Percentage cell widths size their columns from whichever rows give them, taking the largest when rows disagree, sharing a merged cell's percentage, and keeping a grid that already matches.
Originally authored by @pakiat in #5074.
The Paragraph dialog's Special box sets First line and Hanging indents, keeps By when the kind changes, leaves a mixed selection blank, and limits By to 0–22 inches.
Originally authored by @pakiat in #4922.
* #778: take craft-libs integration/wordcraft-778 at beeea924
Every craft-libs git dependency (workspace, ui-egui, wordcraft, wordcraft-web,
krilla patch) moves from 453b0da782 to beeea92451 (craft-libs main 0649a07aa
plus open PRs #2061-#2064). Cargo.lock: only craft-libs sources move, plus
windows-sys edge changes inside craft-libs crates; no crates.io versions change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #778: docx stacking-order test follows the shared reader's one-more z
craft-libs 5f453efa9 (craft-docx: a relativeHeight of 0 stays the lowest
stacking order) reads Float::z as one more than the file's relativeHeight and
writes stored orders one less, so a file's 0 is no longer "none". The test
expected the old values; relative order and the round trip are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #778: take craft-libs integration/wordcraft-778 at 24a15211
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* deps: pin craft-libs to the merged shared cutover prerequisites
* #778: pin craft-libs main at 93dac78c (with #2061-#2064)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #778: picture probes use craft-image-codecs (DIB wrapping, JPEG budget, PNG/WebP) (#6955)
* #778: picture: wrap clipboard DIBs with craft-image-codecs' bmp_probe
probe_dib's local dib_to_bmp and dib_layout are replaced by
craft_image_codecs::bmp_probe::{dib_to_bmp, dib_layout}. The shared copies are
the same code; the only change is that they reserve their output fallibly. A
failed reservation in dib_to_bmp (the BMP's bytes can't be allocated) is
Refusal::TooLarge with the size the DIB header declares, the refusal a picture
too large to hold already gets; dib_layout leaves the DIB unchanged then.
The local dib_to_bmp name stays as a thin wrapper returning Result<_, Refusal>.
The old code is kept in the tests as an oracle: wrapping and layout are
byte-equal on thousands of crafted DIBs (core, info, V4, V5 headers, bit
fields, alpha with BI_RGB, every truncation), and probe_dib agrees.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #778: picture: budget JPEG coefficients with craft-image-codecs
probe's JPEG held-coefficient estimate (jpeg_coefficients, frame_coefficients,
worst_coefficients) is replaced by craft_image_codecs::jpeg_worst_case_coefficients,
which returns the larger of that same estimate and the shared held_coefficients.
So the charge can only grow, never shrink.
It grows for: a SOF0/SOF1 8-bit first frame whose first scan has fewer
components than the frame (the decoder holds the coefficients until the other
scans); and a first-frame SOF2 with sampling factors above 4 (the old rule
clamped them to 4). A real baseline JPEG with all components in its first scan
is charged nothing, as before. With the 512 MiB budget, a 4:4:4 non-interleaved
baseline JPEG is now refused from about 59.6 Mpx instead of 67 Mpx; 4:2:0 and
4:2:2 stay under the RGBA8 conversion and are unchanged.
The old code is kept in the tests as an oracle: the new charge is never less
(crafted headers, every prefix, random damage), equal where the shared rule adds
nothing, and one test pins a case where it grows.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #778: picture: read PNG and WebP probes with craft-image-codecs
The local png_chunks header read is replaced by
craft_image_codecs::png_probe::pixel_model, and the VP8X alpha check in probe
by webp_probe::has_alpha (an error reads as no alpha, as non-RIFF input did).
PNG: for every valid colour type and bit depth, with and without tRNS and an
iCCP chunk, the size, bytes a pixel and alpha are what the old read gave
(PixelModel::bytes_per_pixel is channels * 2 for 16 bits, else channels, with a
palette as RGB and tRNS adding a channel to grey, RGB and palette). pixel_model
is stricter, so these are now refused (probe: Refusal::Broken, dimensions:
None) where the old read measured them: an invalid colour type and bit depth
pair, a zero or past-i32::MAX side, and more than 65,536 chunks before the
image data. None is a valid PNG.
