Compare commits

...
Author SHA1 Message Date
Dylan M. Taylor ba145e9405 security: enforce HTTPS on curl downloads (SonarCloud S6506)
Add --proto '=https' to curl invocations so redirects cannot downgrade to
plaintext HTTP. Resolves SonarCloud S6506 (shell/docker/githubactions).
2026-10-21 17:07:24 -04:00
Dylan M. Taylor f7d7c0288c ci: add sonar-project.properties for CI-based analysis
Marks the project ineligible for SonarCloud Automatic Analysis, resolving the
CI-vs-automatic-analysis conflict so the SonarCloud workflow succeeds.
2026-10-21 16:00:51 -04:00
Dylan M. Taylor e65d188747 ci: add SonarCloud scan via shared reusable workflow
Calls storytold/craft-common-actions/.github/workflows/sonarcloud.yml@main
with secrets: inherit. Adds SonarCloud badge to README where missing and
removes the redundant sonar-project.properties (config now passed by the
reusable workflow).
2026-10-21 15:47:09 -04:00
Dylan M. Taylor e05b7fe86f Merge pull request #2381 from dylanmtaylor/ci/shared-nix-workflow
ci: switch to shared nix workflow from craft-common-actions
2026-10-21 02:39:43 -04:00
skylar taylor-barrick e3bb06a298 Merge pull request #800 from isaac-levine/ci-xtask-ci-workflow
CI: run cargo xtask ci on pull requests and pushes to main (#754)
2026-10-20 22:11:52 -06:00
skylar taylor-barrick cd709ea0a9 Merge current main into CI workflow branch 2026-10-20 21:31:13 -06:00
Brandon ThomasandClaude Opus 5.5 d1b0bfc3f9 Skip CodeRabbit reviews on Dependabot PRs (#3302)
Adds .coderabbit.yaml so CodeRabbit no longer reviews or comments on PRs opened by dependabot[bot].

Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-20 11:39:49 -04:00
Dylan M. Taylor fd24c1ab3f Merge pull request #2935 from dylanmtaylor/docs/add-sonar-coderabbit-badges
docs: add SonarCloud and CodeRabbit badges to README
2026-10-20 11:31:10 -04:00
Brandon ThomasandClaude Opus 5.5 2e6db0bd88 Release 0.9.0
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-19 22:18:50 -04:00
Dylan M. Taylor 9b4b948c6e Merge pull request #2854 from dylanmtaylor/chore/drop-competitor-references
docs: drop references to competing commercial software
2026-10-19 19:22:22 -04:00
e6c5a701dd The caret and clicks land between the letters of a ligature (#1974) (#2006)
* Fix #1974: the caret and clicks land between the letters of a ligature

The glyph's advance is shared evenly among its letters (mirrored in
right-to-left text); ligature letters keep their character spacing.

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Fix #1974: a long word never wraps inside a ligature

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Fix #1974: a hyphenation point never splits a ligature

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #1974: ligature tests use a bundled font that draws fi and ffi as ligatures

They assumed the installed default-font stand-in draws ligatures; Liberation Sans (Linux) draws
none, so they failed there.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-19 13:09:22 -04:00
4e2568e2da RTF import: heading and code formatting compared with the file's own stylesheet (gap 19 follow-up) (#2236)
* gap 19: RTF import compares heading and code run formatting with the file's stylesheet

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gap 19: review round 1: a style chain with formatting words is complete; unnumbered entries are Normal

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-19 12:57:16 -04:00
b8debb3050 Content controls: typed text takes the control's formatting, in any language (#3116)
* Content controls: typed text takes the control's formatting, in any language

Text typed in place of a content control's placeholder takes the
control's own run properties (w:sdtPr/w:rPr, ECMA-376 §17.5.2.27), as
in Word: in a protected form filled in Word for Mac, all 17 entered
values carry the style their control's rPr names. WordCraft kept that
rPr only as XML to write back and gave typed text the placeholder's
formatting minus the placeholder style.

It also knew the placeholder style only by its English id,
PlaceholderText. German Word writes Platzhaltertext (the name stays
"Placeholder Text"), so in forms from German Word every entry stayed
grey, on screen and in the saved file, and restoring a placeholder
added a second Placeholder Text style. The style is now found by its
built-in name.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Content controls: typing drops a showing placeholder's own style

A control without run properties that shows its placeholder: text typed
over it drops the placeholder run's character style whatever its id
and name. Word localizes only the placeholder style's id, but another
program may localize its name too, and the name lookup then misses it.
Restoring a placeholder and new controls keep the name lookup.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-19 12:53:44 -04:00
f1b4be8a04 Restrict Editing: forms protection keeps a form's content controls (#3112)
* Restrict Editing: forms protection keeps a form's content controls

With Filling in forms enforced (w:edit="forms"), Word greys out the
Developer tab's control buttons and Properties, and a field's shortcut
menu offers no Remove Content Control: only the fields' content can be
filled in. WordCraft refused these only outside the fillable fields, so
with the caret in one it removed the control (its content left as
plain, no longer fillable text), inserted new controls inside it, and
renamed it; only lock changes were refused.

Under forms protection, inserting any content control, Properties
(every setting) and Remove Content Control are now refused before they
run. Typing, Toggle Check Box, Choose List Item and Set Date still fill
the fields. Other protection modes are unchanged.

Observed in Word for Mac 16.113 on a protected real-world form
(72 content controls: rich text, check boxes, date pickers); the same
form filled in WordCraft opens in Word still protected, with its values,
ticks and dates.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Restrict Editing: parity row states what forms protection refuses

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Restrict Editing: correct the comment on content-control property edits in forms mode

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-19 12:49:05 -04:00
2304bf6ac6 Fix Paragraph::sentence_at at the first character of a sentence (#3105)
* Fix Paragraph::sentence_at at the first character of a sentence

Found while testing crates/doc/src/para.rs:1397-1427

Co-Authored-By: Claude Haiku 5.5 <[email protected]>
Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* sentence_at: a terminator followed directly by text doesn't end the sentence at the caret

"Hello.World" at offset 6 gave an empty (6,6) range; only treat the caret as just past a
sentence end when the character at the caret is a space.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Claude Haiku 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-19 12:45:25 -04:00
NetterSilverandClaude Haiku 5.5 e8866ea58b Keep an ordered list's start number in Markdown import (#3103)
Found while testing crates/formats/src/markdown.rs:160-161,463 and crates/formats/src/model.rs:111 (ListInfo)

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-19 12:32:33 -04:00
NetterSilverandClaude Haiku 5.5 74ab8d2819 Keep a single leading space in HTML export paragraphs (#3102)
Found while testing crates/formats/src/html.rs (export, around line 1744; paragraph text writer)

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-19 12:32:27 -04:00
NetterSilverandClaude Haiku 5.5 517ba09250 Write large whole numbers without decimals in format_number (#3107)
Found while testing crates/doc/src/locale_number.rs:114

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-19 12:32:21 -04:00
Diego GueriniandClaude Sonnet 5.5 49a934fba7 #3067: script parse test compares paths with Path::join (#3106)
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
2026-10-19 12:29:17 -04:00
Gabriel MatozoandBrandon Thomas 117aea31a8 Navigation pane: a long heading wraps within the pane instead of ending in an ellipsis (#3015) (#3029)
The pane drew each heading with `Button::selectable(..).truncate()`, so a heading wider than the
pane ended in an ellipsis and its text had to be read from the hover tooltip. Word wraps the row
instead.

`nav_heading` now wraps: the button takes `Button::wrap()` and a minimum size of the available width
and one row, so the label continues on further lines inside the pane, the row grows with the text,
and the pane still keeps its width. The trailing grow atom still keeps the text flush left under its
parent (#1199). The hover tooltip goes with the ellipsis: nothing is hidden now, and a word wider
than the pane is broken by the wrap instead of overflowing.

The navigation test asserts the wrap instead of the ellipsis. For a heading well past the width and
for a single word wider than the pane, the label has more than one line, is not elided, stays inside
the pane, does not widen it, and takes a taller row than a one-line heading in the same pane.

`docs/ui-parity.md` and `ROADMAP.md` record the new behaviour.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-19 03:08:47 -04:00
SlavenDj fdf791a8b2 #3045: the empty window's ribbon tabs stay live, so File reaches the Start screen (#3064)
With no document open, the empty window greys the Share / Editing /
Comments buttons out by disabling the whole right-to-left line, and the
tab strip is a child of that line: egui gates interactions on a widget's
enabled state, so every tab was dead. File did nothing, and the empty
window had no way back to the Start screen, or to making a document.

Only those three buttons are disabled now. The test presses Escape on
the Start screen (the empty window, as #1129 has it), clicks the File
tab and asserts the Start screen, with its Blank document, is back.
2026-10-19 03:05:41 -04:00
MarekSpodBiedry dbc5c3407b Size the text cursor by the font, not the line spacing (#2920) (#3011) 2026-10-19 03:05:17 -04:00
MarekSpodBiedry 3f9c19d9f3 Opening a missing recent file keeps File open and shows the error (#2938) (#3012) 2026-10-19 03:05:11 -04:00
Martin BallandClaude Opus 5.5 435be0a4cb #2972: regression tests for File page AccessKit focus paths (mac chevron, Options back, Blank document) (#3019)
* #2972: regression tests for File page / sidebar AccessKit focus paths

The macOS back chevron beside the traffic lights (the report's "sidebar
collapse toggle"; there is no other toggle there), File > New/Home > Blank
document, Escape, and File > Options > back (both layouts, with a document
and from the Start screen) all run under kittest's AccessKit. The chevron
and Options tests panic with "Focused ID ... is not in the node list"
before #2856 and pass on main.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* #2972: review — fresh document per Blank document path; Options back asserts closing and focus in every case

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_012WEhZXSxzS8orX7f9J3qpr

---------

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
2026-10-19 03:05:06 -04:00
Martin BallandClaude Opus 5.5 6950c22d73 #2999: a file dialog that "cancels" at once asks again without a parent, then shows the built-in browser (#3030)
* #2999: a file dialog that "cancels" at once asks again without a parent, then shows the built-in browser

rfd folds the xdg-desktop-portal's failure response into "cancelled", so on desktops whose
FileChooser portal fails to show a dialog, Save As, Export and Open did nothing and logged
nothing. A cancel quicker than anyone can make one (300 ms) is now taken as that failure: the
dialog is asked for once more without the parent window (GridCraft, which passes none, works on
the reporter's machine), and if that too comes back at once the app shows its built-in browser.
Each step logs a warning.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_012WEhZXSxzS8orX7f9J3qpr

* #2999: review — the quick-cancel retry runs only on Linux and the BSDs; Windows and macOS answer a quick cancel as a cancel

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_012WEhZXSxzS8orX7f9J3qpr

---------

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
2026-10-19 03:04:59 -04:00
Juan Camilo Agudelo e0ff4b0c3d Paste: Ctrl+V pastes a clipboard picture even when Ctrl is released before V (#3049) (#3058)
* Paste: Ctrl+V pastes a clipboard picture even when Ctrl is released before V (#3049)

egui-winit keeps the Mod+V press, so a clipboard picture is pasted on V's
release, which required Mod still held. Letting Ctrl up first (the natural
one-hand Left Ctrl+V) pasted nothing. A V release with no V press seen was
Mod+V: use the modifiers held with Mod then, so Mod+Shift+V and Mod+Alt+V
stay apart.

Fixes #3049

* Paste: record the modifiers held with Mod in event order (#3049)

Mod can go down and up within one frame, so the frame's modifiers may no
longer have it when the canvas reads them. Take them from each
ModifiersChanged and key press as they arrive. Review feedback on #3058.

* Paste: document the Mod+V release helpers and tests (#3049)
2026-10-19 03:04:54 -04:00
SlavenDjandBrandon Thomas ebabb57e5e #3039: mini toolbar font colour Automatic returns focus to the document (#3043)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-19 03:04:49 -04:00
SlavenDj 25ad07f868 #3038: mini toolbar highlight swatch returns focus to the document (#3042) 2026-10-19 02:59:34 -04:00
SlavenDj e1e6c31571 #3040: undo-list row returns focus to the document (#3044) 2026-10-19 02:59:29 -04:00
SlavenDj 396344c31e #3031: Tabs dialog says the list is empty instead of a blank box (#3050) 2026-10-19 02:59:24 -04:00
SlavenDj 6e3f800a8b #3032: Styles pane greys out Delete for built-in styles (#3051) 2026-10-19 02:59:19 -04:00
SlavenDj 83014016c6 #3036: AutoCorrect Replace list says when the filter matches nothing (#3053) 2026-10-19 02:59:14 -04:00
570b992947 #3037: file browser says when a folder (or the filter) shows no items (#3054)
* #3037: file browser says when a folder (or the filter) shows no items

* #3037: Czech and Vietnamese catalogs carry the file browser's no-items string

* #3037: show the no-items hint only for a folder that was read

An unreadable folder already shows its error below the list.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-19 02:59:09 -04:00
SlavenDj 76d8347f77 #2892: rules end flat on screen, like the PDF export (#3061) 2026-10-19 02:44:02 -04:00
SlavenDj 19e5bb48cf #3033: file browser Open/Save uses the themed primary button (#3047) 2026-10-19 02:43:57 -04:00
SlavenDj ab968b072d #3034: SmartArt Pyramid separators follow the theme line colour (#3052) 2026-10-19 02:43:52 -04:00
SlavenDj d26b8dcae7 #3035: WordArt reflection swatch follows the theme's dimmed text colour (#3055) 2026-10-19 02:43:46 -04:00
SlavenDj 68bc118a68 Add @SlavenDj to contributors/people.toml (#3048) 2026-10-19 02:43:34 -04:00
Diego GueriniandClaude Opus 5.5 18c7c1d114 #2845 follow-up: clippy passes on Windows and Linux again (#3006)
`graphics.rs` allowed dead code only on macOS (where the Windows items are unused). Since #2845
the two indirect-validation functions are called only on macOS, so on Windows and Linux
`cargo clippy --workspace --all-targets -- -D warnings` stopped with "function is never used".
They now carry `allow(dead_code)` off macOS.

Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-19 02:43:28 -04:00
NetterSilverandClaude Haiku 5.5 60ad86acb3 Keep a single trailing space when exporting a paragraph to ODT (#3066)
Found while testing crates/formats/src/odt.rs:354-390 (plain_inlines)

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-19 02:43:22 -04:00
Fin Warman d588e0b873 Limit table row deferral to genuine widow splits (#2925) 2026-10-18 17:11:52 -04:00
ReesandBrandon Thomas 61e565cd1c fix(ui): keep recent fonts visible after an unmatched query (#2068) (#2929)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 17:10:01 -04:00
Diego GueriniandClaude Opus 5.5 2d0169e718 A Word 2003 XML or web page saved as .doc opens by what it holds (#2853) (#2942)
* #2853: Word XML, web pages and MHTML saved as .doc open by what they hold

A Word 2003 XML Document named `.doc` went to the Word 97-2003 reader, which refused it ("Invalid
CFB file"), and the window showed a blank page. Word XML (2003 or Flat OPC), a web page and a
Single File Web Page (MHTML) under a `.doc`, `.docx` or `.rtf` name now open by what they hold,
as in Word. MHTML is read for the first time (`.mht`/`.mhtml` too): the MIME parts are split and
decoded, the page is imported as HTML in its charset and its pictures load from the other parts.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #2853: leave MHTML to #1104, keep Word XML and web pages saved as .doc

#1104 adds an MHT reader (`wordcraft_formats::mht`), so this branch drops its own MHTML reader and
the `.mht`/`.mhtml` extensions instead of adding a second one. A Word XML Document or a web page
under a `.doc`, `.docx` or `.rtf` name still opens by what it holds (the #2853 fix).

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #2853: a web page's signature needs its tag name to end (`<htmlish>` is other XML)

CodeRabbit: `starts_with("<html")` took XML with a root like `<htmlish>` under a `.doc` name for a
web page, though XML that isn't a Word document stays an error. `<html` and `<!DOCTYPE html` now
need white space, `>` or `/` after them.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #2853: rustfmt

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 17:04:33 -04:00
dhruvLearner8 384d5ef307 Cross-reference: caption labels of your own are Reference types (#2940) (#2957)
* Cross-reference: caption labels of your own are Reference types (#2940)

* Caption labels come from the SEQ field's caption, not a digit (#2940, review)

* Caption label after a chapter number field (#2940, review)
2026-10-18 17:01:41 -04:00
dfgodvananddfgodvan 99ff497180 Layout: a roman page number in the footer stays roman when the body shows the page count (#2966)
A paragraph with a PAGE field is cached by the page number, the totals, the
section and the bookmarks, but not by the number format the field shows the
page number in. Paginating lays the footer out to know where the body ends,
before the section's number format is set in the field context, so as
decimal. When the body shows NUMPAGES or SECTIONPAGES, the pass that has the
right totals made that layout's key equal to the page's own footer's, and the
cached decimal one was drawn: page 2 of a section numbered in lower-case
roman showed "2", not "ii". The number format is now part of the key.

Co-authored-by: dfgodvan <[email protected]>
2026-10-18 16:55:08 -04:00
dfgodvananddfgodvan 4b18b307be Selection rectangles walk only the paragraphs the range reaches (#2967)
Document::paths_between listed every paragraph of the story (every cell
paragraph of every table) and then kept those between the two paths, so each
selection_rects call cost the whole document. The canvas calls it every frame
for each Find reading highlight (up to 5,000), each content control frame,
Read Aloud and the hyperlink under the pointer. It now walks only the
blocks, rows and cells between the two paths: for a one-paragraph range in a
document of 26,000 paragraphs (four 1,500-row tables), 1.0 ms -> 2.6 us.

Co-authored-by: dfgodvan <[email protected]>
2026-10-18 16:53:39 -04:00
dhruvLearner8 972252d0b4 Ruler: margin handles on even mirrored pages edit the visible edge's stored margin (#2914) (#2975) 2026-10-18 16:49:22 -04:00
100dfb37c8 Layout: a word after a slash wraps whole, like Word (#2963) (#2985)
* Layout: a word after a slash wraps whole, like Word (#2963)

UAX #14 (LB13) forbids a break between a space and a slash, and the break
after a slash is dropped on purpose (Word keeps "and/or" whole). Together
"caption /word" had no break opportunity at all, so a line too short for
both filled letter by letter and split the word. Add the break before a
slash that follows whitespace and starts a word.

Co-Authored-By: Claude <[email protected]>

* Layout: no break before a slash after a no-break space (#2963)

The break before a slash word took any whitespace, U+00A0, U+2007 and
U+202F included, so "Nr. /12" could wrap after the no-break space.
Only a breakable space counts now. The parity checklist lists the rule
on its own row as untested against Word.

Co-Authored-By: Claude <[email protected]>

* Layout: test the slash word in a fixed-width table cell (#2963)

The report came from a table cell, where the cell's own width is the
line width.

Co-Authored-By: Claude <[email protected]>

* Docs: gaps and typing parity for the slash-word break (#2963)

Doc wording taken from #2994 (maxm11), with the test names of this PR.

Co-Authored-By: Claude <[email protected]>

---------

Co-authored-by: Claude (ThinkSystem) <[email protected]>
Co-authored-by: Claude <[email protected]>
2026-10-18 16:41:44 -04:00
Max Murphy 00abb59a85 render: keep the legacy font effect test off antialiasing (#2898 x #2899) (#2990)
The shadow copy is solid RGB 128; the old 100-160 grey band also counted the glyph's antialiasing, whose centroid hid the copy's offset with some fonts. This was already failing on origin/main on macOS; count the copy's own grey.
2026-10-18 16:40:22 -04:00
Max Murphy 72328ab521 cli: keep the non-UTF-8 file name test off macOS (#2993)
* cli: keep the non-UTF-8 file name test off macOS

APFS refuses file names that aren't valid UTF-8, so the fixture can't be
made there (as in tests/non_utf8_args.rs); the unit test runs on the
other unixes.

* chore: add @maxm11 to About-window contributors with their own name
2026-10-18 16:37:59 -04:00
Dylan M. Taylor 652a1596aa docs: put SonarCloud and CodeRabbit badges on their own line 2026-10-18 14:47:49 -04:00
Dylan M. Taylor d82f17ac58 ci: configure Dependabot for Cargo and GitHub Actions
Co-authored-by: Dylan M. Taylor <[email protected]>
2026-10-18 12:31:59 -06:00
Dylan M. Taylor 0b3733749d docs: update feature catalog parity link and qualify OpenGL/WARP fallback 2026-10-18 12:50:08 -04:00
3af927501b Ribbon collapse review follow-ups: collapsed groups stay open, Styles kept (#1285) (#1293)
* #662: ribbon groups shrink to fit the window (compact, then one drop-down)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #1285: collapsed ribbon groups stay open while used; Styles gallery trims before it folds

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Ribbon collapse: accessible name for folded groups, Tables/Links icons (#662)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Ribbon: Esc in a collapsed-group popup doesn't leave full screen; tab-strip test ignores same-named ribbon buttons (#1285)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Ribbon tab-strip test: pick the tab, not a same-named folded group (#662)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Ribbon: a folded group's open state is keyed by its place in the row, not just its title (#1285)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* rustfmt: collapsed_group signature (#1285)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 12:03:32 -04:00
c266e5a751 Scroll the selected command into view in Customize Keyboard (#1731)
Fixes #1696

Co-authored-by: Claude Haiku 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:58:10 -04:00
e85fc75808 File (Backstage): an icon beside each side-menu item (#1714) (#1723)
* File (Backstage): an icon beside each side-menu item (#1714)

The side menu (Home, New, Open, Info, Save, Save As, Print, Export,
Close, Options) was text only. Each item now has an 18 pt icon in the
label colour, greyed out with it when no document is open. Existing
icons where they fit (house, new page, open folder, info, save, printer,
close), and three new ones drawn in code: Save As (the save disk with a
pencil), Export (a page with an arrow leaving it) and Options (a gear).
Original artwork, no asset files.

Test: every item names its own icon, and each one is a drawn icon, not
the letter-tile fallback an unknown name gets.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* docs: one Backstage row with both the Close (#1712) and side-menu icon (#1714) notes

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:57:55 -04:00
ambroandBrandon Thomas 12428df221 Italian: fixes from a native speaker review (#1717)
Went through the whole Italian catalog. Fixed the ones that were wrong (Mailings is Lettere in Word, Redo is Ripristina, cursor commands say riga) and made some terms consistent (input penna, Disponi testo, Ombreggiatura).

Also marked Italian as reviewed in the localization doc and added myself to people.toml.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:57:39 -04:00
Fin WarmanandBrandon Thomas 8b789b87d1 Keep fitting exact-height rows together (#1641)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:57:23 -04:00
528da1f6a2 Comments: a draft survives a failed commit (#1348) (#1512)
* #1348: WIP

* #1348: keep comment drafts per document, keep them on a refused edit, write them before save/close/new/open

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1348: a refused comment draft can be dropped; fields read-only where comments can't change

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1348: ui-parity: one Task panes and Context menus row after the rebase

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Docs after rebasing on main

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1348: keep the window open while a web save is unconfirmed; split the roadmap row by operation

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #1348: pending comment drafts count as unsaved changes for the browser; a save error isn't hidden by a draft refusal

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #1348: an agent can't replace the document over a pending comment draft without force

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:57:07 -04:00
Brandon ThomasandClaude Opus 5.5 b7579bfe34 ci: recipient-list test passes fields to RecipientFilter::params (#1295 x #2621)
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-18 11:56:43 -04:00
b8b1ec6141 Mail merge review follow-ups: keep the filter on failure, cap spreadsheet text (#1287) (#1295)
* #778: .xls and legacy-encoding recipient lists; filter and sort recipients

- Recipient lists: legacy Excel 97-2003 .xls (BIFF8 reader on the cfb crate already in the
  tree); Excel 5.0/95 and password-protected .xls refused with a clear message; a CSV named
  .xls reads as text, an HTML one gets an error. .xls added to the picker filters (desktop,
  web, built-in browser) and drop handling.
- Text lists (and plain-text import) detect legacy encodings when not UTF-8/UTF-16: GBK/GB18030,
  Big5, Shift_JIS, EUC-KR, Windows code pages (encoding_rs + chardetng, MIT/Apache); an
  `encoding` param overrides.
- mailings.filterRecipients / mailings.sortRecipients: field, comparison (equal, notEqual,
  less, greater, lessOrEqual, greaterOrEqual, blank, notBlank, contains, notContains), value,
  And/Or joins; sort keys. Preview, Finish & Merge and labels use only the matching records
  in sort order. Edit Recipient List edits every record and keeps the filter.
- Recipient List dialog: Filter and Sort section and a merge-order column; catalog rows;
  strings in every catalog; docs.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* i18n: Hungarian and French strings for this change

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Review fixes for #885: total order for recipient sort, Bulgarian strings

- compare_cells now ranks numbers (compared numerically) before text (compared
  case-insensitively), so it is a total order; a column mixing 9, 10 and 1a
  could make slice::sort_by panic. Regression test sorts mixed columns.
- Bulgarian catalog: the 24 Filter and Sort strings.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1287: placeholder (work in progress)

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1287: keep the filter on a failed Edit Recipients; cap .xls/.xlsx cell text; dialog drops rules on unknown fields and caches the merge order

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1287: rustfmt

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:56:19 -04:00
914e92b163 Ribbon groups shrink to fit the window (#662) (#708)
* #662: ribbon groups shrink to fit the window (compact, then one drop-down)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Ribbon collapse: accessible name for folded groups, Tables/Links icons (#662)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Ribbon tab-strip test: pick the tab, not a same-named folded group (#662)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:56:02 -04:00
8b0f1762a1 Split Cells on merged cells: divisor message and 1x1 top-row cut (#2592) (#2913)
* #2592: Split Cells on a merged cell: divisor message, 1 x 1 cuts the top row off (Word check)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #2592: Split Cells dialog shows the divisor message; docs

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #2592: clippy

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #2592: Split Cells 1 x 1 on a merged cell changes nothing (corrected Word check)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Split Cells on a merged cell follows Word: divisor message, 1 x 1 changes nothing (#2592)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:55:32 -04:00
PanyangYandBrandon Thomas de8a81c1e4 References: Word sources file import in Manage Sources (#2640) (#2897)
* References: Word sources file import in Manage Sources (#2640)

Manage Sources gains Import… and references.sources an "import": path
parameter: it reads a b:Sources document (Word's own Sources.xml) into the
document's sources with the same reader as a document's own bibliography
part. A tag already here is left alone, an untagged entry is matched by
title and author or gets a tag like a new source, the whole import is one
undo step, and a file that isn't b:Sources is refused. Desktop only; the
web build hides the button.

The docx reader now shares the b:Sources parser (crates/docx/src/
bibliography.rs) with the import, with the same dedupe and 10,000-source
cap; the hostile-input cap test still passes.

Tests: docx unit tests (names, kinds, accessed/DOI fallbacks, repeated
tags, empty entries, wrong root), engine import test (add, undo, second
import, untagged matching, refused file), ui tests (the dialog button asks
for a "sources" file; finish_pick imports), wordcraft-cli end to end
(import, cite, bibliography, save, reopen; the package's customXml
carries the sources). The button's label is in every interface catalogue.

ROADMAP progress log and docs/roadmap.md, gaps.md, file-format-parity.md
and target-app-parity.md get their rows.

* References: review fixes for the sources import (#2640)

- read_sources validates the root namespace instead of the local name alone:
  an un-namespaced or unrelated <Sources> is refused, and both bibliography
  namespaces read — OOXML's and Word's own 2004/10 (the VBA Source.XML
  format), the latter mapped onto the canonical b prefix with parse_ns.
  Tests: the 2004/10 namespace is accepted; a no-namespace or unrelated
  root is refused.
- The import loop builds the tag and (title, author-or-editor) sets once and
  keeps a per-stem counter, with the stem rule shared with auto_tag
  (auto_tag_base). Importing thousands of untagged sources no longer costs
  O(n^3). Test: untagged sources sharing a stem get Riv21, Riv211, Src, and
  a second import adds none.
- docs/target-app-parity.md and docs/file-format-parity.md no longer claim
  bibliography sources aren't saved in DOCX: #169 and #532 are closed and
  the round trip exists (#790).

* References: doc comments for the sources import helpers (#2640)

Hoist the untagged-source matching helper (by_author) to module level and
document sources, auto_tag_base and sources_ui, so the functions this PR
touches meet CodeRabbit's docstring-coverage pre-merge check.

* References: re-apply the sources import's progress notes (#2640)

The rebase onto current main took upstream's copies of ROADMAP.md,
docs/file-format-parity.md and docs/gaps.md, because the rows this PR
adds conflicted with upstream's own appends. Re-apply them on the current
text: the progress log and revision rows, the bibliography row's
source-format note, and #15's evidence, with each status line's change
note. The other two documents (docs/roadmap.md, docs/target-app-parity.md)
merged cleanly and keep their rows.

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:54:47 -04:00
2646700ca0 #2573: Picture Format › Size takes sizes in any unit, pixels too, and opens a Size dialog (#2889)
* #2573: Picture Format › Size takes sizes in any unit, pixels too, and opens a Size dialog

The picture size options were too limited (#2573, and #1901's request to enter pixels without
deforming the picture): the ribbon's W/H fields were points only, rejected "300 px", resized on
every keystroke (one undo step per digit), and there was no way to scale from the original
size, lock or unlock the aspect ratio, or reset the size.

- Picture Format › Size: W/H show the interface unit (Options › Advanced) and take a size typed
  in any unit, `px` included; it applies on Enter as one undo step, and the other side follows
  so the picture keeps its proportions. A drag still applies live.
- The Size group's launcher (and `picture.size` run without a size) opens a Size dialog: height
  and width, scale of the original size or of the size it opened with, Lock aspect ratio,
  Relative to original picture size, the original size with its pixels, and Reset.
- `picture.size {"original": true, "scale"?}` sizes from the picture's original size (its pixels
  at 96 ppi, read from the image adjustments started from, less what is cropped). A size past
  4000 pt that keeps the proportions now shrinks as a whole instead of flattening.
- New strings translated in every interface catalog.

The crop tool already exists (Picture Format › Size › Crop) and is being extended in #2027 and
#2440; options cut off in a narrow window are what #708 and #1293 (ribbon groups that shrink to
fit) address.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01JZArmeAT48gF84aKj7Zx4C

* contributors: credit @matteo-andreuzza by their GitHub profile name

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01JZArmeAT48gF84aKj7Zx4C

* #2573: Size keeps the proportions at the 1–4000 pt limits and follows the selection

Review follow-ups (CodeRabbit on #2889):

- Size dialog Reset on a photo larger than 4000 pt (a 6000 px wide one is 4500 pt) set it to
  the original size, which `picture.size` then clamped side by side: the picture was flattened.
  Reset now shrinks both sides by one factor.
- Locked, a side typed past the limit clamped the other on its own (15×30 pt with width 4000
  became 4000×4000). Both sides now scale by one factor.
- `fit_size` (engine) only shrank: a proportional size under 1 pt (scale 1% of 50×200 pt) became
  1×2 pt. It now picks the factor nearest 1 that brings both sides within 1–4000 pt, falling
  back to side-by-side clamping only for shapes too long and thin for any; the dialog uses it.
- Dialogs aren't modal: picking another picture while the Size dialog was open made OK size it
  with the first one's numbers. The form now follows the selection (starts over on the new
  object, closes when nothing is left to size).
- Doc comments on the Size form's methods.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01JZArmeAT48gF84aKj7Zx4C

* #2573: Size dialog reads the picture again after a crop, change or resize

Review follow-ups (CodeRabbit on #2889):

- The open Size form only compared the selected object's position. A crop from the ribbon,
  Change Picture or a handle drag while the dialog is open left Reset and OK working from the
  old crop and size. The form now remembers the picture, crop and size it was read from (bit for
  bit, without comparing whole objects that may hold large embedded data) and reads them again
  when any changes.
- An egui harness test drives the dialog end to end: launcher, a height typed in pixels, nothing
  applied before OK, OK sizes in proportion as one undo step and closes the dialog.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01JZArmeAT48gF84aKj7Zx4C

* #2573: size fields never resize a picture just by showing it

Review follow-up (CodeRabbit on #2889): egui's DragValue clamps a value already outside its
range when it is drawn and reports that as a change. The Size fields act on changes, so:

- Picture Format's W/H resized a picture smaller than 1 pt (a 1×1 px spacer is 0.75 pt) or
  larger than 4000 pt as soon as the tab was shown, an edit and an undo step nobody asked for.
  The code before this PR had the same problem with its 4–2000 pt range.
- The Size dialog's scale fields silently changed a picture enlarged past 10000% (a small icon
  made large), and its lengths a size outside 1–4000 pt, which OK then applied.

The range now bounds only what is typed or dragged (`clamp_existing_to_range(false)`). Both
regression tests fail without it.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01JZArmeAT48gF84aKj7Zx4C

* #2573: Shape Format and Chart Design size fields leave existing sizes alone too

Review follow-up (CodeRabbit on #2889, outside the diff): the same egui clamp as in d56a105.
Shape Format's W/H (4–2000 pt) and Chart Design's (36–2000 pt) clamped a size outside their
range as soon as the tab was shown and resized the object unasked. The new Size dialog makes this
easier to reach, since it takes shapes up to 4000 pt, but a thin line under 4 pt was already cut.
Both now show such a size as it is and act only on a new value. Shape Format's range follows
`picture.size` (1–4000 pt), so dragging a 3000 pt shape no longer snaps it back to 2000.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01JZArmeAT48gF84aKj7Zx4C

* #2573: Esc cancels a size typed in Picture Format; a typed size applies when the field is left

Review follow-up (CodeRabbit on #2889, outside the diff) asked to apply typed sizes on Enter only.
That is not done: a field keeps what was typed when it loses focus, as text fields do, and
otherwise typing a height in the Size dialog and clicking OK would throw it away. A harness test
now pins that path with a real click (press, then release frames later).

Checking it turned up a real bug: Esc did not cancel in the ribbon's W/H fields. egui 0.36 keeps
a DragValue's typed text after Esc and applies it the next frame, when the field still counts as
having just lost focus but Esc is no longer pressed. The fields now remember when Esc ended the
typing and drop that late value. (In the dialog Esc cancels the whole dialog, so nothing applies
there either; both are tested.)

Docs and comments now say "applied once typed (Enter, Tab or a click elsewhere; Esc cancels)"
instead of "applied on Enter".

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01JZArmeAT48gF84aKj7Zx4C

---------

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:54:25 -04:00
9b9eeb8c61 Fit inline text boxes with Resize shape to fit text to their text (#2912)
Fixes #2791

Co-authored-by: Claude Haiku 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:52:32 -04:00
72467d037b Keep the selection's font when typing over a backward selection (#2911)
Fixes #2879

Co-authored-by: Claude Haiku 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:52:17 -04:00
MarekSpodBiedryandBrandon Thomas ca6faa8c1d DOCX: Word 2007 WordArt (VML textpath) in the body is shown as text, not dropped (#2875) (#2903)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:52:01 -04:00
MarekSpodBiedryandBrandon Thomas 0786b2eae5 Keep the Insert SmartArt dialog within the window (#2832) (#2902)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:51:45 -04:00
MarekSpodBiedryandBrandon Thomas ae5da324ad Proofing: say when a dictionary is installed but unreadable, with file and reason (#2830) (#2900)
* Proofing: say when a dictionary is installed but unreadable, with file and reason (#2830)

* proof test: use SHIFT_JIS as the unsupported encoding (ISO 8859-1 is decoded since #2904)

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:51:13 -04:00
MarekSpodBiedryandBrandon Thomas 69302481ad Decode ISO 8859-1 Hunspell dictionaries so the original pt_BR works (#2824) (#2904)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:50:59 -04:00
03a992a46a File page: the back arrow no longer ends WordCraft when accessibility is on (#2856)
* File page: the back arrow no longer ends WordCraft when accessibility is on

A click outside the page hands the keyboard back to it at once
(refocus_after_chrome_click), by the page's widget id from the last frame
that drew it. When the click was the File page's back arrow, the File page
was closed in the same pass, so the page had not been drawn in it: egui kept
focus on a widget missing from that pass, and with AccessKit running (a
screen reader, or any UI Automation client on Windows) AccessKit's tree
rejected the update ("Focused ID ... is not in the node list") and the
program ended without a word. 0.8.0 on Windows vanished after Blank
document, File, then the back arrow.

The page's id is now kept with the pass that drew it, and focus is requested
by id only in that pass; otherwise want_focus hands the keyboard over in the
next frame, as it already did. The same path covered File > New > Blank
document with a document open.

Test: back_arrow_returns_to_the_page_with_accessibility_on clicks the arrow
in the harness (which always runs AccessKit); it panicked with AccessKit's
message before the change.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* control input: widget_id now carries its pass (match the pair)

---------

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:50:41 -04:00
9ce4553899 fix(macos): skip wgpu indirect validation on Intel Macs so launch no longer exits (#2652) (#2845)
wgpu compiles its indirect-call validation compute shaders inside request_device.
On Intel HD Graphics 5000 (Big Sur) Metal's Intel compiler aborts on them, the
device is lost and eframe exits before the window opens. WordCraft issues no
indirect draws or dispatches, so x86_64 macOS builds leave
VALIDATION_INDIRECT_CALL out of the instance flags (WGPU_VALIDATION_INDIRECT_CALL=1
still re-enables it). Same fix as storytold/PhotoCraft#3171.

Fixes storytold/WordCraft#2652

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:50:25 -04:00
Fin WarmanandBrandon Thomas fad83fd876 419: count shared row border clearance (#2789)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:50:03 -04:00
GavinQiErandBrandon Thomas 8bde3b3c25 fix: dropping an object near the top of the next page no longer sends it back (#2829)
move_object anchored the object to the paragraph under the drop point and
computed the offset from that paragraph origin. With a Square-wrapped object
the surrounding text reflows once the object moves away, so the first
paragraph of the target page - the anchor when dropping near its top -
slides back to the previous page and the object follows it.

Re-check the page after placing an object across pages and re-anchor when the
placement did not land on the target page (up to three attempts). Same-page
moves still place once, so resize and nudge are unchanged.

Adds a kittest that presses, moves and releases onto the top of the next page
and asserts the object stays on the page under the pointer.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:49:47 -04:00
GavinQiErandBrandon Thomas fe4faca3f0 fix: do not leave a dragged object on its page (#2720)
While a floating picture, shape or text box is dragged, the page raster is
not repainted (the texture cache key does not include the drag state) and a
translucent preview is pasted at the pointer. The user therefore sees the
original copy at full opacity plus the preview, i.e. two copies of the same
object.

Exclude only the dragged object from that page's display pass, and repaint
that page once when the drag starts and once when it ends; the preview keeps
using the previous texture. The rest of the page is left untouched (no
full-page clear, no per-frame repaint).

