Illustrator sets a paragraph's direction (Left-to-Right / Right-to-Left
Paragraph Direction in the Paragraph panel, with the Middle Eastern
options) and keeps its alignment buttons physical; it has no automatic
alignment toggle. So:
- ParaStyle.direction (None: from each paragraph's first strong
character, as before; written only when set), the Paragraph panel's two
direction buttons with Show Indic Options (original icons), and
text.setFormat {direction: "auto"|"leftToRight"|"rightToLeft"}. The
layout uses it as the base direction.
- The Automatic alignment check box is gone. Justify::Auto stays as the
start of the paragraph's direction, set on new type only (text.create,
the Type tools, with the mojikumi default in one new_type_para()):
TextObject::point is Left again, so importers (PDF, SVG, DXF, EMF) keep
their own alignment. The alignment buttons show what Auto resolves to.
- Bidi only where there is something right to left: plain paragraphs skip
the algorithm (no per-layout allocation), it runs once per paragraph
(shaping takes the paragraph's levels instead of recomputing them; path
type computed it four times), and the shaper keeps Han, Hiragana,
Katakana and Bopomofo in one segment (splitting at every change of
script cost Japanese text a shaper call per kana-kanji switch).
- Vertical type keeps its logical order, so its glyphs and lines are not
marked right to left (the caret no longer runs backwards down a column).
- PDF import puts Hebrew and Arabic, which PDFs (and VectorCraft's own
export) draw in visual order, back in logical order with the direction
that shows it as drawn; with bidi layout it came back mirrored.
- SVG import pins the direction SVG implies (left to right unless
direction: rtl); SVG export also outlines right-to-left paragraphs.
- Docs (development.md), ROADMAP (folded into the Type entry, Type core
row) and README updated; catalog rows for the new labels.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Unicode bidi (unicode-bidi, unicode-script, unicode-general-category):
each paragraph's direction from its first strong character, Hebrew and
Arabic shaped right to left, embedded Latin and numbers keep theirs,
lines wrapped in logical order and reordered per line by whole
clusters; source text, undo and style ranges stay logical.
- Caret, hit testing, selection and the arrow keys follow the visual
order.
- New text aligns automatically: right-to-left paragraphs right, others
left (Justify::Auto, text.setStyle {justify: "auto"}); the Paragraph
panel highlights the resolved alignment and has an Automatic alignment
toggle. Explicit alignments, and older documents', are kept.
- SVG export outlines bidirectional text (with a warning) until it can
be written as directional live text.
- A preview whose menu has gone (the panel isn't drawn: text deselected,
panel closed) or is closing (Escape, a press outside the menu) is undone
at the start of the frame, before the canvas sees the input: a canvas
drag no longer runs inside the preview's interaction and gets rolled
back with it. The menu also ends the preview in the frame it closes.
- The list fits the popup (it was taller than the popup, which then
scrolled too), so the current family is centred in view on opening;
↑/↓ scroll only as far as needed.
- Type › Font (in-window menu bar): each item keeps room for its sample
after the name, so long family names no longer run under it.
- Rows are listed again only when the query, filters, expanded families,
favourites or font list change, not every frame.
- Samples: one budget per frame across every menu (Type › Font gave each
item its own), at most 8 rendering at once, the worker asks for a
repaint when it is done instead of polling, the cache holds 512, and
one function draws a sample for both menus.
- Font menus use menu_family_list (the hidden-family rule lives there);
FontClass::label is gone (the menu's translated labels are the only
ones).
- Tests: a preview left by a menu that is gone is undone; a press
outside the open menu ends the preview in that frame.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Line-end Punctuation Half Width now also spaces Japanese from Latin
letters and digits (JLREQ 3.2.2): a quarter em of the Japanese
characters' size between a kana/kanji (or full-width letter) and a
Latin letter or digit, either way round — 雅楽 2026 年, not 雅楽2026年 —
horizontal and vertical (around sideways Latin).
- Not next to punctuation or symbols, not inside or around a
tate-chu-yoko block (it is set as one Japanese character), and none
left at the end of a line.
- Mojikumi None (older documents, imported text) is unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Japanese full-width punctuation is half a mark and half a space (an
opening bracket's space before it; a closing bracket's, comma's or full
stop's after it). Set solid, 」「 or 。」 show a full em of air between
the marks, and a line ending in 。 stops half an em short of the edge.