WebP: a bare lossless (VP8L) picture whose header sets alpha_is_used now reports
alpha; the old rule only knew the VP8X flag.
The old PNG read and WebP rule are kept in the tests as oracles.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #778: picture tests follow craft-image-codecs' DIB fixes
The shared dib_to_bmp no longer adds 12 appended mask bytes to the pixel
offset of 52/56-byte BI_BITFIELDS headers, and dib_layout treats the bytes
after the header as a mask copy only when the payload is longer than the
pixel rows need. The differential test excludes exactly those inputs by
named predicate, asserts the new output for them, and stays byte-equal
elsewhere.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #778: dib_header_dims uses the shared fallible dib_to_bmp
Calls the shared probe directly so a failed allocation cannot recurse through the refusal wrapper.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* #778: WordCraft's MCP stdio server runs on craft-mcp-server (#6956)
The JSON-RPC framing, bounded line reading, initialize, ping, tools/* and
resources/* now come from craft_mcp_server::Server; WordCraft keeps its tools,
resources, instructions and backends. The old server.rs stays as a test-only
oracle for a differential test. click/key schemas list every argument the app
accepts, since the shared check refuses unlisted ones.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Dylan M. Taylor <dylan@dylanmtaylor.com>
* Fonts: tell where a missing font can be downloaded (Get the Missing Fonts)
A document opened from a file that is missing fonts their publisher gives
away (Microsoft's Aptos family, Cascadia Code and Mono) brings up a box
naming each font, the font drawn instead, the publisher's download page and
how to install the download. Once per document; never for a new document or
one a script opened. "Don't show this again" turns it off (File > Options >
General turns it back on); the status bar's missing-fonts menu opens it by
hand (ui.dialog fontDownload).
font.missing reports the same page as "download", so agents get it without
the box. WordCraft links to the page only: nothing is downloaded or bundled.
The desktop and web apps switch the box on at start-up
(tell_about_font_downloads); tests and ui_shot don't, so what they show
doesn't depend on the fonts installed where they run.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GBVQLv63guWCuWUM3SctPP
* Get the Missing Fonts: the box with nothing to download is one row tall
Its Close button's far-side layout took the rest of the window's height.
The button row is now a row of its own, shared by both forms of the box.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GBVQLv63guWCuWUM3SctPP
* Floating windows: a flat title bar and room around the content
Dialogs, the Read Aloud player and the built-in file browser had a grey
band behind a centred title and 6 pt margins. The title bar is now the
window's own colour with no band or line under it, and the title and the
content have 18 pt of room. The close box follows the new title bar height.
egui fills the front window's title bar with the colour of an open menu
button, so show_window makes that colour the window's while the title bar
is drawn and puts it back before the content is.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GBVQLv63guWCuWUM3SctPP
* Get the Missing Fonts: a headline, a card to each download, numbered steps
The box was a plain table and a paragraph of steps. It now opens with the
warning mark and how many fonts are missing; lists each download once
(Aptos and Aptos Display come in one) with the fonts drawn instead, the
publisher and a Download button that opens the publisher's page; and gives
the two steps that add the download, with Add Font… beside its step.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GBVQLv63guWCuWUM3SctPP
* Fonts: the download pages are a JSON list (font_downloads.json)
The list of fonts their publisher gives away, and each publisher's page,
was a table in the code. It is now crates/fonts/src/font_downloads.json,
built into the app: adding a font is a new entry there. Entries without a
family, a publisher or an https page are left out, and a list that isn't
JSON has no downloads rather than failing.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GBVQLv63guWCuWUM3SctPP
* Get the Missing Fonts: licence notice, and fonts offered with limits
The box says that fonts come with licence terms and that following the
publisher's terms is up to the user. A font whose publisher limits who may
use it is marked "restricted" in font_downloads.json, says so on its card
and in font.missing ("restricted"), and has a card of its own.