Adds three tests: a dragged inline object and a dragged floating shape are
left out of their page, and the view key changes when a drag takes an object
off the page.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:49:30 -04:00
Fin WarmanandBrandon Thomas 596176c794 371: clear header tables for floating body tables (#2660)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:49:00 -04:00
c91c1793a3 #2108: File › Print scales pages to another paper size (#2699)
* #2108: File › Print scales pages to another paper size

Scale to paper size on File › Print (Word: Pages per sheet › Scale to
paper size): No Scaling or any Layout › Size paper (Letter, Legal, A4, …).
Every page is scaled to the paper and centred, the paper turned to the
page's orientation, in the preview, the print job and Save these pages
as PDF. Native jobs also ask the spooler for that media. The document's
page size is unchanged. New `paper` param on ui.print/ui.printSubmit and
file.printPdf; unknown papers are errors.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* pdf: bookmark destinations from #2364 use the scaled Fit too

---------

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:48:17 -04:00
00cd3f3368 Fix stray leading space in RTF list items on export (#2667)
Found while testing crates/formats/src/rtf.rs:269-272 (Writer::para)

Co-authored-by: Claude Haiku 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:48:02 -04:00
Debarghaya MitraandBrandon Thomas cff4cd35b8 Page setup: resolve section correctly when focus is in a header, footer or note (#2072) (#2663)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:47:46 -04:00
Fin WarmanandBrandon Thomas 5909ab642a T372: wrap body around page-relative footer tables (#2659)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:47:30 -04:00
Fin WarmanandBrandon Thomas 60617d9a71 371: count shared section pages in fields (#2654)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:47:03 -04:00
MarekSpodBiedryandBrandon Thomas e7cd8679ad Keep the ODT paragraph 'Keep with next' option on import and export (#2813) (#2825)
* Keep the ODT paragraph 'Keep with next' option on import and export (#2813)

* file-format-parity: keep with next on the ODT row

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:46:48 -04:00
MarekSpodBiedryandBrandon Thomas 4495e2be53 Use the Default Paragraph Style font when importing ODT (#2804) (#2823)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:46:32 -04:00
833f5b64d9 Keep explicit text:s spaces at paragraph ends on ODT import (#2641)
Found while testing crates/formats/src/odt.rs:1480 (finish_para calls p.trim()); crates/formats/src/model.rs:188 (Para::trim)

Co-authored-by: Claude Haiku 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:46:15 -04:00
866b48ac55 Recipient list: find box and merged-only view (#1576) (#2621)
* #1576: Recipient list search and filtered view

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #1576: drive the recipient grid through egui_kittest

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #1576: recipient search treats final sigma as sigma

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #1576: removing a row lets go of the focused cell

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:45:59 -04:00
6e30349bbb #1327: protection checks cover tables, Paste Formatting and recipient refresh (#2618)
* #1327: protection checks cover tables, Paste Formatting, template update and recipient refresh

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #1327: record any content below the protection walk's depth limit

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #1327: formatting-only protection also refuses table properties and row height

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:45:40 -04:00
8c812bf1d6 #1230: Insert Caption on pictures and shapes; Different First Page in Page Setup (#2615)
* #1230: Insert Caption on pictures and shapes; Page Setup header and footer options

- Right-clicking a picture, shape, text box or chart offers Insert Caption…, as in Word.
- references.caption with a picture or shape selected puts the caption next to the
  paragraph holding it, in the same cell or text box (it used to go after the whole table).
- The Caption dialog proposes Figure for a picture selected in a table, not Table.
- Page Setup: Headers and footers check boxes (Different Odd & Even Pages, Different First
  Page), applied on OK only if changed, in one undo step; a title page can go unnumbered.
- Tests: engine caption_for_an_object_in_a_cell_stays_in_the_cell; kittest
  right_click_on_a_picture_offers_insert_caption and
  page_setup_different_first_page_leaves_the_title_page_unnumbered.
- Docs: ui-parity, target-app-parity, ROADMAP progress log.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_01P8veLveuUsJJm6UEcQPEuW

* ui-parity: keep Insert Caption on the context-menu row

---------

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:45:24 -04:00
dec24458cf #1273: equations from Word files keep all their content (#2613)
- Math straight in the body or a cell (m:oMathPara/m:oMath without w:p,
  allowed by EG_BlockLevelElts) gets its own paragraph instead of vanishing.
- mc:AlternateContent Requires="m" picks the equation, not its fallback.
- Inside an equation, runs in w:hyperlink, w:fldSimple, w:dir, w:bdo and
  mc:AlternateContent are kept (EG_PContentMath); depth stays bounded.
- w:sym, w:noBreakHyphen and Symbol-font text in math runs become Unicode.
- Layout: n-ary operators, delimiters, accents and group characters the math
  font lacks use a fallback font instead of a missing-glyph box.
- Coverage test reads every ECMA-376 §22.1 element and checks no argument's
  text is dropped; tests for body-level math, Requires="m", symbols, hostile
  nesting and the font fallback. Parity docs and progress log updated.


Claude-Session: https://claude.ai/code/session_01P8veLveuUsJJm6UEcQPEuW

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:45:01 -04:00
Fin WarmanandBrandon Thomas 473df388f3 Keep the line in an empty section starting a page (#2609)
* 371: retain the line in an empty page section

* 371: resolve hidden section marks

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:44:45 -04:00
beedd1b30d Drop Cap Options dialog (#2601) (#2602)
* #2601: Drop Cap Options dialog

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* gaps: keep the Drop Cap Options note on the dialog-depth row

* i18n: translate the two Drop Cap disabled reasons in the audited catalogs

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:44:30 -04:00
87cab84e4a #1086: ribbon tooltips show why a control is disabled (#2623)
Hover hints give the command's own reason, cover split-button arrows, menu buttons and the font/size/outline boxes, and show on keyboard focus.

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:44:12 -04:00
5fdd0a0e89 Watermarks in header table cells drawn behind the text (#1775) (#2580)
A floating picture or VML shape in a header table cell is placed on the page behind the text.

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:43:55 -04:00
Fin WarmanandBrandon Thomas a7abd346e7 layout: preserve subpoint table grid widths (#2589)
* 356: preserve subpoint table grid widths

* 356: normalize tiny positive grid totals

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:43:27 -04:00
bdb7672c0f Dragging the scrollbar names the page at the top of the window (#2551) (#2575)
* #2551: dragging the scrollbar names the page at the top of the window

While the canvas's vertical scrollbar is dragged, a tip beside the pointer
says which page is at the top of the window, in the status bar's words
("Page 7 of 10"), so a drag through a long document shows where it will land.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #2551: the scrollbar tip names the page whose visible band is at the top, also with Hide White Space

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:43:09 -04:00
Fin WarmanandBrandon Thomas b5759bcd99 351: reserve footer flow-table border clearance (#2585)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:42:54 -04:00
Fin WarmanandBrandon Thomas 5b211d6dc5 345: ignore explicit page breaks in table cells (#2574)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:42:39 -04:00
6dd7c3bc3a #1963: wordcraft-cli runs commands from a script file (#2566)
Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:42:13 -04:00
ReesandBrandon Thomas 1005c59bfc fix(ui): keep note editing text readable (#2545) (#2562)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:41:58 -04:00
11f8881cd2 #537: test that System pins the title bar to the desktop's light/dark answer on Linux (#2559)
Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:41:42 -04:00
Marco ArnulfoandBrandon Thomas 913621c1c4 Font dialog: Kerning for fonts and its Points and above size (#2295)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:41:26 -04:00
divyanshu2074andBrandon Thomas 8c328386fa fix(formats): enforce tag boundary when skipping raw text elements in HTML (#2257)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:41:10 -04:00
63ee34c557 #2105: Highlight No Color sticks to the button (#2208)
No Color removed the highlight from the selection, but the Highlight
split button kept the colour picked before, so the next click on it
highlighted again. As in Word, No Color now becomes the button's choice
and the next click removes highlighting. The ribbon and the mini toolbar
share one highlight menu (the ribbon's copy is gone); No Color is
translated in the mini toolbar and no longer shows Ctrl+Alt+H beside it
in the ribbon.

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:40:55 -04:00
divyanshu2074andBrandon Thomas e04dd0e785 fix(formats): support cell and paragraph borders in RTF export (#2207)
* fix(formats): support cell and paragraph borders in RTF export

* rustfmt

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:40:31 -04:00
divyanshu2074andBrandon Thomas 9cd34ebc80 fix(formats): preserve tilde characters in LaTeX URLs and raw arguments (#2202)
* fix(formats): preserve tilde characters in LaTeX URLs and raw arguments

* rustfmt

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:40:15 -04:00
divyanshu2074andBrandon Thomas d9ca2573a6 fix(engine): count multi-column spanning cells once in table formula ranges (#2200)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:40:00 -04:00
divyanshu2074andBrandon Thomas a114a79a69 fix(doc, formats): support OEM character set 255 in charset_code_page (#2195)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:39:37 -04:00
Antoine BoxhoandBrandon Thomas 61a626820e #1552: no spelling squiggle under the word being typed (#2183)
Word draws no red squiggle under the word being typed. It appears when the word ends or the
caret leaves it by a key or a click. The session records the word after a text command that
edits the caret's paragraph, with apostrophes and hyphens counted as the spelling check counts
them, and drops it once the caret is anywhere else. The layout leaves that one range out of the
spelling squiggles, and a caret move without an edit lays out only that paragraph again. F7 and
the Spelling pane still report the word, and grammar marks are unchanged.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:39:21 -04:00
divyanshu2074andBrandon Thomas 545673d644 fix(formats): resolve sub-location and external URLs in RTF HYPERLINK fields (#2181)
* fix(formats): resolve sub-location and external URLs in RTF HYPERLINK fields

* rustfmt

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:39:05 -04:00
divyanshu2074andBrandon Thomas 973e6e12f5 fix(formats): support quoted and non-latin charset declarations in HTML decode (#2180)
* fix(formats): support quoted and non-latin charset declarations in HTML decode

* html: clippy (question_mark, collapsible_if) after merging main

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:38:50 -04:00
94e3dda6af SmartArt: type into items; open the Text Pane by double-click or frame button (#2170)
* SmartArt: typing with the graphic selected types into an item

Typing with a SmartArt graphic selected replaced the whole graphic with
the typed text. Now the text goes at the end of the current item (the
first before any is picked) and the Text Pane opens, as in Word. The
typing is one undo step.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* SmartArt: double-click or the frame button opens the Text Pane

A double-click on a SmartArt graphic opens its Text Pane. While a
graphic is selected, a Text Pane button sits on the left edge of its
frame, as in Word, and shows or hides the pane.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* SmartArt: the Text Pane closes with the graphic's selection

It opened by itself on the next graphic selected, because the pane flag
outlived the selection. The frame button hides while the pane is open.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

---------

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:38:34 -04:00
divyanshu2074andBrandon Thomas 3031033790 fix(layout): support quoted bookmark names with spaces in PAGEREF fields (#2169)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:38:19 -04:00
95b5c71f80 Save writes the chosen file when its folder refuses a temporary file (macOS privacy protection) (#2138) (#2167)
* #2138: save writes the chosen file when its folder refuses a temporary file (macOS privacy protection)

On macOS a Save-panel pick inside a protected folder (Desktop, Documents…) grants
access to that file only, so the sibling .wordcraft-save-*.tmp was likely refused
with EPERM. On PermissionDenied creating the temporary file, write the destination
directly; a remaining permission error on macOS says where to allow access.
Untested on a Mac: the fallback path is pinned on Unix with a read-only folder.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* #2138: a failed save in a folder refusing the temporary file keeps the original

Review (FredApps): the in-place fallback truncated the destination before
writing, so a failure partway through lost the previous document.

The fallback now keeps the existing file's bytes in memory (up to 512 MB)
before writing it in place and writes them back if the save fails, or
removes a new file it created. If that write-back fails too, the error
says the file may be incomplete. A file over the limit isn't written in
place; the error suggests Save As.

Tests: the reviewer's partial-write test (read-only folder), a new file
removed on failure, and the size limit.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_01P8veLveuUsJJm6UEcQPEuW

* #2138: keep the original untouched when no safe replacement is possible

When the folder refuses the sibling temporary file, an existing destination
is no longer overwritten in place (no in-memory rollback): it stays
byte-identical and the error says to use Save As to another folder or allow
access to the folder (macOS: Privacy & Security > Files and Folders). A
destination that doesn't exist yet is created exclusively and removed again
if the write fails.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_01P8veLveuUsJJm6UEcQPEuW

---------

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:38:04 -04:00
divyanshu2074andBrandon Thomas 6cf7159a71 fix(formats): emit \landscape in RTF export when section is landscape (#2159)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:37:42 -04:00
divyanshu2074andBrandon Thomas 65127c2727 fix(formats): handle negative timezone offsets and flexible day digits in spreadsheet iso_datetime (#2154)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:37:27 -04:00
874e5a76a0 #2052: a second picture style, border or artistic effect replaces the first (#2148)
Picture Styles, Picture Border and Artistic Effects now draw again from the
bitmap under them (kept per drawn picture in Session::picture_looks), so
Thick Frame then Simple Frame is only a simple frame, a border replaces the
border and keeps the style, a style replaces the border, and an effect
replaces the effect ("none" removes it), as in Word. Transparency was already
absolute (#1175); the test pins it.

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:37:12 -04:00
divyanshu2074andBrandon Thomas 720f8acd2c fix(layout): support field format switches without whitespace after \* (#2147)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:36:58 -04:00
2054f8fd88 fix(web): fall back after asynchronous save picker rejection (#2145)
Preserve AbortError cancellation and report download fallback failures without marking saves successful.

Fixes #1246

Assisted-by: OpenAI Codex

Co-authored-by: Codex <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:36:42 -04:00
4c8508cd93 Insert › Shapes menu: an icon for each shape and connector (#2142)
Each item shows a small drawn outline of what it inserts (rectangle,
oval, triangle, star, elbow connector...). The icons are drawn in code in
icons.rs as shape.<kind>; widgets::icon_menu_item draws the item.

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:36:27 -04:00
divyanshu2074andBrandon Thomas e7babf364e fix(formats): trim whitespace when parsing numeric XML attributes in ODT (#2137)
* fix(formats): trim whitespace when parsing numeric XML attributes in ODT

* rustfmt

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:36:12 -04:00
divyanshu2074andBrandon Thomas c9468a13cb fix(docx): support self-closing w:compat in settings merge (#2131)
* fix(docx): support self-closing w:compat in settings merge

* Drop the duplicate divyanshu2074 entry in people.toml (already on main)

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:35:43 -04:00
divyanshu2074andBrandon Thomas e2d9224996 fix(formats): support relative font size values in HTML font tags (#2130)
* fix(formats): support relative font size values in HTML font tags

* Drop the duplicate divyanshu2074 entry in people.toml (already on main)

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:35:28 -04:00
divyanshu2074andBrandon Thomas eedf9c9740 fix(doc): strip negative sign when formatting numbers rounded to zero (#2122)
* fix(doc): strip negative sign when formatting numbers rounded to zero

* Drop the duplicate divyanshu2074 entry in people.toml (already on main)

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:35:13 -04:00
divyanshu2074andBrandon Thomas 5977112a0d render, pdf: fix vertical double and wave border rendering (#2113)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:34:58 -04:00
872b2bc99d DOCX: ink inside groups and drawing canvases survives saving (gap 2) (#2097)
* gap: nested ink WIP

* gap 2: native ink nested in groups and drawing canvases survives DOCX saves

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gap 2: review round 1: ungrouping keeps nested ink; no orphan parts

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gap 2: Impact cell matches the restated impact

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* gap 2: only Word ink pins a group against Ungroup

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* gap 2: alternate ink must hold an a:graphic to be written as standalone

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* gap: nested ink WIP

* gap 2: native ink nested in groups and drawing canvases survives DOCX saves

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gap 2: review round 1: ungrouping keeps nested ink; no orphan parts

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gap 2: Impact cell matches the restated impact

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* gap 2: only Word ink pins a group against Ungroup

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* gap 2: alternate ink must hold an a:graphic to be written as standalone

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:34:42 -04:00
Antoine BoxhoandBrandon Thomas aed0b71786 #1878: Rulers thin out with the zoom and span the caret's page (#2093)
Ruler numbers now thin out as the page shrinks on screen: a number
every 1, 2, 4, 8... units, whichever keeps them at least 30 px apart,
and the small ticks drop out below 4 px, with the text at its usual
size. A number gives way to an indent, tab or column marker sitting on
it, and the margins get ticks only. Both rulers use the caret page's
rect and scale in every layout, including One Page and Multiple Pages,
and follow the caret to another page. Tests cover the density rule, the
marks, the marker rule for indents, tab stops (set or dragged) and
table columns and rows, and the multi-page geometry.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:34:26 -04:00
Antoine BoxhoandBrandon Thomas 0f5e945a69 #1945: Hide the Quick Access Toolbar and show it again (#2089)
The Quick Access Toolbar's menu ends with Hide Quick Access Toolbar, as Word's does. Show Quick Access Toolbar brings it back from the ribbon's right-click menu and from a check box on Customize Ribbon's Quick Access Toolbar page. The setting is kept between runs, and keytips skip the toolbar's digit badges while it is hidden. The UI command ui.quickAccessToolbar drives it from scripts and MCP. Regenerates docs/parity-checklist.md, which was behind the catalog. Adds my contributor entry.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:34:10 -04:00
e50850d4dd RTF: formatting and paragraph-mark revisions are read and written (gap 19) (#2088)
* gap: RTF formatting revisions WIP

* gap 19: RTF formatting and paragraph-mark revisions

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gap 19: review round 1: list kept on Reject; old formatting normalised; section marks; hostile borders

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gap 19: review round 2: \oldcprops holds only the changed properties, as Word writes it

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gap 19: old formatting in styled paragraphs uses the style's size and font

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gap 19: RTF import keeps main's heading and code normalisation

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gap 19: file-format-parity status line keeps main's history

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* gap 19: a paragraph that was a list item gets its old list back on Reject

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* gap 19: a numbered item made bulleted keeps its numbered list for Reject

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* gap 19: format matrix names the tracked section mark

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* rtf test: Writer has the section fields main added; rustfmt

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:33:46 -04:00
Marco ArnulfoandBrandon Thomas 166b95feba Proofing: a hyphenated word added to the dictionary is no longer flagged (fixes #2082) (#2085)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:33:30 -04:00
072acce71b Legacy form fields are drawn: check boxes, and drop-downs' selected entry (#2076)
* Legacy form fields are drawn: check boxes, and drop-downs' selected entry

Legacy form fields keep their w:ffData since the last round, but were
not drawn: a FORMCHECKBOX has an empty result, so a form of check
boxes came out with none, and a FORMDROPDOWN showed its saved result
rather than its selected entry.

- wordcraft-doc: `legacy_form` reads what layout needs from the kept
  w:ffData (kind, name, check box state and size, drop-down entries and
  selection, text field length); `InlineObject::legacy_form()`.
- wordcraft-layout: a check box takes the text line's height of room
  and is drawn as Word draws it; a drop-down shows its selected entry
  (w:result, else w:default, else the first).

Measured on Word's PDF export with a synthetic document (Calibri):

  text size | Word square | WordCraft
  8 pt      | 7.44        | 7.45
  10 pt     | 9.96        | 9.89
  12 pt     | 12.36       | 12.33
  20 pt     | 22.08       | 22.09
  w:size 15 | 16.08       | 15.99

The square sits 1.16 pt inside the line height on every side (room on
the line within 0.2 pt of Word's), has a 0.72 pt border and, checked,
an X from corner to corner in 0.48 pt lines; w:checked beats
w:default. docs/file-format-parity.md is updated.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Draw::Shape takes the dash list main added

---------

Co-authored-by: Daniele Mori <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:33:15 -04:00
70fc00c80f Symbols keep their symbol font when the text font changes (#2036) (#2037)
* Symbols: a w:sym character keeps its symbol font when the text font changes (#2036)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #2036: review round 3: docx test reads sym_font, paste drops the symbol font, astral symbols stay text

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* #2036: insert.symbol decides symbol or text per character

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
2026-10-18 11:33:00 -04:00
20188e4610 A floating object's anchor takes no room on a line (#2033)
A line's first item stays on it even when it is wider than the line;
markers before it don't count. A floating object's anchor (a zero-size
object cluster) did count, so a picture a little wider than the line,
after an anchor, went to a second line and left the anchor an empty
line of its own above it. Word keeps the picture on the first line.
Anchors are now ignored like markers when deciding whether the line
already has content.

Co-authored-by: Daniele Mori <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:32:22 -04:00
1b45615940 Insert › Shapes keeps the selected shape and text (#2019) (#2030)
Inserting a second shape removed the first: after an insert the new
shape stays selected, and `insert.shape` replaced the selection, so each
shape replaced the one before. Word draws a new shape beside whatever is
selected and never deletes it (checked by hand in Word for Windows: a
second shape is drawn over the first, which stays).

The new shape now goes after the selection, which is left alone, and
becomes the selection. Selected text stays too. A picture still replaces
a selected picture, which is what Word does as well (checked in Word for
Windows), so `insert.picture` is unchanged.

Tests: three shapes inserted in a row are all kept, which fails with the
old code (1 of 3 kept); a shape inserted over selected text keeps the
text and is selected itself; a picture inserted over a selected picture
replaces it.

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 11:32:07 -04:00
Dylan M. Taylor ac76375d74 docs: add SonarCloud and CodeRabbit badges to README 2026-10-18 11:01:34 -04:00
Dylan M. Taylor f4c109a0ab docs: clarify that feature percentage is measured against tracked features in docs/gaps.md 2026-10-18 10:44:06 -04:00
Dylan M. Taylor 715f491d29 docs: drop references to competing commercial software 2026-10-18 10:43:43 -04:00
Brandon ThomasandClaude Opus 5.5 7333a302f5 ci: File Close test expects the window kept (#1712); Polish coverage test skips French's newer strings
#2448's imported-Word close test predates #1996 (File › Close keeps the window), and #2492's French catalog carries keys the Polish catalog lacks, as the PR's own test change said.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-18 10:37:08 -04:00
c9a5191b70 i18n: French interface catalog completed (2,615 entries) with native-speaker review (#2492)
Adds the 402 entries the French catalog lacked: equation gallery (matrices, accents, brackets, integrals, operators, functions), built-in style and table-style names, citation styles, paper sizes, watermarks, and the Bookmark, Go To and Print dialog strings. Drops the obsolete 'Style: APA' entry. Terminology reviewed by a native French speaker and Word user. The Polish superset test exempts fr like the other catalogs carrying newer strings. Progress docs and people.toml updated.

Co-authored-by: Vivien JANNY <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:29:17 -04:00
492aceb960 Fonts Word embeds as subsets lay out the text set in them (#2240) (#2540)
* #2240: fonts Word embeds as subsets lay out the text set in them

Embedded fonts were registered under the family inside the font file. Word
embeds a subset ("embed only the characters used") under a made-up family
(___WRD_EMBED_SUB_46) and the text asks for it by the font table's name
(OBRBWW+Century Schoolbook), so it still fell back to a substitute. Each
embedded file now comes with its font table name, and its faces go by that
name when they don't already.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #2240: the black-circle fallback test passes its font's table name

main's new `circle_checks_face_coverage` test (#2582) called `add_document_fonts` with bare font
bytes; with this branch each embedded font comes with the name the document's font table gives
it, so the test passes "Times New Roman" alongside the face it renames to it.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:28:52 -04:00
ReesandBrandon Thomas ff39a5c357 fix(ui): deliver control input in inactive windows (#2420) (#2530)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:28:37 -04:00
ReesandBrandon Thomas 1595343fa0 fix(ui): expose save confirmation context to screen readers (#2489)
* fix(ui): expose save confirmation context (#1015)

* #1015: require alert-dialog semantics in save regression

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:28:20 -04:00
Fin WarmanandBrandon Thomas a5d817a9ef TOC: keep direct number tabs on refresh (task 276) (#2475)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:27:26 -04:00
09b822fdbe Update Field after save and reopen of a table of contents; no extra blank line per cycle (#2182) (#2553)
Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:27:11 -04:00
Bryan MarkandClaude Opus 5.5 8107581df2 Opening a DOCX no longer copies its pictures (#2441)
Each picture, and each embedded object or macro part kept for saving, was
copied out of the unzipped package (bytes.to_vec()), so a document with
pictures held them twice while it opened, and the package's copies were
left behind as freed memory. Package now holds its parts as Arc<Vec<u8>>,
the type the document keeps media and passthrough parts in, and hands them
over with get_shared. keep_unmodelled also searches the main part in place
instead of copying it.

A 92 MB document with 20 pictures on macOS 27: the app's footprint once
open 866 -> 772 MB (freed blocks kept 206 -> 118 MB, peak 915-932 ->
821-823 MB); wordcraft-cli info peak 319 -> 231 MB.

Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 10:26:56 -04:00
Fin WarmanandBrandon Thomas ca45f97ace Layout: exclude floating anchors from line font metrics (task 256) (#2462)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:26:32 -04:00
ReesandBrandon Thomas 9726d6f1cb fix(ui): offer Save As for imported Word files (#2448)
* SCRUM-47: offer Save As for imported Word files

* #2445: reject Save As answers for a replaced document

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:26:16 -04:00
Fin WarmanandBrandon Thomas 57dfacfc07 Layout: preserve equation baselines under exact line spacing (#2435)
* Layout: preserve equation baselines under exact line spacing (task 243)

* layout-parity: keep the cell-margin inheritance note on the Tables row

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:26:01 -04:00
Fin WarmanandBrandon Thomas 4cb4b2d231 244: inherit omitted cell margins (#2434)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:25:38 -04:00
PanyangYandBrandon Thomas a14948a23d DOCX: strict line heights with a unit are points under auto (#2329) (#2439)
Word 2024 saves Strict Open XML with measures written out with their unit:
the issue's file carries `w:spacing w:after="8pt" w:line="13.90pt"
w:lineRule="auto"` in its paragraph defaults. The reader sent `13.90`
straight into the `auto` branch, where a bare number counts 240ths of a
line (240 = single), so 13.90pt became a 13.9 multiple: 191.8 pt lines,
and the issue's document laid out as 10 pages against Word's 1.

Under `lineRule="auto"` a unit-suffixed height now reads through twips, as
Word renders it (ISO 29500 `ST_SignedTwipsMeasure`: 13.90 pt = 278 twips =
the 1.1583 multiple). A bare number still counts 240ths of a line;
`exact`/`atLeast` already read units as points.

Checked with the issue's file: 1 page with baselines on Word's (15.98 pt
line advance against Word's 16.10/15.96; paragraph advance 23.99 against
24.00; table lines 12.65 against 12.60/12.72), and a save/reopen keeps it.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:25:14 -04:00
Fin WarmanandBrandon Thomas 4deb654862 Layout: keep narrow cell line widths (task 245) (#2436)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:24:58 -04:00
Fin WarmanandBrandon Thomas d7bd5bcd3c Layout: preserve additive spacing after section breaks (task 243) (#2433)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:24:41 -04:00
28d6260a81 #1789: Add Font imports every selected font file (#2404)
* Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1789: cap font bytes waiting in the web inbox

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:24:26 -04:00
Fin WarmanandBrandon Thomas a492189a84 DOCX: preserve zero effect extents on rotated drawings (#2432)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:23:59 -04:00
5c4408f341 Paste: plain text keeps the formatting of the paragraph it lands in (#2423) (#2424)
Paste of plain text from another program gave every line after the first
a paragraph with default formatting. The text became a fragment of
unformatted paragraphs (Fragment::from_text), and insert_fragment keeps
the target's formatting only in the first one. Pasting several lines into
a right-to-left paragraph turned all but the first left-to-right, and the
text after the caret lost its paragraph formatting too.

Paste now inserts plain text as Keep Text Only does. Each line after the
first splits the paragraph as Enter does, so it keeps the paragraph's
direction, style, alignment, indents, spacing and list. Empty lines become
empty paragraphs with that formatting. Copies from WordCraft and HTML or
RTF from other programs keep their own formatting as before. A paste is
still one undo step.

Tests cover a right-to-left paragraph, empty lines and a final line break,
and a list. The typing and UI parity docs and the ROADMAP progress log
have rows for the change.

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:23:43 -04:00
Marco ArnulfoandBrandon Thomas 8f0000de63 Text boxes: framed paragraphs run with the box's text, not drawn over each other (part of #2357) (#2383)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:23:27 -04:00
Marco ArnulfoandBrandon Thomas 9bed0156df OpenType ligatures follow w14:ligatures: model, DOCX read and write, shaping (#2351)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:23:12 -04:00
Marco ArnulfoandBrandon Thomas f7de98a711 DOCX: a w:u without w:val draws no underline (fixes #2362) (#2380)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 10:22:57 -04:00
d6f20f93fd Tables: at-least rows taller than a page start on a fresh page; header rows follow (#1742) (#2004)
* WIP for #1742

* Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1742: parity doc separates the Word-observed case from synthetic-test behaviours

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 09:57:47 -04:00
3348e93140 ASK and SET fields show nothing (#1765) (#1785)
* #1765: WIP

* #1765: ASK and SET fields show nothing

Like Word, ASK and SET only assign text to a bookmark: they draw no
result on the page or in a PDF. The stored result is kept and saved.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* file-format-parity: note ASK/SET on the existing Fields row instead of a third one

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 09:57:31 -04:00
11f1d0aee6 Lowercase list items no longer get a grammar underline (#1741) (#1783)
* #1741: WIP

* #1741: list items no longer get the sentence-capital grammar squiggle

The English "Capitalize the first word of a sentence" rule skips the first
word of a list paragraph (own or style numbering); later sentences are still
checked. check_text_with and paragraph_issues take a list_item flag.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1741: review fixes

Thread the capital-first flag into the Polish, Croatian, Italian, Turkish and
Portuguese grammar rules; test the drawn squiggles through layout() for own,
style and num 0 numbering; tidy the proof test.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1741: main's Vietnamese proofing test passes the new list-item argument

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* #1741: the third call in main's Vietnamese proofing test

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 09:57:15 -04:00
b8279c4da5 #1960: add editable table cell margins dialog (#1988)
Co-authored-by: justinjohn0306 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 09:56:48 -04:00
SlavenDjandBrandon Thomas 2caa9db109 #1903: undo-history footer Cancel closes the popup (#1916)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 09:56:34 -04:00
Fin WarmanandBrandon Thomas 4dcd4749ca Layout: align behind-text cell pictures (#1871) (#1871)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 09:56:19 -04:00
d4a3292834 Proofing: -ically adverbs such as "specifically" are no longer flagged (#1694) (#1854)
* #1694: -ically adverbs are no longer flagged as misspelled

English spelling accepts "-ally" adverbs built on an "-ic" adjective in the
word list (specific → specifically, scientific → scientifically), as Word does.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* #1694: only -ic adjectives take -ally

Review (FredApps): the "-ally" → stem rule applied to every known word,
so "desktopally" was accepted. The rule is now "-ically" → "-ic"
(specifically → specific); other words don't take "-ally".

Test: non_ic_words_do_not_take_ally (desktop, house, final, green).

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_01P8veLveuUsJJm6UEcQPEuW

---------

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 09:56:04 -04:00
7f4b649f6f RTF export: sections are saved (gap 19) (#1776)
* rtf-sect-export: WIP

* Gap 19: RTF export writes sections

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Gap 19: RTF export writes sections — review fixes

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* odt-sect-export: WIP

* Gap 19: ODT export writes sections

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Gap 19: ODT export writes sections — review fixes

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* gaps log: RTF and ODT section export both done

* Gap 19: ODT continued lists share the earlier list's style; one RTF and ODT row in the parity table

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 09:55:42 -04:00
0c8399896f cli: open a password-protected Word file with --password (#1770)
info, text, convert, render and inspect took no password, though file.open
understood one, so an agent driving the CLI could not open a protected
document at all. --password, --password-file and WORDCRAFT_PASSWORD, in
that order.

The file form exists because a password on a command line is readable by
every other process on the machine; a trailing newline is dropped so a
file written by echo works. Never logged, never named in an error.

Rebased onto current main, which moved `open` from &str to &Path so that a
file name that isn't valid Unicode no longer crashes the CLI (#1440). Both
changes touch the same function, so the resolution keeps their shape: the
closure that substitutes the real path still stands, and it takes the
password as a second argument.

Co-authored-by: Bra1nZ <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 09:55:26 -04:00
69c2414615 Fonts: bold Arial Black is not emboldened again, as in Word (#1750)
Old releases of heavy faces (the core-fonts Arial Black) claim weight 400 in
OS/2, so bold text in them was drawn with synthetic bold. Current releases say
900 and Word draws bold Arial Black unchanged. A static face whose file says less
than semibold now takes the weight its family's last word names: Black/Heavy 900,
ExtraBold/UltraBold 800. Files that already declare a heavy weight keep it.

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 09:55:11 -04:00
Diego GueriniandClaude Opus 5.5 879e688fec File › Close closes the document and keeps the window (#1712) (#1996)
* File › Close closes the document and keeps the window (#1712)

File › Close (Mod+W, Mod+F4) quit the whole app. In Word it closes the
document and the window stays. WordCraft already has that state from
the Start screen (#1129): no page, the ribbon greyed out but for Help,
File › New and Open to go on.

`WordApp::execute` now handles `file.close` itself. After the usual
Save / Don't Save / Cancel, the document is replaced by the placeholder
blank document that marks "no document open" and the Backstage closes.
The window's close button passes `{"quit": true}`, so it still asks and
then quits. The parameter travels with the prompt, so a Cancel leaves no
state behind. With no document open, Close has nothing to close.

Tests: the Close tests that used `quit_requested` as "Close went through"
now check that the document closed and the window stayed. Closing from
the empty window asks nothing, File › New goes on from there, and the
window's close button still quits after Don't Save.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Close keeps the window for programmatic closes too (#1712, #1391)

#1391 landed on main meanwhile: a programmatic `file.close` on unsaved
work needs `{"force": true}`. Its tests, and the save prompt's focus
test, took "Close went through" to mean `quit_requested`. With #1712 a
close, forced or not, closes the document and keeps the window, so they
now check that the document closed and the app is still running. The
`force` guard itself is unchanged.

docs/control-protocol.md: a closed document leaves the window open, and
`file.new` or `file.open` goes on from there.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1712: a direct file.close never discards unsaved changes unless forced; Don't Save forces it

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 09:38:16 -04:00
.CooCoo f5b3c5d069 Contributors: dotCooCoo's entry shows the GitHub profile name again (#1982)
The merge of main into #1250 (ea0d27a6) resolved its conflict in
contributors/people.toml without the sync_github_name = true line that
#1250 added to [people.dotCooCoo]. The line is back, so the credits
show the name on the GitHub profile as this entry's display name
(crates/ui-egui/build.rs).
2026-10-18 09:38:11 -04:00
Ian Holloway f8ffab72e2 Tests: keep font-cache misses independent of optional Thai fonts (#1981)
Assisted-by: GPT-6

Signed-off-by: IanHollow <[email protected]>
2026-10-18 09:38:06 -04:00
NetterSilverandClaude Haiku 5.5 0b1c957233 Escape block syntax after manual line breaks in Markdown export (#1980)
Found while testing crates/formats/src/markdown.rs:1222

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-18 09:38:01 -04:00
Wade 999e2ae896 PDF tests read ActualText spans, so RTL extraction no longer depends on the installed Arabic font (#1773) (#1973) 2026-10-18 09:37:56 -04:00
.CooCoo 438dec0d61 Fonts: the font scan leaves out faces with no outline or color bitmap tables (#1972)
The system font scan (file_face_names) also catalogs faces with none of
the tables skrifa reads glyph outlines from (glyf, CFF, CFF2, VARC) and
none of the color bitmap tables the renderer draws (sbix, CBDT). On
macOS 27.0.1 the scan of /System/Library/Fonts catalogs "GB18030 Bitmap"
(Supplemental/NISC18030.ttf), and the font list shows it. The face
stores its glyphs as bdat bitmaps and maps 27,544 characters.
FontDb::outline returns an empty path for each of them, and the face has
no color glyph tables. The renderer draws none of them.

file_face_names now leaves out a face whose table directory has none of
those six tables. It reuses OUTLINE_TABLES from the interface CJK
fallback (#1250). VectorCraft's font scan checks the four outline tables
(storytold/vectorcraft#781). WordCraft's renderer also draws sbix and
CBDT bitmaps (#1504). Noto Color Emoji has CBDT and none of the outline
tables, and the scan keeps listing it.
2026-10-18 09:37:50 -04:00
1cea0f7e11 Equations: a script runs until a space or operator, as in Word (V_dd, x_2a) (#1609) (#1951)
* WIP for #1609

* Equations: a script after _ or ^ is the whole run of letters and digits (#1609)

Word for Mac 16.113.4: V_dd and V_(dd) build up the same (dd is all
subscript), likewise x^ab, and Linear format shows V_dd bare.

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* Equations: bare scripts end at a space, digits continue a script run, upright function names (#1609)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* Equations: a bare script also ends before a following period (#1609 review)

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* ROADMAP: drop three progress rows main already has

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 09:37:37 -04:00
NetterSilverandClaude Haiku 5.5 d04d756c4e Grey out and guard ribbon split buttons when disabled (#1918)
Fixes #1908

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-18 09:37:31 -04:00
SlavenDj 85c7d1214b #1895: recovery page buttons keep theme text colors (#1898) 2026-10-18 09:37:27 -04:00
SlavenDj 5e751c848b #1873: Backstage back chevron uses plain inset in macOS full screen (#1879) 2026-10-18 09:37:23 -04:00
Bra1nZandBra1nZ 2707ac403f mcp: let open_document take a password (#1796)
The engine has read passwords since the support landed — file.open pulls
`password` out of its parameters and hands it to open_path_in. The UI asks
and retries, the CLI has --password. The MCP tool never passed it on, so an
agent driving the server or the app could not open a protected document by
any route.

Advertise it in the schema, keep it optional, forward it.

Without a password the parameter goes over as null, not "": file.open reads
an empty string as "no password given", so collapsing the two would open a
protected document with an empty password instead of failing.

Three tests. The first I wrote expecting file.open as the backend method
and it failed — exec wraps everything as engine.execute with the command
under params. The test was wrong about the code.

13 lib + 5 integration tests, clippy -D warnings and fmt clean.

Co-authored-by: Bra1nZ <[email protected]>
2026-10-18 09:37:17 -04:00
3bede06646 Restrict Editing: exceptions don't open page setup or styles; tracked-changes protection keeps revisions (#1881)
With read-only (or comments) protection and an editable exception, the check after a
command only compared text and paragraph formatting outside the exceptions, so page setup
(section properties) and style definitions could change for the whole document. They are
now compared too (styles a command adds are allowed). Protected for tracked changes,
Accept/Reject and the Track Changes toggle are refused, as in Word.

Refs #1327

Co-authored-by: dfgodvan <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 09:37:12 -04:00
da607a00ee ODT export: sections are saved (gap 19) (#1777)
* odt-sect-export: WIP

* Gap 19: ODT export writes sections

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Gap 19: ODT export writes sections — review fixes

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* ODT export: a list cut by a section continues with text:continue-list; drop duplicate parity rows

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 09:37:06 -04:00
JoaoAugusto00andBrandon Thomas 4a251d4b9c docs: how to install spelling dictionaries (pt_BR example) (#2510)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:57:46 -04:00
Marco ArnulfoandBrandon Thomas 4362e6624d Word's Spike: Ctrl+F3 collects cuts, Ctrl+Shift+F3 inserts them (#2507)
* Spike: Ctrl+F3 collects cuts, Ctrl+Shift+F3 inserts them and empties it (gap #24)

* Search test: a lone INSERT's witness within the box's 14 results; Table checked past it

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:57:27 -04:00
Fin WarmanandBrandon Thomas c92d120098 Honor table row keep-next during pagination (#2494)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:57:07 -04:00
86ad3ddb63 DOCX: two lines in one and horizontal in vertical (w:eastAsianLayout) survive a save (#2486)
Read w:eastAsianLayout (ISO/IEC 29500-1 §17.3.2.10) into a new
CharProps::east_asian_layout and write it back after w:lang, in runs and
styles. Values outside the schema are left unset. Layout is unchanged.

