoded glyphs)
2026-11-12 23:53:13 +07:00
dotCooCoo 3bcdbd7e71 Type: letter spacing and Full Justify keep shaped clusters together
Single Word Justification's Full Justify counted and placed a gap after
every drawn glyph, so a combining mark or the vowel sign of a Devanagari
syllable moved away from its letter. Desired Letter Spacing and the
letter spacing range of a justified line did the same between two glyphs
of one cluster.

- The Full Justify gap count and placement share one check that skips
  glyphs with the same cluster byte as the glyph drawn after them.
- Letter spacing goes after the last glyph of a cluster only.
- Letter spacing, Full Justify and CJK inter-character gaps pass over
  every default-ignorable character drawn with no width (soft hyphens,
  bidi marks, embeddings, overrides and isolates, joiners, word joiners,
  variation selectors, tags, the byte order mark) and put the space
  between the visible characters on either side. The Hangul fillers stay
  letters.
- A zero-width space takes no word-space room on a justified line and
  stays a line break opportunity. Letter spacing before it is set after
  it, so a line broken there ends at its last letter.
- A single word that Full Justify spreads and that breaks at a hyphen
  has the same gap before the drawn hyphen as between its letters.
- Kinsoku push-in counts the gaps it can close past format controls, as
  the layout closes them.

Tests cover horizontal, right-to-left and vertical type, Desired and
fitted letter spacing, format controls, zero-width spaces, the drawn
hyphen, push-in, and a Devanagari syllable when a Devanagari face is
installed.
2026-11-12 06:54:44 -08:00
Taha-Mahmoodi ebad5d75c0 Use the unencoded fixture glyph in ligature range regression 2026-11-12 18:21:43 +04:30
Taha-Mahmoodi 4c0c6179c9 Fix contextual glyph face, cluster boundaries and insertion status 2026-11-12 17:22:47 +04:30
Taha-Mahmoodi 9624b29dad Merge main into contextual glyph alternates 2026-11-12 15:56:45 +04:30
Sven Lafebre ab17837cba Merge remote-tracking branch 'origin/main' into fix/5111-kern-paragraph-start
# Conflicts:
#	docs/target-app-parity.md
2026-11-12 09:24:22 +01:00
Taha-Mahmoodi 5144f22976 Merge main 48e066f47 into glyph alternates #6084 2026-11-12 12:34:54 +04:30
dotCooCoo 253c60a23e Type: a line's first word is measured as set when looking for room
In a frame that is not a rectangle, or with text wrap, layout moves a
line down to the first span wide enough for its first word. The word
was measured at full width before Mojikumi set an opening bracket at
the line's start flush, and with its last character at full width, so
a word that fit as set moved down to a wider line. It also ended at any
break opportunity, so an opening bracket counted without the character
kinsoku keeps with it and could be placed alone on a line too narrow
for that character.

- word_width ends the word only where break_line may end a line: at a
  cluster end that kinsoku allows, not inside a No Break run, whose
  spaces count. It leaves out the spaces at the word's end and narrows
  its last glyph by the Mojikumi line-end credit.
- place_flow_line measures the word after the opening-bracket trim, and
  the Every-line Composer's first-line placement takes off the same
  start credit, so the composed widths match the placed lines.
2026-11-11 15:20:50 -08:00
dotCooCoo a5bfa524dc Type: Mojikumi spacing at line ends counts when lines break
With Line-end Punctuation Half Width, a closing bracket, comma or full
stop ending a line is set half width, but both composers measured it at
full width. A closing bracket that fit only at half width went to the
next line with the character before it, and so did a comma or full stop
with Burasagari None.

The single-line and every-line composers now measure each candidate
line with its last character set as narrow as Mojikumi sets it, using
the same rule the line is drawn with. A mark that already lost its half
to the punctuation after it and a line ending in a hyphen take nothing
more off, and the quarter em between Japanese and Latin text counts only
where the line ends. Kinsoku Shori Type measures the pushed-in and
pushed-out lines the same way. Right-to-left lines and Mojikumi None
break as before.