The list gains Selawik (Microsoft's open-source font) and, marked
restricted, Segoe UI Variable, Segoe Fluent Icons and Segoe MDL2 Assets
from Microsoft's Windows design downloads page. Plain Segoe UI is not
offered for download and is not matched.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GBVQLv63guWCuWUM3SctPP
* Get the Missing Fonts: the downloads scroll in a short window
With many downloads in a short window the box ran past the bottom of the
screen, leaving the steps, Add Font…, Don't show this again and Close out
of reach. The cards now scroll in the room the window has left, between
96 and 320 pt, and the rest of the box stays in view.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
* Get the Missing Fonts: all above the Close row scrolls in a short window
Only the download cards scrolled, within fixed limits, so in a window
shorter than about 480 pt the opt-out and Close could still fall off
screen. Everything above the row with Don't show this again and Close now
scrolls in the room the window has left, and that row stays on screen.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
* Floating windows: the flat title bar and wider margins leave this change
Reverts 2d39c54e and its revision-history row. The restyle changes every
dialog, the Read Aloud player and the file browser, and is to be reviewed
on its own.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
* Get the Missing Fonts: the headline counts the fonts that can be downloaded
The box lists only the missing fonts with a download, so "1 font in this
document isn't installed" was wrong for a document missing others too. It
now says "1 missing font can be downloaded", and the line under it no longer
mixes one and many.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
* Header pictures: while editing the header, a click selects its Behind Text picture wherever no header text covers it
Header and footer Behind Text pictures skip the deferred drag-over-text path
where the plain hit test picked them with no text under them, and a right-click
selects them (and sets the page they were hit on) so Reset Picture and Wrap
Text act on them. Tests cover a header and a footer, a drag, and a repeated
picture right-clicked on another page.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #5601: header Behind Text picture: a left-click selects the copy clicked on a repeated page
The press branch of pointer() selected the picture but left page_hint on the page being edited, so on a header picture repeated on page 2 the frame and handles were resolved and drawn on page 1. Set page_hint to the picked object's page, as select_at does for a right-click. Adds a left-click counterpart to the right-click test.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #5601: header text hit test checks each line's own extent
CodeRabbit: hf_text_at took the whole paragraph's right edge from its last line, so
a click on a long first line past a short last line's end counted as clear of text
and selected the Behind Text picture under it. Each line now has its own band and
extent.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #5601: pin that centred and right-aligned header lines cover their whole band
Line::left/right are the edges available to a line, not where its ink sits,
so a centred or right-aligned header line already covers the Behind Text
picture across its band (indent and floats narrow it) plus 40pt. Test pins
that, and the parity row now says what counts as covering the picture.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* #5601: pin Line.left/right as band edges in the header band test
Assert the band's left edge is the 150pt indent, the centred/right ink
starts well inside it, and a click between the indent and the ink is
text. Fails if Line.left were the ink extent.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* File › Print: opened again, the page doesn't repeat the last job's message
"Print job submitted…" (or a failure) stayed on the Print page the next
time it was opened. The job's outcome is said once in the status bar, so a
fresh open drops it; a failure arriving while the page is open stays, and
printer status (finding printers, none found) still shows.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCYPZSei56X4AGFPSu5
* Print page: the last job's marker is kept per window, since egui counts passes per viewport
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCYPZSei56X4AGFPSu5
* #5597: decide the Print page's hidden job message per window
The shared printing.message is no longer cleared on reopening Print (ui.printSettings
and other windows still report it). Each window remembers what it said (status bar or
Print page) and hides only that; a result that arrived under File's other pages is
not counted as said. Tests: failure-on-open-page now depends on reopen detection,
never-said and two-window state.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* #5597: follow the job message by generation, not by text
A retry that fails again with the same words is a new result and shows. A generation
(WordApp::print_gen) is bumped when a job starts and when the message changes; each
window's Print memo stores generations, so cross-window comparison is by identity too.
Also: update the print_memo doc comment, drop an empty branch, align the docs row date.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCXtSjB6e5jmy9Uq7u8
---------
Co-authored-by: Martin Ball <martin.ball@ucsmedical.co.uk>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Save As and Export: Browse in WordCraft when the system dialog doesn't appear (#3535) (#3801)
Ported from #3801 by VoiceLessQ onto current main.
Re-resolved against main:
- file_browser.rs tests: rebuilt whole, keeping main's #3859 tests (browser
above the dialog, Escape, dialog closing into the browser, modeless
keyboard, no keytips) and the PR's #3535 export tests.
- print_page.rs: EXPORT_FORMATS now carries main's OpenDocument Text
Template (*.ott) and Flat OpenDocument Text (*.fodt) in main's order;
main's tests module kept.