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:56:48 -04:00
Fin WarmanandBrandon Thomas fc08b0e4f6 Fix PDF outline targets after leading page breaks (#2484)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:56:29 -04:00
Fin WarmanandBrandon Thomas 455e1f419f Include table-cell headings in regenerated TOCs (#2482)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:56:10 -04:00
Fin WarmanandBrandon Thomas 56d0a162b3 Share AutoFit donor reductions equally (#2464)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:55:37 -04:00
3a2cea6ca4 #1843: Read Aloud speaks Russian text with a Russian voice (#2430)
Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:55:18 -04:00
NetterSilverandClaude Haiku 5.5 5a82eafa5a Refresh DATE/TIME fields when opening .doc/.dot by extension (#2418)
Found while testing crates/engine/src/io.rs:183-187 (read_bytes, the "doc" | "dot" arm returns before opened() is called)

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-18 04:55:00 -04:00
divyanshu2074 1880f6162c fix(formats): support html entities, wbr zero-width break, and page-break-after (#2400) 2026-10-18 04:54:50 -04:00
Marco ArnulfoandBrandon Thomas 61ad110e9f Layout: automatic space between Asian text and Latin text or digits (autoSpaceDE, autoSpaceDN) (#2358) (#2397)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:54:39 -04:00
fa755e9de4 Fonts: restore missing strokes in hint-reliant Chinese fonts (#2389)
Co-authored-by: xyz9516 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:54:20 -04:00
Marco ArnulfoandBrandon Thomas d2321d05d9 DOCX: positional tabs (w:ptab) keep their alignment, base and leader and go to the margins or indents, as in Word (#2367) (#2387)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:54:01 -04:00
7c95f02050 #1743: Ctrl+wheel zoom no longer scrolls the document (#2386)
Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:53:42 -04:00
2db55b5498 fix(smart_art): normalize layout and color identifiers and preserve item text with leading whitespace (#2365)
* fix(smart_art): normalize layout and color identifiers and preserve item text with leading whitespace

* Use Iterator::find (clippy manual_find)

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:45:17 -04:00
341f370368 fix(pdf): scale bookmark destination coordinates with page Fit and support case-insensitive bookmark lookup (#2364)
* fix(pdf): scale bookmark destination coordinates with page Fit and support case-insensitive bookmark lookup

* cargo fmt

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:45:06 -04:00
divyanshu2074 09f14d5a5e fix(math): parse array column specs, support starred operatorname and tag, preserve delim sep in latex (#2353) 2026-10-18 04:44:55 -04:00
NetterSilverandClaude Haiku 5.5 088156d4ca Keep spaces that follow a tab in the ODT round trip (#2317)
Found while testing crates/formats/src/odt.rs:347 (writer, `plain_inlines`) and crates/formats/src/odt.rs:1550 (reader, `text`)

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-18 04:44:44 -04:00
divyanshu2074 e1b39cd41d fix(docbin): correct sprmTCellWidth operand parsing and handle percentage widths (#2312) 2026-10-18 04:44:34 -04:00
Marco Arnulfo d32119c377 Copying a picture or shape replaces the clipboard's old text, so Paste duplicates it (#2306) 2026-10-18 04:44:23 -04:00
6b3de1c4dc fix(docbin): handle default codepage, VT_I4, multiple property sets, and FILETIME dates (#2304)
* fix(docbin): handle default codepage, VT_I4, multiple property sets, and FILETIME dates

* cargo fmt

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:44:13 -04:00
Marco ArnulfoandBrandon Thomas 5e2b63b710 Contributors: add Marco Arnulfo (#2302)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:44:03 -04:00
a3aab66bfa fix(formats): support none and dual scripts in mmultiscripts, mpadded phantoms, menclose borders, and mtable column align (#2300)
* fix(formats): support none and dual scripts in mmultiscripts, mpadded phantoms, menclose borders, and mtable column align

* cargo fmt

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Collapse the nested if (clippy collapsible_if)

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:43:44 -04:00
8fcaa66e19 Turkish case mapping: i/İ and ı/I capitalise correctly (#1923) (#2009)
* WIP for #1923

* AutoCorrect: Turkish and Azerbaijani case mapping (#1923)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* AutoCorrect: Turkic language detection shares proofing's tag normalisation, I to dotless i covered (#1923)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:43:32 -04:00
559f16601d A mouse drag on a table no longer scrolls the view to the caret (#1738) (#2008)
* Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

* Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:43:13 -04:00
0254f3afc7 Hebrew: field results like STYLEREF draw right to left; smart quotes match Word (#1670) (#2007)
* Field results (STYLEREF, REF, PAGE text, note marks) shown as one atomic
cluster now place their inner clusters in UAX #9 visual order, in the
direction of the text around the field, so "פרק א" in a header no longer
draws reversed. Smart quotes already matched Word's logical-order result
in Hebrew (” in acronyms, not gershayim); pinned with a typing test.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1670: fix round 1: field results shaped per bidi run, so brackets mirror correctly

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:42:54 -04:00
da8b7b7feb Tables: Backspace removes a selected table, rows or columns; Delete clears them (part of #1844) (#2005)
* WIP for #1844

* Tables: Backspace removes a selected table, rows or columns; Delete clears them (#1844)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* Tables: delete_range edge-case tests, single-cell Backspace clears, deep tables kept (#1844)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:42:35 -04:00
3787af0a67 Rendering and PDF check image limits before allocating (#1321) (#2001)
* WIP for #1321

* Images: check size budgets before decoding and cap page rasters (#1321)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* Images: charge decode peak, valid-CRC header tests, cap text-line rasters (#1321)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:42:17 -04:00
c9f3da2fc5 DOCX: mixed Hebrew/Latin keeps its order in Word (runs split by direction, w:rtl and language) (#1543) (#1950)
* WIP for #1543

* DOCX: split mixed-direction runs and mark right-to-left pieces w:rtl so Word orders them as WordCraft does (#1543)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* DOCX: right-to-left pieces of split runs say their language (w:lang w:bidi) (#1543)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* DOCX: no phantom tracked formatting on split runs; typed text doesn't inherit run direction (#1543)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* Engine: text typed or inserted anywhere takes its direction from its own letters, not the caret's run (#1543)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:41:59 -04:00
82bf806118 Web: input-method rewrites erase before retyping, so non-Latin typing no longer repeats (part of #1480) (#1949)
* WIP for #1480

* Web: an input method's DeleteSurrounding erases around the caret, so non-Latin typing no longer repeats (#1480)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* Web: input-method rewrites keep formatting and the typing undo group; AutoCorrect resets the web text field (#1480)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:41:39 -04:00
NetterSilverandClaude Haiku 5.5 6b192f58ce Clamp negative margins before checking they leave text room (#1941)
Found while testing crates/engine/src/cmd/page.rs:392 (margins) and :424 (margins_fit)

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-18 04:41:20 -04:00
eff1444f08 Fonts: bounded family caches; math follows font changes (#1354) (#1929)
* WIP for #1354

* Fonts: bound the family caches and refresh math resolution on font changes (#1354)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* Fonts: test math font lookup against a really installed and removed font (#1354)

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:41:10 -04:00
SlavenDj 228a3eae3e #1905: empty keys list shows (none) (#1919) 2026-10-18 04:26:52 -04:00
SlavenDj db9d9d488a #1904: bookmark dialog placeholder when no bookmarks (#1917) 2026-10-18 04:26:41 -04:00
SlavenDj af90833305 #1902: clipboard and styles pane actions return focus to the document (#1914) 2026-10-18 04:26:30 -04:00
SlavenDjandBrandon Thomas 31cb2c8714 #1861: searchable Interface Theme command (View, light/dark/system + toggle) (#1868)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:26:19 -04:00
bcdad07d00 macOS: dialogs put Cancel left of the default button, and ⌘. cancels (#1835)
Mac order, [Cancel] [OK] with the default button rightmost, in every
dialog with OK and Cancel; Cmd-period cancels wherever Escape does.



Claude-Session: https://claude.ai/code/session_01JTqaRSzgjKv75TzSC55HEY

Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:25:59 -04:00
SlavenDj 0c4558cd08 #1906: content-control Properties OK inherits theme text (#1920) 2026-10-18 04:25:04 -04:00
6d83d13257 Update Field works in headers, footers, notes and text boxes; printing updates every field (#1706) (#1780)
* #1706: WIP

* #1706: Update Fields refreshes fields in headers, footers and text boxes

Co-Authored-By: Claude Opus 5.5 <[email protected]>

# Conflicts:
#	ROADMAP.md

# Conflicts:
#	ROADMAP.md

* #1706: review fixes

Revision-history row, update_fields doc comment, test covers body field and locked header field.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1706: update the fields of the part being edited, as in Word

Word check: F9 with the caret in the body leaves header fields alone; with the caret in a header it updates that header. SEQ, citations and TOC run only from the body.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1706: printing updates every field in the document

references.updateFields takes {"all"?: bool}: the body (with SEQ, citations, TOC) plus every part story, wherever the caret is. Printing passes it.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1706: ROADMAP status line after rebase

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1706: tests build the field form value added on main

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1706: the test's fields take main's new dirty flag

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:24:53 -04:00
c5a9c954f7 Track Changes Options: Moves (move tracking, Moved from/to marks and colours) (gap 5) (#1779)
* track-move-options: WIP

* track-move-options: Moves options in Track Changes Options and w:doNotTrackMoves

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* track-move-options: Moved from Hidden leaves the layout; per-revision hiding

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* track-move-options: Word's mark lists, (none) moves are plain, no borders on collapsed paragraphs

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

* track-move-options: hr and sv translations for Moved from:/Moved to:

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:24:34 -04:00
1f0aeb4626 Ink to Shape: two-stroke arrows and closed-triangle heads (gap 12) (#1778)
* ink-arrows-2: WIP

* ink-arrows-2: Ink to Shape makes two-stroke arrows and closed-triangle heads

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* ink-arrows-2: review fixes — recognise each stroke once, drawing-order pairing, arrow rect

Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:24:15 -04:00
mbaksa 211d6c0c06 UI: Font dropdown list height doesn't update when filter results increase (#1732)
* Fix searchable combo box height when filter results increase

* Clean up unnecessary blank lines
2026-10-18 04:24:05 -04:00
9acc34f22c i18n(pl): citation style names, audit coverage and attribution (#750) (#1720)
* i18n(pl): citation style names, audit coverage and attribution (#750)

Follow-up to #871.
- pl.tsv: add the 12 citation style names (APA, Chicago, GOST, ISO 690,
  ...) that hu.tsv and bg.tsv already translate.
- tests.rs: Polish also checked against the citation styles; Italian and
  Hungarian may differ only by the obsolete "Style: APA".
- audit.rs: Polish joins the catalogs that must cover every looked-up string.
- ATTRIBUTION.md: credit Aleksander Kutycki (@olekkut) with Gemini 3.8 Flash for pl.tsv.
- contributors/people.toml: add [people.olekkut] entry.

Co-Authored-By: Gemini 3.8 Flash <[email protected]>

* i18n(pl): one-line equation gallery pieces (Large, Limit and, Log) so Polish matches Italian

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Gemini 3.8 Flash <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:23:54 -04:00
Hamza Naveed c2b97f1180 DOCX: read every VML shape in a w:pict, and a v:group's members (#1662)
* DOCX: read every VML shape in a `w:pict`, and a `v:group`'s members (#1643)

`read_vml` took `pict.els().find(..)`, the first `v:*` child that is not a
`v:shapetype`, and read only that one. So a `w:pict` holding several shapes
kept the first and dropped the rest, and a `v:group` was read as a single
shape -- or, with no text box in it, as nothing at all -- losing every member.
Both are gone after a save, since saving writes the model back as DrawingML.

Read all of them instead:

- `read_vml_all` walks every `v:*` child of a `w:pict` and emits one object
  each; `read_vml` keeps returning the first for the two places that show a
  single drawing (an OLE object's picture, a `mc:Fallback` preview).
- a `v:group` becomes an `InlineObject::Group`, its members mapped from the
  space `coordsize`/`coordorigin` give it onto the group's own box, nested
  groups flattened into it -- what `group_members` already does for a
  DrawingML `wpg:wgp`, so the model, the renderer and the writer need no
  change. Nesting is bounded at `MAX_VML_GROUP_DEPTH`, members at
  `MAX_GROUP_CHILDREN`.
- `read_vml_shape` is the old per-shape body, taking the group's scale so a
  member's `width`/`height` (and a `v:line`'s `from`/`to`) come through in
  the group's coordinates.

`boxes()` in the text-box tests now asks for `text_boxes()` rather than
`text_box()`, which only reports an object that is itself a box and so could
not see one inside a group; its assertions are unchanged.

* DOCX: cover a `v:group` holding several text boxes (#1643)

@KNUFFIXX reported the same loss from the other end: an inline `v:group` with
two `_x0000_t202` boxes side by side came through a convert as box A only, in
both a tall-B and a too-short-B variant, against Word reading both.

A text box is the one member a group used to keep -- `vml_textbox` recurses to
the first box that has content -- so this case looked half-working rather than
broken, and it is worth pinning separately from the shape cases. On `main` the
group collapses to a single `Shape { kind: TextBox, w: 400, h: 100 }` wearing
the group's own size and holding box A's story; here both boxes read, keep
their own stories and positions, and survive a save.

Built in code from the reported fixture rather than committing the .docx.

* DOCX: cover the shapetype lookup for a group's member (#1643)

Word writes each `v:shapetype` once, at its first use, as a child of the
`w:pict` -- so a group's member refers to one that is its parent's sibling,
not its own. Reading all shapes means `read_vml_shape` is now called for a
member as well as for a top-level shape, and both need the picture to find it,
which is why `pict` is threaded through the group walk.

Two tests: a member with no geometry and no colours of its own, taking its
diamond from the shapetype's `o:spt` and its fill from its `fillcolor`, and the
same for a shape outside a group, which is the case the lookup was written for.
2026-10-18 04:23:34 -04:00
FredApps 47c2eda22c Focus the Insert Hyperlink address box (#1610) (#1642)
* chore: reserve draft implementation for #1610

* fix(ui): focus hyperlink address when opening the dialog (#1610)
2026-10-18 04:23:24 -04:00
8701f4e28d Theme: default to system appearance (#718)
* Theme: default to system appearance

Sets Appearance::System as the default variant for ThemePreference,
so fresh installs and unconfigured preference files follow the host
OS appearance dynamically.

* contributors: add the missing blank line

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* contributors: drop the duplicate entry (already listed on main)

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Web prefs test: junk storage keeps the new default (System) appearance

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:23:13 -04:00
dhruvLearner8 6fefeb72e6 Hover tip on a table of contents link says Current Document (#2177) (#2281) 2026-10-18 04:23:02 -04:00
6311f9d411 AutoCorrect: replace the double-hyphen arrow (#2223) (#2263)
Co-authored-by: FredApps <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:22:50 -04:00
divyanshu2074andBrandon Thomas 963df12a11 fix(docx): trim marker id, bound bookmarks map, and support bookmarkEnd name fallback (#2252)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:22:31 -04:00
e7f5d32c35 Numbered headings: heading-linked multilevel list presets keep the heading styles (#1793) (#2239)
* Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1793: wording: other paragraph styles become Heading 1, not only body text

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:22:12 -04:00
divyanshu2074 8cdd85f7e7 fix(engine): support none in wordart reflection and outline color (#2238) 2026-10-18 04:21:53 -04:00
1d37043128 AutoCorrect: reversing a capitalisation adds an automatic exception, as in Word (#1857) (#2227)
* Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Co-Authored-By: Claude Opus 5.5 <[email protected]>

* #1857: forget the capital when a command fails; INitial CAps auto-add checked in Word

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:21:42 -04:00
divyanshu2074andBrandon Thomas 35cbdb8002 fix(docx): support hex without hash, rgb(), and case-insensitive color names in VML and DrawingML (#2222)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:21:24 -04:00
dhruvLearner8 498d9aaa9b Context menu: Remove Table of Contents on a TOC entry (#2179) (#2217) 2026-10-18 04:21:05 -04:00
Antoine BoxhoandBrandon Thomas 72cea2daf4 #1885: dialog close box fills red on hover, whole box clickable (#2090)
Dialogs, the Read Aloud player and the file browser used egui's small close glyph: the hit area was the glyph alone and hovering only changed its stroke. The close control is now a box at the right end of the title bar, the full title-bar height and 1.6 times as wide, as a Windows caption button on a Word dialog. The whole box closes the window. It fills red with a white cross while the pointer is over it and darker red while pressed and still under the pointer, with colours from the theme tokens in both themes. The box is pointer-only, so Tab never lands on it; dragging from it does not move the window and Escape closes dialogs as before. Its accessibility name is translated like its tooltip.

docs/parity-checklist.md is the regenerated checklist.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:04:12 -04:00
ab57696ba7 Draw undrawable private-use list bullets as the level's default bullet, not an icon-font glyph (#1683) (#2206)
A bullet stored as a private-use code (U+F0xx) in a font that isn't
installed, and that isn't a mapped Symbol/Wingdings code, fell back to
any installed face with a glyph at that code - an icon font such as
Font Awesome / Nerd Fonts, where U+F081 is the Twitter logo. Such a
bullet now shows the level's default bullet (• ○ ▪), so level 2 is a
hollow circle.

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:03:35 -04:00
a30a1866e2 Web: keep preferences such as dark mode between visits (#2132) (#2205)
The web shell never loaded or saved the UiState the desktop app keeps in
ui.json, so a reload reset the theme and every other preference. It now
restores it from localStorage at startup and writes it back whenever it
changes (a reload runs no exit code).

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:03:16 -04:00
Antoine BoxhoandBrandon Thomas 04eadcb2ca #2068: Recently Used Fonts at the top of the font list (#2192)
The font box (ribbon and mini toolbar) opens with a Recently Used Fonts
section: the fonts applied from the box, the mini toolbar or the Font
dialog, most recent first, once each, ten at most, kept in ui.json. All
Fonts follows. The sections show while nothing is typed; a query lists
its matches only. Up/Down walk the recent entries then the list and Enter
applies the highlighted one; a click on a heading keeps the list open.
The typed list is sized again when its length changes, in a discarded
pass so it never blanks; it used to stay one row tall after a one-match
query. Names over 128 bytes are not kept, and reading a saved list stops
at ten names.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:02:57 -04:00
f354e7fc9b fix(engine): support nested quotations and opening delimiters in smart quotes (#2176)
* fix(engine): support nested quotations and opening delimiters in smart quotes

* docs: typing-parity line for smart quotes after guillemets, low quotes and nesting

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:02:37 -04:00
Antoine BoxhoandBrandon Thomas 1c8c352cf5 #1549: one Thai face for a run in a missing Thai font (#2175)
A paragraph in a Thai font that isn't installed (TH Sarabun New, Angsana
New, Cordia New, the UPC families…) drew its Thai letters in one fallback
face and the Latin letters, digits and baht sign in a Latin substitute at
a visibly different size. Word substitutes one font (Cordia New) for the
whole run.

Thai families now have their own substitute lists (Cordia New and
Leelawadee UI on Windows, Thonburi on macOS, Noto Sans Thai and the TLWG
fonts on Linux; serif families prefer Angsana New, Sathu or Noto Serif
Thai), and the first listed face with Latin letters and digits as well
as Thai wins. When the substitute still lacks some of the run's Thai or
Latin text, the run moves to one face covering it through
FontDb::fallback_for_run, the generalised East Asian run fallback. Only
Thai fonts get this: a missing Latin font keeps Latin text in its own
substitute, a run with its own complex-script font keeps it for the Thai
letters, and a missing roman-classed Thai font is no longer drawn in
Cambria.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:02:18 -04:00
7fc2bbf248 fix(doc): preserve overparen, underparen, and brackets in LaTeX group characters (#2163)
* fix(doc): preserve overparen, underparen, and brackets in LaTeX group characters

* cargo fmt

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:01:59 -04:00
540692515b fix(engine): resolve section and section pages context when unlinking fields (#2156)
* fix(engine): resolve section and section pages context when unlinking fields

* cargo fmt

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:01:48 -04:00
Marco ArnulfoandBrandon Thomas 986d7e2a5a Layout: the caret after spaces hanging past the margin stays at the margin (#2152)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:01:38 -04:00
Marco ArnulfoandBrandon Thomas cdb13144b3 Lists: Enter on an empty nested item numbered by its paragraph style moves it up a level (#2143) (#2151)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:01:19 -04:00
8812fe068d fix(engine): support switches and quoted bookmarks when updating REF and PAGEREF fields (#2149)
* fix(engine): support switches and quoted bookmarks when updating REF and PAGEREF fields

* cargo fmt

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:01:00 -04:00
Marco ArnulfoandBrandon Thomas 7bd3eb8ca7 Backspace and Delete remove a section break, imported or not (#2133)
* Backspace and Delete remove a section break, imported or not (fixes #2127)

* Delete removes a section break on the last paragraph too (#2127)

A continuous break on the mark of the document's last paragraph, with the
last section holding no paragraph (as in the file attached to #2127), had
nothing for Delete to join, so it stayed. Delete at the end now removes it.
The text takes the last section's setup, as in Word, and the headers and
footers that section showed by linking to the removed one stay, here and
when the break goes by joining two paragraphs. Undo brings it back.

* Deleting over several section breaks keeps the headers they linked to (#2127)

A range across two or more breaks dropped the paragraphs in between with
their sections, so only the first section's headers and footers were carried
over. The joined section now takes those it leaves out from every removed
section, nearest first, as "link to previous" resolved them before the delete.
The typing-parity row says that Delete with Track Changes on still leaves a
break on the last paragraph alone.

---------

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:00:49 -04:00
357deba3c2 fix(engine): use table_formula parse_number for table sorting (#2126)
* fix(engine): use table_formula parse_number for table sorting

* contributors: drop the duplicate entry (already listed)

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* cargo fmt

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 04:00:30 -04:00
MarekSpodBiedry ecd94e0f8a Ribbon tab clicks no longer scroll the page to the caret while a footer is edited (#2054) (#2121) 2026-10-18 04:00:18 -04:00
1cc3d190a7 Fix Find navigation to show search results (#2096) (#2118)
Co-authored-by: Codex <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 04:00:07 -04:00
divyanshu2074 35fb78f0cc fix(engine): exclude carriage returns from word count character statistics (#2116) 2026-10-18 03:59:49 -04:00
a93932d491 fix(engine): ensure nextFootnote navigates only footnotes and skips endnotes (#2114)
* engine: filter note_refs in nextFootnote to skip endnotes

* Collapse the nested if (clippy collapsible_if)

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 03:59:37 -04:00
6670b9c867 Tab strip never draws tabs over Comments / Editing / Share (#2026) (#2084)
The strip kept File, the active tab and contextual tabs however narrow the window got, so
they were drawn under the buttons on the right. The strip is now planned first: the buttons
keep their room and the tabs give way in order (tab padding, Comments and Editing labels,
Share label, plain tabs into », active and contextual labels cut with …, last the active tab
into »). File always stays. A regression test checks widths 300-1280 in en, pl and de.

Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:59:19 -04:00
AlikandBrandon Thomas d8b11c69d3 Columns on selected text go in their own section (#2024) (#2071)
With text selected, Layout > Columns used to change the whole section. As in
Word (observed black-box), a continuous section break now goes where the
selection starts and one where it ends, splitting a paragraph there, unless that
end already meets a section boundary (the document's start or end, or an
existing break); the columns go on the section between, in one undo step.
`layout.columns` gains `apply: "selection"`, its default when text is selected
in the main text. More Columns... offers "Selected text" and picks it then.

Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:59:01 -04:00
NetterSilverandClaude Haiku 5.5 7f94643154 Give the status bar zoom percentage a fixed width (#2059)
Fixes #1975

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-18 03:58:42 -04:00
Brandon ThomasandClaude Opus 5.5 edee285c36 docs: fix ROADMAP tables (progress-log separator under its header, stray progress row out of the headline table)
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-18 03:50:50 -04:00
2e062168f0 Text effects: legacy outline, shadow, emboss and imprint are drawn (#2534) (#2548)
The legacy run properties w:outline, w:shadow, w:emboss and w:imprint were
read, resolved and saved but never drawn. Layout now turns them into text
effects (hollow outlined letters; a sharp grey copy down-right for shadow
and emboss, up-left for imprint, as Word's PDFs show them) and draws them
through the existing WordArt path, so screen and PDF both get them and the
PDF keeps the text. WordArt w14 effects on the same run take precedence.

Text drawn from outlines keeps synthetic bold (a widened outline, or one
in a solid fill's colour and transparency). Colour glyphs (emoji, no
outlines) and top-to-bottom runs with upright characters are drawn as
plain glyphs, and tracked insertions and deletions keep their legacy
effects.

Co-authored-by: Naxvog <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:50:15 -04:00
8fb4d44d2c Different First Page: page 1's header can be edited and inserted (#2826) (#2833)
With Different First Page on, page 1 shows the first-page header, but
Insert > Header/Footer/Page Number always wrote the default one (which
page 1 no longer shows), Edit Header without a page went to the default
too, and a double-click in page 1's header area was ignored while
another header was being edited (the usual case: edit the header, tick
Different First Page, double-click page 1). Now hf_part and edit_hf use
the slot of the caret's page, and a double-click in a header/footer
zone switches to the one that page shows even from another header.

Fixes #2826
Fixes #2811


Claude-Session: https://claude.ai/code/session_013kKPnx8r1vcYumjyL3kcuz

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:47:13 -04:00
abb36aceb2 #1270: pasted HTML blockquote is an indented paragraph, not Quote (#2737)
LibreOffice puts a paragraph indented on both sides on the clipboard as
<blockquote>, its margins in the page's stylesheet. Paste mapped it to the
centred, italic Quote style. On paste only, it now becomes a Normal
paragraph indented by the blockquote's stylesheet margins (else 0.5 in each
side). Opening .html files still maps blockquote to Quote.

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:46:54 -04:00
da11d6634b Shapes: preserve DrawingML outline dashes (#2661)
* Shapes: preserve DrawingML outline dashes

* Shapes: scale dashes to effective path width (#380)

* Keep the outline dash with noFill; cap the raster dashed stroke width at 200 pt

Addresses the two CodeRabbit review notes.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Use clamp for the dashed stroke width (clippy)

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Brandon Thomas <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 03:46:35 -04:00
ce38b0c372 #1297: copied equations go in the MathML clipboard formats Word reads (#2614)
A copy or cut of one equation on its own now also puts its MathML in the
platform's MathML clipboard formats, beside the HTML (with MathML, #816) and
the linear text: "MathML" and "MathML Presentation" (UTF-16LE) on Windows,
public.mathml.presentation and public.mathml on macOS. These are the names
W3C MathML 3 section 6.2.4 registers, and Microsoft's "MathML Support in
Microsoft 365" doc says Office checks them first when it pastes math. No
public clipboard format carries OMML. A copy with text around the equation
doesn't offer them, so Word keeps the text.

Not yet verified in Word: the parity docs say MathML clipboard formats, not
an editable paste.


Claude-Session: https://claude.ai/code/session_01P8veLveuUsJJm6UEcQPEuW

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:46:15 -04:00
Fin WarmanandBrandon Thomas 81098c52f6 CLI: allow explicit field updates during conversion (#2014)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:45:57 -04:00
Fin WarmanandBrandon Thomas a3fcb9ed47 Fonts: keep Arial Unicode MS fallbacks proportional (#2657)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:45:37 -04:00
Alik ff61b32177 Tests: the Alt+0233 test expects the system code page's character (#2040)
alt_numpad_code_types_its_character_without_keytips always expected é, but on
Windows the app (rightly) reads 0233 in the system locale's ANSI code page, so
a Russian system (1251) types й and the test failed. The test now works out the
expected character the same way, as its Alt+233 half already did.
2026-10-18 03:45:07 -04:00
da84f48040 A cell emptied by widow control takes its row to the next page (#2035)
split_row already keeps widow/orphan control in each cell paragraph.
When that moves all of a cell's lines on although its first line would
fit, the other cells stayed on the page and the row was split beside
an empty cell. Word moves the whole row on instead. Observed with
synthetic documents (exact 14 pt lines, a one-line cell next to a
longer one near the bottom of the page):

  long cell | room for | Word      | before          | now
  4 lines   | 2        | 2 + 2     | 2 + 2           | 2 + 2
  3 lines   | 2        | row moves | short cell kept | row moves
  3 lines   | 1        | row moves | short cell kept | row moves
  4 lines   | 1        | row moves | short cell kept | row moves
  5 lines   | 3        | 3 + 2     | 3 + 2           | 3 + 2
  3, control off | 2   | 2 + 1     | 2 + 1           | 2 + 1

Co-authored-by: Daniele Mori <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 03:44:56 -04:00
795cad3326 Automatic text in a styled text box takes the shape style's font colour (#2032)
* Automatic text in a styled text box takes the shape style's font colour

A text box shape whose style has a:fontRef with a colour (Word's
default shape style uses lt1, white on the accent fill) shows its
automatic-colour text in that colour. WordCraft drew it black. Word,
observed with a synthetic document: only text whose resolved colour is
automatic takes the fontRef colour; a run's own colour, a paragraph or
character style colour and a colour from the document defaults all win
over it. A box without a fontRef keeps black.

- wordcraft-doc: TextBody gets `auto_color`.
- wordcraft-docx: read from wps:style/a:fontRef (theme colours resolved
  as for SmartArt).
- wordcraft-layout: lines in a text box story resolve automatic colour
  to it.

Saving does not write a shape style yet, so the colour is not kept
through a save; the reader and layout parts stand on their own.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

* Remove stray doc-comment lines from the new test

Co-Authored-By: Claude Opus 5.5 <[email protected]>

---------

Co-authored-by: Daniele Mori <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:44:45 -04:00
ba8dc7bf1c DOCX: a style ID defined twice is its last definition (#2031)
Some generators write a placeholder Heading 1 and then their own,
under the same w:styleId. The reader kept the first definition and
skipped the rest, so headings came out in the placeholder's font,
size and colour. Word uses the last definition, whole: a synthetic
document with an italic red 20 pt first definition and a bold blue
12 pt second one shows bold blue 12 pt, with nothing of the first
(no italic, no centring), for a custom style and for Heading 1 alike.

The reader now replaces an earlier definition in place instead of
skipping the later one.

Co-authored-by: Daniele Mori <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
2026-10-18 03:44:34 -04:00
EncodeE55 f6115ea7fa Update ROADMAP.md (#2020)
Fixxed the Markdown table formatting—specifically, adding the header separator row (| --- | --- | --- |).
2026-10-18 03:44:23 -04:00
MarekSpodBiedry ace1adc624 Format Painter: double-click keeps it on, armed and sticky look different (#2785) (#2822) 2026-10-18 03:44:13 -04:00
4d8204c2f2 #2634: a table inserted in right-to-left text is right to left, so Tab moves left (#2688)
Tab already follows reading order (next logical cell) and right-to-left
tables (w:bidiVisual) already lay the first cell out at the right. The
bug was that insert.table ignored the caret paragraph's direction, so a
table made while writing RTL text was a left-to-right table and Tab went
rightwards. Like Word, insert.table now makes a right-to-left table when
the caret's paragraph reads right to left.

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:43:47 -04:00
Charan Rathore 5b322e1530 fix(commands): Export PDF and PNG without a path ask where to save (#2665)
Signed-off-by: Charan Rathore <[email protected]>
2026-10-18 03:43:27 -04:00
NetterSilverandClaude Haiku 5.5 70c5c44f01 Promote the cell below when deleting a vertical merge's first row (#2649)
Found while testing crates/doc/src/table.rs:129 (Table::delete_row), called from crates/engine/src/cmd/table.rs:526 (del_row)

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-18 03:43:16 -04:00
NetterSilverandClaude Haiku 5.5 33f00e8dfe Keep custom footnote/endnote marks as text inside the note on DOCX save (#2647)
Found while testing crates/docx/src/write/story.rs:512-530 (InlineObject::NoteRef, the `current_note == Some(..)` branch)

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-18 03:43:04 -04:00
4950690479 #498: Indonesian (id-ID) spelling proofing (#2576)
Indonesian spelling reads the installed Hunspell id_ID dictionary at runtime
(Debian/Ubuntu hunspell-id, LibreOffice dict-id, the Dictionaries folder or
WORDCRAFT_DICTIONARIES), like Czech, Polish and Portuguese: nothing is bundled,
and without a dictionary Indonesian text gets no spelling marks.

- ProofLang::Id for id, id-ID, ind and the legacy in tag
- suggestions reuse the dictionary-driven Czech ranking (MAP/REP/TRY/KEY of
  the loaded dictionary), nothing Indonesian hard-coded
- hyphenated reduplication accepted whole or by parts (kupu-kupu, anak-anak);
  hyphenated clitics after capitals/abbreviations (KTP-nya, se-Indonesia)
- generic grammar only: repeated word (offers anak-anak first), space before
  punctuation, double space, with Indonesian messages
- Review > Language, status bar, Set Proofing Language, File > Options >
  Proofing list Indonesian; name translated in every catalog
- tests use a hand-written synthetic dictionary; engine end-to-end test

Co-authored-by: Claude Opus 5.5 (1M context) <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:42:52 -04:00
4442390213 DOCX: keep the glossary document (building blocks) (#2631)
Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:23:25 -04:00
Fin WarmanandBrandon Thomas bd0521d476 Prefer a covering text face for missing black-circle glyphs (#2582)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:23:05 -04:00
55bef52d38 Comments readable in the dark theme (#2045) (#2563)
Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:22:46 -04:00
df9c29f673 Read Mode keeps the keyboard and shows whole, unobscured columns (#1772) (#2591)
Co-Authored-By: Claude Opus 5.5 <[email protected]>

Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:22:26 -04:00
Fin WarmanandBrandon Thomas 03e713b72f Preserve cached field result formatting (#2605)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:22:06 -04:00
b2b8ad8f52 Font list: entries of tall-ascent fonts are no longer cut off (#2570) (#2598)
Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Opus 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:21:48 -04:00
Fin Warman fc15682fb0 Preserve REF field hyperlinks in PDF exports (#2597) 2026-10-18 03:21:29 -04:00
aa665b16a9 #2592: Split Cells and Convert Text to Table dialogs (#2593)
Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:21:14 -04:00
8c1a39ba95 A new source's Tag follows Word's format; editing keeps the Tag (#2094) (#2571)
Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:20:56 -04:00
16622f0853 #1876: Web Layout text fills the window width (#2539)
Co-authored-by: Dmitrii Khlapov <[email protected]>
Co-authored-by: Claude Sonnet 5.5 <[email protected]>
2026-10-18 03:20:37 -04:00
NetterSilverandClaude Haiku 5.5 75122c1178 Keep leading U+3000/NBSP in Markdown import and round trip (#2538)
Found while testing crates/formats/src/markdown.rs:253 (join_lines); also the trim_start() calls at lines 229, 311, 426 and 432, against escape_line_start at line 1077