- `ParaStyle.mojikumi` (Paragraph panel › Mojikumi): None, or Line-end
Punctuation Half Width, which (JLREQ 3.1.4) lets consecutive
punctuation share one half-em space — a closing mark followed by
punctuation loses the space after it, an opening bracket after another
the space before it — and sets a closing mark ending a line half
width. Only full-width marks (advance an em) are touched, never inside
a tate-chu-yoko block; vertical type the same way down the column.
- New type (`text.create`, Area/Path Type) takes Line-end Punctuation
Half Width, the usual default of Japanese layout; documents from
before it and imported text keep None (written only when set, so
older files and older apps are unaffected). `text.setFormat
{mojikumi: "none"|"lineEndHalf"}`.
- Builds on #344: on justified lines the room it frees is spread
between the characters.
- Catalog rows in ja and zh-hant.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Justify (last line left, centred or right) widened only word spaces, so
a Japanese line, which has none, stayed set solid and the right edge of
a justified Japanese paragraph was ragged by up to almost a character
(only Justify All spread the characters). Following JLREQ 3.8, the room
left on a line now goes to the gaps next to a CJK character, shared
with any word spaces:
- not inside a Latin word or a tate-chu-yoko block, nor next to a tab;
- lines without CJK characters are justified as before;
- vertical type too (the same code, in line space).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Upright glyphs in vertical type advanced one em (or their horizontal
advance, if wider) and turned about a fixed point 0.38 em above the
baseline. Fonts that carry vertical metrics now place them as they ask:
- `FontFace::vertical_glyph`: a glyph's advance down the column (`vmtx`)
and the height of its vertical origin (`VORG`, else the glyph's top
plus its top side bearing, as in TrueType fonts); `None` without
`vhea`/`vmtx` or with values outside a few ems.
`FontFace::ideographic_centre`: the em box centre from an ideograph's
cell, else 0.38.
- The layout advances an upright glyph by its vertical advance and hangs
it from its vertical origin; tate-chu-yoko blocks and point type's
anchor use the font's em box centre. Fonts without vertical metrics
are laid out exactly as before.
- Hiragino's tables give the values the layout assumed (one em, origin
0.88 em), so Japanese vertical type in it renders pixel for pixel the
same; this is the base for proportional vertical metrics (`vpal`) and
for fonts whose em box sits elsewhere.
- `test_fonts::vertical_font`: Source Sans 3 with `vhea`/`vmtx` (and
optionally `VORG`) built in code, one glyph 1.4 em tall hanging from
1.1 em; tested with and without `VORG`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Character and Properties panels' font menu was a plain searchable
list of names. With hundreds of installed families (and Japanese names
that say little to someone who can't read them) a name isn't enough.
- Each row shows the family's name and a sample of the selected text set
in it (a generic sample when the font lacks its characters). Samples
are rendered off the UI thread, a few per frame, and only for the rows
in view (the list is virtualised).
- The row under the pointer or reached with ↑/↓ previews its font on the
selected text, as one live interaction with nothing in the history; a
click or Enter applies it as one step, Escape or closing the menu puts
the text back. The current family is highlighted and scrolled into
view, also when the document names it in another language.
- Filters by script (Japanese, Latin) and kind (serif/Mincho, sans
serif/Gothic, rounded, script/brush, monospaced, decorative), from the
font's OS/2 classification and, for CJK fonts whose tables seldom say,
its names. `FontTraits`/`FontClass` in the text crate; installed
families get them from the same scan as their names.
- ▸ (or →/←) lists a family's styles; a style row previews and applies
that style.
- ☆ stars a family; the ★ filter shows only the favourites (kept with
the UI state).
- Preferences › Type › Enable in-menu font previews and Font Preview
Size now turn the samples off and size the rows; the in-window Type ›
Font menu (Windows, Linux) shows a sample beside each name.
- Families whose name starts with a dot (the system's own) are hidden.
- Catalog rows for the new labels in zh-hant (complete) and ja.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The font database made one style per font file, so a variable font
such as Figtree[wght].ttf listed only its default instance (Light).
Faces now also yield their fvar named instances, each a style of the
family named by its subfamily name (the default instance is the face
itself; unnamed or duplicate instances are left out; axes and
instances read are capped). The system font scan reads the fvar table
beside the name table, so installed instances are listed before the
file is loaded and found by name and by PostScript name.