- docs/control-protocol.md: ui.fileBrowser purpose "export" added to main's
current row (keeps ui.displayServer).
- docs/ui-parity.md: File dialogs row merged with main's Linux display
server row, #3955 history row kept, status line and revision history
bumped.
Added a test that the ott and fodt export types get their extension and
write the file.
Co-Authored-By: VoiceLessQ <44067308+VoiceLessQ@users.noreply.github.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCVCCEo7u5wetHaJRU4
* #3535: Word XML export from the system dialog adds .xml to a typed name
CodeRabbit: the system Save dialog accepts any name, and the engine takes the
explicit Flat OPC format only for an .xml path, so `letter` failed to export. The
extension is added as the built-in browser does; an existing file at the new name
is refused, since the dialog never confirmed replacing it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #3535: an existing file at the .xml name is refused on the status bar
CodeRabbit: export_dialog drops ask_file's result, so the refusal was silent with a
blocking Save dialog. finish_pick now puts it on the status bar as well.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #3535: Export refusals show on the Export page; a typed document extension gives way to the chosen type's
A refused or failed export is said on the Export page (Backstage hides the status bar), as Save As does. A known document extension on the typed name is replaced by the chosen Export type's, and any other is kept with the extension added.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
* #3535: Print page shows export errors; cancel clears them; export_name edge cases
The export error is drawn inside export_formats, so the Print page shows it as well as the Export page. Cancelling an export drops an earlier error. export_name treats a name that is only a known extension (.txt) as that extension and drops a trailing dot before adding the type's.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCXtSjB6e5jmy9Uq7u8
* #3535: Export error is per window; a dotfile export name gets a real extension
The export error moves from a shared Context-data id into per-window state (swapped with the window like missing_fonts), so one window's frame outside Backstage no longer erases another's message. export_name treats a dotfile name such as .docx as a stem, so the engine's explicit-format guard (Path::extension) accepts the result; Word 2003 XML exports are named the same way.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCXtSjB6e5jmy9Uq7u8
---------
Co-authored-by: Martin Ball <martin.ball@ucsmedical.co.uk>
Co-authored-by: VoiceLessQ <44067308+VoiceLessQ@users.noreply.github.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* #3518: Review › Language can set the whole document's proofing language
review.language takes scope "selection" (default, unchanged) or "document":
every run and paragraph mark of every story (body, headers, footers, notes,
comments, text boxes, table cells, glossary) takes lang/noProof in one undo
step, selection unchanged, pending typing format follows. The Language dialog
gets an "Apply to the whole document" check box.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCVCCEo7u5wetHaJRU4
* #3518: whole-document OK sends only what was changed in the Language dialog
With "Apply to the whole document" ticked, OK sends noProof only if the user
changed it from the dialog's opening value, and lang if changed or if neither
changed (never the engine's en-US default). Heading reads "Mark the whole
document as:" while ticked. Glossary walker uses doc's PARA_WALK_DEPTH (now pub).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCVCCEo7u5wetHaJRU4
* #3518: whole-document OK with Do not check ticked marks the whole document
CodeRabbit: with the caret's text already not checked, the box was unchanged, so a
whole-document OK never sent noProof and the rest of the document stayed checked.
A ticked box is now always sent; an unticked, unchanged one is still omitted.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #3518: refuse whole-document language changes under protection
Protection leaves only its exceptions editable, so document scope is refused before anything changes. The post-check now compares the glossary, which no story-level check reached.
* #3518: whole-document language keeps per-run proofing, returns selection, skips glossary while tracking
- Language dialog: whole-document OK sends noProof only when the box was changed, so a language-only change no longer marks everything no-proof.
- review.language scope "document" returns the caret/selection result (plus scope).
- Whole-document change skips glossary blocks while Track Changes and Track Formatting are on (tracked formatting does not cover them).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #3518: whole-document Do not check is sent when the box is clicked; report skipped glossary
- Language dialog: a whole-document OK sends noProof iff the Do-not-check box was clicked (proof_touched), so ticking it with the caret in a no-proof run still marks the whole document, while a language-only change leaves other runs alone.
- review.language document scope reports "glossarySkipped": true when the glossary was skipped under Track Changes + Track Formatting; doc comment reworded to state only the limitation.