A test font sets ideographs and Japanese punctuation one em wide, so
the layout tests run without system fonts.
2026-11-11 15:20:34 -08:00
Sven Lafebre d9f961b0c1 Fix #5111: check spaced harakat with the in-memory Arabic font
The harakat test now runs on the in-memory Arabic test font in every
environment, as well as on an installed Arabic face, and checks that
the line is the space longer.
2026-11-11 22:38:30 +01:00
Sven Lafebre bd9fef69b8 Fix #5111: keep coloured harakat attached when the first letter is spaced
paint_only ignores kerning_before, as drawn_alike does: a space before
a line's first letter, set on the letter alone, no longer splits it from
its harakat coloured apart.
2026-11-11 22:24:49 +01:00
Sven Lafebre 7844657096 Fix #5111: keep ligatures whole and place line starts in SVG by value
- drawn_alike ignores kerning_before, so a space before the "f" of "fi"
  no longer splits the ligature.
- SVG export takes a line's starting space from kerningBefore and the em
  shaping sets the character in (vectorcraft_text::line_start_space),
  not from where the glyph is drawn, which can include a shaping offset;
  the reader layout for justified lines leaves the space out, so those
  lines start there too.
- Type on a path writes the space as the first character's dx; a space
  after a line break, which <textPath> can't write, is outlined.
2026-11-11 22:09:20 +01:00
Sven Lafebre 6ae7051e4e Fix #5111: kerning at the caret before a line's first character moves it
At the Type tool's caret, kerning goes on the character before the caret.
At the start of the text there is none, so Alt+arrows, the Kerning field
and Reset Tracking and Kerning did nothing; at a later paragraph's start
the kerning landed on the invisible line break.

CharStyle::kerning_before (kerningBefore) is the space before a character
that starts a line by itself: the first character of the text, of a
paragraph, or after a forced line break. Shaping gives its cluster the
space and draws it that much later along the line (right to left the
space is at its right; in vertical type it moves down the column). The
caret at such a line's start sets it (edit::caret_kerning_target), in
type.step, type.resetTracking and the Character panel; text.setRangeStyle
takes it. SVG export starts the line where the character is drawn. The
Eyedropper and character styles leave it alone.
2026-11-11 21:43:31 +01:00
BFlat f9ff1caf54 Merge pull request #6648 from dotCooCoo/feat/dynamic-text-on-path
Type: Dynamic Text sizes type on a path to fill its brackets
2026-11-12 02:45:39 +07:00
dotCooCoo 2606b70604 Type: Dynamic Text sizes type on a path to fill its brackets
Type › Dynamic Text (`type.dynamicText`, also on the Contextual Task Bar
for type on a path) scales the text by one factor so it fills the span
between its start and end brackets, and the layout fits it again after
each edit to the text, the path or a bracket. The first character's
size moves in 0.1 pt steps within 0.1 to 1,296 pt. Turning it off, or
threading the type, keeps the sizes it was drawn at.

- The model saves `dynamicText` only when on.
- SVG, the legacy .ai writer and the .ai editing data write the drawn
  sizes through `vectorcraft_text::drawn`; the editing data warns that
  the setting is not written.
- Shrink Text to Fit and Dynamic Text step the size the Character panel
  shows when a size measures the cap height, x-height or ICF box.

Docs: docs/mcp.md, docs/type-parity.md, docs/target-app-parity.md.
Part of #4134.
2026-11-11 10:42:49 -08:00
jmvergesandClaude Opus 5.5 9f80b4daae Type: aligned lines leave the tracking after their last character out (#6610)
A centred or right-aligned line of tracked type was measured with the
tracking after its last character, so it sat half that tracking
(centred) or the whole tracking (right-aligned) left of where the page
of users' .ai files draws it. That page aligns a line as wide as its
characters without that tracking.

Shaping now records the tracking each glyph's advance holds
(SGlyph::tracking), and a line that isn't justified leaves its last
character's out of the width it is aligned by. The last character is
the last one the line draws before a paragraph's end, a forced line
break or the spaces a wrap leaves; spaces or a tab typed before a
paragraph's end or a forced line break keep the letter's tracking, as
the page does. Forced line breaks and wraps aren't in the files: they
follow the rule so that turning a wrap into a paragraph or a forced line
break keeps the line in place. Hyphenated lines, lines whose last
character hangs, vertical and right-to-left lines and type on a path
keep it. The tracking stays in the glyph's advance, so the caret,
selections, line breaking and Fit Headline are unchanged.
LineInfo::trailing gives the amount. Right-aligned area type sets the
text after its last tab the same way.