An instance face carries its axis settings: skrifa draws its outlines
and reads its metrics and advances at that location, harfrust shapes
with a ShaperInstance, so layout, the canvas, Create Outlines and hit
testing all use the instance. PDF export embeds it through krilla's
Font::new_variable, SVG @font-face through subsetter's
subset_with_variations: both write a static font of that instance.
Tests build a small variable font in code (test_fonts, via
write-fonts; no font file is committed) and check listing, resolving,
glyphs and advances, PDF and SVG export and damaged fvar/gvar tables;
Bahnschrift and Segoe UI Variable are checked where installed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
In vertical type every character now sits in its own cell down the column:
"10" and "22" in 「10月22日」 stand side by side in one em, centred on the
column, and 月/日 start right after them. Justified (Justify All) columns
no longer split a tate-chu-yoko block, and tracked or justified upright
characters stay on the column's centre line instead of drifting across it.
Root cause: in 0.3.1 (the reporter's build) upright glyphs were placed by
the font's ascent rather than the ideographic em box, so a font with a tall
ascent (Noto Sans CJK JP, 1.16 em) drew each upright character ~0.28 em
down into the next cell, onto the sideways digits; #208 (after 0.3.1)
replaced that with em-box placement and tate-chu-yoko, which this commit
locks in with regression tests. Two defects of the same kind remained:
Justify All added its inter-character gap inside a tate-chu-yoko block
(the block's second digit slid down the column, away from the first), and
upright glyphs turned about the middle of their advance, which tracking and
justification lengthen, instead of the middle of their own cell.
Tests: the issue's layout with bundled glyphs (cells one after the other,
every ink inside its cell and no overlaps, 2-3 digit blocks one em and
centred, single digits and 2026 on their side advancing by their length,
spaces), the issue's string with craft-fonts' Japanese face (skips without
craft-fonts), Justify All with a block, tracking.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Conflicts with #214 (fonts by PostScript name): this branch's name index (FaceKeys,
aliases) stays, and #214's `by_postscript_name` (used by PDF import) reads a PostScript
map filled from it, keyed by the same normalized names.
Removed: scanning ~/Library/Application Support/Adobe (CoreSync/plugins/livetype and
Fonts), and the preference that switched it (Type › Include Fonts of Other Design Apps'
Folders). Fonts activated through another vendor's subscription are licensed for use
through that vendor's own sync; reading their files from VectorCraft would go around that
licence, and the project's clean-room rules keep VectorCraft out of other vendors' files.
Fonts the user has installed in the system or user font folders are still found by every
name, as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
macOS keeps faces for its own interface under family names that start with
"." (".SF NS", ".LastResort", ".Aqua Kana"…); Mac apps don't list them. Since
the font menus list every installed family (#36), they showed 74 of them on a
stock macOS, at the top of every list.
- FontDb::menu_family_list: family_list without names that start with ".",
cached like it. Only a leading dot hides a family.
- Type › Font, the family dropdowns (Character, Properties), text.fontList and
Refresh Font List's count use it.
- Lookups by name still see every family: documents and fallbacks that name
a hidden family still find it (face, has_family, find_family, families,
text.fontList {family}, Fonts in Document's missing check).
- Test: tests_sysfonts::font_lists_leave_out_hidden_system_families_which_still_resolve
(a temporary font folder); the engine and UI font-list tests compare with
the listed families.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Shippori Mincho (assets/fonts, from #145) moves to storytold/craft-fonts. crates/text/build.rs
embeds the fonts of a craft-fonts checkout when CRAFT_FONTS_DIR is set (web: BIZ UDPGothic
Regular only) as vectorcraft_text::CRAFT_FONTS; unset, it is empty and Japanese falls back to the
installed fonts. Document text gets the Japanese faces after the bundled fonts (Mincho first); the
UI adds them at the end of every egui family (BIZ UDPGothic first), before ui_fonts' system-font
discovery. Release and FreeBSD CI builds check craft-fonts out at a pinned commit; packages carry
the fonts' OFL texts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Builds on the vertical type from #145 (same data, commands and tools).
- Kinsoku (the strict set, horizontal Japanese too): no line starts with a
closing bracket, 、。, a middle dot, colon, !?, the long vowel mark,
iteration marks or small kana; no line ends with an opening bracket.