- Docs status line and revision row bumped to 2026-11-16.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Martin Ball <martin.ball@ucsmedical.co.uk>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-authored-by: Dylan M. Taylor <dylan@dylanmtaylor.com>
* #1537: Ctrl/Cmd+Shift+V pastes text only when no formatting was copied, as in Word for Windows
With formatting picked up by Ctrl/Cmd+Shift+C it still pastes the formatting.
Closes#1537, #4943; takes over #1562.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: review — a browser's Mod+Shift+V paste event pastes text only, once (CodeRabbit)
On the web the key no longer pastes as well: the paste event that follows carries the text.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: review — copied formatting stays for every Mod+Shift+V; a browser's paste after it is ignored (wanikhawar)
Copy Formatting keeps its formatting apart from the one-shot Format Painter brush, so a second
Mod+Shift+V pastes formatting again instead of text, and Mod+Shift+C no longer arms the brush.
After Mod+Shift+V pastes formatting, the browser's paste event for the same keystroke is ignored;
paste events read the modifiers held in event order. Only Mod+Shift+V itself falls back to Keep
Text Only.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCYPZSei56X4AGFPSu5
* #1537: review — one format buffer, formatting-only copy, browser paste suppression only where it exists
- ignore_next_paste armed only for Mod+Shift+V on platforms that send a paste event after it (web); never for a rebound key
- the ignored paste is dropped before equation and picture-file handling
- Event::Paste takes paste_mods so they don't leak into later pastes
- Copy Formatting and Format Painter replace each other; a used-up brush no longer hands over to older copied formatting
- copied formatting carries no hyperlink or revision marks
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: review — shared format buffer, no tracked paragraph changes, pointer press ends the paste wait
- picked_format strips ParaProps fmt_change/num_change
- Format Painter pick is stored in copied_format too; a used-up or cancelled brush leaves it
- a pointer press clears ignore_next_paste
- docs: repeat-until-replaced wording
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: review — a caret Paste Formatting keeps the target's own link and tracked marks
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: review — an unrelated key's modifiers can't steer a later paste event
A separate paste_event_mods is set by Mod changes and Mod+V's press, and cleared by any other key, a pointer press or focus change; the release path keeps paste_mods.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: review — the text-only paste test checks the destination's own formatting stays (CodeRabbit)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: review — the docs say a Format Painter pick counts as picked-up formatting too (CodeRabbit)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: review — a dropped paste uses up its modifiers; range Paste Formatting keeps the link (alehel)
- The browser paste event dropped after a formatting paste now clears paste_event_mods too, so a
later menu-bar paste isn't read as Keep Text Only. Note the no-text clipboard case.
- Test that Paste Formatting over a selection keeps the target's hyperlink.
- session.rs: copied_format holds a Format Painter pick too.
- ROADMAP: entry moved up with the other 2026-11-04 entries; stray blank line dropped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: review — the wait for Mod+Shift+V's browser paste ends after a second, so a later menu-bar paste isn't dropped (FredApps)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCX8G9hbs6XYZFsgjbL
* #1537: review — an expired paste wait drops the chord's noted Mod+Shift too (CodeRabbit)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FypgCX8G9hbs6XYZFsgjbL
* #1537: review — dropping the paired paste uses up its modifiers; range Paste Formatting keeps the target's link
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: review — restore ui-parity status line; add Paste Formatting row to typing-parity
docs/ui-parity.md: the rebase had replaced main's status line with an older one;
restore it and prepend this PR's entry. docs/typing-parity.md: add the
Ctrl/Cmd+Shift+V row to the Paste section, bump status line and history.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* #1537: Ctrl/Cmd+Shift+V's V release never pastes a second time
On desktop the chord's key-down path pastes (text, or the copied formatting) but didn't mark the
paste as done, so releasing Shift before V left a plain Mod+V release that ran Paste again through
the clipboard-picture hook, inserting the text twice. The canvas's chord now marks the paste done;
the global-shortcut path (canvas unfocused, no release seen) doesn't, and a focus change clears
the mark so a lost release can't swallow a later Mod+V picture.