What aligns by line widths follows. Convert To Area Type makes its
frame end where the lines are aligned, with room for the tracking, so
tracked type converted either way stays in place. Affinity import lines
up differently aligned paragraphs of artistic text by the aligned
widths. SVG import moves a centred or right-aligned letter-spaced line
back by its share of the last letter spacing, so it stays where a reader
draws it and SVG export, which anchors lines at the end of their
advances, reads back unchanged.

Centred and right-aligned tracked type lays out differently than before
in every document.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-11 16:07:40 +00:00
Taha-Mahmoodi 253fb7637c Merge remote-tracking branch 'origin/main' into feat/glyph-alternates
# Conflicts:
#	docs/type-parity.md
2026-11-11 13:36:04 +04:30
Taha-Mahmoodi 422412e389 Merge main b37241f5a into glyph alternates 2026-11-11 13:15:56 +04:30
jmverges af29e7b71a Merge remote-tracking branch 'upstream/main' into fix/tracking-horizontal-scale
# Conflicts:
#	docs/file-format-parity.md
2026-11-11 08:10:00 +00:00
jmvergesandClaude Opus 5.5 df50bbf0f3 Type: tracking and manual kerning take the Horizontal Scale
In horizontal type, tracking and manual kerning are now thousandths of
the em as the Horizontal Scale draws it: at 50%, a tracking of 100 adds
0.05 em after each character. They were added at the full em while the
glyph advances took the scale, so narrow tracked type drifted from the
page of users' .ai files letter by letter. That page writes such type's
character spacing as the tracking itself in a text space as wide as the
scale, which PDF scales by it (ISO 32000-1 9.4.4). Vertical type keeps
the full em: its advances down the column don't all take the scale.

vectorcraft_text::spacing_scale gives the factor. Shaping, inline
graphics and the Optical pairs use it, and so do the conversions from
points back to thousandths of an em: the automatic kerning the Character
panel reads and Touch Type's limits (glyph_em), Combine Point Text, and
PDF import, whose tracking is now the file's Tc. Scale Text Attributes
now keeps tracked type where a horizontal stretch drew it.
SVG export writes letter-spacing in thousandths of the font size it
writes (the vertical scale), so uniformly scaled type keeps its spacing
in a viewer.

Type with both a Horizontal Scale other than 100% and tracking or manual
kerning lays out differently than before, in every document.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-11 07:16:55 +00:00
Taha-Mahmoodi f7ce2f7bcb Merge remote-tracking branch 'origin/main' into feat/glyph-alternates
# Conflicts:
#	docs/gaps.md
#	docs/target-app-parity.md
#	docs/type-parity.md
2026-11-11 11:18:10 +04:30
dotCooCoo 5a8a57d5f6 Type: upright glyphs on a vertical path keep vertical point type's cells, and the caret spans the column
On a vertical path, an upright glyph is now placed in its column as
vertical point type places it, then the column is centered on the path.

- Its cell starts at its place on the path; it started 0.12 em earlier
  with fonts whose cell top is less than a full cell above the baseline.
- A baseline shift moves it across the column instead of along the path.
- A horizontal scale widens it evenly about the column's center.
- A combining mark stays on its upright letter.
- The caret and the selection highlight span the column, half its em on
  each side of the path, instead of the font's ascent and descent.

With Skew, 3D Ribbon and Stair Step on a path that runs to the left,
letters are a quarter turn counterclockwise from upright glyphs: Skew and
3D Ribbon mirror both kinds there, and Stair Step keeps letters level. The
test, the progress doc and the params text of Tate-chu-yoko Settings now
state these conditions and the block's directions on a path.
2026-11-10 22:27:53 -08:00
dotCooCoo f8307b19cd Type: on a vertical path, letters lie a quarter turn clockwise from upright glyphs, in a column centered on the path
Follows #5557. Letters and digits lying on their side in vertical type on
a path kept their top toward the path's end, opposite the upright glyphs.
They are now set as horizontal type on the path sets them: baseline along
the path, top to the path's left, a quarter turn clockwise from upright
glyphs (CSS Writing Modes 3, Unicode UAX #50).