Hyphenation breaks obey it too.
- Tate-chu-yoko: a run of two or three half-width digits (10月, 第100回)
stands upright across the column as one block one em long, squeezed to
the em when wider; longer numbers stay on their side.
- Upright or on its side by Unicode's Vertical_Orientation: CJK and
symbols upright, marks that need a vertical form on their side when the
font has none; upright glyphs advance at least one em, centred, and turn
about the ideographic em box centre so they sit on the column's centre
line. Point type keeps its anchor on the first column's centre line.
- `vrt2` is no longer requested: its glyphs are already on their side, so
with the layout turning Latin and digits too they came out upside down
(and tate-chu-yoko blocks broke) in fonts that have it, e.g. Hiragino.
- The line→text mapping is one affine (`TextLayout::line_xf`, an exact
quarter turn plus the point-type offset); `logical_point` and
`physical_point` use it. Pixel snapping skips vertical type.
- Tests: kinsoku in both directions, tate-chu-yoko blocks, upright vs
sideways, anchor on the centre line, arrows along columns.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Documents made elsewhere often name a font differently from the family
name VectorCraft lists: a localized family name (ヒラギノ角ゴシック), a
legacy family/style pair (Hiragino Sans + W3) or a PostScript name
(HiraginoSans-W3). FontDb now indexes every name in a face's name table
and resolves a request to the right face, telling the caller whether it
matched exactly, by another name, by substitution or not at all
(FontMatch).
- text: FaceKeys/aliases, FontDb::resolve, canonical, find_postscript;
fix a race where a face loaded by another thread was reported missing.
- engine: text.fonts reports status, the resolved face and missing
glyphs; Replace Font resolves the same way.
- render: substituted-font highlighting uses resolve, so a font found
by another name is no longer shown as missing.
- Find Font notes how each font was found.
- Preferences › Type › Include Fonts of Other Design Apps' Folders
(macOS, on by default).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A document names its fonts by PostScript name, which the importer split at the
first hyphen into family and style. A family with hyphens in its name was never
found: "Rounded-X-Mplus-1c-black" became family "rounded", style
"x-mplus-1c-black", and the text fell back to another font. Common Japanese
families have hyphens in their names.
The system font scan now also indexes each installed face's PostScript name
(read from the same name table it already reads). The importer asks for the face
with that exact name first (`FontDb::by_postscript_name`) and splits the name only
when no installed face has it.
Test: installed faces are found by PostScript name, in any case.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The system font scan reads collection offsets and table records with
chunks_exact; clippy 1.99 asks for as_chunks there as well.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01U9TrG4w49oSRYbP3cqxA3D
Layer and object names, document tabs and any other UI text in scripts the
UI fonts lack now draw in an installed font that covers them, as the canvas
already did.
Root cause: egui only had the bundled UI fonts (Source Sans 3, JetBrains
Mono) and its defaults, none with CJK glyphs. After each frame the UI now
looks at the text it painted; the first time a character the UI fonts lack
appears, a thread finds an installed font covering it through the font
database (the canvas's fallback search, so a font the canvas loaded is
reused, its bytes shared rather than copied) and adds it to every egui
family as the last fallback. Nothing is bundled, Latin-only UIs load no
extra font, and on the web, without system fonts, the boxes remain.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Opening, placing, pasting or importing a file, the font menus and MCP now find
the installed system fonts (Arial, Times New Roman...) in a fresh session, so a
file renders the same whatever happened earlier, and PDF import keeps the real
family names (Microsoft YaHei, not "Microsoft Ya Hei").
Root cause: FontDb only consulted the system font catalog, which stayed empty
until a CJK fallback or Character panel > Show System Fonts happened to scan
the font folders. Every lookup by family name (face, families, styles,
has_family) now waits for one cataloging scan per session: the app starts it
in the background, other processes run it on first use. The scan reads only
each file's table directory and name table (about 70 ms for 780 faces here,
instead of reading every font file whole), the catalog is indexed by family,
the family list is cached, and an installed style of a loaded family (a
bundled family also installed with more weights) loads when asked for.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With Convert Text to Outlines off (advanced.outlineText: false), type is
written as selectable, searchable text: each glyph in its font, embedded
as a subset with a ToUnicode map, where its outline would be (rotated,
scaled, skewed and on-path type too; gradients keep their place). Fonts
whose licence forbids embedding, or allows only whole fonts, fall back to
outlines with a warning; a subset threshold below 100% is accepted with a
warning. Laid-out glyphs now carry their glyph id and transform.