Docs: web text paste needs the browser to send its paired Paste event; with formatting copied that
event is dropped only within about a second of the key.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Fixes#4030
On a small screen (an 11-inch 1366x768 panel) the system can report a very large display scale. The Auto interface size followed it blindly, so the window and ribbon came out far bigger than the screen and File > Options, where the size is set, could not be reached.
What this changes:
- Auto interface size is now capped so a window of at least 900x460 points still fits the screen (about 152% on 1366x768). Scales that already fit (for example 1920x1080 at 100-150%, or 1366x768 at up to 150%) are unchanged. If the screen size is unknown there is no cap.
- A size chosen in File > Options is still taken as is. If a large size pushes the options out of reach, Ctrl+Shift+0 (Cmd+Shift+0 on Mac) returns to Auto from anywhere. Ctrl+Alt+0 is not used because it is AltGr+0 on some layouts and would type a character.
- `ui.interfaceSize` and the "Auto (n%)" label in Options report the capped value. The cap is never saved to `ui.json`.
- Tests: the cap for several screen sizes, the shortcut, AltGr+0 still typing, and the reported percent. Docs: ui-parity, control-protocol, ROADMAP progress log.
Not covered:
- The root cause of the oversized scale (X11 DPI detection) is handled by #721 and #3240. This change is separate and works with or without them.
- A manually chosen large size can still push controls off screen, but there is now a way back.
- Not tested on a real 1366x768 screen.
`cargo xtask ci` passed locally on the pushed code; GitHub CI has not run yet.
<!-- This is an auto-generated comment: release notes by coderabbit.ai -->
## Summary by CodeRabbit
* **New Features**
* Auto interface sizing now scales down on smaller displays to keep a window of at least 900×460 points within the screen. Sizes selected in Options remain uncapped, and the minimum Auto scale is 100%.
* Press Ctrl+Shift+0 (⌘⇧0 on Mac) to restore Auto sizing. When Auto sizing is capped, the displayed zoom percentage reflects the screen-fit limit; the cap does not change the saved size.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
Register Latin Bosnian as an interface language, with catalog-backed locale, plural and font regressions. Preserve the existing system-locale detector; POSIX script-modifier handling is outside this change.
Original contribution by @nine-codes.
Review follow-up. The break after '?' is dropped only before an ASCII letter,
digit or underscore, so a '?' in CJK text keeps the break UAX #14 gives it.
The break before a slash that starts a word now also takes '/_': with the
break after '/' before '_' gone, "caption /_back" had no break left there
and split mid-word when it did not fit.
Word keeps "/portal/_back/?type=..." whole: a web address goes to the next line whole, and
a long one breaks only where it no longer fits. UAX #14 allows a break after the '?' of a
query and after a slash before an underscore, and the slash rule that keeps "and/or" whole
covered only a letter or digit after the slash.
crates/metafile is deleted. formats, layout, render and pdf take their
WMF, EMF and EMF+ reader from craft-metafile (craft-libs 9dcbd90c, the
rev craft-docx is already pinned at), so EMF+ records are read: an EMF+
only file's DrawImage of a 32bppPARGB bitmap draws its pixels.
render and pdf draw what the shared model adds: pattern and hatch fills,
clip regions and clip paths, pen caps, joins, miter limits and dashes,
turned, skewed or mirrored bitmaps, and the opaque background behind
text. The text layout and the default font for a file that names none
(Arial, a 16-unit cell) move to layout, next to metafile_text_run.
layout/src/metafile.rs holds the rules both outputs follow: a negative
crop side draws as no crop, as before; a cosmetic pen's shortest dash
or gap is one output pixel; a dash pattern whose period is under a
pixel, or that would split a path into more than 100,000 pieces, draws
solid; and an invalid miter limit is 10. On screen, a pattern tile whose
sides are both under 1.5 device pixels fills with its average color,
computed once a tile, and the vector cache counts each item's clip
region, clip list and dash pattern.
Fixes#6636
* docx: ignore w:lnNumType without w:countBy (#6209)
* DOCX: record the w:lnNumType rule in the parity docs (#6209)
A row for `w:lnNumType` (§17.6.8), its status line and a revision entry in
`docs/file-format-parity.md`, the progress-log bullet in `ROADMAP.md`, a comment in
the reader saying why a missing `w:countBy` leaves numbering off, and a save check in
the zero-interval test.