Co-authored-by: Claude Haiku 5.5 <[email protected]>
2026-10-18 03:20:26 -04:00
dhruvLearner8 ad820dce6c Remember Navigation Pane visibility across launches (#2526) (#2528) 2026-10-18 03:20:14 -04:00
Fin WarmanandBrandon Thomas 3507a34430 Retain cell top insets across table row page breaks (#2524)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:19:59 -04:00
Fin WarmanandBrandon Thomas db0affc0d3 Recognize the Symbol PCL6 font alias (#2521)
Co-authored-by: Brandon Thomas <[email protected]>
2026-10-18 03:19:40 -04:00
Fin Warman 74c1f26c4e Resolve exact OpenType full font names (#2516) 2026-10-18 03:19:21 -04:00
Dylan M. Taylor 72453d9def ci: switch to shared nix workflow from craft-common-actions 2026-10-15 23:57:51 -04:00
Isaac Levine 76f5d4b817 CI: run cargo xtask ci on pull requests and pushes to main (#754) 2026-10-11 22:30:39 -04:00
273 changed files with 36288 additions and 2612 deletions
+6
View File
@@ -0,0 +1,6 @@
# yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json
reviews:
auto_review:
# Dependabot bumps are checked by CI; skip CodeRabbit reviews on them.
ignore_usernames:
- "dependabot[bot]"
+30
View File
@@ -0,0 +1,30 @@
version: 2
updates:
# Maintain Rust / Cargo dependencies
- package-ecosystem: "cargo"
directory: "/"
schedule:
interval: "daily"
time: "06:00"
timezone: "America/New_York"
# Consolidate updates into a single PR rather than one PR per crate
groups:
cargo-dependencies:
patterns:
- "*"
commit-message:
prefix: "chore(deps)"
# Maintain GitHub Actions workflows
- package-ecosystem: "github-actions"
directory: "/"
schedule:
interval: "daily"
time: "06:00"
timezone: "America/New_York"
groups:
actions:
patterns:
- "*"
commit-message:
prefix: "ci(deps)"
+62
View File
@@ -0,0 +1,62 @@
# The CLAUDE.md quality gate (`cargo xtask ci`: fmt, clippy -D warnings, tests, assets, layers,
# wasm) on every pull request and every push to main, so a merge burst can't land a red tree that
# only the next contributor's local run discovers.
name: CI
on:
push:
branches: [main]
pull_request:
workflow_dispatch:
permissions:
contents: read
concurrency:
group: ci-${{ github.ref }}
cancel-in-progress: true
env:
# Same craft-fonts commit as release.yml and freebsd.yml, so layout tests see the fonts a
# release build embeds (the checkout is an optional build input, see crates/fonts/build.rs).
CRAFT_FONTS_REF: 8dcdacd5153e64560d109541a47d806f26f048c0
jobs:
gate:
name: cargo xtask ci
runs-on: ubuntu-24.04
timeout-minutes: 60
env:
CARGO_TERM_COLOR: always
CARGO_INCREMENTAL: "0"
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
- name: Check out craft-fonts
uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
with:
repository: storytold/craft-fonts
ref: ${{ env.CRAFT_FONTS_REF }}
path: craft-fonts
- name: Install Linux build dependencies (winit, wgpu, rfd)
run: |
sudo apt-get update
sudo apt-get install -y --no-install-recommends \
pkg-config libxkbcommon-dev libwayland-dev libx11-dev libxcursor-dev libxrandr-dev \
libxi-dev libgtk-3-dev libfontconfig1-dev libfreetype-dev mesa-vulkan-drivers
- name: Install Rust (stable + clippy, rustfmt, wasm32)
run: |
rustup toolchain install stable --profile minimal --component clippy,rustfmt \
--target wasm32-unknown-unknown
rustup default stable
- uses: Swatinem/rust-cache@6323deb102c322ba6fcbdcafc7e3dddab59af2b6 # v2.9.2
with:
save-if: ${{ github.ref == 'refs/heads/main' }}
- name: Quality gate
env:
CRAFT_FONTS_DIR: ${{ github.workspace }}/craft-fonts
run: cargo xtask ci
+3 -13
View File
@@ -18,16 +18,6 @@ concurrency:
cancel-in-progress: true
jobs:
eval:
runs-on: ubuntu-latest
timeout-minutes: 15
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
- uses: cachix/install-nix-action@13d8dd58da0234aa297dedd986986ccb8e7f3e24 # v31.11.1
- name: nix flake check --no-build
run: |
if [ -f flake.nix ]; then
nix flake check --no-build --all-systems
else
echo "No flake.nix present in this repository; skipping check."
fi
nix:
# Intentionally tracked at @main (not SHA-pinned) so org-wide CI improvements, security fixes, and build optimizations propagate without PR churn across all craftapp repos.
uses: storytold/craft-common-actions/.github/workflows/nix.yml@main
+20
View File
@@ -0,0 +1,20 @@
name: SonarCloud
on:
push:
branches: [main]
pull_request:
types: [opened, synchronize, reopened]
permissions:
contents: read
concurrency:
group: sonarcloud-${{ github.ref }}
cancel-in-progress: true
jobs:
sonarcloud:
# Tracked at @main so org-wide CI improvements propagate without PR churn.
uses: storytold/craft-common-actions/.github/workflows/sonarcloud.yml@main
secrets: inherit
+1 -1
View File
@@ -67,7 +67,7 @@ Generated-in-code art is original and has no file to list: the UI icon set (`cra
| craft-fonts (optional build input, not files in this repo): BIZ UDPGothic, BIZ UDMincho, Shippori Mincho, Noto Sans CJK SC, Noto Sans Arabic | see https://github.com/storytold/craft-fonts/blob/main/ATTRIBUTION.md | embedded only when built with `CRAFT_FONTS_DIR` (all official releases) | OFL-1.1 | Japanese and Chinese UI, Japanese document fallback |
| `crates/ui-egui/src/i18n/uk.tsv` | @dmatviichuk with Codex | Original translation from the English interface labels in this repository; no other product localisation copied | MIT OR Apache-2.0 | Ukrainian interface catalog; spelling checked with external VESUM and ULIF dictionaries, neither bundled |
| `crates/ui-egui/src/i18n/et.tsv` | @henno with Codex | Original translation from the English interface labels in this repository; no other product localisation copied | MIT OR Apache-2.0 | Estonian interface catalog |
| `crates/ui-egui/src/i18n/pl.tsv` | Konrad Witkowski with Claude (AI agent) | Original translation from the English interface labels in this repository; no other product localisation copied | MIT OR Apache-2.0 | Polish interface catalog; not yet reviewed by a second native speaker |
| `crates/ui-egui/src/i18n/pl.tsv` | Konrad Witkowski with Claude (AI agent); Aleksander Kutycki (@olekkut) with Gemini 3.8 Flash | Original translation from the English interface labels in this repository; no other product localisation copied | MIT OR Apache-2.0 | Polish interface catalog; not yet reviewed by a second native speaker |
| `crates/ui-egui/src/i18n/pt-br.tsv` | @edurbs with ZCode | Original translation from the English interface labels in this repository; no other product localisation copied | MIT OR Apache-2.0 | Brazilian Portuguese interface catalog; spelling checked with the external pt_BR Hunspell dictionary, not bundled |
| `crates/ui-egui/src/i18n/cs.tsv` | @DanielKnourek with Antigravity for VS Code | Original translation from the English interface labels in this repository; no other product localisation copied | MIT OR Apache-2.0 | Czech interface catalog; review suggestions by @Losminek |
| `crates/ui-egui/src/i18n/de.tsv` | @arekown with Kimi | Original translation from the English interface labels in this repository; no other product localisation copied | MIT OR Apache-2.0 | German interface catalog |
Generated
+23 -20
View File
@@ -5131,7 +5131,7 @@ dependencies = [
[[package]]
name = "wordcraft"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"arboard",
"clipboard-win",
@@ -5144,6 +5144,7 @@ dependencies = [
"log",
"objc2 0.6.5",
"objc2-app-kit 0.3.2",
"objc2-foundation 0.3.2",
"open",
"pollster 0.4.0",
"rfd",
@@ -5163,7 +5164,7 @@ dependencies = [
[[package]]
name = "wordcraft-cli"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"serde_json",
"sys-locale",
@@ -5176,11 +5177,11 @@ dependencies = [
[[package]]
name = "wordcraft-control-key"
version = "0.8.0"
version = "0.9.0"
[[package]]
name = "wordcraft-doc"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"proptest",
"serde",
@@ -5192,7 +5193,7 @@ dependencies = [
[[package]]
name = "wordcraft-docbin"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"cfb",
"image",
@@ -5208,7 +5209,7 @@ dependencies = [
[[package]]
name = "wordcraft-docx"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"aes",
"base64",
@@ -5222,6 +5223,7 @@ dependencies = [
"sha1",
"sha2",
"thiserror 2.0.21",
"unicode-bidi",
"wordcraft-doc",
"wordcraft-geom",
"zeroize",
@@ -5230,7 +5232,7 @@ dependencies = [
[[package]]
name = "wordcraft-engine"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"cfb",
"encoding_rs",
@@ -5259,7 +5261,7 @@ dependencies = [
[[package]]
name = "wordcraft-fonts"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"harfrust 0.13.3",
"kurbo",
@@ -5272,7 +5274,7 @@ dependencies = [
[[package]]
name = "wordcraft-formats"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"cfb",
"chardetng",
@@ -5289,14 +5291,14 @@ dependencies = [
[[package]]
name = "wordcraft-geom"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"serde",
]
[[package]]
name = "wordcraft-layout"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"image",
"js-sys",
@@ -5315,7 +5317,7 @@ dependencies = [
[[package]]
name = "wordcraft-mcp"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"log",
"serde_json",
@@ -5327,7 +5329,7 @@ dependencies = [
[[package]]
name = "wordcraft-metafile"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"kurbo",
"thiserror 2.0.21",
@@ -5335,7 +5337,7 @@ dependencies = [
[[package]]
name = "wordcraft-pdf"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"craft-pdf",
"hayro-cmap",
@@ -5355,7 +5357,7 @@ dependencies = [
[[package]]
name = "wordcraft-proof"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"log",
"miniz_oxide 0.8.9",
@@ -5364,7 +5366,7 @@ dependencies = [
[[package]]
name = "wordcraft-render"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"image",
"log",
@@ -5378,8 +5380,9 @@ dependencies = [
[[package]]
name = "wordcraft-ui-egui"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"cfb",
"craft-ui",
"egui",
"egui_kittest",
@@ -5404,7 +5407,7 @@ dependencies = [
[[package]]
name = "wordcraft-web"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"eframe",
"egui",
@@ -5423,7 +5426,7 @@ dependencies = [
[[package]]
name = "wordcraft-zotero"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"log",
"proptest",
@@ -5536,7 +5539,7 @@ checksum = "9440ea3e5aeabb0ac63af70daf835274065238cdd0cec83418f417eae38bacee"
[[package]]
name = "xtask"
version = "0.8.0"
version = "0.9.0"
dependencies = [
"serde_json",
]
+1 -1
View File
@@ -4,7 +4,7 @@ members = ["crates/*", "apps/*", "xtask"]
default-members = ["crates/*", "apps/wordcraft", "apps/wordcraft-cli", "xtask"]
[workspace.package]
version = "0.8.0"
version = "0.9.0"
edition = "2024"
license = "MIT OR Apache-2.0"
rust-version = "1.95"
+25 -19
View File
@@ -10,11 +10,11 @@
<h1 align="center">WordCraft</h1>
<p align="center">
<b>Writing and document design; an open-source, clean-room reimplementation of Microsoft Word, rebuilt in pure Rust.</b>
<b>Writing and document design; an open-source, modern word processor, built in pure Rust.</b>
</p>
<p align="center">
A fast, open-source word processor with the Word workflow you already know: the ribbon, styles,
A fast, open-source word processor with a familiar workflow: the ribbon, styles,
tables, track changes, references and mail merge. It reads and writes .docx, runs natively on
macOS, Windows, Linux and BSD, and in the browser via WebAssembly.<br>
<i>By the ArtCraft team.</i>
@@ -25,6 +25,9 @@
<img alt="Runs on macOS, Windows, Linux, BSD and the web" src="https://img.shields.io/badge/runs%20on-macOS%20%C2%B7%20Windows%20%C2%B7%20Linux%20%C2%B7%20BSD%20%C2%B7%20Web-3b5bdb?style=flat-square">
<img alt="License: MIT OR Apache-2.0" src="https://img.shields.io/badge/license-MIT%20%2F%20Apache--2.0-2b47b5?style=flat-square">
<img alt="Agent-drivable over MCP" src="https://img.shields.io/badge/agents-MCP%20%C2%B7%20CLI-3b5bdb?style=flat-square">
<br>
<a href="https://sonarcloud.io/project/overview?id=storytold_wordcraft"><img alt="Scanned by SonarCloud" src="https://img.shields.io/badge/Scanned_by-SonarCloud-4b9fd5?style=flat-square&logo=sonarqubecloud&logoColor=white"></a>
<a href="https://coderabbit.ai"><img alt="Reviewed by CodeRabbit" src="https://img.shields.io/coderabbit/prs/github/storytold/wordcraft?style=flat-square&label=Reviewed%20by%20CodeRabbit&logo=coderabbit"></a>
</p>
<p align="center">
@@ -86,10 +89,10 @@ Every screenshot below is WordCraft itself, rendered offscreen by its own UI tes
## Why WordCraft
- **Familiar.** Word's ribbon tabs, groups, shortcuts and behaviour: Enter continues a list,
- **Familiar.** Ribbon tabs, groups, shortcuts and behaviour: Enter continues a list,
Tab demotes it, Ctrl/⌘+B bolds the word under the caret, the Styles gallery previews styles live,
F4 repeats, F7 checks spelling, F8 extends the selection.
- **Your files.** Opens and saves .docx (OOXML), and also .odt, .rtf, .html, .md, .tex (LaTeX), .txt; opens Word 97-2003 .doc; exports PDF
- **Your files.** Opens and saves .docx (OOXML), and also .odt, .rtf, .html, .md, .tex (LaTeX), .txt; opens legacy 97-2003 .doc; exports PDF
with real, selectable text, links and bookmarks.
- **Fast.** Paragraph layout is cached, so typing in a 188-page document re-lays it out in about
1.4 ms; pages render on demand.
@@ -118,7 +121,7 @@ Every screenshot below is WordCraft itself, rendered offscreen by its own UI tes
| **Review** | Spelling and grammar with suggestions (English and Polish, by each run's language), thesaurus, word count, comments in margin balloons or a pane, track changes, accept/reject, compare documents, restrict editing, accessibility checker, document inspector |
| **Mailings** | Mail merge from CSV, merge fields, address block, greeting line, rules, preview, finish to a document; envelopes and labels |
| **View** | Print layout, web layout, draft, read mode, focus, zoom, one/multiple pages, page width, Navigation pane, rulers, gridlines, dark mode; interface in English, 简体中文, 繁體中文, 日本語, Português (Brasil), Español, Українська, Српски / Srpski, Eesti, Türkçe, Русский, Italiano, Deutsch, Polski, Čeština, Magyar, Français, Български, Tiếng Việt, 한국어 or Svenska (follows the system language by default) |
| **Files** | .docx read/write (opens in Word), Word 97-2003 .doc import, PDF export, .odt, .rtf, .html, .md, .tex, .txt import/export, page images |
| **Files** | .docx read/write, legacy 97-2003 .doc import, PDF export, .odt, .rtf, .html, .md, .tex, .txt import/export, page images |
The honest picture, area by area, is in [ROADMAP.md](ROADMAP.md), the
[parity assessment](docs/target-app-parity.md) and the generated
@@ -141,6 +144,7 @@ wordcraft-cli text report.docx # plain text
wordcraft-cli inspect report.docx # structure as JSON
wordcraft-cli run --template sample \
--cmd 'select.text={"text":"Membership"}' --cmd format.bold --save out.docx
wordcraft-cli run --script job.json # the same from a file: commands, then save to docx and pdf
```
Web: `cd apps/wordcraft-web && trunk serve`, then open <http://127.0.0.1:8771/?sample>.
@@ -189,9 +193,9 @@ covers every target starting with it (`wordcraft*=debug`). The logger is
| L1 | `wordcraft-doc`, `wordcraft-fonts`, `wordcraft-proof` | document model and editing; fonts and shaping; spelling, grammar, hyphenation |
| L2 | `wordcraft-layout`, `wordcraft-docx`, `wordcraft-formats` | line breaking, pagination, tables, notes, hit testing; OOXML; ODT/RTF/HTML/Markdown/TXT |
| L3 | `wordcraft-render`, `wordcraft-pdf` | rasteriser (vello_cpu); PDF ([craft-pdf](docs/shared-pdf.md), Krilla) |
| L4 | `wordcraft-engine` | session, undo, 422 commands, Word feature catalog |
| L4 | `wordcraft-engine` | session, undo, 422 commands, document feature catalog |
| L5 | `wordcraft-mcp` | MCP server |
| L6 | `wordcraft-ui-egui` | the Word-style front end (swappable) |
| L6 | `wordcraft-ui-egui` | the word processor front end (swappable) |
| apps | `wordcraft`, `wordcraft-cli`, `wordcraft-web` | desktop, command line, browser |
`cargo xtask ci` runs formatting, clippy, ~800 tests, the asset-attribution check, the layering
@@ -199,12 +203,13 @@ check and the wasm build. Contributor and agent instructions: [AGENTS.md](AGENTS
## Roadmap
WordCraft is in **alpha**. It covers 88% of Word's ribbon features with commands today; counting
depth, file fidelity, stability and the rest, we estimate it is about 60% of the way to replacing
all of Word for real work, about 55% for a typical professional's weekly work and about 63% for
everyday essentials. Beta needs testing against real-world .docx files, keeping charts and SmartArt
through a round trip, pagination closer to Word's, a stability sweep and native printing. Details
and estimates: [ROADMAP.md](ROADMAP.md) and [where we fall short](docs/gaps.md).
WordCraft is in **alpha**. It has commands for 97% of the ribbon and menu features in its feature catalog
([`docs/parity-checklist.md`](docs/parity-checklist.md)) today; counting
depth, file fidelity, stability and the rest, we estimate it is about 60% ready for real work,
about 55% for a typical professional's weekly work and about 63% for everyday essentials. Beta
needs testing against real-world .docx files, keeping charts and SmartArt through a round trip,
closer pagination fidelity, a stability sweep and native printing. Details and estimates:
[ROADMAP.md](ROADMAP.md) and [where we fall short](docs/gaps.md).
## Downloads
@@ -247,9 +252,10 @@ WordCraft compiles DirectX 12 shaders with FXC, the compiler built into Windows,
also avoids DirectX 12 altogether.
**If the app is slow on an older Windows PC:** when DirectX 12 finds no graphics card with a
driver for it (for example Intel 4th-generation graphics on Windows 11), it only offers Microsoft's
software renderer, which draws every frame on the processor. In that case WordCraft uses OpenGL
instead, without setting anything. The log file records the choice (lines starting with
driver for it (for example Intel 4th-generation graphics on Windows 11), it only offers the built-in
software renderer, which draws every frame on the processor. WordCraft normally uses OpenGL instead,
without setting anything. If OpenGL has no usable adapter, it can use DirectX 12's WARP software
adapter. The log file records the choice (lines starting with
`graphics:`). `WGPU_BACKEND` still overrides it, e.g. `$env:WGPU_BACKEND = "gl"` or `"dx12"`.
### macOS
@@ -366,7 +372,9 @@ Bundled fonts, icons, images and other assets keep their own open licenses; each
with its author, source and license in [ATTRIBUTION.md](ATTRIBUTION.md).
The English spelling dictionary and hyphenation come from Grady Ward's public-domain Moby
Hyphenator II word list. Polish spelling reads the Hunspell `pl_PL` dictionary installed on the system (not bundled) and
Hyphenator II word list. Polish spelling reads the Hunspell `pl_PL` dictionary installed on the system (not bundled), as
do Russian, Portuguese, Czech, Italian, Croatian and Turkish spelling (how to install them:
[docs/spelling-dictionaries.md](docs/spelling-dictionaries.md)), and
Polish hyphenation the TeX patterns by Hanna Kołodziejska, Bogusław Jackowski and Marek Ryćko (MIT
option). The sample documents and templates are original text written for WordCraft.
@@ -375,8 +383,6 @@ ArtCraft Team and are not covered by this license. They may be used only unmodif
part of this repository and WordCraft, under [`docs/brand/LICENSE-brand.txt`](docs/brand/LICENSE-brand.txt).
Forks and modified versions must remove them.
<sub>Microsoft and Microsoft Word are trademarks of the Microsoft group of companies. Adobe, Photoshop, Illustrator, Premiere Pro, Lightroom, Acrobat, After Effects and InDesign are trademarks or registered trademarks of Adobe Inc. in the United States and/or other countries. WordCraft is an independent, open-source project and is not affiliated with, sponsored by or endorsed by Microsoft Corporation or Adobe Inc.; these names are used only to describe the workflows it is compatible with.</sub>
<p align="center">
<a href="https://getartcraft.com/"><img alt="ArtCraft" src="docs/brand/artcraft-mark.svg" width="28"></a><br>
<sub>Made by the <a href="https://getartcraft.com/">ArtCraft</a> team and community.</sub>
+190 -6
View File
@@ -2,8 +2,8 @@
**Stage: alpha** · next: beta, ~15 points (60% → 75% ready for real work) and ~150–230 h away
> **Last reviewed:** 2026-10-10 · **Last updated:** 2026-10-17 · **Change:** trivial (Language dialog keeps inherited proofing language and no-proof choices (#2717)); previously trivial (table style editor formats every part, #714); previously trivial: VML freeforms read and kept on save; previously trivial: progress log: consistent preferred cell widths; previously trivial: multi-paragraph field instructions no longer add blank lines; previously trivial: page-break-only list paragraphs hide their labels; previously trivial: progress log: grouped `flowChartOr` shapes draw their circle and crossing spokes; previously trivial: progress log: copies keep their formatting in other programs, #1523; previously trivial: progress log: privacy options on save, #1062; previously trivial: progress log: equation pictures in theme colours on Windows, #1002; previously trivial: symbol fonts, #776; previously trivial: the save prompt answers from the keyboard, #861; previously trivial: table column and row markers on the rulers, #700; previously trivial: table borders and the resize handle drag on the page, #217; previously trivial: formatting revisions landed, #41; previously trivial: Draw tab ink, #307, and Draw Table and Eraser, #303, landed: catalog 402/434; previously minor: readiness table with hours per audience; full number back to the additive weighted sum, 60%; mainstream and essentials numbers added; alpha gate checked; previously major: full re-measure against Word for Mac 16.113.4; restructured to the craftrules progress-docs standard) · **Target:** Microsoft Word (Microsoft 365) for Mac 16.113.4
> **Last reviewed:** 2026-10-10 · **Last updated:** 2026-10-14 · **Change:** trivial (progress log: print-area strings translated in every catalog; previously trivial: progress log: ODT sections, #1379; previously trivial: progress log: table deletes, Change Case and Unlink Fields respect Track Changes and locked content controls, #1329; previously trivial: progress log: rounded rectangle corners, #1476; previously trivial: progress log: Save As type list, #1399; previously trivial: progress log: privacy options on save, #1062; previously trivial: progress log: equation pictures in theme colours on Windows, #1002; previously trivial: symbol fonts, #776; previously trivial: the save prompt answers from the keyboard, #861; previously trivial: table column and row markers on the rulers, #700; previously trivial: table borders and the resize handle drag on the page, #217; previously trivial: formatting revisions landed, #41; previously trivial: Draw tab ink, #307, and Draw Table and Eraser, #303, landed: catalog 402/434; previously minor: readiness table with hours per audience; full number back to the additive weighted sum, 60%; mainstream and essentials numbers added; alpha gate checked; previously major: full re-measure against Word for Mac 16.113.4; restructured to the craftrules progress-docs standard) · **Target:** Microsoft Word (Microsoft 365) for Mac 16.113.4
> **Last reviewed:** 2026-10-10 · **Last updated:** 2026-10-19 · **Change:** trivial (progress log: caret inside ligatures, #1974; previously trivial: text typed into a content control takes the control's formatting; localized placeholder styles recognised; previously trivial: forms protection keeps its content controls: no inserting, Properties or Remove Content Control; previously trivial: long navigation headings wrap onto further lines within the pane, #3015; previously trivial: clearing a no-match font query keeps the recent heading visible, #2068; previously trivial: a word after a slash wraps whole, like Word, #2963; previously trivial: progress log: a header-cell picture watermark sits behind the text, #1775; previously trivial: symbols keep their font when the text font changes, #2036; previously trivial: opening a DOCX no longer copies its pictures; previously trivial: Add Font imports every selected font file, #1789); previously trivial (RTF export writes sections (gap 19); previously trivial: ODT export writes sections (gap 19); previously trivial: Read Aloud speaks each sentence in its own language, #1843; previously trivial: hint-reliant CJK outlines; previously trivial: Ctrl+wheel zoom no longer scrolls the document, #1743; previously trivial: progress log: Turkish case mapping in AutoCorrect, #1923; previously trivial: a mouse drag on a table leaves the view where it is, #1738; previously trivial: progress log: Hebrew field results and smart quotes, #1670; previously trivial: Backspace removes a selected table, rows or columns, #1844; previously trivial: progress log: mixed Hebrew/Latin text keeps its order in Word, #1543; previously trivial: progress log: web typing in non-Latin text no longer repeats, #1480; previously trivial: progress log: font family caches bounded and math font lookups follow font changes, #1354; previously trivial: `-->` AutoCorrect replacement, #2223; previously trivial: numbered headings, #1793; previously trivial: Navigation Find exposes Results, #2096; previously trivial: progress log: Columns on selected text, #2024; previously trivial: progress log: legacy outline, shadow, emboss and imprint text effects drawn, #2534; previously trivial: progress log: Indonesian proofing, #498; previously trivial: progress log: DOCX glossary documents (building blocks); previously trivial: progress log: Read Mode keeps the keyboard and shows whole, unobscured columns, #1772); previously trivial: progress log: font list entries of tall-ascent fonts no longer cut off, #2570); previously trivial: progress log: Split Cells and Convert Text to Table dialogs follow the Word check, #2592; previously trivial: a new source's Tag is made like Word's, three letters of the author and two digits of the year with a number added when taken, and an edit keeps the Tag, #2094; previously trivial: Web Layout fills the window width, #1876; previously trivial: progress log: Different First Page headers reachable from page 1, #2826; previously trivial: Language dialog keeps inherited proofing language and no-proof choices (#2717); previously trivial: progress log: web preferences kept between visits, #2132; previously trivial: progress log: undrawable private-use list bullets show the level's default bullet, #1683; previously trivial: progress log: `w:eastAsianLayout` kept on save; previously trivial: table style editor formats every part, #714); previously trivial: VML freeforms read and kept on save; previously trivial: progress log: consistent preferred cell widths; previously trivial: multi-paragraph field instructions no longer add blank lines; previously trivial: page-break-only list paragraphs hide their labels; previously trivial: progress log: grouped `flowChartOr` shapes draw their circle and crossing spokes; previously trivial: progress log: copies keep their formatting in other programs, #1523; previously trivial: progress log: privacy options on save, #1062; previously trivial: progress log: equation pictures in theme colours on Windows, #1002; previously trivial: symbol fonts, #776; previously trivial: the save prompt answers from the keyboard, #861; previously trivial: table column and row markers on the rulers, #700; previously trivial: table borders and the resize handle drag on the page, #217; previously trivial: formatting revisions landed, #41; previously trivial: Draw tab ink, #307, and Draw Table and Eraser, #303, landed: catalog 402/434; previously minor: readiness table with hours per audience; full number back to the additive weighted sum, 60%; mainstream and essentials numbers added; alpha gate checked; previously major: full re-measure against Word for Mac 16.113.4; restructured to the craftrules progress-docs standard) · **Target:** Microsoft Word (Microsoft 365) for Mac 16.113.4
> **Last reviewed:** 2026-10-10 · **Last updated:** 2026-10-18 · **Change:** trivial (long navigation headings wrap onto further lines within the pane, #3015; previously trivial: clearing a no-match font query keeps the recent heading visible, #2068; previously trivial: a word after a slash wraps whole, like Word, #2963; previously trivial: RTF export writes sections (gap 19); previously trivial: ODT export writes sections (gap 19); previously trivial: progress log: pasted blockquotes are indented paragraphs, #1270; previously trivial: progress log: print-area strings translated in every catalog; previously trivial: progress log: comments readable in the dark theme, #2045; previously trivial: progress log: ODT sections, #1379; previously trivial: progress log: table deletes, Change Case and Unlink Fields respect Track Changes and locked content controls, #1329; previously trivial: progress log: rounded rectangle corners, #1476; previously trivial: progress log: Save As type list, #1399; previously trivial: progress log: privacy options on save, #1062; previously trivial: progress log: equation pictures in theme colours on Windows, #1002; previously trivial: symbol fonts, #776; previously trivial: the save prompt answers from the keyboard, #861; previously trivial: table column and row markers on the rulers, #700; previously trivial: table borders and the resize handle drag on the page, #217; previously trivial: formatting revisions landed, #41; previously trivial: Draw tab ink, #307, and Draw Table and Eraser, #303, landed: catalog 402/434; previously minor: readiness table with hours per audience; full number back to the additive weighted sum, 60%; mainstream and essentials numbers added; alpha gate checked; previously major: full re-measure against Word for Mac 16.113.4; restructured to the craftrules progress-docs standard) · **Target:** Microsoft Word (Microsoft 365) for Mac 16.113.4
> **Printing update:** 2026-10-12 · direct desktop printing through craft-print; details below.
WordCraft aims for complete parity with Microsoft Word, then goes further on speed, openness and
@@ -15,7 +15,7 @@ agent control. This page is the summary; the assessment is
## Headline numbers (2026-10-10)
| Number | Value | Kind |
| 2026-10-12 | Table formulas: cell references and ranges (`=B2+B3`, `=SUM(B2:C3)`), arithmetic and comparisons, IF, AND, OR, NOT, TRUE, FALSE, DEFINED and the `\#` number format; ABOVE and LEFT sum the run of numbers next to the cell as Word does; an unknown function or a syntax error is an error instead of a sum; Update Field (F9) recalculates formula fields |
| --- | --- | --- |
| Ribbon/menu catalog coverage | **487 / 502 (97.0%)** | measured (`cargo xtask parity` → [`docs/parity-checklist.md`](docs/parity-checklist.md)) |
| **Feature breadth** (weighted, incl. dialog options, styles, languages beyond the ribbon) | **~80%** | estimated |
| **Ready for real work** (full target) | **~60%** (55–63%) | estimated, additive weighted sum over the dimensions |
@@ -104,7 +104,7 @@ Weights and evidence per area: [docs/target-app-parity.md](docs/target-app-parit
| Spanish | es | partial | ~95% |
| Hindi | hi | none | 0% |
| Arabic | ar | none (document shaping and bidi work; no mirrored UI) | 0% |
| French | fr | partial (interface) | ~95% |
| French | fr | partial (interface complete, native-speaker reviewed; no proofing) | 100% (2,615 entries) |
| Portuguese | pt-br (pt-PT none) | partial | ~95% |
| Indonesian | id | none | 0% |
| Japanese | ja | partial | ~95% |
@@ -116,7 +116,7 @@ Also shipped: Traditional Chinese (`zh-hant`), Ukrainian (`uk`), Serbian (`sr`,
system-locale selection and a saved language choice (Italian also has 51 strings none of the others had yet); Estonian, Turkish and Italian proofing are not included. Hungarian (`hu`) has a
complete 2,419-entry catalog, also without proofing. Vietnamese (`vi`, #1434) has a complete
2,448-entry catalog, also without proofing.
Proofing: English, Polish, Portuguese and Russian (spelling, grammar, hyphenation), plus Czech (system Hunspell spelling, generic checks), chosen per run from its language. Word ships
Proofing: English, Polish, Portuguese and Russian (spelling, grammar, hyphenation), plus Czech and Indonesian (system Hunspell spelling, generic checks), chosen per run from its language. Word ships
30 interface languages and proofing for ~50. Detail: [docs/localization-parity.md](docs/localization-parity.md).
## Upcoming
@@ -134,20 +134,115 @@ Ranked, to beta (detail and remaining milestones in [docs/roadmap.md](docs/roadm
## Progress log
Imported Word 97–2003 `.doc`/`.dot` files now ask for a supported DOCX target on interactive Save (#2445). Save before Close waits for that write; cancel or failure keeps the draft and original file. This does not add binary Word output. Late Save As answers are rejected before writing if automation has replaced the document.
- 2026-10-19: Restrict Editing: with Filling in forms enforced, inserting content controls, their Properties and Remove Content Control are refused, as in Word (it greys them out), so a form's fields stay as they are and only their content is filled in.
- 2026-10-19: Content controls: text typed over a placeholder takes the control's own run properties (`w:sdtPr/w:rPr`, ECMA-376 §17.5.2.27), as in Word, and the placeholder style is recognised by its built-in name "Placeholder Text" whatever its localized id (German Word: `Platzhaltertext`), so entries in forms from localized Word no longer stay grey.
- 2026-10-15: Find opens Navigation on Results; editing the query restores Results after manual Headings/Pages navigation. Typed and selected-text searches retain document contents (#2096).
| Date | What landed |
| 2026-10-19 | Save As, Export and Open no longer do nothing on Linux desktops whose file chooser portal fails to show its dialog (#2999): a "cancel" quicker than anyone could make one asks again without the parent window, then opens the built-in file browser. |
| 2026-10-18 | Long navigation headings wrap onto further lines inside the resizable pane instead of ending in an ellipsis: the row grows with the text and the pane keeps its width (#759, #3015). |
| 2026-10-18 | Clearing a font query with no matches keeps Recently Used Fonts visible at the top of a scrollable list. The initial scroll position no longer skips its heading; measured keyboard scrolling is unchanged (#2068). |
| 2026-10-17 | Edit Recipient List finds records (#1576): a Find in list box hides rows with no cell containing the text (any case) and counts the matches, and "Show only records that will be merged" hides the rows the filter leaves out; both only change the view. Rows are striped, and edits and removes go to the right record however the grid is narrowed |
| 2026-10-14 | DOCX row table border exceptions render and survive saving. Explicit cell borders win; a row exception does not change other rows. |
| 2026-10-14 | Tab stops edited on the ruler: the corner selector picks Left/Center/Right/Decimal/Bar, a click sets a stop (Alt skips the snap), dragging moves it with a guide line and position tooltip, dragging off the ruler clears it; markers draw per kind and mirror for RTL; one `para.tabs` call and one undo step per gesture |
| 2026-10-12 | Fields keep `w:dirty` through DOCX saves (#877): a begin marker flagged by a generator that couldn't compute a result (pandoc-style TOC placeholders) keeps the flag on open/save — as a buffered field or a range field — so Word still refreshes it on open; computing a result (Update Field, Update Table) clears it, like Word's F9 |
| 2026-10-18 | Symbols keep their font (#2036): a `w:sym` or Insert › Symbol character in Wingdings, Symbol or Webdings remembers its symbol font apart from the text's font, so selecting all and changing the font (or Clear Formatting) no longer turns the symbols into boxes; saved back to DOCX as `w:sym`, so Word shows them too. Typed text and the Format Painter don't pick up the symbol font. |
| 2026-10-16 | `wordcraft-cli run --script FILE.json` (`-` for stdin) runs a JSON list of commands, then saves to one or more outputs (docx and pdf in one run); errors name the command number and id and nothing is saved, #1963 |
| 2026-10-17 | Restrict Editing holds for structure too (#1327): an exception inside a table cell no longer lets table, row or cell formatting change the protected rest of the table; formatting limited to styles refuses Paste Formatting, table shading, borders, cell alignment, cell margins, text direction, table-style edits and locked table styles; loading, filtering or sorting recipients no longer rewrites merge-field results in a protected document. |
| 2026-10-18 | Manage Sources › Import… (#2640) reads a Word bibliography file (`b:Sources`: Word's own `Sources.xml` master list) into the document: the tags, names, kinds and fields Word saves are read by the same code that reads the bibliography custom XML part of a `.docx`, a tag already in the document is left alone (an import never overwrites an edited source), an untagged source gets a tag the way a new one does, the whole import is one undo step, and a file that isn't `b:Sources` is refused. |
| 2026-10-14 | RTF export writes every section (gap 19): `\sect` ends each section and `\sectd` carries its break type, page size and orientation, margins, columns, page numbering and title page, so multi-section documents survive an RTF round trip |
| 2026-10-14 | `ASK` and `SET` fields no longer show their stored result on the page or in PDFs (#1765): like Word, they only assign text to a bookmark and draw nothing; the stored result is kept and saved |
| 2026-10-15 | Table rows with an at-least height start the next page whole when less than their minimum is left on a page that already holds something, so one-row-per-page booklets keep their pages; rows taller than a page split from the top of the next page and repeated header rows go with the row (#1742) |
| 2026-10-16 | Strict OOXML line heights render as Word renders them (#2329): `w:line="13.90pt"` under `lineRule="auto"` is a height read through twips (13.90 × 20 = 278; the 1.1583 multiple Word shows), not the 13.9 multiple the bare transitional number counts, so a Word 2024 file with 13.9 lines of line spacing opens at Word's 1 page with matching baselines instead of 10 |
| 2026-10-15 | RTF keeps more tracked changes (gap 19): tracked paragraph marks (a `\par` inside a `\revised`/`\deleted` group) and character and paragraph formatting changes (`\crauth`/`\crdate` with `\oldcprops`, `\prauth`/`\prdate` with `\oldpprops`) are read and written, so their balloons and Accept/Reject survive an RTF round trip |