- The column is centered on the path: upright glyphs, letters on their
  side and tate-chu-yoko blocks have their em box centers on the path
  line instead of a column one em wide to the path's right.
- Upright glyphs are turned in the path effect's glyph space, so with
  Skew on a curve their top runs back along the tangent and their x axis
  stays up the page, as letters keep theirs.
- Tate-chu-yoko blocks on a path stand upright, stay whole on a curve and
  move by Tate-chu-yoko Settings toward the path's start and left.
- A tate-chu-yoko block of mixed sizes sets every glyph in its first
  character's em, on a path and off it.

Existing vertical type on a path redraws with these placements.
2026-11-10 21:36:30 -08:00
dotCooCoo 3487e1ad80 Type: vertical-only settings keep horizontal ligatures whole
Shaping split a cluster wherever two characters' styles differed, so
setting Standard Vertical Roman Alignment or a tate-chu-yoko offset on the
f of "fi" in horizontal type broke the ligature. Horizontal type does not
read those settings, so they no longer split a cluster there. Vertical
type still shapes such characters apart.
2026-11-10 15:17:59 -08:00
Christian Lappin 78a72a523b test(text): preserve duplicate glyph occurrence order (#6323) 2026-11-09 13:04:03 -05:00
Taha-Mahmoodi 367131d73d Merge origin/main 2026-11-09 19:43:16 +04:30
Ahmed Gomaa a3e626600c Merge remote-tracking branch 'upstream/main' into fix/arabic-mark-colour-ownership 2026-11-09 12:36:54 +02:00
Taha-Mahmoodi 2726446198 Merge origin/main 2026-11-09 11:47:43 +04:30
M Balazs PC b64cfed610 Merge remote-tracking branch 'origin/main' into fix/ai-right-aligned-tabs
# Conflicts:
#	crates/text/src/layout/flow.rs
#	docs/file-format-parity.md
#	docs/roadmap.md
#	docs/target-app-parity.md
2026-11-09 04:48:48 +01:00
dotCooCoo 6e8e1e213e Type: a pushed-in line leaves the gap after inline art open
Kinsoku push-in room does not count a gap after inline art or inside a list marker, but the squeeze also closed those gaps. The squeeze now applies only to the gaps the room counts.
2026-11-08 18:47:54 -08:00
BFlat 636ffe78ce Merge pull request #5070 from dotCooCoo/feat/kinsoku-shori-type
Paragraph: Kinsoku Shori Type chooses Push In First, Push Out First or Push Out Only
2026-11-09 09:22:52 +08:00
dotCooCoo 0ce616da4b Merge main into feat/kinsoku-shori-type
# Conflicts:
#	crates/engine/src/cmd/panelcmds.rs
#	crates/text/src/layout/breaking.rs
#	crates/ui-egui/src/panels/paragraph.rs
#	docs/target-app-parity.md
2026-11-08 16:13:35 -08:00
dotCooCoo 9a312e8ada Type: type in a composite font is measured in its own document's definition
- A file read for opening is laid out in the composite fonts registered
  then. Opening a document and reverting the front one now measure again
  the type set in a font named like a composite font that an open document
  defines, the font database holds or the session registered earlier. Type
  in other fonts, missing ones included, keeps the bounds it was read with.
- Switching to a document or closing the front one measures again the type
  set in a composite font whose definition changed, and any type without
  bounds.
- A document reverted while in the back measures its type in composite
  fonts when it comes to the front, not in the front document's.
- Text holding an inline graphic whose symbol's type was measured again
  takes the graphic's new size.
- A failed or panicked command puts the font database's composite fonts
  back with the documents.
- Setting a composite font to the definition an export lends keeps it after
  the export ends.
2026-11-08 14:23:43 -08:00
M Balazs PC b3fb745575 Merge remote-tracking branch 'origin/main' into fix/ai-right-aligned-tabs
# Conflicts:
#	crates/text/src/layout/flow.rs
2026-11-08 22:25:23 +01:00
dotCooCoo c94269ff72 Type: Composite Fonts… sets each class of character in a font of its own
Squashed from #6244 onto main.
2026-11-09 03:08:23 +07:00
M Balazs PC 7f17207d67 Merge remote-tracking branch 'origin/main' into fix/ai-type-leading-forced-breaks
# Conflicts:
#	docs/roadmap.md
#	docs/target-app-parity.md
2026-11-08 19:33:41 +01:00
dotCooCoo 8b4ef958ac Type: a ligature's font caret on a vertical path divides the proportional cell
Proportional Metrics changes an upright glyph's cell on a vertical path, and its width runs along that changed cell. Its font ligature carets were still divided by the unchanged cell, so the caret, a click and the selection missed the drawn stop. They now divide the same length the width uses.
2026-11-08 23:51:42 +07:00
dotCooCoo 4882488145 Type: upright glyphs on a vertical path face the path's start
In vertical type on a path, every glyph was turned a quarter turn with
its top toward the path's end. An upright glyph's characters then read
backwards along the path, so a ligature's font caret other than its
middle did not line up with the glyph as drawn.