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>
File > Package (saved documents; unsaved changes are saved first) writes
<folder>/<name>/<document>.vectorcraft, the linked files in Links/ (relinked in the
packaged copy), the fonts its type uses in Fonts/ (leaving out fonts whose licence
doesn't allow embedding) and <document> Report.txt, then offers Show Package. Without a
folder (the web, agents) the same files come back as a zip.
Document Info's flyout picks one category (document, objects, styles, spot colours,
patterns, gradients, symbols, fonts, font details, linked and embedded images) and Save…
writes the text report Package shares.
Engine: file.package {folder?, name?, copyLinks?, linksFolder?, relink?, copyFonts?,
report?}; document.info gains category, format: text and sections. Text: FontFace
path, fs_type and embeddable. UI: ui.packageDialog (dialog `package`), file.showPackage,
docInfo.save.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
PNG is written by an in-code chunk writer with the resolution in pHYs and optional Adam7
interlacing (JPEG stores the resolution in JFIF). document.export takes ppi (72/150/300/other,
wins over scale), background (transparent, white, black or a colour) and antiAlias (none, art,
type: text snapped to whole pixels); Export for Screens rows take them too (plus one top-level
antiAlias, and a row's ppi names its @Nx suffix), and so does object.rasterize. Hard edges stay
hard in every offscreen pass (groups, masks, filters), whatever the thread count.
RenderOptions.anti_alias and render::encode::RasterExportOptions; PNG/JPEG/WebP Options dialog
(pngOptions/jpgOptions/webpOptions) after File > Export As PNG.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
File > Document Setup is a tabbed dialog over one engine command,
document.setup (no params: a query; any keys: one undo step;
document.setUnits stays as the units alias).
- General: units, bleed (four sides with a link toggle), Edit Artboards,
Show Images in Outline Mode (greyscale pixels under the frame),
Highlight Substituted Fonts/Glyphs (pink boxes on screen only).
- Transparency: grid size and colour presets or custom colours, Simulate
Colored Paper, the flattener preset (built-in or saved) and Discard
White Overprint (Overprint Preview keeps white overprints visible).
- Type: language with its quotes, Use Typographer's Quotes for the Type
tool and Smart Punctuation, superscript/subscript size and position and
small caps size (Character panel position/smallCaps; changing the
setup restyles the text in the same undo step), text export (SVG text
as text or outlines by default).
- The canvas draws a red bleed outline and the grid in the document's
colours and cell size as one textured quad, also in rotated views; a
white Background Contents exports raster formats opaque.
Model: Document.setup (DocSetup, defaulted, skipped when default) and
CharStyle position/small_caps; old files load unchanged.
tests_transparencygrid's helper now finds the grid by its texture (the
grid is one textured mesh instead of per-cell rectangles).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Shipped code no longer unwraps, expects or panics: every one of the 124
unwrap()/expect()/panic!/unreachable! sites outside tests now handles the
failure (Result/Option, ?, let-else, defaults) and reports errors to the user.
- Lints: workspace clippy denies unwrap_used, expect_used, panic, todo,
unimplemented and unreachable; clippy.toml allows them in tests, and
integration tests, examples and testkit opt out explicitly.
- Safety net (vectorcraft_engine::guard): a panic in a top-level command rolls
the document back and returns EngineError::Internal; a panicking tool is
reset; the MCP server answers with an internal error and keeps serving; the
UI frame and each control request are guarded.
- APIs that hid failures now return them: Rendered::to_png/to_jpeg/to_webp
return Result (JPEG/WebP errors used to write empty files),
FontDb::face returns Option, Region::top returns Option.
- Fuzzing: engine/tests/import_fuzz.rs feeds garbage, hostile and mutated SVG
and PDF through import → render → export (30k cases locally, no panics).
- Docs: AGENTS.md/CLAUDE.md rule, docs/development.md "Robustness" section,
README pointer and a PR template checklist.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
bflatastic
Daniel Efrat
Brandon Thomas
imachus
EthanSuquet
Hiroshi Inukai
midasdf
theironchef