| 2026-10-14 | ODT export writes sections (gap 19): every new-page section gets its own page layout and master page (the first stays "Standard"), named by its first paragraph or table with the page-numbering restart; a continuous section with columns is written as a columned `text:section`, so a landscape, two-column or restarted part survives a save as .odt |
| 2026-10-14 | A bulleted or numbered list item that starts with a lower-case word no longer gets the grammar squiggle for a missing sentence capital (#1741): list items are often fragments, so the sentence-capital rule (every proofed language) skips the first word of a list paragraph (own or style numbering); later sentences in the item are still checked |
| 2026-10-15 | The caret, clicks and selection edges fall between the letters of a ligature (fi, ffi, lam-alef), as in Word: the glyph's advance is shared evenly among its letters, mirrored in right-to-left text (#1974) |
| 2026-10-16 | The Spike, as in Word: Ctrl+F3 (⌘F3 on macOS) cuts the selection and adds it to the end of the Spike, each piece followed by a paragraph mark; Ctrl+Shift+F3 (⌘⇧F3) inserts everything collected, in order and with its formatting, and empties the Spike. The clipboard is left alone; each is one undo step. The Spike is the AutoText entry "Spike", so Insert › Quick Parts inserts it without emptying it. It lasts for the session (Word keeps it in Normal.dotm) |
| 2026-10-16 | Update Field is offered on a table of contents after a save and reopen, and repeated save and update adds no blank line (#2182; the flow in that report already worked since #1197) |
| 2026-10-16 | DOCX two lines in one and horizontal in vertical (`w:eastAsianLayout`: id, combine, brackets, vert, vertCompress) are read and kept on save, in runs and character styles, instead of being dropped. Layout doesn't apply them yet. |
| 2026-10-16 | Read Aloud speaks Russian, Ukrainian, Greek, Arabic, Hebrew, CJK and Thai text with a voice for its language instead of spelling out letters (#1843): the sentence's language comes from its run's tag (`w:bidi` for right-to-left text) or, when that disagrees with the script, from its letters, even when English letters outnumber them; Windows reads the text as UTF-8 and picks an installed voice of that language, macOS `say -v` and Linux `espeak -v` get the same language. The user's own voice stays whenever it already speaks the language (English documents read exactly as before), and a voice espeak lacks falls back to the default voice instead of silence. Not run on Windows or macOS here. |
| 2026-10-16 | Hint-reliant CJK fonts execute TrueType contour assembly; Windows DFKai-SB no longer loses strokes. Synthetic local-font regression; no font binaries bundled. |
| 2026-10-16 | Ctrl/⌘+wheel only zooms: a coast or wheel notch still gliding no longer scrolls the page on top of the zoom, which keeps the page point under the pointer (#1743). |
| 2026-10-15 | AutoCorrect follows Turkish and Azerbaijani case mapping: `istanbul` → `İstanbul`, `ılık` → `Ilık`, `İNgilizce` → `İngilizce` with no combining dot (#1923) |
| 2026-10-16 | Releasing a table drag with the mouse (column or row border, move handle, selection areas) leaves the view where it is instead of scrolling back to the caret (#1738) |
| 2026-10-16 | Hebrew in fields and quotes (#1670): a STYLEREF (or any field) showing right-to-left text such as "פרק א" reads right to left in headers and the body instead of reversed; smart quotes in Hebrew pinned to Word's logical-order result (“ before, ” after, ” in acronyms like צה"ל) |
| 2026-10-15 | Backspace with a whole table selected removes it, as in Word, and selected whole rows or columns go the same way; Delete clears them and keeps the table. A selection that covers a table and the paragraph before or after it removes the table, also when it starts inside the table or the table is nested (#1844) |
| 2026-10-15 | Image size limits are checked before allocating (#1321): a picture whose header declares more RGBA bytes than the 512 MiB budget is refused before decoding (render and PDF share one helper in `wordcraft-geom`), and page rasters are capped at 100 million pixels in total, scaled down rather than only per side. |
| 2026-10-15 | Saved DOCX files keep the order of mixed right-to-left and left-to-right text in Word: Hebrew around a Latin name, numbers, brackets or an e-mail address no longer changes places, and Word reports the Hebrew as Hebrew rather than Arabic (#1543). |
| 2026-10-15 | Web: typing in Cyrillic or another non-Latin script no longer repeats text. WordCraft now honours the input method's request to erase around the caret before retyping (#1480). Chrome can still move the hidden text field's caret, so later letters may still land out of order. That cause is still open |
| 2026-10-15 | Equations: typing `V_dd` or `x^ab` and pressing Space keeps `dd` and `ab` whole in the subscript or superscript, like `V_(dd)`, as Word for Mac does; Linear format shows the bare form |
| 2026-10-15 | Plain text pasted from another program keeps the formatting of the paragraph it lands in: Paste inserts it as Keep Text Only does, so every pasted line, empty lines too, becomes a paragraph with that paragraph's direction, style, alignment, indents and list, and the text after the caret keeps them. The lines after the first had default formatting, so pasting several lines into a right-to-left paragraph turned all but the first left-to-right |
| 2026-10-14 | Add Font… imports every file selected in the file chooser (#1789), not just the first: the desktop dialog and the web picker are multi-select, each file is added, a bad file is reported ("Added 11 fonts; 1 couldn't be read: X.ttf") without stopping the rest, and one pick adds at most 2,000 (the status line counts the skipped). On the web each font is added and reported on its own, with no single summary line. The built-in file browser still picks one file. |
| 2026-10-15 | Additive paragraph spacing retains the full before-space after a section page break; ordinary overlapping spacing is unchanged. |
| 2026-10-14 | DOCX row table border exceptions render and survive saving. Explicit cell borders win; a row exception does not change other rows. |
| 2026-10-14 | Tab stops edited on the ruler: the corner selector picks Left/Center/Right/Decimal/Bar, a click sets a stop (Alt skips the snap), dragging moves it with a guide line and position tooltip, dragging off the ruler clears it; markers draw per kind and mirror for RTL; one `para.tabs` call and one undo step per gesture |
| 2026-10-12 | Fields keep `w:dirty` through DOCX saves (#877): a begin marker flagged by a generator that couldn't compute a result (pandoc-style TOC placeholders) keeps the flag on open/save — as a buffered field or a range field — so Word still refreshes it on open; computing a result (Update Field, Update Table) clears it, like Word's F9 |
| 2026-10-15 | Exact line spacing keeps the run font baseline around tall equations; automatic and minimum spacing still grow for math. |
| 2026-10-16 | Zero-extent drawing anchors contribute no inline font height. |
| 2026-10-16 | Control-channel keyboard and text input reaches the focused widget while the native window is inactive, without activating it; normal widget targets, dialogs and document protection remain in force (#2420) |
| 2026-10-15 | Hide Quick Access Toolbar (#1945): the toolbar's menu ends with Hide Quick Access Toolbar, as Word's does; the ribbon's right-click menu and Customize Ribbon › Quick Access Toolbar offer Show Quick Access Toolbar; kept between runs; `ui.quickAccessToolbar {"value"?}` for scripts and MCP |
| 2026-10-15 | Rulers: the numbers thin out as the page shrinks on screen (every 1, 2, 4… units, at least 30 px apart, the small ticks at least 4 px apart), give way to the indent, tab and column markers, and the margins get ticks only; both rulers span the caret's page at its own scale in One Page and Multiple Pages and follow the caret to another page (#1878) |
| 2026-10-15 | Spelling as you type no longer flags the word being typed; the squiggle comes when the word ends or the caret leaves it, as in Word (#1552) |
| 2026-10-15 | Text Highlight Color: No Color becomes the Highlight button's choice, as in Word, so the next click on the button removes highlighting instead of applying the colour picked before; the ribbon and the mini toolbar share one menu (#2105) |
| 2026-10-16 | Footnote and endnote editing keeps the body and active note text at normal opacity (#2545), without changing document colours or print/PDF output. |
| 2026-10-16 | A picture watermark in a header that starts with a table is drawn centred on the page behind the body text, on every page, instead of in the header's cell at the top of the page (#1775); `layoutInCell="0"` and VML `o:allowincell="f"` survive save and reopen |
| 2026-10-17 | Ribbon hover hints: a disabled control says why (the command's own reason), split-button arrows, menu buttons and the font, size and outline boxes have hints, and hints show on keyboard focus (#1086). |
| 2026-10-17 | Insert › Drop Cap is a menu (None, Dropped, In margin, Drop Cap Options…); the dialog sets position, font, lines to drop and distance; letters in the margin; DOCX keeps them (#2601). |
| 2026-10-16 | An otherwise empty section starting a new page retains its paragraph line before a continuous section, matching an original native Word control. |
| 2026-10-16 | Equations from Word files keep all their content (#1273): math straight in the body or a table cell, math behind `mc:AlternateContent Requires="m"`, runs inside hyperlinks, simple fields and markup-compatibility blocks within an equation, and `w:sym`/Symbol-font characters in math; operators, delimiters and accents the math font lacks come from a fallback font. Every ECMA-376 §22.1 element is pinned by a coverage test. |
| 2026-10-16 | Captions and title pages (#1230): Insert Caption… on the right-click menu of pictures, shapes, text boxes and charts, the caption going under (or over) the object in its own cell or text box; Page Setup's Different First Page and Different Odd & Even Pages check boxes, so a title page can go without its page number |
| 2026-10-17 | SECTIONPAGES includes shared continuous-section pages. A native original control confirms a section starting on page 1 and continuing on page 2 reports 2. |
| 2026-10-17 | File › Print's Scale to paper size (#2108): pick Letter, Legal, A4 or any Layout › Size paper and every page is scaled to it and centred, the paper turned to each page's orientation, in the preview, the print job (with the spooler's media) and Save these pages as PDF; the document keeps its page size. `paper` param on `ui.print`/`file.printPdf` |
| 2026-10-17 | Non-overlapping body floating tables clear header tables. Original native controls distinguish never-overlap from allowed overlap. |
| 2026-10-14 | Comment text being typed is no longer lost or misplaced (#1348): an edit or reply the document refuses (for example read-only protection) keeps the typed text in its field and says why on the status line; drafts belong to their document, so after New or Open a comment with the same number shows its own text; Save, Export, Print, Close, New and Open write pending comment text first; if the document refuses it, Save, Export and Print carry on with the stored comments and show why on the status line (a refused draft is dropped when Save is the only change), while Close, New and Open show the Save Changes prompt (Don't Save drops text the document refuses); Escape discards what was typed; in a document protected as read only or for forms the comment fields can't be typed in |
|---|---|
| 2026-10-18 | Picture Format › Size (#2573): width and height in the interface unit, and a size typed in any unit works, pixels too (`300 px` keeps the picture's proportions, #1901), applied once typed (Enter, Tab or a click elsewhere; Esc cancels) in one undo step. The group's launcher opens a Size dialog: height and width, scale of the original size (or of the size it opened with), Lock aspect ratio, the original size with its pixels, and Reset. `picture.size {"original": true, "scale"?}` sizes from the original; sizes past 4000 pt that keep the proportions shrink as a whole. |
| 2026-10-16 | Save Changes gives assistive technology a named alert dialog with the loss warning and any save failure; visual layout and save policy unchanged (#1015). Native screen-reader verification remains pending. |
| 2026-10-15 | The font family caches are bounded (4096 families for text, 1024 for math, names over 256 bytes aren't remembered), so a document naming thousands of missing fonts can't grow memory; equation fonts are looked up again after a font is added or removed instead of keeping the old choice (#1354) |
| 2026-10-15 | AutoCorrect recognizes `-->` as `→` (#2223), respecting replacement settings and undo without changing paragraph-start arrow bullets. |
| 2026-10-15 | Numbered headings (#1793): the Multilevel List gallery offers four lists linked to Heading 1–9 (1 / 1.1 / 1.1.1, Article I. / Section 1.01 / (a), I. / A. / 1., Chapter 1; `list.multilevelPreset`). The heading styles carry the numbers, so paragraphs already in a Heading style keep their style and level, any other selected paragraph (Normal, Quote, Title) becomes Heading 1 (checked in Word), a new Heading 2 is numbered 2.1 by itself and body text stays unnumbered; Numbering or typing `1. ` keeps a heading a heading |
| 2026-10-15 | Reversing an automatic sentence capital (Undo, or Backspace and retype) adds the word before it to AutoCorrect's First Letter exceptions, as Word does; "Automatically add words to list" switches for First Letter and INitial CAps (#1857). |
| 2026-10-15 | Dialog close control: a Windows-style box at the title bar's right end, the whole box clickable, red on hover, in both themes (#1885) |
| 2026-10-15 | Font box: a Recently Used Fonts section at the top of the list (the fonts applied lately, most recent first, ten at most, kept between runs), then All Fonts, as in Word (#2068) |
| 2026-10-15 | A missing Thai document font (TH Sarabun New, Angsana New, Cordia New, the UPC families…) is substituted by one Thai face for the whole run, Latin letters and digits included, as Word substitutes Cordia New; Thai no longer sits in one font next to larger Latin in another (#1549) |
|---|---|
| 2026-10-17 | Paste from LibreOffice: a paragraph indented on both sides (HTML `<blockquote>` on the clipboard) pastes as an indented Normal paragraph with LibreOffice's margins, not centred and italic in the Quote style (#1270). |
| 2026-10-17 | DrawingML shape outlines retain explicit preset/custom dashes on screen and in PDF, including zero-width path strokes. |
| 2026-10-17 | Missing Arial Unicode MS tries Arial and its proportional substitutes before code-font heuristics. |
| 2026-10-16 | Web Layout text fills the window width and re-wraps once a resize settles, or at once on a zoom, instead of a fixed 800 pt column; Draft and Outline keep their 800 pt column (#1876). |
| 2026-10-16 | Create Source: the automatic Tag is three letters of the first author and the last two digits of the year (`Smi20`), with 1, 2… added when taken, as Word for Mac 16.113.4 does (user-checked). Editing Author or Year keeps the Tag, as in Word, and the citations in the text show the new author and year (#2094; the reporter expected the Tag to follow). |
| 2026-10-18 | Split Cells and Convert Text to Table dialogs (#2592), matched to Word: Split Cells starts from the selection (twice its columns, its rows; Merge cells before split offered for several cells or one cell merged across columns); `table.split` gains `rows` and `mergeFirst`, a merged cell splits into 1 to its height rows (its height cuts it into rows, 1 × 1 changes nothing, a number that doesn't divide the height or more than it is refused with Word's messages) and column splits make equal parts without moving other rows' grid lines. Convert Text to Table guesses columns live, shows the row count, and offers Initial column width (Auto or up to 22 in, kept as each column's preferred width), AutoFit to contents/window and Paragraphs/Commas/Tabs/Other separators; `table.fromText` gains `paragraph`/`other` + `char`, `columns`, `autofit` and `width`. |
| 2026-10-16 | Font list entries of fonts with a very tall ascent (Amiri, Amiri Quran) sit in their row like the others instead of low or cut in half (#2570) |
| 2026-10-16 | Read Mode draws documents cleanly (#1772): the arrows, Page Up/Down and Escape work as soon as it opens (the canvas kept a stable widget id, so it no longer lost the keyboard), only whole columns show, centred, at any column width, text size or window size, and the columns end above the page-turn bar instead of under it. |
| 2026-10-16 | Comment and revision balloons are readable in the dark theme (#2045): they were drawn on a fixed white card under light text; the card, markup area and author colours are now theme tokens with a contrast test. |
| 2026-10-17 | DOCX glossary documents (building blocks) open into the model: Insert › Quick Parts and AutoText list and insert the document's own entries, and saving an entry with `"in": "document"` keeps it in the file (undoable). An untouched glossary is saved back byte for byte; an edited one is written from the model |
| 2026-10-16 | Indonesian proofing (#498): id-ID spelling reads the installed Hunspell `id_ID` dictionary (`hunspell-id` on Debian/Ubuntu, LibreOffice's `dict-id`, or the Dictionaries folder), suggestions use its MAP/REP/TRY/KEY tables, generic repeated-word (offering `anak-anak`) and spacing checks explain in Indonesian; Review › Language, Set Proofing Language and File › Options › Proofing offer Indonesian. No dictionary data bundled; hyphenation keeps the English fallback; parity estimates unchanged |
| 2026-10-17 | A table inserted in a right-to-left paragraph is a right-to-left table, like Word, so Tab moves right to left through its cells (#2634) |
| 2026-10-16 | French interface catalog completed: 402 entries added (equation gallery, built-in style and table-style names, citation styles, paper sizes, watermarks, Bookmark / Go To / Print dialogs), 2,615 in all, with a native-speaker review of the terminology; no French proofing yet |
| 2026-10-14 | DOCX row table border exceptions render and survive saving. Explicit cell borders win; a row exception does not change other rows. |
| 2026-10-15 | Layout › Columns with text selected puts only that text in columns, as in Word (#2024): a continuous section break goes where the selection starts and one where it ends (splitting a paragraph there; none at the document's start or end, or where a section already begins or ends), and the columns go on the section between, one undo step; More Columns… offers "Selected text" and picks it. Before, the whole section took the columns |
| 2026-10-14 | Tab stops edited on the ruler: the corner selector picks Left/Center/Right/Decimal/Bar, a click sets a stop (Alt skips the snap), dragging moves it with a guide line and position tooltip, dragging off the ruler clears it; markers draw per kind and mirror for RTL; one `para.tabs` call and one undo step per gesture |
| 2026-10-12 | Table formulas: cell references and ranges (`=B2+B3`, `=SUM(B2:C3)`), arithmetic and comparisons, IF, AND, OR, NOT, TRUE, FALSE, DEFINED and the `\#` number format; ABOVE and LEFT sum the run of numbers next to the cell as Word does; an unknown function or a syntax error is an error instead of a sum; Update Field (F9) recalculates formula fields |
| 2026-10-12 | Fields keep `w:dirty` through DOCX saves (#877): a begin marker flagged by a generator that couldn't compute a result (pandoc-style TOC placeholders) keeps the flag on open/save — as a buffered field or a range field — so Word still refreshes it on open; computing a result (Update Field, Update Table) clears it, like Word's F9 |
| 2026-10-18 | Different First Page works from page 1 (#2826): Insert › Header and Edit Header with the caret on page 1, and a double-click there while another header is being edited, go into page 1's own header instead of the default one page 1 no longer shows. |
| 2026-10-15 | Track Changes Options › Moves (gap #5): Moved from and Moved to each have a mark and a colour (by default green double strikethrough and double underline, as before and as in Word for Mac), saved per user; `review.trackingOptions` takes `moveFromMark`, `moveFromColor`, `moveToMark`, `moveToColor`, and `trackMoves`, the document's Track moves setting: off, a tracked cut and paste is an ordinary deletion and insertion. DOCX `w:doNotTrackMoves` round-trips |
| 2026-10-17 | Language dialog keeps inherited proofing language and no-proof choices (#2717). Opening and accepting unchanged no longer replaces style-derived values with English US and proofing enabled. |
| 2026-10-16 | Text with the legacy Outline, Shadow, Emboss or Engrave font effects (`w:outline`, `w:shadow`, `w:emboss`, `w:imprint`) is drawn with them on screen and in PDFs, no longer as plain text: hollow outlined letters, or a grey copy behind the letters; the PDF keeps the text (#2534) |
| 2026-10-15 | Table Layout › Cell Margins opens an editable dialog initialized from the table defaults; confirming an unchanged dialog preserves imported margins, and changing one side retains the others (#1960) |
| 2026-10-15 | Word's native ink inside groups and drawing canvases survives DOCX saves (gap 2): the ink member keeps its InkML part, relationship, content type and picture fallback, stays in its place among the members, moves with the group and is drawn from its fallback picture; deleting the group drops its part |
| 2026-10-14 | Consistent preferred cell widths override a stale table grid when both totals agree, for fixed and AutoFit tables. Conflicting widths, spans and percentages remain unchanged. |
| 2026-10-14 | Legacy VML freeforms (`v:shape` with its own `path`), such as the rule lines in many older headers, are read as freeform shapes: they show on screen and in PDFs and are saved back as DrawingML custom geometry instead of vanishing |
| 2026-10-14 | Field instructions spanning paragraphs no longer add blank lines during pagination; empty result paragraphs still take a line. |
| 2026-10-16 | Opening a DOCX no longer copies its pictures: each picture, and each embedded object or macro part kept for saving, was copied out of the unzipped package, so a document with pictures held them twice while it opened and left the package's copies behind as freed memory. The document now shares the package's bytes: a 92 MB document with 20 pictures opens in about 90 MB less (866 → 772 MB in the app on macOS; the command line's peak, 319 → 231 MB) |
| 2026-10-14 | Page-break-only list paragraphs no longer print stray bullets or numbers. The next item keeps its number, matching the reference export. |
| 2026-10-12 | Backspace at the start of the paragraph after a table nested in a cell moves into the nested table's last cell, no longer into the last cell of the table around it |
| 2026-10-14 | Grouped `flowChartOr` shapes draw their circle and crossing spokes, as in Word |
| 2026-10-14 | Deleting a comment removes its anchors from headers, footers, notes and text boxes too, not just the body, and deletes the whole thread below it, replies to replies included; no anchor or reply is left pointing at a deleted comment (#1345) |
| 2026-10-14 | Legacy DOC DIB conversion checks the BMP header and copy before allocation, uses the caller's decode allowance, and caps PNG output while encoding; a BMP fallback is kept when it fits. A generated 1-pixel regression covers 57/58/512-byte limits. Resource-exhaustion consequences remain unmeasured. Related: #1334 and #1554 |
| 2026-10-15 | Draw: Ink to Shape makes two-stroke arrows and arrows with a closed triangle head. A line plus a separate head stroke (an open V or a small closed triangle, drawn after the line or before it) whose apex sits at one of the line's ends becomes one straight arrow connector, and the head stroke goes, in one undo step; this works while drawing with Ink to Shape on and on lasso-selected strokes with Convert to Shapes. A shaft with a small triangle drawn round its tip in the same stroke is an arrow too; a triangle without a shaft stays a triangle. Our own design (Word for Mac has no Ink to Shape). |
| 2026-10-13 | Justified vertical page alignment (#1320): a section set to Justified spreads each page's paragraphs and tables from the top margin to the bottom one, as Word does, including the last page and pages ending in a manual page break. DOCX `w:vAlign="both"` opens as Justified (it showed centred) and saves back; `layout.verticalAlign` takes `justify`, and Page Setup has a Vertical alignment box |
| 2026-10-13 | macOS shortcuts as in Word for Mac: ⌘⇧S Save As, ⌘⌥E Insert Endnote, ⌘⌥A New Comment, ⌃H Replace, ⌘⌥R collapse the ribbon |
| 2026-10-13 | The status bar flags a proofing language whose Hunspell dictionary isn't installed (Polish, Portuguese): "Polish (no dictionary)", with a tooltip naming the dictionary and where it is looked up (#1167) |
@@ -159,6 +254,7 @@ Ranked, to beta (detail and remaining milestones in [docs/roadmap.md](docs/roadm
| 2026-10-13 | Cut copies to the system clipboard however it is run (#1265): the ribbon, the context menu and the native menu cut deleted the words without copying them (only the Cmd+X event wrote the clipboard, so Paste re-inserted the previous copy); Copy/Cut publication is centralised in one sink and every entry point is tested |
| 2026-10-13 | Insert › Bookmark deletes bookmarks (#1236): the dialog is laid out as Word's, with Add, Delete and Go To beside the list, Sort by Name or Location and Hidden bookmarks; it opens on the bookmark around the caret, Delete and Go To keep it open, and Go To selects the bookmark's text. `insert.deleteBookmark {name}` removes both ends of a bookmark, also one across paragraphs (adding a bookmark under a name in use left such a one's end behind) |
| 2026-10-12 | Password-protected Word 97-2003 `.doc` files open after a password prompt (XOR obfuscation, RC4, RC4 CryptoAPI; wrong passwords are rejected by the stored verifier, #1079); written from the public specs and tested on synthetic files only, AES-encrypted `.doc` files are still refused |
| 2026-10-14 | Update Field works inside headers, footers, notes and text boxes, on the part being edited, as in Word (#1706); with the caret in the body it still leaves them alone, and printing updates every field in the document |
| 2026-10-12 | Comments can show only for the selected text (#1055): Track Changes Options › Comments › "Only for the selected text" drops the margin column, and the comment under the caret, or the clicked one, floats beside the page (over its edge in a narrow window). The default stays "In the margin", as in Word; `review.trackingOptions {"commentDisplay"}` for agents |
| 2026-10-12 | Clicking inside a selected comment's balloon moves the caret in its text (#1013); the click used to leave the balloon, so the caret could only be moved with the arrow keys. A press on any balloon now belongs to the balloon, not the page behind it |
| 2026-10-12 | Page background pictures and textures (Design › Page Color › Fill Effects) are drawn (tiled or stretched) and kept on save instead of being dropped from the package (#999); PDF export paints them too |
@@ -191,6 +287,7 @@ Ranked, to beta (detail and remaining milestones in [docs/roadmap.md](docs/roadm
| 2026-10-13 | Interface size means what it says (#822): a chosen size is absolute, as in PhotoCraft and PdfCraft, so "100%" is one screen pixel per point even where the system reports 200 % (a bare window manager on X11, where winit guesses the scale from the monitor's DPI). The new default, Auto, keeps the system's display scale and its menu entry shows the size it draws at; sizes run 75–300 %. A saved size from before stays the same if it was the default (now Auto) and becomes absolute otherwise |
| 2026-10-14 | Copied text keeps its formatting in other programs (#1523): Copy and Cut put HTML beside the plain text on the system clipboard for every selection, not only one holding an equation, so a browser, mail client or another word processor pastes bold, italic, fonts and sizes. RTF on the clipboard is not written yet |
| 2026-10-13 | Copying an equation also puts MathML in HTML on the clipboard beside the linear text like `y=2x_s` (#816; Word/LibreOffice paste not yet verified; OMML is #1297). HTML export writes equations as MathML too, and opening or pasting HTML with MathML gives equations back |
| 2026-10-16 | Copying one equation on its own also puts it in the clipboard's MathML formats, which Microsoft documents Word reading first when it pastes math. Whether it pastes into Word as an editable equation hasn't been confirmed yet; testers welcome (#1297) |
| 2026-10-13 | The Equation tab's galleries, structure buttons and symbols keep the theme's colours on Windows: their pictures share a few textures written in place (no wipe of pictures already on screen, room for the built-ins at high display scales), instead of hundreds of small textures freed in bulk, which on DirectX 12 showed inverted, red or teal tiles (#1002) |
| 2026-10-13 | `file.close` over the control channel or MCP bridge no longer quits and discards unsaved changes (#1268): it returns an error unless the document is saved or `{"force": true}` is passed; the user's Save / Don't Save / Cancel prompt is unchanged |
| 2026-10-14 | Text copied out of an exported PDF reads as shown (#1473): All Caps and Small Caps text comes out in capitals, as from Word's PDFs, and plain text after it keeps its own case (before, a capital first drawn for a lowercase letter made later capitals extract lowercase) |
@@ -204,6 +301,7 @@ Ranked, to beta (detail and remaining milestones in [docs/roadmap.md](docs/roadm
| 2026-10-14 | Hebrew and Arabic in exported PDFs copy and search as whole words (#1383): right-to-left glyphs are written left to right as they appear, the order Chrome and Edge (PDFium) and poppler read text back from (checked by hand), instead of in reading order, which they reversed (`שלום` came back as `םולש`, Arabic in pieces). Lam-alef stays whole and vowel marks follow their letters, though Chrome and poppler can still break a fully vowelled word with spaces. The page looks the same. Readers that take glyphs strictly in file order get only plain Hebrew runs right, through `/ReversedChars`; screen readers are unverified |
| 2026-10-14 | HTML import (#1381): header cells (`<th>`) are bold unless their own CSS says otherwise, and `text-align` on a cell, its row, row group or table aligns the cell's paragraphs (a paragraph's own alignment wins), so `wordcraft-cli convert page.html out.docx` writes the bold header and `w:jc` right-aligned cells. HTML, Markdown and plain text, which give no page size, open on the region's paper (A4 in Italy, Letter in the US) like a new blank document, in the app and in `wordcraft-cli`; documents with their own page size keep it |
| 2026-10-14 | Custom page sizes (#1433): Layout › Size ends with More Paper Sizes…, which opens Page Setup's new Paper group — a Paper size list (the named sizes, or Custom size) and Width and Height in the measurement unit; OK sets the size only when it changed, and a wider-than-tall size comes out landscape. `layout.size` now clamps each side to Word's 0.1"–22" and refuses a page the margins don't fit (the margins rule: 36 pt of text room), and Page Setup checks the size and margins together before applying either |
| 2026-10-14 | Bold text in Arial Black and other Black or Heavy fonts looks as in Word: older copies of these fonts (Arial Black in the core fonts) call themselves regular weight, so bold was drawn with an extra synthetic bold; a family named Black/Heavy now counts as that heavy |
| 2026-10-13 | Highlighted text reads on a dark page (View › Switch Modes, #1252): highlights and character shading are dimmed, keeping their hue (yellow turns a dim olive, as in Word), and the text on them is drawn light (coloured text below 4.5:1 is drawn white, and underlines and strikethroughs follow the same rule), instead of white text on bright yellow; light page, print and exports unchanged |
| 2026-10-13 | Full screen (#980): F11, or View › Window › Full Screen, fills the screen with the window (⌃⌘F on macOS); F11 again or Escape gives back the window as it was. Next Field, which Word puts on F11, keeps Word's Alt+F1 and Alt+Shift+F1 (and Shift+F11). The web app leaves full screen to the browser's own F11 |
| 2026-10-13 | Double-clicking the top or bottom margin of a page creates its header or footer when it has none and puts the caret in it, as in Word (#960): the one that page shows (different first page, odd and even pages), only in Print Layout; one left empty is taken away again when you leave it, so nothing empty is saved. `insert.editHeader`/`insert.editFooter` take `{"page"}` for agents |
@@ -284,7 +382,6 @@ Ranked, to beta (detail and remaining milestones in [docs/roadmap.md](docs/roadm
| 2026-10-12 | Symbol fonts (#776): text in Wingdings, Wingdings 2, Wingdings 3, Webdings and Symbol (proprietary, never bundled) shows its intended symbols — full code tables from Unicode's published correspondence, both storage forms (`w:sym` private-use codes and plain codes); PDF text, copy and TXT export use the Unicode characters; the saved text is unchanged |
| 2026-10-11 | Hide White Space (#645): `view.hideWhiteSpace` (View › Show checkbox, or double-click the gap between pages) collapses margins, headers and footers between pages to a thin line in Print Layout; screen only, page count unchanged |
| 2026-10-11 | Fonts: Palatino Linotype and Book Antiqua fall back to Palatino (macOS) or TeX Gyre Pagella / P052 / URW Palladio (Linux) before generic serif (#670) |
|---|---|
| 2026-10-12 | Czech proofing: cs-CZ spelling reads the installed Hunspell dictionary (`hunspell-cs` on Debian/Ubuntu), suggestions use runtime MAP/REP/TRY/KEY tables, generic repeated-word and spacing checks explain in Czech; Review › Language offers Czech/Čeština; the Czech interface writes new documents and opened TXT/Markdown in cs-CZ. Affix headers accept trailing comments. No system dictionary data or Czech hyphenation patterns bundled; parity estimates unchanged |
| 2026-10-12 | Tables: dragging the border of one selected cell (Select Cell, or a click along its left edge) moves it in that cell's row only, as in Word, like the border of cells selected across (#962) |
| 2026-10-13 | Korean interface catalog completed: all 2,463 entries translated; the ko-KR system locale and saved language choice select the interface. The UI prefers Korean embedded/system font faces so Hangul renders instead of boxes, and print settings and printer feedback are translated. Korean proofing remains unavailable; macOS Korean IME fixes are still pending (#163, #164) |
@@ -399,16 +496,98 @@ Ranked, to beta (detail and remaining milestones in [docs/roadmap.md](docs/roadm
| 2026-10-05 | M0–M12 first versions: model, layout, render, engine with 200+ commands, Word-style UI, DOCX I/O, PDF and other formats, references, review, mailings, CLI, MCP, web |
| 2026-10-12 | Mail-merge recipients (#778): legacy Excel .xls workbooks; CSV/TXT lists in GBK, Big5, Shift_JIS, UTF-16 and other encodings read correctly (plain-text import too); Edit Recipient List filters (equal, not equal, less, greater, blank, contains… joined by And/Or) and sorts, and preview, Finish & Merge and labels use only the matching records in that order (`mailings.filterRecipients`, `mailings.sortRecipients`) |
| 2026-10-13 | The interface-translation audit runs in CI again (#1128): the size labels of Envelopes, Labels and Layout › Size are translated (`4,13 × 9,5 дюйма`) and use the region's decimal separator, and status-bar messages (table split and merge, ink replay, citations, SmartArt, Researcher, Design defaults), the Reading Mode toolbar, Go To and Search's Suggested heading are translated in every interface language, and the Read Aloud player no longer shows a speech error a frame before the status bar does, the race behind the audit's rare failure under load |
| 2026-10-15 | Lists (#1683): a bullet stored as a private-use character its font can't draw (the font isn't installed) shows the level's default bullet (• ○ ▪); it used to be whatever an installed icon font has at that code, such as Font Awesome's Twitter logo |
| 2026-10-15 | The web app keeps its preferences between visits (#2132): theme (dark mode), interface size, editing options and the rest of what the desktop app saves in `ui.json` are kept in the browser's `localStorage`, written as soon as they change |
| 2026-10-17 | Saving into a folder that refuses a temporary file (macOS privacy protection, read-only folders, #2138) never overwrites an existing document in place: it is left untouched and the error says to use Save As to another folder or allow WordCraft access to the folder (System Settings › Privacy & Security › Files and Folders on macOS). A new file is created directly. Unit tests on Linux; not yet confirmed on a Mac |
Footer page-table height and body wrapping are corrected. Other pagination gaps remain open.
## Revision history
| Date | Change | Summary |
|---|---|---|
| 2026-10-19 | trivial | Progress log: forms protection keeps its content controls: no inserting, Properties or Remove Content Control |
| 2026-10-19 | trivial | Progress log: text typed into a content control takes the control's formatting; localized placeholder styles recognised |
| 2026-10-18 | trivial | Progress log: long navigation headings wrap within the pane instead of ending in an ellipsis (#3015) |
| 2026-10-18 | trivial | Font list starts at its first row after clearing a no-match query; recent heading remains visible (#2068) |
| 2026-10-18 | trivial | A word after a slash wraps whole, like Word (#2963) |
| 2026-10-17 | trivial | Progress log: recipient list search and merged-only view (#1576) |
| 2026-10-18 | trivial | Progress log: symbols keep their font when the text font changes (#2036) |
| 2026-10-17 | trivial | Progress log: Restrict Editing guards tables, Paste Formatting and merge fields (#1327) |