An upright glyph whose top faces the path's end now turns half a turn
about the centre of its cell, so its top faces the path's start. It
keeps the same stretch of the path. Glyphs lying on their side, and
upright glyphs that the Skew, 3D Ribbon and Stair Step effects already
turn this way on a path running to the left, are unchanged. The ligature
stops follow the top of the cell, so the caret, a click and the
selection meet the drawn caret line.
2026-11-08 23:51:42 +07:00
M Balazs PC f5c3f47612 Merge remote-tracking branch 'origin/main' into fix/ai-type-leading-forced-breaks
# Conflicts:
#	crates/eps/src/import/ate.rs
#	crates/text/src/layout.rs
2026-11-08 14:57:14 +01:00
Taha-Mahmoodi ff6a4be475 Glyphs: Entire Font applies a glyph no feature reaches to the selected letter as a glyph choice; text tests for ligatures, word probe, candidate cap and aalt order 2026-11-08 15:06:29 +04:30
dotCooCoo 1d32aff63f Tests: the CSS, glow and .ai Zstandard guards count work instead of timing it
Three tests failed on a loaded machine because they asserted elapsed
time. Each now counts the work that the slow behavior it guards
against would repeat, with a counter compiled only into tests:

- The CSS cascade test counts selector tests, at most two per element
  (testing every rule against every element tests N x N). It styles
  5,000 elements, so the slow behavior fails on the count rather than
  on the hang guard, which now waits 300 s.
- The stacked-glow test counts the silhouettes drawn, 30 for 30 glows
  with and without the shadow cache.
- The Zstandard frames test counts the bytes looked at to find the
  zero padding.

The hang guards for invalid dash patterns, text layout and script
checks of a pipe wait 300 s instead of 20 s.
2026-11-08 12:21:06 +07:00
Ahmed Gomaa f3e5cfad26 Type: reuse harakat ownership shaping metadata
Reuse resolved features and borrowed GDEF classes for mark ownership,
skip the character-level pass without eligible mark glyphs, and reuse
paint-only boundary classification. Compare non-paint style attributes
without cloning while preserving the previous PartialEq behavior.

Add deterministic font variants and a field-by-field style oracle
to preserve glyph geometry, conservative ownership and NaN handling.
2026-11-08 06:55:03 +02:00
dotCooCoo dbcdbf2466 Open: the first frame uses the layouts the open made for type objects
Opening a document laid out every type object for its bounds and
baselines and threw the layouts away, and the first frame laid the same
objects out again. The open now keeps them, up to 128 MiB, and a frame
takes each one once.

A frame takes a layout only for a text object equal to the one laid out,
and only while the fonts, the system fallback setting and the hyphenation
exceptions are unchanged. When the last frame being drawn ends, the rest
are dropped if a frame took one, and all of them are dropped after eight
frames that take none. A layout with bitmap or color glyphs and a text
object with a gradient are not kept.
2026-11-08 11:42:45 +07:00
Ahmed Gomaa 11cfde2661 Type: harakat tests tell letter dots from authored marks
Noto Sans Arabic draws a letter's dots as zero-advance GDEF marks, so the tests expected a beh's dot to take its kasra's colour. Compare against the glyphs of the bare letters instead. Strict colour, geometry and kashida ownership assertions stay as they were.
2026-11-07 22:24:41 +02:00