| 2026-10-18 | trivial | Progress log: Picture Format › Size, sizes typed in any unit and the Size dialog (#2573); readiness estimates unchanged |
| 2026-10-18 | trivial | Progress log: Manage Sources imports Word's `Sources.xml` (#2640) |
| 2026-10-14 | trivial | Progress log: a list item's first word is not flagged for a missing capital (#1741) |
| 2026-10-18 | trivial | #2445: pending Save As answers keep their originating document identity, so a replacement document is never written to an old picker destination; real Save-button and focused-editor shortcut controls verified |
| 2026-10-16 | trivial | #2445: interactive Save on import-only .doc/.dot paths uses the existing DOCX Save As flow; cancel and failed writes keep the draft, and Close waits for a successful save. Explicit/script saves still report unsupported legacy output |
| 2026-10-16 | trivial | Save Changes exposes an alert-dialog role, the document-specific question and loss/failure context to accessibility; full-app focus/ancestry regression. Native screen-reader verification remains pending. |
| 2026-10-16 | trivial | Progress log: the Spike |
| 2026-10-16 | trivial | Progress log: `w:eastAsianLayout` kept on save |
| 2026-10-15 | trivial | Progress log: undrawable private-use list bullets show the level's default bullet (#1683) |
| 2026-10-15 | trivial | Progress log: web preferences kept between visits (#2132) |
| 2026-10-15 | trivial | Progress log: Recently Used Fonts in the font box (#2068) |
| 2026-10-15 | trivial | Progress log: one Thai face for a run in a missing Thai font (#1549) |
| 2026-10-18 | trivial | Progress log: Different First Page headers reachable from page 1 (#2826) |
| 2026-10-18 | trivial | Progress log: bold Black/Heavy fonts are not emboldened again |
| 2026-10-16 | trivial | Progress log: Web Layout fills the window width (#1876) |
| 2026-10-17 | trivial | Progress log: DOCX glossary documents (building blocks) |
| 2026-10-15 | trivial | Progress log: move options in Track Changes Options (gap #5) |
| 2026-10-15 | trivial | Progress log: image size limits checked before allocation (#1321) |
| 2026-10-17 | trivial | Progress log: Read Aloud speaks each sentence in its own language, #1843 |
| 2026-10-18 | trivial | Progress log: `wordcraft-cli run --script` (#1963) |
| 2026-10-18 | trivial | Progress log: header-cell picture watermarks drawn behind the text (#1775); readiness estimates unchanged |
| 2026-10-14 | trivial | Progress log: comment drafts kept and written before saving or closing (#1348) |
| 2026-10-17 | trivial | Language dialog keeps inherited proofing language and no-proof choices (#2717); readiness estimates unchanged |
| 2026-10-14 | trivial | Progress log: ASK and SET fields show nothing (#1765) |
| 2026-10-15 | trivial | Progress log: at-least table rows start the next page whole (#1742) |
| 2026-10-14 | trivial | Progress log: ODT sections on export (gap 19) |
| 2026-10-17 | trivial | Explicit DrawingML preset/custom outline dashes survive import/save and paint in raster/PDF, scaled to the effective stroke width even for zero-width paths. Synthetic controls; theme-inherited dashes and line caps remain unverified. |
| 2026-10-15 | trivial | `wordcraft-cli convert --update-fields` explicitly refreshes unlocked fields before conversion; ordinary conversion keeps cached results. Field refresh can change pagination |
| 2026-10-17 | trivial | Arial Unicode MS uses Arial and its proportional substitutes; glyph coverage and exact Word line height remain font-dependent. |
| 2026-10-18 | trivial | Progress log: Split Cells and Convert Text to Table dialogs, matched to the Word check (#2592) |
| 2026-10-16 | trivial | Progress log: tall-ascent font list entries no longer cut off (#2570) |
| 2026-10-16 | trivial | Progress log: Read Mode keeps the keyboard, whole columns, page-turn bar clear (#1772) |
| 2026-10-16 | trivial | Progress log: comments readable in the dark theme (#2045) |
| 2026-10-16 | trivial | Progress log: Indonesian proofing (#498) |
| 2026-10-16 | trivial | Progress log: legacy outline, shadow, emboss and imprint text effects drawn (#2534) |
| 2026-10-15 | trivial | Progress log: Columns on selected text (#2024) |
| 2026-10-15 | trivial | Navigation Find and query edits expose Results (#2096) |
| 2026-10-15 | trivial | Progress log: numbered headings (#1793) |
| 2026-10-15 | trivial | Add the missing `-->` → `→` AutoCorrect entry (#2223), with typing, undo and preference regression coverage; existing arrow bullets unchanged |
| 2026-10-15 | trivial | Progress log: Update Field updates the part being edited, as in Word (#1706) |
| 2026-10-14 | trivial | Progress log: Update Fields reaches headers, footers and text boxes (#1706) |
| 2026-10-15 | trivial | Progress log: font family caches bounded; math font resolution follows font changes (#1354) |
| 2026-10-15 | trivial | Progress log: web typing in non-Latin text no longer repeats (#1480) |
| 2026-10-15 | trivial | Progress log: Backspace removes a selected table, rows or columns (#1844) |
| 2026-10-16 | trivial | Progress log: Hebrew field results and smart quotes (#1670) |
| 2026-10-16 | trivial | Progress log: Ctrl+wheel zoom no longer scrolls the document (#1743) |
| 2026-10-16 | trivial | Hint-reliant CJK fonts execute TrueType contour assembly; Windows DFKai-SB no longer loses strokes. Synthetic local-font regression; no font binaries bundled. |
| 2026-10-14 | trivial | Progress log: RTF export writes sections (gap 19) |
| 2026-10-15 | trivial | Progress log: multi-letter equation scripts (#1609) |
| 2026-10-15 | trivial | Progress log: plain-text paste keeps the paragraph formatting where it lands |
| 2026-10-14 | trivial | Progress log: Add Font imports every selected font file (#1789) |
| 2026-10-15 | trivial | Additive paragraph spacing retains the full before-space after a section page break; ordinary overlapping spacing is unchanged. |
| 2026-10-15 | trivial | Exact line spacing keeps the run font baseline around tall equations; automatic and minimum spacing still grow for math. |
| 2026-10-16 | trivial | Zero-extent drawing anchors contribute no inline font height. Synthetic regressions. |
| 2026-10-16 | trivial | Control keyboard/text input works in the inactive native window (#2420) |
| 2026-10-15 | trivial | Progress log: Hide Quick Access Toolbar (#1945) |
| 2026-10-15 | trivial | Progress log: native ink nested in groups and drawing canvases kept on save |
| 2026-10-15 | trivial | Progress log: the word being typed gets no squiggle until the caret leaves it (#1552) |
| 2026-10-15 | trivial | Progress log: Highlight No Color sticks to the button (#2105) |
| 2026-10-16 | trivial | Body and active note text keep normal opacity while editing footnotes or endnotes (#2545); header/footer dimming is unchanged |
| 2026-10-17 | minor | Progress log: Drop Cap Options (#2601) |
| 2026-10-16 | trivial | An otherwise empty section at page start retains its line before continuous content; later closing marks stay quiet |
| 2026-10-16 | trivial | Progress log: equations from Word files keep their content (#1273) |
| 2026-10-16 | trivial | Progress log: Insert Caption… on objects and Page Setup header and footer options (#1230) |
| 2026-10-17 | trivial | SECTIONPAGES includes shared continuous-section pages in body and header/footer fields |
| 2026-10-17 | trivial | Footer page tables no longer inflate footer text height; body lines wrap at their physical bounds |
| 2026-10-17 | minor | Progress log: Scale to paper size on File › Print, #2108 |
| 2026-10-17 | trivial | Non-overlapping body floating tables clear header tables; overlapping tables keep their requested position |
| 2026-10-15 | trivial | Progress log: caret inside ligatures (#1974) |
| 2026-10-14 | trivial | Progress log: row table border exceptions |
| 2026-10-12 | minor | Czech proofing review: short-word splits rank after deletion, NOSUGGEST excludes suggestions while accepting typed forms, affix comments apply only to headers, ratios keep spaces and repeated-word messages use Czech quotes; synthetic engine end-to-end and latency checks |
| 2026-10-12 | minor | Czech spelling from the installed Hunspell dictionary, runtime-table suggestions and generic checks; Czech interface writes cs-CZ documents; no Czech hyphenation patterns bundled; estimates unchanged |
| 2026-10-12 | trivial | Progress log: one selected cell's border drags in its row only (#962) |
| 2026-10-16 | trivial | Progress log: opening a DOCX no longer copies its pictures |
| 2026-10-12 | minor | Progress log: Word default shortcuts expanded to 140 commands on macOS (169 combinations) and 158 on Linux/Windows (188 combinations), addressing gap #24 |
| 2026-10-13 | trivial | Web/Draft long-page blur fixed with visible full-resolution raster tiles |
| 2026-10-14 | trivial | Progress log: VML freeforms read and kept on save |
@@ -418,6 +597,7 @@ Ranked, to beta (detail and remaining milestones in [docs/roadmap.md](docs/roadm
| 2026-10-14 | trivial | Floating-table overlap checks use visible line content; empty line-break lines no longer move a table |
| 2026-10-14 | trivial | Progress log: deleting a comment leaves no dangling anchors or replies (#1345) |
| 2026-10-14 | minor | Legacy DOC DIB conversion byte limits only; related #1334/#1554, stage and readiness estimates unchanged |
| 2026-10-15 | trivial | Progress log: Ink to Shape two-stroke and closed-head arrows |
| 2026-10-13 | trivial | Progress log: justified vertical page alignment (#1320) |
| 2026-10-14 | trivial | Progress log: file names that aren't valid Unicode (#1440) |
| 2026-10-13 | trivial | Behind Text pictures and shapes drag over overlapping text in the body and headers, selected or not, while ordinary clicks keep editing text; In Front of Text drag covered too |
@@ -592,3 +772,7 @@ Ranked, to beta (detail and remaining milestones in [docs/roadmap.md](docs/roadm
| 2026-10-09 | minor | Recently landed: RTL, Ukrainian and Brazilian Portuguese |
| 2026-10-06 | major | First honest estimate: 88% catalog, ~62% real parity, alpha ~85% of the way |
| 2026-10-12 | trivial | Progress log: recipient filter and sort, .xls and legacy-encoding recipient lists (#778) |
| 2026-10-15 | trivial | Progress log: dialog close box (#1885) |
| 2026-10-15 | trivial | Progress log: mixed Hebrew/Latin text keeps its order in Word (#1543) |
| 2026-10-15 | trivial | Progress log: ruler numbers by zoom, rulers follow the caret's page (#1878) |
| 2026-10-17 | trivial | Progress log: saving into a folder that refuses a temporary file (#2138) |
+369 -41
View File
@@ -2,6 +2,7 @@
#![deny(clippy::unwrap_used, clippy::expect_used, clippy::panic, clippy::unimplemented, clippy::todo, clippy::unreachable)]
mod automation_args;
mod script;
use std::ffi::{OsStr, OsString};
use std::fmt;
@@ -16,9 +17,10 @@ const USAGE: &str = "\
wordcraft-cli — WordCraft from the command line
USAGE:
wordcraft-cli convert <in> <out> [--markup] [--format wordml2003|flatopc]
wordcraft-cli convert <in> <out> [--markup] [--update-fields] [--format wordml2003|flatopc]
convert between formats (docx, doc [read], pdf, odt, rtf, html, md, tex, txt, json, png,
xml); an .xml file is a Word XML Document (Flat OPC), or with
xml); --update-fields refreshes unlocked fields before saving;
an .xml file is a Word XML Document (Flat OPC), or with
--format wordml2003 a Word 2003 XML Document;
--markup (Print Markup): comments beside each page and
tracked changes in pdf/png, a comment list in html/md/txt
@@ -26,9 +28,15 @@ USAGE:
wordcraft-cli text <file> plain text
wordcraft-cli inspect <file> document structure (JSON)
wordcraft-cli render <file> <out.png> [--page N] [--scale S]
wordcraft-cli run [--file F | --template T] --cmd 'id={json}' [--cmd …] [--save OUT] [--print]
[--author NAME] run commands headlessly, then save (--author names
who tracked changes and comments are by)
wordcraft-cli run [--file F | --template T] [--script FILE|-] --cmd 'id={json}' [--cmd …]
[--save OUT] [--print] [--author NAME]
run commands headlessly, then save (--author names
who tracked changes and comments are by).
--script runs a JSON file (- is stdin) first:
{\"file\"?, \"template\"?, \"commands\": [{\"command\": id,
\"params\"?: {…}} | id, …], \"save\"?: [OUT, …]}; save
formats follow the extension (docx and pdf in one run);
relative paths resolve against the script's folder
wordcraft-cli commands [--json] list every command
wordcraft-cli parity [--markdown] feature-catalog parity
wordcraft-cli mcp [--connect HOST:PORT | --author NAME] [--automation-read-root DIR] [--automation-write-root DIR]
@@ -47,6 +55,11 @@ USAGE:
wordcraft-cli [<command>] -h | --help this help
wordcraft-cli --version
A password-protected Word file is opened with --password PASS, --password-file PATH or the
WORDCRAFT_PASSWORD environment variable; a password given on the command line is readable by
every other process, so the other two are better. A file opened with a password is written with
it again, as the app does. Exports that cannot carry a password (pdf, html, txt) stay plain.
Paper: documents created here, and HTML, Markdown and text files (which give no page size), use
the system region's paper (A4, or Letter in the US and other Letter regions), read from
LC_ALL / LC_PAPER / LANG or the platform. Set WORDCRAFT_REGION=en-US (or another tag) for the
@@ -60,13 +73,24 @@ fn wants_help(cmd: &str, rest: &[String]) -> bool {
SUBCOMMANDS.contains(&cmd) && rest.iter().any(|a| a == "--help" || a == "-h")
}
fn open(path: &Path) -> Result<Session, String> {
/// The error every subcommand gives when a required path is not there.
fn missing_argument() -> Failure {
format!("missing argument\n\n{USAGE}").into()
}
/// Open a document for a subcommand. A password-protected Word file needs its password
/// ([`password_from`]). As in the app, a document opened with one is saved with it again, so
/// `convert` does not quietly write a copy the password no longer protects.
fn open(path: &Path, password: Option<&str>) -> Result<Session, String> {
// A file that gives no page size (HTML, Markdown, plain text) opens on the system region's
// paper, as in the app: A4 in Italy, Letter in the US (#1381).
let region = system_region();
let (doc, _) = wordcraft_engine::io::open_path_in(path, None, &region)?;
let (doc, encrypted) = wordcraft_engine::io::open_path_in(path, password, &region)?;
let mut s = session(doc);
s.path = Some(path.into());
if encrypted {
s.password = password.map(wordcraft_engine::io::Password::new);
}
Ok(s)
}
@@ -85,18 +109,57 @@ fn text_args<S: AsRef<OsStr>>(raw: &[S]) -> Result<(Vec<String>, Option<PathBuf>
None if cmd == "run" => args.iter().position(|a| a == "--file").map(|i| i + 1),
None => None,
};
// `run --script FILE` may also name its script with such a file name.
let script_at = if cmd == "run" { args.iter().position(|a| a == "--script").map(|i| i + 1) } else { None };
let mut file = None;
for (i, a) in raw.iter().map(AsRef::as_ref).enumerate() {
if a.to_str().is_none() {
if Some(i) != file_at {
if Some(i) != file_at && Some(i) != script_at {
return Err(format!("argument {} is not valid Unicode: {}", i + 1, a.to_string_lossy()));
}
file = Some(PathBuf::from(a));
if Some(i) == file_at {
file = Some(PathBuf::from(a));
}
}
}
Ok((args, file))
}
/// The password for a protected Word file: `--password`, `--password-file`, or
/// `WORDCRAFT_PASSWORD`, in that order. The file form exists because a password on a command line
/// is readable by every other process on the machine; a trailing newline is dropped so a file
/// written by `echo` works. Never logged, and never named in an error.
fn password_from(args: &[String]) -> Result<Option<String>, String> {
if let Some(password) = arg_value(args, "--password") {
return Ok(Some(password));
}
if let Some(path) = arg_value(args, "--password-file") {
let raw = std::fs::read_to_string(&path).map_err(|e| format!("--password-file {path}: {e}"))?;
return Ok(Some(raw.trim_end_matches(['\r', '\n']).to_string()));
}
Ok(std::env::var("WORDCRAFT_PASSWORD").ok().filter(|p| !p.is_empty()))
}
/// The positional arguments, skipping every option's value so an option value is never read as a
/// path. Mirrors [`validate_options`], which has already checked the options are known.
fn positionals<'a>(args: &'a [String], specs: &[OptionSpec]) -> Vec<&'a String> {
let mut out = Vec::new();
let mut i = 0;
while let Some(arg) = args.get(i) {
let takes_value =
arg.starts_with("--") && specs.iter().any(|s| s.name == arg && s.takes_value) && args.get(i + 1).is_some_and(|v| !v.starts_with("--"));
if takes_value {
i += 2;
} else {
if !arg.starts_with("--") {
out.push(arg);
}
i += 1;
}
}
out
}
/// A headless session in the system region: every way the CLI starts one gives `file.new` the
/// same paper (#1381).
fn session(doc: wordcraft_doc::Document) -> Session {
@@ -222,13 +285,30 @@ fn automation_roots(
Ok((roots, left))
}
const NO_OPTIONS: &[OptionSpec] = &[];
const CONVERT_OPTIONS: &[OptionSpec] = &[OptionSpec { name: "--markup", takes_value: false }, OptionSpec { name: "--format", takes_value: true }];
const RENDER_OPTIONS: &[OptionSpec] = &[OptionSpec { name: "--page", takes_value: true }, OptionSpec { name: "--scale", takes_value: true }];
/// The password a protected Word file is opened with: `--password` (which any other process on
/// the machine can read from the command line), `--password-file`, or `WORDCRAFT_PASSWORD`.
const PASSWORD_OPTIONS: &[OptionSpec] =
&[OptionSpec { name: "--password", takes_value: true }, OptionSpec { name: "--password-file", takes_value: true }];
const CONVERT_OPTIONS: &[OptionSpec] = &[
OptionSpec { name: "--markup", takes_value: false },
OptionSpec { name: "--format", takes_value: true },
OptionSpec { name: "--update-fields", takes_value: false },
OptionSpec { name: "--password", takes_value: true },
OptionSpec { name: "--password-file", takes_value: true },
];
const RENDER_OPTIONS: &[OptionSpec] = &[
OptionSpec { name: "--page", takes_value: true },
OptionSpec { name: "--scale", takes_value: true },
OptionSpec { name: "--password", takes_value: true },
OptionSpec { name: "--password-file", takes_value: true },
];
const RUN_OPTIONS: &[OptionSpec] = &[
OptionSpec { name: "--file", takes_value: true },
OptionSpec { name: "--password", takes_value: true },
OptionSpec { name: "--password-file", takes_value: true },
OptionSpec { name: "--template", takes_value: true },
OptionSpec { name: "--cmd", takes_value: true },
OptionSpec { name: "--script", takes_value: true },
OptionSpec { name: "--save", takes_value: true },
OptionSpec { name: "--page", takes_value: true },
OptionSpec { name: "--scale", takes_value: true },
@@ -291,32 +371,37 @@ fn write_stdout(write: impl FnOnce(&mut std::io::StdoutLock<'static>) -> std::io
Ok(())
}
/// The operand after `--script`, by its real name (it may not be valid Unicode).
fn script_operand<S: AsRef<OsStr>>(raw: &[S]) -> Option<PathBuf> {
let at = raw.iter().position(|a| a.as_ref() == OsStr::new("--script"))?;
raw.get(at + 1).map(|a| PathBuf::from(a.as_ref()))
}
fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
let script_path = script_operand(args);
let (args, file) = text_args(args)?;
// Opens the file by its real name, which its text in `args` may have lost.
let open = |text: &str| open(file.as_deref().unwrap_or(Path::new(text)));
// Opens the file by its real name, which its text in `args` may have lost, and with the
// password of a protected document (`password_from`).
let open = |text: &str, password: Option<&str>| open(file.as_deref().unwrap_or(Path::new(text)), password);
let cmd = args.first().map(String::as_str).unwrap_or("");
let rest: Vec<String> = args.iter().skip(1).cloned().collect();
// Before options are checked or files and folders opened.
if wants_help(cmd, &rest) {
return write_stdout(|o| write!(o, "{USAGE}"));
}
let pos = |i: usize| rest.iter().filter(|a| !a.starts_with("--")).nth(i).cloned().ok_or_else(|| format!("missing argument\n\n{USAGE}"));
match cmd {
"convert" => {
validate_options("convert", &rest, CONVERT_OPTIONS)?;
// `--format wordml2003` (or `flatopc`) picks the kind of `.xml` file; its value isn't a path.
let format = arg_value(&rest, "--format");
let paths: Vec<&String> = rest
.iter()
.enumerate()
.filter(|(n, a)| !a.starts_with("--") && !(*n > 0 && rest.get(n - 1).is_some_and(|p| p == "--format")))
.map(|(_, a)| a)
.collect();
let missing = || format!("missing argument\n\n{USAGE}");
let (i, o) = (paths.first().ok_or_else(missing)?.to_string(), paths.get(1).ok_or_else(missing)?.to_string());
let paths = positionals(&rest, CONVERT_OPTIONS);
let (i, o) = (paths.first().ok_or_else(missing_argument)?.to_string(), paths.get(1).ok_or_else(missing_argument)?.to_string());
let markup = rest.iter().any(|a| a == "--markup");
let mut s = open(&i)?;
let password = password_from(&rest)?;
let mut s = open(&i, password.as_deref())?;
if rest.iter().any(|a| a == "--update-fields") {
s.run("references.updateFields", &json!({})).map_err(|e| e.to_string())?;
}
let r = if o.to_ascii_lowercase().ends_with(".png") {
s.run("file.exportPng", &json!({"path": o, "page": 1, "markup": markup}))
} else {
@@ -331,29 +416,40 @@ fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
Ok(())
}
"info" => {
validate_options("info", &rest, NO_OPTIONS)?;
let mut s = open(&pos(0)?)?;
validate_options("info", &rest, PASSWORD_OPTIONS)?;
let paths = positionals(&rest, PASSWORD_OPTIONS);
let password = password_from(&rest)?;
let mut s = open(paths.first().ok_or_else(missing_argument)?, password.as_deref())?;
let info = serde_json::to_string_pretty(&s.run("file.info", &json!({})).map_err(|e| e.to_string())?).unwrap_or_default();
write_stdout(|o| writeln!(o, "{info}"))
}
"text" => {
validate_options("text", &rest, NO_OPTIONS)?;
let s = open(&pos(0)?)?;
validate_options("text", &rest, PASSWORD_OPTIONS)?;
let paths = positionals(&rest, PASSWORD_OPTIONS);
let password = password_from(&rest)?;
let s = open(paths.first().ok_or_else(missing_argument)?, password.as_deref())?;
let text = listed_text(&s.doc)?;
write_stdout(|o| writeln!(o, "{text}"))
}
"inspect" => {
validate_options("inspect", &rest, NO_OPTIONS)?;
let mut s = open(&pos(0)?)?;
validate_options("inspect", &rest, PASSWORD_OPTIONS)?;
let paths = positionals(&rest, PASSWORD_OPTIONS);
let password = password_from(&rest)?;
let mut s = open(paths.first().ok_or_else(missing_argument)?, password.as_deref())?;
let tree = serde_json::to_string_pretty(&s.run("document.inspect", &json!({})).map_err(|e| e.to_string())?).unwrap_or_default();
write_stdout(|o| writeln!(o, "{tree}"))
}
"render" => {
validate_options("render", &rest, RENDER_OPTIONS)?;
let (i, o) = (pos(0)?, pos(1)?);
let paths = positionals(&rest, RENDER_OPTIONS);
let password = password_from(&rest)?;
let (Some(i), Some(o)) = (paths.first(), paths.get(1)) else {
return Err(missing_argument());
};
let (i, o) = (i.to_string(), o.to_string());
let page = parse_page_arg("render", &rest)?.unwrap_or(1);
let scale = parse_scale_arg("render", &rest)?.unwrap_or(2.0);
let mut s = open(&i)?;
let mut s = open(&i, password.as_deref())?;
let r = s.run("file.exportPng", &json!({"path": o, "page": page, "scale": scale})).map_err(|e| e.to_string())?;
write_stdout(|o| writeln!(o, "{r}"))
}
@@ -361,8 +457,20 @@ fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
validate_options("run", &rest, RUN_OPTIONS)?;
let page = parse_page_arg("run", &rest)?;
let scale = parse_scale_arg("run", &rest)?;
let mut s = match (arg_value(&rest, "--file"), arg_value(&rest, "--template")) {
(Some(f), _) => open(&f)?,
let password = password_from(&rest)?;
let script = match arg_value(&rest, "--script") {
Some(path) if !path.starts_with("--") => Some(script::load(script_path.as_deref().unwrap_or(Path::new(&path)))?),
_ if rest.iter().any(|a| a == "--script") => return Err("run: --script requires a value".into()),
_ => None,
};
// A source named on the command line wins over both of the script's.
let (mut from_file, mut from_template) = (arg_value(&rest, "--file"), arg_value(&rest, "--template"));
if from_file.is_none() && from_template.is_none() {
from_file = script.as_ref().and_then(|c| c.file.clone());
from_template = script.as_ref().and_then(|c| c.template.clone());
}
let mut s = match (from_file, from_template) {
(Some(f), _) => open(&f, password.as_deref())?,
(None, Some(t)) => {
let mut s = session(wordcraft_doc::Document::new());
s.run("file.new", &json!({"template": t})).map_err(|e| e.to_string())?;
@@ -373,6 +481,13 @@ fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
if let Some(name) = arg_value(&rest, "--author") {
set_author(&mut s, &name)?;
}
let print = rest.iter().any(|a| a == "--print");
for (n, (id, params)) in script.iter().flat_map(|c| c.commands.iter()).enumerate() {
let r = s.run(id, params).map_err(|e| format!("command {} ({id}): {e}", n + 1))?;
if print {
write_stdout(|o| writeln!(o, "{id}: {r}"))?;
}
}
let mut i = 0;
while i < rest.len() {
if rest.get(i).map(String::as_str) == Some("--cmd") {
@@ -382,7 +497,7 @@ fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
None => (spec.clone(), json!({})),
};
let r = s.run(&id, &params).map_err(|e| format!("{id}: {e}"))?;
if rest.iter().any(|a| a == "--print") {
if print {
write_stdout(|o| writeln!(o, "{id}: {r}"))?;
}
i += 2;
@@ -390,7 +505,21 @@ fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
i += 1;
}
}
if let Some(out) = arg_value(&rest, "--save") {
let outputs: Vec<String> = script.iter().flat_map(|c| c.save.iter().cloned()).chain(arg_value(&rest, "--save")).collect();
// Check every output can be made before writing any, so a bad one leaves no half-done job.
for out in &outputs {
let parent = Path::new(out).parent().filter(|p| !p.as_os_str().is_empty());
if let Some(p) = parent.filter(|p| !p.is_dir()) {
return Err(format!("{out}: folder {} does not exist", p.display()).into());
}
if Path::new(out).is_dir() {
return Err(format!("{out}: is a folder, not a file").into());
}
if !out.to_ascii_lowercase().ends_with(".png") {
wordcraft_engine::io::save_bytes(out, &s.doc)?;
}
}
for out in outputs {
let r = if out.to_ascii_lowercase().ends_with(".png") {
let mut v = json!({"path": out});
if let Some(n) = page {
@@ -478,10 +607,12 @@ fn run<S: AsRef<OsStr>>(args: &[S]) -> Result<(), Failure> {
}
"zotero" => {
validate_options("zotero", &rest, ZOTERO_OPTIONS)?;
let name = pos(0)?;
let command = wordcraft_zotero::Command::from_name(&name).ok_or_else(|| format!("unknown Zotero command `{name}`\n\n{USAGE}"))?;
let file = pos(1)?;
let mut s = open(&file)?;
let paths = positionals(&rest, ZOTERO_OPTIONS);
let (Some(name), Some(file)) = (paths.first(), paths.get(1)) else {
return Err(missing_argument());
};
let command = wordcraft_zotero::Command::from_name(name).ok_or_else(|| format!("unknown Zotero command `{name}`\n\n{USAGE}"))?;
let mut s = open(file, None)?;
for spec in rest.iter().zip(rest.iter().skip(1)).filter(|(a, _)| *a == "--cmd").map(|(_, v)| v) {
let (id, params) = match spec.split_once('=') {
Some((id, p)) => (id.to_string(), serde_json::from_str::<Value>(p).map_err(|e| format!("{id}: bad JSON params: {e}"))?),
@@ -569,6 +700,7 @@ fn main() -> ExitCode {
#[cfg(test)]
mod tests {
use super::*;
use std::io::Read;
fn args(values: &[&str]) -> Vec<String> {
values.iter().map(|value| (*value).into()).collect()
@@ -576,7 +708,7 @@ mod tests {
#[test]
fn validators_accept_only_the_documented_options() {
assert!(validate_options("info", &args(&["file.docx"]), NO_OPTIONS).is_ok());
assert!(validate_options("info", &args(&["file.docx"]), PASSWORD_OPTIONS).is_ok());
assert!(validate_options("render", &args(&["in.docx", "out.png", "--page", "2", "--scale", "1.5"]), RENDER_OPTIONS).is_ok());
assert!(validate_options("run", &args(&["--cmd", "text.insert={}", "--cmd", "edit.undo", "--print"]), RUN_OPTIONS).is_ok());
assert!(validate_options("commands", &args(&["--json"]), COMMANDS_OPTIONS).is_ok());
@@ -593,6 +725,41 @@ mod tests {
);
}
#[test]
fn an_option_value_is_never_read_as_a_path() {
// `--password`'s value sits between the paths and would be taken for one without
// `positionals`, so `render` would write its picture over a file called after the password.
assert_eq!(positionals(&args(&["in.docx", "--password", "secret", "out.png"]), RENDER_OPTIONS), vec!["in.docx", "out.png"]);
assert_eq!(positionals(&args(&["in.docx", "out.png", "--page", "2"]), RENDER_OPTIONS), vec!["in.docx", "out.png"]);
// A bare flag takes no value, so the next argument stays a path.
assert_eq!(positionals(&args(&["--print", "in.docx"]), RUN_OPTIONS), vec!["in.docx"]);
// `convert` is the one with a value-taking flag between the two paths.
assert_eq!(positionals(&args(&["in.docx", "--format", "flatopc", "out.docx"]), CONVERT_OPTIONS), vec!["in.docx", "out.docx"]);
}
#[test]
fn the_password_comes_from_an_option_a_file_or_the_environment() {
assert_eq!(password_from(&args(&["f.docx"])), Ok(None));
assert_eq!(password_from(&args(&["f.docx", "--password", "s3cret"])), Ok(Some("s3cret".into())));
// `--password` wins over the other two sources.
assert_eq!(password_from(&args(&["--password", "a", "--password-file", "/nonexistent"])), Ok(Some("a".into())));
}
#[test]
fn a_password_file_loses_only_its_trailing_newline() {
let path = std::env::temp_dir().join("wordcraft-cli-password-test.txt");
std::fs::write(&path, "s3cret\r\n").unwrap();
let got = password_from(&args(&["f.docx", "--password-file", &path.to_string_lossy()]));
std::fs::remove_file(&path).ok();
assert_eq!(got, Ok(Some("s3cret".into())));
}
#[test]
fn a_missing_password_file_is_named_without_the_password() {
let err = password_from(&args(&["f.docx", "--password-file", "/nonexistent/secret.txt"])).unwrap_err();
assert!(err.contains("--password-file"), "{err}");
}
#[test]
fn unknown_options_are_rejected_before_file_access() {
assert_eq!(run(&args(&["info", "missing.docx", "--passwrod", "x"])), Err("info: unknown option --passwrod".into()));
@@ -893,4 +1060,165 @@ mod tests {
let _ = std::fs::remove_dir_all(dir);
}
/// A scratch folder that removes itself, even when the test fails.
struct ScriptDir(PathBuf);
impl std::ops::Deref for ScriptDir {
type Target = Path;
fn deref(&self) -> &Path {
&self.0
}
}
impl Drop for ScriptDir {
fn drop(&mut self) {
let _ = std::fs::remove_dir_all(&self.0);
}
}
fn script_dir(tag: &str) -> ScriptDir {
let dir = std::env::temp_dir().join(format!("wc-cli-script-{tag}-{}", std::process::id()));
let _ = std::fs::remove_dir_all(&dir);
std::fs::create_dir_all(&dir).unwrap();
ScriptDir(dir)
}
fn write_script(dir: &Path, text: &str) -> String {
let p = dir.join("job.json");
std::fs::write(&p, text).unwrap();
p.to_string_lossy().into_owned()
}
#[test]
fn script_builds_a_document_and_saves_docx_and_pdf_relative_to_the_script() {
let dir = script_dir("two");
let script = write_script(
&dir,
r#"{"template":"blank","commands":[{"command":"text.insert","params":{"text":"Hello script"}},{"command":"format.bold"}],"save":["out.docx","out.pdf"]}"#,
);
run(&args(&["run", "--script", &script])).unwrap();
assert!(std::fs::read(dir.join("out.pdf")).unwrap().starts_with(b"%PDF"));
let doc = open(&dir.join("out.docx"), None).unwrap().doc;
let json = serde_json::to_string(&doc).unwrap();
assert!(json.contains("Hello script") && json.contains("\"bold\":true"));
}
#[test]
fn script_with_a_bad_output_writes_no_output() {
let dir = script_dir("badout");
let script = write_script(&dir, r#"{"commands":["caret.docEnd"],"save":["a.docx","b.xyz"]}"#);
let e = run(&args(&["run", "--script", &script])).unwrap_err().to_string();
assert_eq!(e.matches("b.xyz").count(), 1, "{e}");
assert!(!dir.join("a.docx").exists());
let script = write_script(&dir, r#"{"commands":["caret.docEnd"],"save":["a.docx","nope/b.docx"]}"#);
assert!(run(&args(&["run", "--script", &script])).is_err());
assert!(!dir.join("a.docx").exists());
}
#[test]
fn script_with_a_folder_as_output_writes_no_output() {
let dir = script_dir("folderout");
std::fs::create_dir(dir.join("b.pdf")).unwrap();
let script = write_script(&dir, r#"{"commands":["caret.docEnd"],"save":["a.docx","b.pdf"]}"#);
let e = run(&args(&["run", "--script", &script])).unwrap_err().to_string();
assert!(e.contains("b.pdf"), "{e}");
assert!(!dir.join("a.docx").exists());
}
/// APFS refuses file names that aren't valid UTF-8, so the fixture can't be made on macOS
/// (as in `tests/non_utf8_args.rs`).
#[cfg(all(unix, not(target_os = "macos")))]
#[test]
fn script_file_name_need_not_be_unicode() {
use std::os::unix::ffi::OsStringExt;
let dir = script_dir("latin1");
let mut name = dir.as_os_str().to_os_string().into_vec();
name.extend_from_slice(b"/caf\xe9.json");
let path = PathBuf::from(OsString::from_vec(name));
std::fs::write(&path, r#"{"commands":["caret.docEnd"],"save":["o.txt"]}"#).unwrap();
let raw = vec![OsString::from("run"), OsString::from("--script"), path.into_os_string()];
run(&raw).unwrap();
assert!(dir.join("o.txt").exists());
}
#[test]
fn command_line_template_wins_over_the_scripts_file() {
let dir = script_dir("source");
let input = dir.join("in.docx");
run(&args(&["run", "--template", "blank", "--cmd", "text.insert={\"text\":\"FROM FILE\"}", "--save", &input.to_string_lossy()])).unwrap();
let script = write_script(&dir, r#"{"file":"in.docx","commands":["caret.docEnd"]}"#);
let (a, b) = (dir.join("a.txt"), dir.join("b.txt"));
run(&args(&["run", "--script", &script, "--template", "blank", "--save", &a.to_string_lossy()])).unwrap();
assert!(!std::fs::read_to_string(&a).unwrap().contains("FROM FILE"));
run(&args(&["run", "--script", &script, "--save", &b.to_string_lossy()])).unwrap();
assert!(std::fs::read_to_string(&b).unwrap().contains("FROM FILE"));
}
#[test]
fn script_value_missing_is_reported() {
let e = run(&args(&["run", "--script", "--cmd", "caret.docEnd"])).unwrap_err().to_string();
assert!(e.contains("--script requires a value"), "{e}");
}
#[test]
fn script_reads_from_a_stream_and_caps_its_size() {
let s = script::load_from(&br#"{"commands":["caret.docEnd"],"save":["o.docx"]}"#[..], "-", None).unwrap();
assert_eq!((s.commands.len(), s.save), (1, vec!["o.docx".to_string()]));
let big = std::io::repeat(b' ').take(script::MAX_SCRIPT_BYTES + 10);
let e = script::load_from(big, "-", None).unwrap_err();
assert!(e.contains("too large"), "{e}");
assert!(script::load_from(&b"{"[..], "-", None).is_err());
}
#[test]
fn script_commands_run_before_cmd_flags() {
let dir = script_dir("order");
let script = write_script(&dir, r#"{"commands":[{"command":"text.insert","params":{"text":"first"}}]}"#);
let out = dir.join("o.txt");
run(&args(&["run", "--script", &script, "--cmd", "text.insert={\"text\":\" second\"}", "--save", &out.to_string_lossy()])).unwrap();
assert!(std::fs::read_to_string(&out).unwrap().contains("first second"));
}
#[test]
fn script_failure_names_the_command_and_writes_nothing() {
let dir = script_dir("fail");
let script =
write_script(&dir, r#"{"commands":[{"command":"text.insert","params":{"text":"x"}},{"command":"no.such.command"}],"save":["out.docx"]}"#);
let e = run(&args(&["run", "--script", &script])).unwrap_err().to_string();
assert!(e.contains("command 2") && e.contains("no.such.command"), "{e}");
assert!(!dir.join("out.docx").exists());
}
#[test]
fn script_rejects_junk_empty_missing_and_huge_input() {
let dir = script_dir("junk");
for text in [
"not json",
"[]",
"{}",
r#"{"commands":[]}"#,
r#"{"commands":[{"params":{}}]}"#,
r#"{"commands":[1]}"#,
r#"{"commands":[{"command":"a"}],"save":[1]}"#,
] {
let script = write_script(&dir, text);
assert!(run(&args(&["run", "--script", &script])).is_err(), "{text}");
}
assert!(run(&args(&["run", "--script", &dir.join("missing.json").to_string_lossy()])).is_err());
let huge = write_script(&dir, &" ".repeat(5 * 1024 * 1024));
let e = run(&args(&["run", "--script", &huge])).unwrap_err().to_string();
assert!(e.contains("too large"), "{e}");
}
#[test]
fn script_parse_resolves_paths_against_the_base() {
let s = script::parse(r#"{"file":"in.docx","commands":[{"command":"edit.undo"}],"save":["/abs/o.pdf","o.docx"]}"#, Some(Path::new("/base")))
.unwrap();
let joined = |name: &str| Path::new("/base").join(name).to_string_lossy().into_owned();
assert_eq!(s.file, Some(joined("in.docx")));
assert_eq!(s.save, vec!["/abs/o.pdf".to_string(), joined("o.docx")]);
let s = script::parse(r#"{"commands":["caret.docEnd"],"save":["o.docx"]}"#, None).unwrap();
assert_eq!(s.save, vec!["o.docx".to_string()]);
assert_eq!(s.commands.len(), 1);
}
}
+112
View File
@@ -0,0 +1,112 @@
//! `run --script FILE`: commands read from a JSON file (`-` is stdin), so a job can be kept in a
//! file instead of a long chain of `--cmd` flags (#1963).
//!
//! ```json
//! {"file": "in.docx", "template": "sample",
//! "commands": [{"command": "text.insert", "params": {"text": "Hi"}}, "caret.docEnd"],
//! "save": ["out.docx", "out.pdf"]}
//! ```
//!
//! `file` and `template` are optional (`--file` / `--template` on the command line win), and a
//! command is an object with `command` and optional `params`, or just the id as a string. Relative
//! `file` and `save` paths resolve against the script's folder (the working folder for stdin).
use std::io::Read;
use std::path::Path;
use serde_json::Value;
/// The largest script read. A job is a list of commands; anything bigger is a mistake.
pub const MAX_SCRIPT_BYTES: u64 = 4 * 1024 * 1024;
/// The most commands a script may hold.
const MAX_COMMANDS: usize = 100_000;
/// The most outputs a script may save to.
const MAX_SAVES: usize = 64;
#[derive(Debug)]
pub struct Script {
pub file: Option<String>,
pub template: Option<String>,
pub commands: Vec<(String, Value)>,
pub save: Vec<String>,
}
/// Read the script at `arg` (`-` for stdin) and parse it.
pub fn load(arg: &Path) -> Result<Script, String> {
let name = arg.display().to_string();
if arg == Path::new("-") {
return load_from(std::io::stdin().lock(), &name, None);
}
let f = std::fs::File::open(arg).map_err(|e| format!("--script {name}: {e}"))?;
load_from(f, &name, arg.parent().filter(|p| !p.as_os_str().is_empty()))
}
/// Read and parse a script from `r`, called `name` in errors; relative paths resolve against `base`.
pub fn load_from(r: impl Read, name: &str, base: Option<&Path>) -> Result<Script, String> {
let text = read_capped(r, name)?;
parse(&text, base).map_err(|e| format!("--script {name}: {e}"))
}
fn read_capped(r: impl Read, name: &str) -> Result<String, String> {
let mut bytes = Vec::new();
r.take(MAX_SCRIPT_BYTES + 1).read_to_end(&mut bytes).map_err(|e| format!("--script {name}: {e}"))?;
if bytes.len() as u64 > MAX_SCRIPT_BYTES {
return Err(format!("--script {name}: script too large (over {} MiB)", MAX_SCRIPT_BYTES / 1024 / 1024));
}
String::from_utf8(bytes).map_err(|_| format!("--script {name}: not valid UTF-8"))
}
fn resolve(path: &str, base: Option<&Path>) -> String {
match base {
Some(b) if Path::new(path).is_relative() => b.join(path).to_string_lossy().into_owned(),
_ => path.to_string(),
}
}
fn opt_string(obj: &serde_json::Map<String, Value>, key: &str) -> Result<Option<String>, String> {
match obj.get(key) {
None | Some(Value::Null) => Ok(None),
Some(Value::String(s)) => Ok(Some(s.clone())),
Some(_) => Err(format!("`{key}` must be a string")),
}
}
pub fn parse(text: &str, base: Option<&Path>) -> Result<Script, String> {
let v: Value = serde_json::from_str(text).map_err(|e| format!("bad JSON: {e}"))?;
let obj = v.as_object().ok_or("a script is a JSON object with a `commands` list")?;
let list = match obj.get("commands") {
Some(Value::Array(a)) => a,
_ => return Err("`commands` must be a list".into()),
};
if list.is_empty() {
return Err("`commands` is empty".into());
}
if list.len() > MAX_COMMANDS {
return Err(format!("too many commands (over {MAX_COMMANDS})"));
}
let mut commands = Vec::with_capacity(list.len());
for (i, c) in list.iter().enumerate() {
let n = i + 1;
let entry = match c {
Value::String(id) => Some((id.clone(), Value::Object(Default::default()))),
Value::Object(o) => match o.get("command") {
Some(Value::String(id)) => Some((id.clone(), o.get("params").cloned().unwrap_or_else(|| Value::Object(Default::default())))),
_ => None,
},
_ => None,
};
match entry {
Some((id, _)) if id.is_empty() => return Err(format!("command {n}: empty command id")),
Some(e) => commands.push(e),
None => return Err(format!("command {n}: expected a command id or {{\"command\": id, \"params\": {{…}}}}")),
}
}
let save = match obj.get("save") {
None | Some(Value::Null) => Vec::new(),
Some(Value::String(s)) => vec![resolve(s, base)],
Some(Value::Array(a)) if a.len() <= MAX_SAVES => {
a.iter().map(|s| s.as_str().map(|s| resolve(s, base)).ok_or("`save` entries must be strings")).collect::<Result<_, _>>()?
}
Some(_) => return Err(format!("`save` must be a path or a list of at most {MAX_SAVES} paths")),
};
Ok(Script { file: opt_string(obj, "file")?.map(|f| resolve(&f, base)), template: opt_string(obj, "template")?, commands, save })
}
@@ -0,0 +1,49 @@
//! Explicit field refresh for headless conversions; ordinary conversion keeps cached results.
use std::process::Command;
use wordcraft_doc::para::InlineObject;
use wordcraft_doc::{Document, Path, Pos, StoryRef};
#[test]
fn convert_refresh_is_opt_in() {
let dir = std::env::temp_dir().join(format!("wc-convert-fields-{}", std::process::id()));
std::fs::create_dir_all(&dir).unwrap();
let mut doc = Document::from_text("Contents\nTarget");
doc.para_mut(StoryRef::Body, &Path::top(1)).unwrap().props.page_break_before = Some(true);
doc.insert_object(&Pos::body(1, 0), InlineObject::BookmarkStart { name: "target".into() }, &Default::default()).unwrap();
for locked in [true, false] {
doc.insert_object(
&Pos::body(0, 0),
InlineObject::Field { instr: "PAGEREF target".into(), result: "104".into(), locked, dirty: true, code: false, form: String::new() },
&Default::default(),
)
.unwrap();
}
let input = dir.join("in.json");
std::fs::write(&input, serde_json::to_vec(&doc).unwrap()).unwrap();
let docx = dir.join("in.docx");
let make = Command::new(env!("CARGO_BIN_EXE_wordcraft-cli")).arg("convert").arg(&input).arg(&docx).output().unwrap();
assert!(make.status.success(), "{}", String::from_utf8_lossy(&make.stderr));
for refresh in [false, true] {
let output = dir.join(if refresh { "refreshed.json" } else { "cached.json" });
let mut cmd = Command::new(env!("CARGO_BIN_EXE_wordcraft-cli"));
cmd.arg("convert").arg(&docx).arg(&output);
if refresh {
cmd.arg("--update-fields");
}
let run = cmd.output().unwrap();
assert!(run.status.success(), "{}", String::from_utf8_lossy(&run.stderr));
let out: Document = serde_json::from_slice(&std::fs::read(output).unwrap()).unwrap();
let results: Vec<_> = out
.para(StoryRef::Body, &Path::top(0))
.unwrap()
.objects
.iter()
.filter_map(|o| match o {
InlineObject::Field { result, .. } => Some(result.as_str()),
_ => None,
})
.collect();
assert_eq!(results, if refresh { vec!["2", "104"] } else { vec!["104", "104"] });
}
std::fs::remove_dir_all(dir).unwrap();
}
+1 -1
View File
@@ -29,4 +29,4 @@ rfd = { workspace = true }
wasm-bindgen = "0.2"
wasm-bindgen-futures = "0.4"
js-sys = "0.3"
web-sys = { version = "0.3", features = ["Window", "Document", "Element", "HtmlElement", "HtmlCanvasElement", "HtmlAnchorElement", "HtmlIFrameElement", "Blob", "BlobPropertyBag", "Url", "Location", "CssStyleDeclaration", "Event", "EventTarget", "BeforeUnloadEvent", "ClipboardEvent", "DataTransfer", "FileList", "File", "DomStringList", "IdbFactory", "IdbOpenDbRequest", "IdbRequest", "IdbDatabase", "IdbObjectStore", "IdbTransaction", "IdbTransactionMode", "MediaQueryList"] }
web-sys = { version = "0.3", features = ["Window", "Document", "Element", "HtmlElement", "HtmlCanvasElement", "HtmlAnchorElement", "HtmlIFrameElement", "Blob", "BlobPropertyBag", "Url", "Location", "CssStyleDeclaration", "Event", "EventTarget", "BeforeUnloadEvent", "ClipboardEvent", "DataTransfer", "FileList", "File", "DomStringList", "IdbFactory", "IdbOpenDbRequest", "IdbRequest", "IdbDatabase", "IdbObjectStore", "IdbTransaction", "IdbTransactionMode", "MediaQueryList", "Storage"] }
+6
View File
@@ -6,9 +6,15 @@
//!
//! Differences from the desktop app: no TCP control channel; Open and Insert › Pictures use the
//! browser file picker (bytes arrive through `Services::inbox`); Save and Export download.
//! Preferences (theme, interface size, editing options…) are kept in `localStorage` (`prefs`).
//! URL flags: `?webgl` forces WebGL2; `?sample` opens the sample document.
#![deny(clippy::unwrap_used, clippy::expect_used, clippy::panic, clippy::unimplemented, clippy::todo, clippy::unreachable)]
#[cfg(any(target_arch = "wasm32", test))]
mod prefs;
#[cfg(any(target_arch = "wasm32", test))]
mod save_fallback;
#[cfg(target_arch = "wasm32")]
mod web;
+88
View File
@@ -0,0 +1,88 @@
//! Preferences between visits (#2132): the same [`UiState`] the desktop app keeps in `ui.json`
//! (theme, interface size, recent symbols, editing options…), kept in the browser's
//! `localStorage` under [`KEY`]. Restored at startup; written back whenever it changes, so a
//! reload or a closed tab keeps the last setting (browsers don't reliably run exit code).
use wordcraft_ui_egui::{UiState, WordApp};
/// The `localStorage` key.
#[cfg_attr(not(target_arch = "wasm32"), allow(dead_code))]
pub const KEY: &str = "wordcraft.ui";
/// Largest stored value read back (`localStorage` holds about 5 MB per site).
const MAX_STORED: usize = 4 << 20;
/// Apply preferences stored on an earlier visit. Missing, oversized or unreadable values leave
/// the defaults.
pub fn restore(app: &mut WordApp, stored: Option<&str>) {
if let Some(s) = stored.filter(|s| s.len() <= MAX_STORED)
&& let Some(ui) = UiState::from_json(s.as_bytes())
{
app.apply_prefs(ui);
}
}
/// Tracks what was last written, so the store is written only when a preference changed.
#[derive(Default)]
pub struct PrefsSync {
last: Option<String>,
}
impl PrefsSync {
/// Starts from what is already stored, so restoring doesn't write it straight back.
pub fn new(stored: Option<String>) -> Self {
Self { last: stored }
}
/// The JSON to store when the app's preferences differ from what was last written.
pub fn changed(&mut self, app: &WordApp) -> Option<String> {
let json = match serde_json::to_string(&app.prefs()) {
Ok(j) => j,
Err(e) => {
log::warn!("preferences: cannot serialize: {e}");
return None;
}
};
if self.last.as_deref() == Some(json.as_str()) {
return None;
}
self.last = Some(json.clone());
Some(json)
}
}
#[cfg(test)]
mod tests {
use super::*;
use serde_json::json;
use wordcraft_ui_egui::Services;
fn app() -> WordApp {
WordApp::new(wordcraft_engine::Session::new(wordcraft_engine::sample::sample_document()), Services::default())
}
/// #2132: dark mode chosen on one visit is there after a reload.
#[test]
fn dark_mode_survives_reload() {
let mut first = app();
let mut sync = PrefsSync::new(None);
let initial = sync.changed(&first);
assert!(initial.is_some(), "first frame stores the preferences");
assert_eq!(sync.changed(&first), None, "unchanged preferences are not rewritten");
first.run("ui.theme", json!({"value": "dark"})).unwrap();
let stored = sync.changed(&first).expect("the theme change is stored");
let mut reloaded = app();
restore(&mut reloaded, Some(&stored));
assert_eq!(reloaded.ui.theme, wordcraft_ui_egui::theme::Appearance::Dark);
assert_eq!(PrefsSync::new(Some(stored)).changed(&reloaded), None, "restoring doesn't write back");
}
#[test]
fn junk_storage_keeps_defaults() {
for junk in [None, Some(""), Some("not json"), Some("[1,2]"), Some("{\"theme\":42}")] {
let mut a = app();
restore(&mut a, junk);
assert_eq!(a.ui.theme, wordcraft_ui_egui::theme::Appearance::System, "junk keeps the default appearance");
}
}
}
+46
View File
@@ -0,0 +1,46 @@
//! Save-picker rejection policy, shared with native tests of the browser fallback.
use wordcraft_ui_egui::async_saves::SaveOutcome;
/// A cancelled picker ends the save. Other picker failures try the browser download path.
/// Write/close errors do not pass through this function: they retain their own failure.
pub(crate) fn picker_rejected(name: &str, download: impl FnOnce() -> Result<(), String>) -> SaveOutcome {
if name == "AbortError" {
SaveOutcome::Cancelled
} else {
match download() {
Ok(()) => SaveOutcome::Saved,
Err(error) => SaveOutcome::Failed(error),
}
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn unavailable_picker_downloads_once() {
for name in ["SecurityError", "NotSupportedError", "TypeError", ""] {
let mut calls = 0;
let outcome = picker_rejected(name, || {
calls += 1;
Ok(())
});
assert_eq!(outcome, SaveOutcome::Saved);
assert_eq!(calls, 1);
}
}
#[test]
fn cancelled_picker_never_downloads() {
let outcome = picker_rejected("AbortError", || panic!("cancel must not download"));
assert_eq!(outcome, SaveOutcome::Cancelled);
}
#[test]
fn failed_fallback_remains_a_failed_save() {
let outcome = picker_rejected("SecurityError", || Err("download unavailable".into()));
assert_eq!(outcome, SaveOutcome::Failed("download unavailable".into()));
}
}
+57 -14
View File
@@ -65,11 +65,21 @@ pub fn start() {
// Pasted pictures wait apart from dropped files: the app inserts one only when the page is the edit target (#1349).
app.services.pasted = Some(pasted);
app.autosave = false;
// Preferences from an earlier visit (#2132), as the desktop app reads `ui.json`.
let stored = stored_prefs();
crate::prefs::restore(&mut app, stored.as_deref());
// A region chosen in File › Options replaces the browser's for the first blank
// document too (#838).
if !sample && app.session.prefs.region.is_some() {
let doc = app.session.blank_document();
app.session.set_document(doc);
}
let prefs = crate::prefs::PrefsSync::new(stored);
// No document to show: start on File › Home, as the desktop app does (#766).
if !sample {
app.start_screen();
}
Ok(Box::new(WebShell { app, inbox, messages }))
Ok(Box::new(WebShell { app, inbox, messages, prefs }))
}),
)
.await;
@@ -92,6 +102,16 @@ struct WebShell {
app: WordApp,
inbox: Inbox,
messages: Rc<RefCell<Vec<String>>>,
prefs: crate::prefs::PrefsSync,
}
fn local_storage() -> Option<web_sys::Storage> {
web_sys::window()?.local_storage().ok()?
}
/// The preferences stored on an earlier visit, if the browser allows storage.
fn stored_prefs() -> Option<String> {
local_storage()?.get_item(crate::prefs::KEY).ok()?
}
impl eframe::App for WebShell {
@@ -116,6 +136,17 @@ impl eframe::App for WebShell {
self.app.status(msg);
}
self.app.logic(ctx);
// Store changed preferences straight away: a reload or closed tab runs no exit code (#2132).
if let Some(json) = self.prefs.changed(&self.app) {
match local_storage() {
Some(store) => {
if let Err(e) = store.set_item(crate::prefs::KEY, &json) {
log::warn!("preferences: cannot save: {e:?}");
}
}
None => log::warn!("preferences: browser storage is unavailable"),
}
}
}
fn raw_input_hook(&mut self, _ctx: &egui::Context, raw: &mut egui::RawInput) {
@@ -193,13 +224,23 @@ fn services(inbox: Inbox, ctx: egui::Context, dirty: Rc<Cell<bool>>, saves: Asyn
} else {
rfd::AsyncFileDialog::new().add_filter("Documents", DOC_EXTS)
};
let many = purpose == "font";
wasm_bindgen_futures::spawn_local(async move {
let Some(file) = dialog.pick_file().await else {
return;
};
let bytes = file.read().await;
inbox.lock().unwrap_or_else(|e| e.into_inner()).push((file.file_name(), bytes));
ctx.request_repaint();
// Add Font takes every selected file (#1789); the others one.
let files = if many { dialog.pick_files().await.unwrap_or_default() } else { dialog.pick_file().await.into_iter().collect() };
for file in files.into_iter().take(MAX_PICKED_FILES) {
let mut bytes = file.read().await;
let mut pending = inbox.lock().unwrap_or_else(|e| e.into_inner());
// Fonts waiting for the app to drain them stay under a byte cap. A file
// over it is queued empty, so Add Font reports that file on its own.
let held: usize = pending.iter().map(|(_, b)| b.len()).fold(0, usize::saturating_add);
if many && held.saturating_add(bytes.len()) > MAX_PENDING_PICKED_BYTES {
bytes = Vec::new();
}
pending.push((file.file_name(), bytes));
drop(pending);
ctx.request_repaint();
}
});
})),
// Exports ask for a name; the browser decides where the download goes.
@@ -219,6 +260,10 @@ fn services(inbox: Inbox, ctx: egui::Context, dirty: Rc<Cell<bool>>, saves: Asyn
}
}
/// Most files one multi-select pick (Add Font) reads.
const MAX_PICKED_FILES: usize = 2000;
/// Most bytes of picked fonts held in the inbox before the app drains them.
const MAX_PENDING_PICKED_BYTES: usize = 256 << 20;
/// Most local fonts one Load System Installed Fonts installs (a designer's machine has thousands).
const MAX_SYSTEM_FONTS: usize = 500;
/// Most font bytes one Load System Installed Fonts reads into memory (wasm memory is small).
@@ -408,6 +453,7 @@ fn trigger_save_file_picker(
let prom: js_sys::Promise = prom_val.dyn_into().map_err(|_| "not a promise".to_string())?;
let bytes_vec = bytes.to_vec();
let window = window.clone();
let saves = saves.clone();
let ctx = ctx.clone();
let name = name.to_string();
@@ -429,13 +475,10 @@ fn trigger_save_file_picker(
Err(e) => {
let err_name =
js_sys::Reflect::get(&e, &wasm_bindgen::JsValue::from_str("name")).ok().and_then(|v| v.as_string()).unwrap_or_default();
if err_name == "AbortError" {
log::debug!("save picker cancelled");
SaveOutcome::Cancelled
} else {
log::error!("showSaveFilePicker error: {e:?}");
SaveOutcome::Failed(format!("{e:?}"))
}
// A browser can expose the picker but reject its promise (for example in
// a cross-origin iframe). Use the same download path as an absent API;
// cancelling a picker must never start an unwanted download (#1246).
crate::save_fallback::picker_rejected(&err_name, || trigger_anchor_download(&window, &name, &bytes_vec))
}
};
saves.finish(ticket, outcome);
+5 -1
View File
@@ -38,8 +38,12 @@ fmv-macos-events = "=0.1.0"
# NSWorkspace: a safe call in these bindings. Already in the build through eframe, rfd and arboard,
# at this version; this adds the `NSAccessibility` bindings and, to fit the title bar around the
# traffic lights, the window, button and view ones.
objc2-app-kit = { version = "0.3", default-features = false, features = ["std", "NSWorkspace", "NSAccessibility", "NSApplication", "NSWindow", "NSButton", "NSControl", "NSView", "NSResponder"] }
objc2-app-kit = { version = "0.3", default-features = false, features = ["std", "NSWorkspace", "NSAccessibility", "NSApplication", "NSWindow", "NSButton", "NSControl", "NSView", "NSResponder", "NSPasteboard"] }
objc2 = "0.6"
# A copied equation's MathML goes on the pasteboard as `public.mathml.presentation` beside
# arboard's HTML and text (#1297), through NSPasteboard's safe `setString:forType:`. Already in
# the build through arboard, at this version; this adds only the NSString bindings.
objc2-foundation = { version = "0.3", default-features = false, features = ["std", "NSString"] }
# Closing the window must quit through NSApplication rather than by dropping the NSWindow
# inside the AppKit run loop (see `schedule_macos_terminate`).
dispatch2 = "0.3"
+88 -8
View File
@@ -10,9 +10,14 @@
//! Everything read here comes from other programs: sizes are capped as it is read.
//!
//! Written here: a copy puts HTML beside its text, so other programs keep its formatting (#1523),
//! with equations as MathML (#816; pasting into Word and LibreOffice not yet verified). arboard sets HTML with a
//! text alternative on every desktop platform; formats of other names (Word's "Office Open XML"
//! with OMML, #1297) it can't, so they aren't offered.
//! with equations as MathML (#816). arboard sets HTML with a text alternative on every desktop
//! platform. A copy of one equation alone also goes in the platform's MathML formats (#1297),
//! added beside arboard's: on Windows "MathML" and "MathML Presentation" (UTF-16LE), on macOS
//! `public.mathml.presentation` and `public.mathml` — the names W3C MathML 3 §6.2.4 registers
//! for transferring MathML, and the ones Microsoft documents its Office apps reading first when
//! they paste math (learn.microsoft.com/office/math/mathml, "Clipboard support"). Pasting into
//! Word and LibreOffice is not yet verified. Linux and the BSDs have no Word, and arboard can't
//! add MIME types of other names, so there the HTML carries the MathML alone.
use wordcraft_engine::cmd::paste::MAX_PASTE_BYTES;
use wordcraft_ui_egui::paste_picture::ClipboardPicture;
@@ -26,13 +31,24 @@ thread_local! {
static WRITER: std::cell::RefCell<Option<arboard::Clipboard>> = const { std::cell::RefCell::new(None) };
}
/// Put `html` on the clipboard with `text` as its plain-text alternative. `false` when it couldn't
/// (the caller copies the text alone), including a Wayland session without the data-control
/// protocol, where arboard would fall back to X11 and native Wayland programs might not see it.
pub fn set_html(html: &str, text: &str) -> bool {
/// Put `html` on the clipboard with `text` as its plain-text alternative, and `mathml` (a copy of
/// one equation alone; empty for any other copy) in the platform's MathML formats. `false` when
/// the HTML and text couldn't be set (the caller copies the text alone), including a Wayland
/// session without the data-control protocol, where arboard would fall back to X11 and native
/// Wayland programs might not see it. MathML that can't be added is logged: the HTML still
/// carries it.
pub fn set_html(html: &str, text: &str, mathml: &str) -> bool {
if !platform::can_set() {
return false;
}
let set = set_html_text(html, text);
if set && !mathml.is_empty() && mathml.len() <= MAX_PASTE_BYTES {
platform::add_mathml(mathml);
}
set
}
fn set_html_text(html: &str, text: &str) -> bool {
WRITER.with(|w| {
let Ok(mut slot) = w.try_borrow_mut() else { return false };
if slot.is_none() {
@@ -146,6 +162,9 @@ mod platform {
)
}
/// arboard serves its own MIME types only; Word, which reads the MathML formats, isn't here.
pub fn add_mathml(_mathml: &str) {}
pub fn rtf() -> Option<Vec<u8>> {
if !wayland() {
return None;
@@ -193,9 +212,68 @@ mod platform {
pub fn can_set() -> bool {
true
}
/// Add the MathML to what arboard just set, as "MathML" and "MathML Presentation": UTF-16LE
/// text (with its terminating NUL, as Windows text formats have).
pub fn add_mathml(mathml: &str) {
let Ok(_open) = Clipboard::new_attempts(10) else {
log::warn!("clipboard: couldn't open it to add MathML");
return;
};
let bytes: Vec<u8> = mathml.encode_utf16().chain(std::iter::once(0)).flat_map(u16::to_le_bytes).collect();
for name in super::MATHML_FORMATS {
let Some(format) = raw::register_format(name) else { continue };
if let Err(e) = raw::set_without_clear(format.get(), &bytes) {
log::warn!("clipboard: {name}: {e}");
}
}
}
}
#[cfg(not(any(target_os = "windows", all(unix, not(any(target_os = "macos", target_os = "android", target_os = "emscripten"))))))]
/// The clipboard format names for MathML that Windows programs read (W3C MathML 3 §6.2.4; Office
/// reads and writes both).
#[cfg(target_os = "windows")]
const MATHML_FORMATS: [&str; 2] = ["MathML", "MathML Presentation"];
#[cfg(target_os = "macos")]
mod platform {
use objc2_app_kit::NSPasteboard;
use objc2_foundation::NSString;
pub fn formats() -> Vec<String> {
Vec::new()
}
pub fn rtf() -> Option<Vec<u8>> {
None
}
pub fn encoded_picture() -> Option<Vec<u8>> {
None
}
pub fn can_set() -> bool {
true
}
/// Add the MathML to the pasteboard item arboard just wrote, as the MathML types (W3C MathML 3
/// §6.2.4): presentation markup, and the generic type for readers that only know that one.
pub fn add_mathml(mathml: &str) {
let board = NSPasteboard::generalPasteboard();
let text = NSString::from_str(mathml);
for name in ["public.mathml.presentation", "public.mathml"] {
if !board.setString_forType(&text, &NSString::from_str(name)) {
log::warn!("clipboard: couldn't add {name}");
}
}
}
}
#[cfg(not(any(
target_os = "windows",
target_os = "macos",
all(unix, not(any(target_os = "macos", target_os = "android", target_os = "emscripten")))
)))]
mod platform {
pub fn formats() -> Vec<String> {
Vec::new()
@@ -212,4 +290,6 @@ mod platform {
pub fn can_set() -> bool {
true
}
pub fn add_mathml(_mathml: &str) {}
}
+84 -15
View File
@@ -12,6 +12,12 @@
//! thread first checks that one of them is there and otherwise answers with an error, and the app
//! shows its built-in file browser. `WORDCRAFT_FILE_DIALOG=builtin` always uses the built-in one.
//!
//! On Linux and the BSDs, rfd reports a portal that failed to show its dialog as a cancel (#2999).
//! There a cancel quicker than anyone could dismiss a dialog is taken as that failure: the dialog
//! is asked for once more without a parent window, and if that also comes back at once the
//! built-in browser shows. Windows and macOS always show their dialog, so a quick cancel there is
//! the person's and is answered as one.
//!
//! [`Services::file_dialog`]: wordcraft_ui_egui::Services::file_dialog
use std::future::Future;
@@ -27,7 +33,7 @@ type Request = (FileDialogRequest, Sender<FileDialogAnswer>);
pub type Hook = Box<dyn Fn(FileDialogRequest) -> Receiver<FileDialogAnswer>>;
/// A picker's answer, waited for off the UI thread.
type Picking = Pin<Box<dyn Future<Output = Option<rfd::FileHandle>> + Send>>;
type Picking = Pin<Box<dyn Future<Output = Vec<rfd::FileHandle>> + Send>>;
/// Shows the dialogs the app queued; lives in the eframe app, which has the window.
pub struct Launcher {
@@ -54,23 +60,22 @@ impl Launcher {
pub fn show(&self, parent: &eframe::Frame) {
while let Ok((req, answer)) = self.queue.try_recv() {
// Built here, on the UI thread: macOS shows the panel as a sheet on the window now.
let dialog = rfd::AsyncFileDialog::new().set_parent(parent);
let picking: Picking = match req {
FileDialogRequest::Open { purpose } => Box::pin(open_filters(dialog, &purpose).pick_file()),
FileDialogRequest::Save { name } => {
let mut dialog = dialog.set_file_name(name.clone());
for (label, exts) in save_filters(&name) {
dialog = dialog.add_filter(label, &exts);
}
Box::pin(dialog.save_file())
}
};
let picking = picker(rfd::AsyncFileDialog::new().set_parent(parent), &req);
#[cfg(all(unix, not(target_os = "macos")))]
let unparented = picker(rfd::AsyncFileDialog::new(), &req);
let ctx = self.ctx.clone();
let waiting = std::thread::Builder::new().name("file dialog".into()).spawn(move || {
// The picker only starts when first polled: without a system dialog it never does.
// The pickers only start when first polled: without a system dialog they never do.
let picked = match no_system_dialog() {
Some(reason) => Err(reason),
None => Ok(pollster::block_on(picking).map(|f| f.path().to_string_lossy().to_string())),
None => {
// Only the portal folds a failure into "cancelled" (#2999).
#[cfg(all(unix, not(target_os = "macos")))]
let picked = pick(picking, unparented, TOO_QUICK);
#[cfg(not(all(unix, not(target_os = "macos"))))]
let picked: Result<_, String> = Ok(pollster::block_on(picking));
picked.map(|f| f.iter().map(|f| f.path().to_string_lossy().to_string()).collect())
}
};
let _ = answer.send(picked);
ctx.request_repaint();
@@ -83,6 +88,53 @@ impl Launcher {
}
}
/// The picker for `req` on `dialog`.
fn picker(dialog: rfd::AsyncFileDialog, req: &FileDialogRequest) -> Picking {
match req.clone() {
FileDialogRequest::Open { purpose } => Box::pin(async move { open_filters(dialog, &purpose).pick_file().await.into_iter().collect() }),
FileDialogRequest::OpenMany { purpose } => Box::pin(async move { open_filters(dialog, &purpose).pick_files().await.unwrap_or_default() }),
FileDialogRequest::Save { name } => {
let mut dialog = dialog.set_file_name(name.clone());
for (label, exts) in save_filters(&name) {
dialog = dialog.add_filter(label, &exts);
}
Box::pin(async move { dialog.save_file().await.into_iter().collect() })
}
}
}
/// Quicker than anyone can see a dialog and dismiss it.
// ponytail: a timing heuristic, as rfd folds the portal's failure code into "cancelled". A portal
// failing slower than this still reads as a cancel; ask rfd for the response code if it matters.
#[cfg(all(unix, not(target_os = "macos")))]
const TOO_QUICK: std::time::Duration = std::time::Duration::from_millis(300);
/// Wait for `picking`. Nothing picked within `too_quick` is a portal that failed to show its
/// dialog (#2999), not a person cancelling: try `unparented`, as a portal may fail on the parent
/// window; if that too comes back at once, an error, so the app shows its built-in browser.
/// Linux and the BSDs only (built everywhere for its test).
#[cfg(any(test, all(unix, not(target_os = "macos"))))]
fn pick<T>(
picking: impl Future<Output = Vec<T>>,
unparented: impl Future<Output = Vec<T>>,
too_quick: std::time::Duration,
) -> Result<Vec<T>, String> {
fn timed<T>(f: impl Future<Output = Vec<T>>, too_quick: std::time::Duration) -> (Vec<T>, bool) {
let start = std::time::Instant::now();
let picked = pollster::block_on(f);
let quick = picked.is_empty() && start.elapsed() < too_quick;
(picked, quick)
}
match timed(picking, too_quick) {
(_, true) => log::warn!("the system file dialog closed at once; asking again without a parent window"),
(picked, false) => return Ok(picked),
}
match timed(unparented, too_quick) {
(_, true) => Err("the system file dialog closed at once".into()),
(picked, false) => Ok(picked),
}
}
/// The file types Open offers (shared with the built-in browser).
fn open_filters(mut d: rfd::AsyncFileDialog, purpose: &str) -> rfd::AsyncFileDialog {
for (name, exts) in wordcraft_ui_egui::open_filters(purpose) {
@@ -193,7 +245,24 @@ fn format_name(ext: &str) -> String {
#[cfg(test)]
mod tests {
use super::{save_exts, save_filters};
use super::{pick, save_exts, save_filters};
use std::time::Duration;
/// #2999: a dialog that "cancels" at once is a portal that failed: asked again without a
/// parent, then the built-in browser. A pick, or a cancel after a while, is the answer.
#[test]
fn an_instant_cancel_retries_then_falls_back() {
let quick = Duration::from_millis(200);
let slow_cancel = || async {
std::thread::sleep(Duration::from_millis(250));
Vec::<u8>::new()
};
assert_eq!(pick(async { vec![1u8] }, async { vec![2] }, quick), Ok(vec![1]));
assert_eq!(pick(async { vec![] }, async { vec![2u8] }, quick), Ok(vec![2]), "asked again without a parent");
assert!(pick(async { Vec::<u8>::new() }, async { vec![] }, quick).is_err(), "the built-in browser");
assert_eq!(pick(slow_cancel(), async { vec![2u8] }, quick), Ok(vec![]), "a real cancel isn't asked again");
assert_eq!(pick(async { vec![] }, slow_cancel(), quick), Ok(vec![]), "the second dialog cancelled for real");
}
/// #1399: the Save dialog's type list has one named entry per format, the suggested one first,
/// every savable extension exactly once.
+55 -3
View File
@@ -1,12 +1,37 @@
//! Windows graphics instance configuration (backends and the DirectX 12 shader compiler), applied
//! before wgpu can initialize any driver.
//! Graphics instance configuration applied before wgpu can initialize any driver: Windows
//! backends and the DirectX 12 shader compiler, and the Intel Mac indirect-validation skip.
// On macOS only the indirect-validation items are used outside tests. Elsewhere it is the other
// way round, so those two carry their own `allow`.
#![cfg_attr(target_os = "macos", allow(dead_code))]
use eframe::{
NativeOptions,
egui_wgpu::WgpuSetup,
wgpu::{Backends, DeviceType, Dx12Compiler},
wgpu::{Backends, DeviceType, Dx12Compiler, InstanceFlags},
};
/// wgpu compiles its indirect-call validation compute shaders inside `request_device`. On old
/// Intel Macs (HD Graphics 5000 on Big Sur, #2652) Metal's Intel
/// compiler aborts on them, the device is lost and eframe exits before the window opens. WordCraft
/// issues no indirect draws or dispatches, so Intel Macs skip them; every other GPU keeps wgpu's
/// default. Same fix as PhotoCraft's storytold/PhotoCraft#3171.
#[cfg_attr(not(target_os = "macos"), allow(dead_code))]
pub fn skips_indirect_validation(macos: bool, intel: bool) -> bool {
macos && intel
}
/// Leaves wgpu's indirect-call validation out of the instance eframe creates when `skip`.
/// `WGPU_VALIDATION_INDIRECT_CALL=1` still turns it back on.
#[cfg_attr(not(target_os = "macos"), allow(dead_code))]
pub fn configure_indirect_validation(options: &mut NativeOptions, skip: bool) {
if !skip {
return;
}
if let WgpuSetup::CreateNew(create) = &mut options.wgpu_options.wgpu_setup {
create.instance_descriptor.flags = create.instance_descriptor.flags.difference(InstanceFlags::VALIDATION_INDIRECT_CALL).with_env();
}
}
/// What the start-up probe learned about one DirectX 12 adapter.
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
pub struct Dx12Adapter {
@@ -153,6 +178,33 @@ fn probe_dx12(options: &NativeOptions) -> Vec<Dx12Adapter> {
mod tests {
use super::*;
#[test]
fn intel_macs_skip_the_indirect_validation_shaders() {
// #2652: Metal's Intel compiler aborted on them, losing the device before the window opened.
assert!(skips_indirect_validation(true, true));
// Apple silicon, Windows and Linux keep wgpu's validation.
for (macos, intel) in [(true, false), (false, true), (false, false)] {
assert!(!skips_indirect_validation(macos, intel), "macos={macos} intel={intel}");
}
let mut options = NativeOptions::default();
let WgpuSetup::CreateNew(before) = &options.wgpu_options.wgpu_setup else {
panic!("default native options must create a wgpu instance");
};
let before = before.instance_descriptor.flags;
configure_indirect_validation(&mut options, false);
let WgpuSetup::CreateNew(create) = &options.wgpu_options.wgpu_setup else {
panic!("default native options must create a wgpu instance");
};
assert_eq!(create.instance_descriptor.flags, before, "flags untouched when not skipping");
if std::env::var_os("WGPU_VALIDATION_INDIRECT_CALL").is_none() {
configure_indirect_validation(&mut options, true);
let WgpuSetup::CreateNew(create) = &options.wgpu_options.wgpu_setup else {
panic!("default native options must create a wgpu instance");
};
assert!(!create.instance_descriptor.flags.contains(InstanceFlags::VALIDATION_INDIRECT_CALL));
}
}
fn configured_backends(backends: Backends) -> Backends {
let mut options = NativeOptions::default();
configure_backends(&mut options, backends);
+4 -1
View File
@@ -16,7 +16,7 @@ mod clipboard;
mod control_server;
mod default_apps;
mod file_dialogs;
#[cfg(any(target_os = "windows", test))]
#[cfg(any(target_os = "windows", target_os = "macos", test))]
mod graphics;
mod logging;
mod print;
@@ -389,6 +389,9 @@ fn main() -> eframe::Result {
// FXC, which needs no DLL of ours (#603; see graphics.rs).
#[cfg(target_os = "windows")]
graphics::configure(&mut options, eframe::wgpu::Backends::from_env());
// Intel Macs: Metal's compiler aborts on wgpu's indirect-validation shaders (#2652).
#[cfg(target_os = "macos")]
graphics::configure_indirect_validation(&mut options, graphics::skips_indirect_validation(true, cfg!(target_arch = "x86_64")));
load_user_fonts();
let restored = saved_window();
if let Some(window) = restored {
+31
View File
@@ -119,6 +119,7 @@ display_name = "The Legend Of MTS Dev"
[people.dotCooCoo]
real_name = "Robert Lee"
sync_github_name = true
[people.rbr48]
real_name = "M S Rana"
@@ -135,3 +136,33 @@ sync_github_name = false
real_name = "Andrej Škvorc"
display_name = "Andrej Škvorc"
sync_github_name = false
[people.olekkut]
real_name = "Aleksander Kutycki"
display_name = "Aleksander Kutycki"
sync_github_name = true
[people.marcoarnulfo]
real_name = "Marco Arnulfo"
display_name = "Marco Arnulfo"
sync_github_name = false
[people.rem5411]
real_name = "Vivien Janny"
sync_github_name = true
[people.C00kieMonsta]
real_name = "Antoine Boxho"
sync_github_name = true
[people.matteo-andreuzza]
sync_github_name = true
[people.lucAmbr0]
sync_github_name = true
[people.maxm11]
real_name = "Max Murphy"
[people.SlavenDj]
sync_github_name = true
+3
View File
@@ -38,6 +38,9 @@ pub struct ShapeExtra {
/// 100000ths of the shorter side, at most [`MAX_CORNER`]; `None` draws [`DEFAULT_CORNER`].
#[serde(skip_serializing_if = "Option::is_none")]
pub corner: Option<u32>,
/// The outline dash pattern; absent in older saved documents means solid.
#[serde(skip_serializing_if = "crate::line_dash::LineDash::is_solid")]
pub dash: crate::line_dash::LineDash,
}
fn is_zero(v: &u32) -> bool {
+4
View File
@@ -242,6 +242,10 @@ pub struct ContentControl {
/// The run properties of `w:sdtPr` (`w:rPr`), kept verbatim (XML).
#[serde(skip_serializing_if = "String::is_empty")]
pub rpr_xml: String,
/// The same run properties, read: the formatting text typed in place of the placeholder
/// takes (ECMA-376 §17.5.2.27). `None` when the control names none.
#[serde(skip_serializing_if = "Option::is_none")]
pub run_props: Option<crate::props::CharProps>,
/// Other `w:sdtPr` children we don't model, kept verbatim (XML) in file order.
#[serde(skip_serializing_if = "Vec::is_empty")]
pub extra: Vec<String>,
+398 -31
View File
@@ -40,11 +40,76 @@ impl Fragment {
}
found
}
/// The equation (its linear format and whether it is a display equation) when the fragment
/// is that one equation and nothing else but white space and empty paragraphs: a copy of it
/// also goes on the system clipboard in the MathML formats that equation editors read (#1297).
pub fn sole_equation(&self) -> Option<(&str, bool)> {
let mut found = None;
for b in &self.blocks {
let Block::Para(p) = b else { return None };
for c in p.text.chars().filter(|c| !c.is_whitespace()) {
if c != crate::para::OBJ || found.is_some() {
return None;
}
// The paragraph's only visible character, so its first object.
match p.objects.first() {
Some(InlineObject::Equation { linear, display, .. }) => found = Some((linear.as_str(), *display)),
_ => return None,
}
}
}
found
}
/// The text a copy puts on the clipboard: [`Self::plain_text`] with symbol-font characters
/// (Wingdings…) as their Unicode equivalents.
pub fn clipboard_text(&self) -> String {
self.to_document().shown_text(crate::StoryRef::Body)
}
/// Add `piece` at the end, followed by a paragraph mark unless it already ends with one: how
/// Word's Spike collects what it cuts. The piece's text box stories get ids after this
/// fragment's, so pieces copied from different documents (or after an id was freed and reused)
/// never share one.
pub fn append_piece(&mut self, mut piece: Fragment) {
let Some(last) = piece.blocks.last() else { return };
let ends_with_mark = piece.blocks.len() >= 2 && matches!(last, Block::Para(p) if p.is_empty());
if !ends_with_mark {
let mut mark = Paragraph::new();
if let Block::Para(p) = last {
mark.props = p.props.clone();
}
piece.blocks.push(Block::Para(mark));
}
let next = self.parts.keys().next_back().map_or(1, |k| k.saturating_add(1));
let ids: BTreeMap<u32, u32> = piece.parts.keys().zip(next..=u32::MAX).map(|(old, new)| (*old, new)).collect();
let renumber = |b: &mut Block| {
each_para_mut(b, 0, &mut |p| {
let mut changed = false;
for o in p.objects.iter_mut() {
for slot in o.text_box_slots() {
if let Some(new) = slot.and_then(|old| ids.get(&old)) {
*slot = Some(*new);
changed = true;
}
}
}
if changed {
p.touch();
}
})
};
piece.blocks.iter_mut().for_each(renumber);
for (old, mut part) in std::mem::take(&mut piece.parts) {
part.blocks.iter_mut().map(Arc::make_mut).for_each(renumber);
if let Some(new) = ids.get(&old) {
self.parts.insert(*new, part);
}
}
// The empty paragraph after the last mark is where the next piece starts.
if self.blocks.len() >= 2 && matches!(self.blocks.last(), Some(Block::Para(p)) if p.is_empty()) {
self.blocks.pop();
}
self.blocks.append(&mut piece.blocks);
}
pub fn from_text(s: &str) -> Fragment {
let lines: Vec<&str> = s.split('\n').collect();
Fragment {
@@ -54,6 +119,19 @@ impl Fragment {
}
}
/// Whether block `b`, in a block list the paragraph walk reaches at `depth`, has cells the walk
/// skips ([`crate::PARA_WALK_DEPTH`]). Bounded by that depth.
fn deeper_than_walk(b: &Block, depth: usize) -> bool {
match b {
Block::Para(_) => false,
Block::Table(t) => t
.rows
.iter()
.flat_map(|r| &r.cells)
.any(|c| depth + 1 > crate::PARA_WALK_DEPTH || c.blocks.iter().any(|x| deeper_than_walk(x, depth + 1))),
}
}
fn order(a: &Pos, b: &Pos) -> (Pos, Pos) {
if a <= b { (a.clone(), b.clone()) } else { (b.clone(), a.clone()) }
}
@@ -314,7 +392,12 @@ impl Document {
if a.story != b.story {
return Vec::new();
}
self.para_paths(a.story).into_iter().filter(|p| *p >= a.path && *p <= b.path).collect()
// Only the blocks the range reaches are walked (not every paragraph of the story).
let mut out = Vec::new();
if let Some(bl) = self.story(a.story) {
crate::walk_between(bl, &mut Vec::new(), &mut out, 0, (&a.path.0, &b.path.0));
}
out
}
/// Delete `a..b`. Returns the position where the deletion collapsed to. On error the document
@@ -372,18 +455,42 @@ impl Document {
pb.clone()
};
let bl = self.container_mut(a.story, &a.path)?;
// Section breaks on the marks in between go with their paragraphs.
let between: Vec<_> = bl
.get(ia + 1..ib)
.unwrap_or_default()
.iter()
.filter_map(|b| match &**b {
Block::Para(p) => p.section.clone(),
Block::Table(_) => None,
})
.collect();
if ib < bl.len() && ia < ib {
bl.drain(ia + 1..=ib);
}
tail.props = ParaProps::default();
// The deleted marks include `a`'s, so a section break there goes too: the joined
// paragraph ends where `b`'s mark did, and its text takes the following section's
// setup, as in Word (#2127).
let sect = tail.section.take();
let pa = self.para_mut(a.story, &a.path)?;
let keep_props = pa.props.clone();
pa.append(tail);
pa.props = keep_props;
let before = std::mem::replace(&mut pa.section, sect);
if a.story == crate::StoryRef::Body && a.path.0.len() == 1 {
// Nearest first, as "link to previous" resolved them before the delete.
let following = self.section_mut(a.path.last());
for s in between.iter().rev().chain(before.iter()) {
following.absorb(s);
}
}
return Ok(a);
}
// Different containers (range enters or leaves a table): clear text in every paragraph
// inside; remove whole tables that are entirely covered when at the same level as `a`.
// inside, then remove the tables it covers whole (below).
// Whether `b` is at its paragraph's end, before that paragraph is trimmed.
let b_at_end = self.para(b.story, &b.path).is_some_and(|p| p.len() == b.off);
let paths = self.paths_between(&a, &b);
for p in paths.iter().rev() {
let (from, to) = {
@@ -394,41 +501,68 @@ impl Document {
};
self.para_mut(a.story, p)?.delete_except(from, to, &kept_in(p))?;
}
// Drop top-level tables strictly between a and b at a's level.
if a.path.depth() == 0 && b.path.depth() > 0 {
let first_b = b.path.0.first().copied().unwrap_or(0) as usize;
let ia = a.path.last();
let bl = self.container_mut(a.story, &a.path)?;
let mut i = first_b;
while i > ia + 1 {
i -= 1;
// Tables the range covers whole, in the innermost container holding both ends, go (#1198,
// #1844): those strictly between the ends, the one `a` is in when it starts at that
// table's very start, and the one `b` is in when it ends at that table's very end. A range
// that is one table and nothing else only empties it (Word's Delete keeps the table).
let common = a.path.0.iter().zip(&b.path.0).take_while(|(x, y)| x == y).count();
let level = common - common % 3;
let (Some(&ia), Some(&ib)) = (a.path.0.get(level), b.path.0.get(level)) else { return Ok(a) };
let (ia, ib) = (ia as usize, ib as usize);
if ia >= ib {
return Ok(a);
}
let prefix = a.path.0.get(..level).unwrap_or_default().to_vec();
let block_path = |i: usize| {
let mut v = prefix.clone();
v.push(i as u32);
Path(v)
};
// Paragraphs inside the block at index `i` of the container, in document order.
let paths = self.para_paths(a.story);
let inside = |i: usize| -> Vec<&Path> {
let pre = block_path(i);
paths.iter().filter(|q| q.0.len() > pre.0.len() && q.0.starts_with(&pre.0)).collect()
};
// A table with cells deeper than the paragraph walk goes has paragraphs `inside` never
// lists, so where it starts or ends isn't known: it is never taken as covered.
let walked = |i: usize| self.container(a.story, &block_path(i)).and_then(|bl| bl.get(i)).is_some_and(|b| !deeper_than_walk(b, level / 3));
let a_covers = a.path.0.len() > level + 1 && a.off == 0 && walked(ia) && inside(ia).first() == Some(&&a.path);
let b_covers = b.path.0.len() > level + 1 && walked(ib) && inside(ib).last() == Some(&&b.path) && b_at_end;
let first = if a_covers { ia } else { ia + 1 };
let last = if b_covers { ib } else { ib - 1 };
let mut removed = 0;
{
let bl = self.container_mut(a.story, &block_path(ia))?;
for i in (first..=last.min(bl.len().saturating_sub(1))).rev() {
if matches!(bl.get(i).map(|b| &**b), Some(Block::Table(_))) {
bl.remove(i);
removed += 1;
}
}
}
// From the very start of a top-level table to a top-level paragraph (Select All in a
// document that starts with a table, #1198): the tables are covered whole and go too,
// and the paragraphs left between join.
if a.path.depth() > 0 && b.path.depth() == 0 && a.off == 0 {
let ta = a.path.0.first().copied().unwrap_or(0) as usize;
let first = self.para_paths(a.story).into_iter().find(|q| q.0.first().is_some_and(|t| *t as usize == ta));
if first.as_ref() == Some(&a.path) {
let ib = b.path.last();
let bl = self.container_mut(a.story, &b.path)?;
let mut removed = 0;
for i in (ta..ib).rev() {
if matches!(bl.get(i).map(|b| &**b), Some(Block::Table(_))) {
bl.remove(i);
removed += 1;
}
}
let at = Pos { story: a.story, path: Path::top(ta), off: 0 };
let end = Pos { story: a.story, path: Path::top(ib.saturating_sub(removed)), off: 0 };
return self.delete_range(&at, &end);
if removed == 0 {
return Ok(a);
}
// A story or cell always ends with a paragraph.
if bl.is_empty() || matches!(bl.last().map(|b| &**b), Some(Block::Table(_))) {
bl.push(para_block(Paragraph::new()));
}
}
Ok(a)
// The paragraphs left between the ends join, as in a range within one container.
let at = if a_covers { Pos { story: a.story, path: block_path(ia), off: 0 } } else { a.clone() };
let end = if b_covers {
Pos { story: a.story, path: block_path((ib + 1).saturating_sub(removed)), off: 0 }
} else {
Pos { story: a.story, path: block_path(ib.saturating_sub(removed)), off: 0 }
};
let paras = |p: &Pos| p.path.0.len() == level + 1 && self.para(p.story, &p.path).is_some();
if paras(&at) && paras(&end) && at.path <= end.path {
return self.delete_range(&at, &end);
}
if self.para(at.story, &at.path).is_some() {
return Ok(at);
}
Ok(self.clamp(&at))
}
/// Apply `f` to the character formatting of `a..b`. A collapsed range changes nothing (the
@@ -694,6 +828,27 @@ mod tests {
Document::from_text(&lines.join("\n"))
}
/// Paths between two positions are those of all the story's paragraphs between them.
#[test]
fn paths_between_matches_every_path_filtered() {
let mut d = doc(&["a", "b", "c", "d", "e"]);
let mut inner = Table::new(2, 2, 100.0);
inner.rows[1].cells[0].blocks.push(crate::para_block(crate::Paragraph::new()));
let mut t = Table::new(3, 2, 200.0);
t.rows[1].cells[1].blocks.insert(0, Arc::new(Block::Table(inner)));
d.insert_block(StoryRef::Body, &Path::top(2), Block::Table(t)).unwrap();
d.insert_block(StoryRef::Body, &Path::top(4), Block::Table(Table::new(1, 3, 200.0))).unwrap();
let all = d.para_paths(StoryRef::Body);
for a in &all {
for b in &all {
let (pa, pb) = (Pos { story: StoryRef::Body, path: a.clone(), off: 0 }, Pos { story: StoryRef::Body, path: b.clone(), off: 0 });
let (lo, hi) = if a <= b { (a, b) } else { (b, a) };
let want: Vec<Path> = all.iter().filter(|p| *p >= lo && *p <= hi).cloned().collect();
assert_eq!(d.paths_between(&pa, &pb), want, "{a:?} {b:?}");
}
}
}
#[test]
fn split_and_join() {
let mut d = doc(&["Hello world"]);
@@ -737,6 +892,29 @@ mod tests {
assert_eq!(d.body.len(), 1);
}
/// #2127: a range over two section breaks joins three sections into the last; the headers
/// and footers it showed by linking to earlier ones stay, each from the nearest section that
/// named it, as they resolved before the delete.
#[test]
fn delete_over_two_section_breaks_keeps_the_linked_headers() {
let mut d = doc(&["A", "B", "C"]);
let set = |default| crate::section::HeaderSet { default: Some(default), ..Default::default() };
let s1 = crate::section::SectionProps { headers: set(1), footers: set(5), ..Default::default() };
let s2 = crate::section::SectionProps { headers: set(2), ..Default::default() };
let s3 = crate::section::SectionProps { margin_left: 99.0, ..Default::default() };
for (i, s) in [(0, s1), (1, s2), (2, s3.clone())] {
if let Some(Block::Para(p)) = d.body.get_mut(i).map(Arc::make_mut) {
p.section = Some(Box::new(s));
}
}
assert_eq!(d.sections().len(), 4);
d.delete_range(&Pos::body(0, 1), &Pos::body(2, 0)).unwrap();
assert_eq!(d.plain_text(StoryRef::Body), "AC");
let sections = d.sections();
assert_eq!(sections.len(), 2);
assert_eq!(*sections[0].1, crate::section::SectionProps { headers: set(2), footers: set(5), ..s3 });
}
#[test]
fn format_across() {
let mut d = doc(&["abc", "def"]);
@@ -914,6 +1092,42 @@ mod tests {
assert_eq!(Path(vec![1, 0, 1, 0]).cell(), Some((Path::top(1), 0, 1)));
}
fn para(text: &str) -> Block {
Block::Para(Paragraph::with_text(text, CharProps::default()))
}
#[test]
fn appended_pieces_are_each_followed_by_one_paragraph_mark() {
let mut spike = Fragment::default();
spike.append_piece(Fragment { blocks: vec![para("a")], ..Default::default() });
spike.append_piece(Fragment { blocks: vec![para("b")], ..Default::default() });
assert_eq!(spike.plain_text(), "a\nb\n");
// A piece that already ends with its paragraph mark gets no second one.
let mut spike = Fragment::default();
spike.append_piece(Fragment { blocks: vec![para("a"), para("")], ..Default::default() });
assert_eq!(spike.plain_text(), "a\n");
spike.append_piece(Fragment { blocks: vec![para("b"), para("c")], ..Default::default() });
assert_eq!(spike.plain_text(), "a\nb\nc\n");
}
#[test]
fn appended_pieces_keep_their_own_text_boxes() {
// Two documents give their first text box the same id.
let mut one = doc(&["x"]);
let id1 = text_box(&mut one, &Pos::body(0, 1), "One");
let mut two = doc(&["y"]);
let id2 = text_box(&mut two, &Pos::body(0, 1), "Two");
assert_eq!(id1, id2);
let mut spike = Fragment::default();
spike.append_piece(one.copy_range(&Pos::body(0, 0), &Pos::body(0, 1 + 3)));
spike.append_piece(two.copy_range(&Pos::body(0, 0), &Pos::body(0, 1 + 3)));
let mut dst = doc(&[""]);
dst.insert_fragment(&Pos::body(0, 0), &spike).unwrap();
let texts: Vec<String> =
[0, 1].iter().flat_map(|i| boxes_in(&dst, &Path::top(*i))).map(|b| dst.plain_text(StoryRef::Part(b.unwrap()))).collect();
assert_eq!(texts, ["One", "Two"]);
}
#[test]
fn bad_positions_error() {
let mut d = doc(&["é"]);
@@ -924,4 +1138,157 @@ mod tests {
assert_eq!(c, Pos::body(0, 2));
assert!(d.delete_range(&Pos { story: StoryRef::Part(3), path: Path::top(0), off: 0 }, &Pos::body(0, 0)).is_err());
}
// delete_range over whole tables (#1844), in the body, a header and a footnote.
/// A document whose story (the body, or a header or footnote part) holds `lines`.
fn story_doc(kind: Option<PartKind>, lines: &[&str]) -> (Document, StoryRef) {
match kind {
None => (doc(lines), StoryRef::Body),
Some(k) => {
let mut d = doc(&["body"]);
let blocks = lines.iter().map(|l| para_block(Paragraph::with_text(l, CharProps::default()))).collect();
let id = d.add_part(k, blocks);
(d, StoryRef::Part(id))
}
}
}
const STORIES: [Option<PartKind>; 3] = [None, Some(PartKind::Header), Some(PartKind::Footnote)];
fn at(story: StoryRef, path: &[u32], off: usize) -> Pos {
Pos { story, path: Path(path.to_vec()), off }
}
/// Insert an `rows`×`cols` table at `path` whose cells hold `texts`, row by row.
fn put_table(d: &mut Document, story: StoryRef, path: &[u32], rows: usize, cols: usize, texts: &[&str]) {
d.insert_block(story, &Path(path.to_vec()), Block::Table(Table::new(rows, cols, 200.0))).unwrap();
for (k, t) in texts.iter().enumerate() {
let mut p = path.to_vec();
p.extend([(k / cols) as u32, (k % cols) as u32, 0]);
d.insert_text(&at(story, &p, 0), t, &CharProps::default()).unwrap();
}
}
/// The story's top-level blocks: a paragraph's text, or `[table]`.
fn shape(d: &Document, story: StoryRef) -> Vec<String> {
d.story(story).unwrap().iter().map(|b| b.as_para().map_or("[table]".to_string(), |p| p.text.clone())).collect()
}
fn cell_text(d: &Document, story: StoryRef, table: &[u32], r: u32, c: u32) -> String {
let mut p = table.to_vec();
p.extend([r, c, 0]);
d.para(story, &Path(p)).map(|p| p.text.clone()).unwrap_or_else(|| "<none>".into())
}
#[test]
fn a_range_ending_at_the_end_of_a_cell_that_isnt_the_last_keeps_the_table() {
for kind in STORIES {
for (r, c) in [(0, 1), (1, 0)] {
let (mut d, st) = story_doc(kind, &["before", "after"]);
put_table(&mut d, st, &[1], 2, 2, &["a", "b", "c", "d"]);
d.delete_range(&at(st, &[0], 0), &at(st, &[1, r, c, 0], 1)).unwrap();
assert_eq!(shape(&d, st), ["", "[table]", "after"], "{kind:?} ({r},{c})");
assert_eq!(cell_text(&d, st, &[1], 1, 1), "d", "{kind:?} ({r},{c})");
}
}
}
#[test]
fn a_range_from_the_start_of_one_table_into_the_middle_of_the_next_removes_only_the_first() {
for kind in STORIES {
let (mut d, st) = story_doc(kind, &["x", "mid", "y"]);
put_table(&mut d, st, &[2], 2, 2, &["e", "f", "g", "h"]);
put_table(&mut d, st, &[1], 2, 2, &["a", "b", "c", "d"]);
assert_eq!(shape(&d, st), ["x", "[table]", "mid", "[table]", "y"]);
let to = d.delete_range(&at(st, &[1, 0, 0, 0], 0), &at(st, &[3, 0, 1, 0], 1)).unwrap();
assert_eq!(shape(&d, st), ["x", "", "[table]", "y"], "{kind:?}");
assert_eq!([cell_text(&d, st, &[2], 0, 0), cell_text(&d, st, &[2], 0, 1), cell_text(&d, st, &[2], 1, 0)], ["", "", "g"], "{kind:?}");
assert!(d.para(st, &to.path).is_some(), "{kind:?}: {to:?}");
}
}
#[test]
fn a_range_from_a_cell_paragraph_to_the_end_of_a_table_nested_in_that_cell_removes_only_the_nested_table() {
for kind in STORIES {
let (mut d, st) = story_doc(kind, &["before", "after"]);
put_table(&mut d, st, &[1], 1, 1, &["p"]);
put_table(&mut d, st, &[1, 0, 0, 1], 1, 1, &["n"]);
d.insert_block(st, &Path(vec![1, 0, 0, 2]), Block::Para(Paragraph::with_text("q", CharProps::default()))).unwrap();
let to = d.delete_range(&at(st, &[1, 0, 0, 0], 1), &at(st, &[1, 0, 0, 1, 0, 0, 0], 1)).unwrap();
assert_eq!(shape(&d, st), ["before", "[table]", "after"], "{kind:?}");
let cell = d.container(st, &Path(vec![1, 0, 0, 0])).unwrap();
let cell: Vec<String> = cell.iter().map(|b| b.as_para().map_or("[table]".to_string(), |p| p.text.clone())).collect();
assert_eq!(cell, ["pq"], "{kind:?}");
assert_eq!(to, at(st, &[1, 0, 0, 0], 1), "{kind:?}");
}
}
#[test]
fn two_whole_tables_at_the_end_of_a_story_leave_one_empty_paragraph() {
for kind in STORIES {
for trailing in [true, false] {
let (mut d, st) = story_doc(kind, &[""]);
put_table(&mut d, st, &[0], 1, 2, &["a", "b"]);
put_table(&mut d, st, &[1], 2, 1, &["c", "d"]);
if !trailing {
// A story that ends in a table (a hostile file).
d.story_mut(st).unwrap().pop();
}
let to = d.delete_range(&at(st, &[0, 0, 0, 0], 0), &at(st, &[1, 1, 0, 0], 1)).unwrap();
assert_eq!(shape(&d, st), [""], "{kind:?} trailing {trailing}");
assert_eq!(to, at(st, &[0], 0), "{kind:?} trailing {trailing}");
}
}
}
/// Content controls pair their markers within one block list (a story, or a cell), so a
/// marker in a table never has its partner outside it: a control inside a table removed whole
/// goes with it, and one the range only enters keeps both markers (the control is emptied).
#[test]
fn removing_a_whole_table_keeps_a_control_the_range_only_partly_covers() {
use crate::control::ContentControl;
let start = || InlineObject::ControlStart { control: Box::new(ContentControl::default()) };
for kind in STORIES {
let (mut d, st) = story_doc(kind, &["before", "after"]);
put_table(&mut d, st, &[1], 1, 1, &["c"]);
// One control around "efo" in "before", one around the cell's text.
d.insert_object(&at(st, &[0], 4), InlineObject::ControlEnd, &CharProps::default()).unwrap();
let in_before = d.insert_object(&at(st, &[0], 1), start(), &CharProps::default()).unwrap();
d.insert_object(&at(st, &[1, 0, 0, 0], 1), InlineObject::ControlEnd, &CharProps::default()).unwrap();
d.insert_object(&at(st, &[1, 0, 0, 0], 0), start(), &CharProps::default()).unwrap();
assert_eq!(d.control_ranges(st).len(), 2, "setup");
let end = d.para(st, &Path(vec![1, 0, 0, 0])).unwrap().len();
// From inside the first control to the end of the table.
d.delete_range(&Pos { off: in_before.off + 1, ..in_before }, &at(st, &[1, 0, 0, 0], end)).unwrap();
assert_eq!(shape(&d, st), ["b\u{FFFC}e\u{FFFC}after"], "{kind:?}: the table goes, the paragraphs join");
let ranges = d.control_ranges(st);
assert_eq!(ranges.len(), 1, "{kind:?}: {ranges:?}");
assert!(ranges.iter().all(|r| r.start.path == Path::top(0) && r.end.path == Path::top(0)), "{kind:?}");
let objects = d.para(st, &Path::top(0)).map_or(0, |p| p.objects.len());
assert_eq!(objects, 2, "{kind:?}: no orphan markers left");
}
}
#[test]
fn a_table_nested_deeper_than_the_paragraph_walk_is_never_removed_whole() {
// 20 tables, each in the first cell of the one around it, ahead of that cell's paragraph:
// the innermost paragraphs come first in the document but the paragraph walk stops short.
let mut t = Table::new(1, 1, 200.0);
if let Some(cell) = t.rows.get_mut(0).and_then(|r| r.cells.get_mut(0)) {
cell.blocks = vec![para_block(Paragraph::with_text("deep", CharProps::default()))];
}
for _ in 0..20 {
let mut outer = Table::new(1, 1, 200.0);
if let Some(cell) = outer.rows.get_mut(0).and_then(|r| r.cells.get_mut(0)) {
cell.blocks.insert(0, Arc::new(Block::Table(t)));
}
t = outer;
}
let mut d = doc(&["before", "after"]);
d.insert_block(StoryRef::Body, &Path::top(1), Block::Table(t)).unwrap();
let first = d.para_paths(StoryRef::Body).into_iter().find(|p| p.0.first() == Some(&1)).unwrap();
d.delete_range(&Pos { story: StoryRef::Body, path: first, off: 0 }, &Pos::body(2, 0)).unwrap();
assert!(d.body.get(1).is_some_and(|b| b.as_table().is_some()), "a table with paragraphs the walk never saw was removed");
}
}
+3 -2
View File
@@ -287,12 +287,12 @@ pub fn code_page_byte(code_page: u16, b: u8) -> char {
}
/// The code page of an RTF or Word font charset (`\fcharset`): 204 Russian → 1251, 238 Eastern
/// European → 1250, 77 Mac → 10000, 254 PC → 437… `None` for the ANSI and default charsets (the
/// European → 1250, 77 Mac → 10000, 254 PC / 255 OEM → 437… `None` for the ANSI and default charsets (the
/// document's code page applies), symbol fonts and the double-byte East Asian charsets.
pub fn charset_code_page(charset: i32) -> Option<u16> {
Some(match charset {
77 => MAC_ROMAN_CODE_PAGE,
254 => 437,
254 | 255 => 437,
161 => 1253,
162 => 1254,
163 => 1258,
@@ -323,6 +323,7 @@ mod tests {
assert_eq!(decode(10000, &[0x80, 0x8E, 0xDB, 0xF1, 0xF2, 0xF3, 0xD2, 0xD3, 0x41]), "Äé€ÒÚÛ“”A");
assert_eq!(charset_code_page(77), Some(10000));
assert_eq!(charset_code_page(254), Some(437));
assert_eq!(charset_code_page(255), Some(437));
assert_eq!(charset_code_page(204), Some(1251));
assert_eq!(charset_code_page(0), None);
for cp in SINGLE_BYTE_CODE_PAGES {
+163
View File
@@ -0,0 +1,163 @@
//! Legacy form fields (`FORMCHECKBOX`, `FORMTEXT`, `FORMDROPDOWN`): what their `w:ffData` says,
//! read from the raw XML a [`crate::InlineObject::Field`] keeps in its `form` (ECMA-376 Part 1
//! §17.16.17). Only what layout and PDF export need; the XML itself stays the source of truth
//! for saving.
/// Most drop-down entries kept (Word allows 25).
pub const MAX_ENTRIES: usize = 64;
/// A legacy form field's kind and state.
#[derive(Clone, Debug, PartialEq)]
pub enum LegacyForm {
/// A check box: `checked` is its state (`w:checked`, else `w:default`), `size` its explicit
/// size in points (`w:size`), `None` when it follows the text (`w:sizeAuto`).
CheckBox { name: String, checked: bool, size: Option<f32> },
/// A text field; its text is the field's result.
Text { name: String, max_len: Option<u32> },
/// A drop-down list showing entry `selected` (`w:result`, else `w:default`, else the first).
DropDown { name: String, entries: Vec<String>, selected: usize },
}
impl LegacyForm {
/// The field's definition, or `None` when `ff_data` holds none of the three.
pub fn parse(ff_data: &str) -> Option<LegacyForm> {
let name = attr_of(ff_data, "w:name").unwrap_or_default();
if let Some(cb) = element(ff_data, "w:checkBox") {
let default = attr_of(cb, "w:default").is_some_and(|v| on(&v));
// `<w:checked/>` with no value means checked.
let checked = match tag(cb, "w:checked") {
Some(t) => attr_in(t).is_none_or(|v| on(&v)),
None => default,
};
let size =
attr_of(cb, "w:size").and_then(|v| v.trim().parse::<f32>().ok()).filter(|v| v.is_finite() && *v > 0.0).map(|v| (v / 2.0).min(1584.0));
return Some(LegacyForm::CheckBox { name, checked, size });
}
if let Some(dd) = element(ff_data, "w:ddList") {
let entries: Vec<String> = tags(dd, "w:listEntry").filter_map(attr_in).take(MAX_ENTRIES).collect();
let pick = |n: &str| attr_of(dd, n).and_then(|v| v.trim().parse::<usize>().ok());
let selected = pick("w:result").or_else(|| pick("w:default")).filter(|i| *i < entries.len()).unwrap_or(0);
return Some(LegacyForm::DropDown { name, entries, selected });
}
if ff_data.contains("<w:textInput") {
let max_len =
element(ff_data, "w:textInput").and_then(|t| attr_of(t, "w:maxLength")).and_then(|v| v.trim().parse::<u32>().ok()).filter(|n| *n > 0);
return Some(LegacyForm::Text { name, max_len });
}
None
}
/// The text a drop-down shows (its selected entry); `None` for other kinds or an empty list.
pub fn shown_text(&self) -> Option<&str> {
match self {
LegacyForm::DropDown { entries, selected, .. } => entries.get(*selected).map(String::as_str),
_ => None,
}
}
/// The field's bookmark-style name (`w:name`), possibly empty.
pub fn name(&self) -> &str {
match self {
LegacyForm::CheckBox { name, .. } | LegacyForm::Text { name, .. } | LegacyForm::DropDown { name, .. } => name,
}
}
}
fn on(v: &str) -> bool {
!matches!(v.trim(), "0" | "false" | "off")
}
/// The element `name` in `xml` with its content: from its start tag to its end tag, or the
/// self-closing tag alone.
fn element<'a>(xml: &'a str, name: &str) -> Option<&'a str> {
let open = tag(xml, name)?;
let start = xml.find(open)?;
if open.ends_with("/>") {
return Some(open);
}
let close = format!("</{name}>");
let end = xml.get(start..)?.find(&close)? + start + close.len();
xml.get(start..end)
}
/// The first start (or empty) tag `<name ...>` in `xml`.
fn tag<'a>(xml: &'a str, name: &str) -> Option<&'a str> {
tags(xml, name).next()
}
/// Every start (or empty) tag `<name ...>` in `xml`, in order.
fn tags<'a>(xml: &'a str, name: &str) -> impl Iterator<Item = &'a str> + use<'a> {
let pat = format!("<{name}");
let mut from = 0;
std::iter::from_fn(move || {
loop {
let at = from + xml.get(from..)?.find(&pat)?;
let after = at + pat.len();
from = after;
// `<w:checkBox` must not match `<w:checkBoxes`: the name ends at a space, `/` or `>`.
if xml.get(after..).and_then(|s| s.chars().next()).is_some_and(|c| c == ' ' || c == '/' || c == '>') {
let end = after + xml.get(after..)?.find('>')? + 1;
return xml.get(at..end);
}
}
})
}
/// The `w:val` of the first tag `name` in `xml`.
fn attr_of(xml: &str, name: &str) -> Option<String> {
tag(xml, name).and_then(attr_in)
}
/// The `w:val` of a single tag, XML entities resolved.
fn attr_in(tag: &str) -> Option<String> {
let at = tag.find("w:val=")? + "w:val=".len();
let quote = tag.get(at..)?.chars().next()?;
if quote != '"' && quote != '\'' {
return None;
}
let rest = tag.get(at + 1..)?;
let end = rest.find(quote)?;
Some(unescape(rest.get(..end)?))
}
fn unescape(s: &str) -> String {
s.replace("&lt;", "<").replace("&gt;", ">").replace("&quot;", "\"").replace("&apos;", "'").replace("&amp;", "&")
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn check_boxes_text_fields_and_drop_downs() {
let cb = |inner: &str| format!(r#"<w:ffData><w:name w:val="Box1"/><w:enabled/><w:checkBox>{inner}</w:checkBox></w:ffData>"#);
assert_eq!(
LegacyForm::parse(&cb(r#"<w:sizeAuto/><w:default w:val="0"/>"#)),
Some(LegacyForm::CheckBox { name: "Box1".into(), checked: false, size: None })
);
assert_eq!(
LegacyForm::parse(&cb(r#"<w:sizeAuto/><w:default w:val="1"/>"#)).map(|f| matches!(f, LegacyForm::CheckBox { checked: true, .. })),
Some(true)
);
// `w:checked` wins over `w:default`; without a value it means checked.
assert!(matches!(
LegacyForm::parse(&cb(r#"<w:default w:val="1"/><w:checked w:val="0"/>"#)),
Some(LegacyForm::CheckBox { checked: false, .. })
));
assert!(matches!(LegacyForm::parse(&cb(r#"<w:default w:val="0"/><w:checked/>"#)), Some(LegacyForm::CheckBox { checked: true, .. })));
assert!(matches!(
LegacyForm::parse(&cb(r#"<w:size w:val="30"/><w:default w:val="0"/>"#)),
Some(LegacyForm::CheckBox { size: Some(15.0), .. })
));
let dd = r#"<w:ffData><w:name w:val="List1"/><w:ddList><w:result w:val="1"/><w:listEntry w:val="Alpha"/><w:listEntry w:val="B &amp; C"/></w:ddList></w:ffData>"#;
let f = LegacyForm::parse(dd).unwrap();
assert_eq!(f.shown_text(), Some("B & C"));
assert_eq!(f.name(), "List1");
let dd0 = r#"<w:ffData><w:ddList><w:listEntry w:val="Only"/></w:ddList></w:ffData>"#;
assert_eq!(LegacyForm::parse(dd0).and_then(|f| f.shown_text().map(str::to_string)).as_deref(), Some("Only"));
let tx = r#"<w:ffData><w:name w:val="Name1"/><w:textInput><w:maxLength w:val="40"/></w:textInput></w:ffData>"#;
assert_eq!(LegacyForm::parse(tx), Some(LegacyForm::Text { name: "Name1".into(), max_len: Some(40) }));
assert_eq!(LegacyForm::parse(""), None);
assert_eq!(LegacyForm::parse("<w:ffData><w:name w:val=\"x\"/></w:ffData>"), None);
}
}
+109 -1
View File
@@ -24,6 +24,8 @@ pub mod eq;
pub mod fields;
pub mod freeform;
pub mod graphic;
pub mod legacy_form;
pub mod line_dash;
pub mod link;
pub mod locale_number;
pub mod math;
@@ -310,6 +312,11 @@ pub struct TextBody {
/// Resize shape to fit text (`a:spAutoFit`, VML `mso-fit-shape-to-text`): Word grows the
/// box to its text, so none of the text is cut.
pub auto_fit: bool,
/// The colour the shape's style gives its text (`wps:style/a:fontRef`): what automatic text
/// colour means inside this box, instead of black. Text whose colour comes from its runs,
/// styles or the document defaults keeps it (Word, observed with synthetic documents).
#[serde(skip_serializing_if = "Option::is_none")]
pub auto_color: Option<props::Rgb>,
}
/// Word's default text box margins (left, top, right, bottom), EMU: 0.1" and 0.05".
@@ -387,6 +394,40 @@ impl Document {
}
}
/// A building block (Insert › Quick Parts, AutoText…) kept in the document's glossary
/// (ECMA-376 §17.12: `w:docPart`).
#[derive(Clone, Debug, PartialEq, Serialize, Deserialize, Default)]
#[serde(default, rename_all = "camelCase")]
pub struct BuildingBlock {
/// The entry's name (`w:docPartPr/w:name`).
pub name: String,
/// Its gallery (`w:category/w:gallery`: `docParts` for Quick Parts, `autoTxt` for AutoText…).
pub gallery: String,
/// Its category within the gallery (`w:category/w:name`, e.g. "General").
pub category: String,
#[serde(skip_serializing_if = "String::is_empty")]
pub description: String,
/// The entry's identity (`w:guid`), kept as read.
#[serde(skip_serializing_if = "String::is_empty")]
pub guid: String,
/// The paragraph style it's associated with (`w:style`), kept as read.
#[serde(skip_serializing_if = "String::is_empty")]
pub style: String,
/// Its types (`w:types/w:type`: `normal`, `bbPlcHdr` for a content control's placeholder…).
#[serde(skip_serializing_if = "Vec::is_empty")]
pub types: Vec<String>,
/// How it's inserted (`w:behaviors/w:behavior`: `content`, `p`, `pg`).
#[serde(skip_serializing_if = "Vec::is_empty")]
pub behaviors: Vec<String>,
/// The content (`w:docPartBody`).
pub blocks: Blocks,
}
/// The gallery Quick Parts saves to (`w:gallery w:val="docParts"`).
pub const QUICK_PARTS_GALLERY: &str = "docParts";
/// The AutoText gallery (`w:gallery w:val="autoTxt"`).
pub const AUTO_TEXT_GALLERY: &str = "autoTxt";
/// A bibliography source.
#[derive(Clone, Debug, PartialEq, Serialize, Deserialize, Default)]
#[serde(default, rename_all = "camelCase")]
@@ -448,6 +489,10 @@ pub struct CoreProps {
#[serde(default, rename_all = "camelCase")]
pub struct Settings {
pub track_changes: bool,
/// Record a tracked cut and paste of a sentence or paragraph as a move (Track moves); off
/// when the file has `w:doNotTrackMoves` (ECMA-376 §17.15.1.29), and then it is an
/// ordinary deletion and insertion.
pub track_moves: bool,
/// Default tab stop interval, points.
pub default_tab: f32,
pub even_odd_headers: bool,
@@ -614,6 +659,7 @@ impl Default for Settings {
fn default() -> Self {
Settings {
track_changes: false,
track_moves: true,
default_tab: 36.0,
even_odd_headers: false,
mirror_margins: false,
@@ -682,6 +728,9 @@ pub struct Document {
/// Source-format parts we keep verbatim for round-trip (e.g. custom XML), by part name.
#[serde(skip_serializing_if = "BTreeMap::is_empty")]
pub passthrough: BTreeMap<String, Arc<Vec<u8>>>,
/// The document's own building blocks (its glossary document), in file order.
#[serde(skip_serializing_if = "Vec::is_empty")]
pub glossary: Vec<BuildingBlock>,
}
impl Default for Document {
@@ -709,6 +758,7 @@ impl Document {
endnote_separators: NoteSeparators::default(),
media: BTreeMap::new(),
passthrough: BTreeMap::new(),
glossary: Vec::new(),
}
}
@@ -978,6 +1028,18 @@ impl Document {
&mut self.last_section
}
/// Remove the section break on the mark of top-level body paragraph `block`: its section
/// joins the following one ([`SectionProps::absorb`]). False when there is no break there.
pub fn remove_section_break(&mut self, block: usize) -> bool {
if !matches!(self.body.get(block).map(|b| &**b), Some(Block::Para(p)) if p.section.is_some()) {
return false;
}
let Some(Block::Para(p)) = self.body.get_mut(block).map(Arc::make_mut) else { return false };
let Some(before) = p.section.take() else { return false };
self.section_mut(block).absorb(&before);
true
}
/// Allocate a new part id and insert the part. Part ids run out only past `u32::MAX` parts, more
/// than memory holds; then nothing is added and the id is 0, which [`free_id`] never gives a part.
pub fn add_part(&mut self, kind: PartKind, blocks: Blocks) -> u32 {
@@ -1387,8 +1449,11 @@ pub fn count_words(s: &str) -> usize {
s.split(|c: char| c.is_whitespace() || c == para::OBJ).filter(|w| w.chars().any(char::is_alphanumeric)).count()
}
/// How deep in nested tables [`Document::para_paths`] goes: cells below this are skipped.
pub(crate) const PARA_WALK_DEPTH: usize = 16;
fn walk(bl: &Blocks, prefix: &mut Vec<u32>, out: &mut Vec<Path>, depth: usize) {
if depth > 16 {
if depth > PARA_WALK_DEPTH {
return;
}
for (i, b) in bl.iter().enumerate() {
@@ -1411,6 +1476,49 @@ fn walk(bl: &Blocks, prefix: &mut Vec<u32>, out: &mut Vec<Path>, depth: usize) {
}
}
/// [`walk`] for the paragraphs from path `a` to `b` inclusive, skipping the blocks outside.
pub(crate) fn walk_between(bl: &Blocks, prefix: &mut Vec<u32>, out: &mut Vec<Path>, depth: usize, (a, b): (&[u32], &[u32])) {
use std::cmp::Ordering::{Greater, Less};
if depth > PARA_WALK_DEPTH {
return;
}
// Where the paths starting with `q` lie against `x`: all before, all after, or either.
let side = |q: &[u32], x: &[u32]| {
let m = q.len().min(x.len());
q[..m].cmp(&x[..m])
};
for (i, blk) in bl.iter().enumerate() {
prefix.push(i as u32);
if side(prefix, b) == Greater {
prefix.pop();
break;
}
if side(prefix, a) != Less {
match &**blk {
Block::Para(_) => {
if prefix.as_slice() >= a && prefix.as_slice() <= b {
out.push(Path(prefix.clone()));
}
}
Block::Table(t) => {
for (r, row) in t.rows.iter().enumerate() {
for (c, cell) in row.cells.iter().enumerate() {
prefix.push(r as u32);
prefix.push(c as u32);
if side(prefix, a) != Less && side(prefix, b) != Greater {
walk_between(&cell.blocks, prefix, out, depth + 1, (a, b));
}
prefix.pop();
prefix.pop();
}
}
}
}
}
prefix.pop();
}
}
/// Which text of a paragraph [`text_of`] collects.
#[derive(Clone, Copy)]
enum TextMode {
+124
View File
@@ -0,0 +1,124 @@
//! DrawingML outline dashes, measured in multiples of the stroke width.
use serde::{Deserialize, Serialize};
pub const DASH_PERCENT: u32 = 100_000;
pub const MAX_DASH_STOPS: usize = 64;
const MIN_DASH_POINTS: f32 = 0.01;
const MAX_DASH_POINTS: f32 = 20_000.0;
#[derive(Clone, Debug, Default, PartialEq, Serialize, Deserialize)]
#[serde(rename_all = "camelCase")]
pub enum LineDash {
#[default]
Solid,
Dot,
Dash,
LongDash,
DashDot,
LongDashDot,
LongDashDotDot,
SystemDash,
SystemDot,
SystemDashDot,
SystemDashDotDot,
/// Dash/space pairs in 100000ths of the line width (`a:ds/@d`, `@sp`).
Custom(Vec<(u32, u32)>),
}
impl LineDash {
pub fn from_ooxml(value: &str) -> Self {
match value {
"dot" => Self::Dot,
"dash" => Self::Dash,
"lgDash" => Self::LongDash,
"dashDot" => Self::DashDot,
"lgDashDot" => Self::LongDashDot,
"lgDashDotDot" => Self::LongDashDotDot,
"sysDash" => Self::SystemDash,
"sysDot" => Self::SystemDot,
"sysDashDot" => Self::SystemDashDot,
"sysDashDotDot" => Self::SystemDashDotDot,
_ => Self::Solid,
}
}
pub fn ooxml(&self) -> Option<&'static str> {
Some(match self {
Self::Solid => "solid",
Self::Dot => "dot",
Self::Dash => "dash",
Self::LongDash => "lgDash",
Self::DashDot => "dashDot",
Self::LongDashDot => "lgDashDot",
Self::LongDashDotDot => "lgDashDotDot",
Self::SystemDash => "sysDash",
Self::SystemDot => "sysDot",
Self::SystemDashDot => "sysDashDot",
Self::SystemDashDotDot => "sysDashDotDot",
Self::Custom(_) => return None,
})
}
pub fn is_solid(&self) -> bool {
*self == Self::Solid
}
/// Page-point lengths for the painters. Bound hostile custom lists and tiny spans.
pub fn pattern(&self, width: f32) -> Vec<f32> {
let preset: &[f32] = match self {
Self::Solid => return Vec::new(),
Self::Dot => &[1.0, 3.0],
Self::Dash => &[4.0, 3.0],
Self::LongDash => &[8.0, 3.0],
Self::DashDot => &[4.0, 3.0, 1.0, 3.0],
Self::LongDashDot => &[8.0, 3.0, 1.0, 3.0],
Self::LongDashDotDot => &[8.0, 3.0, 1.0, 3.0, 1.0, 3.0],
Self::SystemDash => &[3.0, 1.0],
Self::SystemDot => &[1.0, 1.0],
Self::SystemDashDot => &[3.0, 1.0, 1.0, 1.0],
Self::SystemDashDotDot => &[3.0, 1.0, 1.0, 1.0, 1.0, 1.0],
Self::Custom(stops) => {
return stops
.iter()
.take(MAX_DASH_STOPS)
.flat_map(|&(d, sp)| [d, sp])
.map(|v| dash_points(v as f32 / DASH_PERCENT as f32, width))
.collect();
}
};
preset.iter().map(|v| dash_points(*v, width)).collect()
}
}
fn dash_points(ratio: f32, width: f32) -> f32 {
let width = if width.is_finite() { width.clamp(0.25, 200.0) } else { 0.75 };
(ratio * width).clamp(MIN_DASH_POINTS, MAX_DASH_POINTS)
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn dash_units_follow_line_width() {
assert_eq!(LineDash::Dash.pattern(1.0), [4.0, 3.0]);
assert_eq!(LineDash::Dash.pattern(2.0), [8.0, 6.0]);
assert_eq!(LineDash::Custom(vec![(200_000, 150_000)]).pattern(2.0), [4.0, 3.0]);
assert!(LineDash::Solid.pattern(2.0).is_empty());
}
#[test]
fn old_shapes_default_to_solid() {
let extra: crate::connector::ShapeExtra = serde_json::from_str("{}").unwrap();
assert!(extra.dash.is_solid());
assert!(!serde_json::to_string(&extra).unwrap().contains("dash"));
}
#[test]
fn custom_dashes_are_bounded() {
let pattern = LineDash::Custom(vec![(0, u32::MAX); MAX_DASH_STOPS + 1]).pattern(f32::NAN);
assert_eq!(pattern.len(), MAX_DASH_STOPS * 2);
assert!(pattern.iter().all(|v| v.is_finite() && *v >= MIN_DASH_POINTS && *v <= MAX_DASH_POINTS));
}
}
+14 -1
View File
@@ -111,10 +111,11 @@ pub fn format_number(v: f64, lang: Option<&str>) -> String {
if !v.is_finite() {
return "0".into();
}
if v.fract() == 0.0 && v.abs() < 1e15 {
if v.fract() == 0.0 && v.abs() < 9.2e18 {
return format!("{}", v as i64);
}
let s = format!("{v:.2}");
let s = if s == "-0.00" { "0.00".to_string() } else { s };
if decimal_separator(lang) == ',' { s.replace('.', ",") } else { s }
}
@@ -147,6 +148,7 @@ mod tests {
assert_eq!(parse_number(".", pl), None);
assert_eq!(format_number(5.75, pl), "5,75");
assert_eq!(format_number(1350.0, pl), "1350");
assert_eq!(format_number(-0.001, pl), "0,00", "a rounded zero has no sign");
}
#[test]
@@ -161,6 +163,8 @@ mod tests {
assert_eq!(parse_number("1.", en), Some(1.0));
assert_eq!(parse_number("0,125", en), None, "never 125");
assert_eq!(format_number(5.75, en), "5.75");
assert_eq!(format_number(-0.001, en), "0.00", "a rounded zero has no sign");
assert_eq!(format_number(-0.0, None), "0");
assert_eq!(decimal_separator(Some("de-CH")), '.');
assert_eq!(decimal_separator(Some("de-DE")), ',');
}
@@ -174,4 +178,13 @@ mod tests {
}
assert_eq!(format_number(f64::NAN, None), "0");
}
#[test]
fn large_whole_numbers_have_no_decimals() {
assert_eq!(format_number(1e15, None), "1000000000000000");
assert_eq!(format_number(2.5e15, None), "2500000000000000");
assert_eq!(format_number(-1e15, None), "-1000000000000000");
assert_eq!(format_number(9.0e18, None), "9000000000000000000");
assert_eq!(format_number(1e15 + 0.5, None), "1000000000000000.50");
}
}
+108 -10
View File
@@ -470,8 +470,27 @@ fn lin_arg(a: &[MNode], s: &mut String, depth: usize) {
if depth > MAX_DEPTH {
return;
}
for n in a {
for (i, n) in a.iter().enumerate() {
lin_node(n, s, depth + 1);
// A bare script or limit ends at a space: a letter, digit or period after it is its own item.
if matches!(n, MNode::Script { kind, .. } if *kind != ScriptKind::Pre) || matches!(n, MNode::Lim { .. }) {
let ends_bare = s.chars().next_back().is_some_and(|c| c.is_alphanumeric() || c == '.');
if ends_bare && a.get(i + 1).and_then(first_lin_char).is_some_and(|c| c.is_alphanumeric() || c == '.') {
s.push(' ');
}
}
}
}
/// The first character `n` is written with, when that is cheap to tell (letters and digits).
fn first_lin_char(n: &MNode) -> Option<char> {
match n {
MNode::Run(r) => r.text.chars().next(),
MNode::Func { name, .. } => name.first().and_then(first_lin_char),
MNode::Script { kind, base, .. } if *kind != ScriptKind::Pre => base.first().and_then(first_lin_char),
MNode::Lim { e, .. } => e.first().and_then(first_lin_char),
MNode::Frac { num, .. } => num.first().and_then(first_lin_char),
_ => None,
}
}
@@ -499,19 +518,25 @@ fn lin_operand(a: &[MNode], s: &mut String, depth: usize) {
}
}
/// A script or limit operand. The parser reads a script as one letter or a number (`x_10`), so a
/// longer run is grouped: `V_(dd)`, not `V_dd` (which reads back as V_d then d).
/// A script or limit operand. The parser reads a run of letters and digits up to the next space or
/// operator, so `V_dd` is written bare as Word's Linear format shows it (`V_(dd)` reads back the same).
fn lin_script(a: &[MNode], s: &mut String, depth: usize) {
// Bare only when the parser reads the whole run back, in the same style.
if let [MNode::Run(r)] = a
&& r.text.chars().nth(1).is_some()
&& !(r.text.starts_with(|c: char| c.is_ascii_digit()) && r.text.chars().all(|c| c.is_ascii_digit() || c == '.'))
{
s.push('(');
s.push_str(&r.text);
s.push(')');
} else {
lin_operand(a, s, depth);
let upright_name = r.sty == Some(MSty::Plain) && crate::math_linear::FUNCTION_NAMES.contains(&r.text.as_str());
let number = r.text.starts_with(|c: char| c.is_ascii_digit()) && r.text.chars().all(|c| c.is_alphanumeric() || c == '.');
let word = r.text.starts_with(char::is_alphabetic) && r.text.chars().all(char::is_alphanumeric);
let upright_other = r.sty == Some(MSty::Plain) && !upright_name;
if upright_other || (!number && !word) {
s.push('(');
s.push_str(&r.text);
s.push(')');
return;
}
}
lin_operand(a, s, depth);
}
fn lin_node(n: &MNode, s: &mut String, depth: usize) {
@@ -962,7 +987,8 @@ mod tests {
#[test]
fn multi_letter_scripts_round_trip() {
// A script reads one letter or a number, so `V_dd` is V_d then d: longer scripts are grouped.
// Word for Mac 16.113.4 (#1609): a script is the whole run up to a space or operator, and Word's
// Linear format shows it bare (`V_dd`); the parenthesised form reads back identically.
let script = |kind, sub: &str, sup: &str| MNode::Script { kind, base: vec![run("V")], sub: parse_linear(sub), sup: parse_linear(sup) };
for n in [
vec![script(ScriptKind::Sub, "dd", "")],
@@ -985,6 +1011,78 @@ mod tests {
// Numbers and single letters stay bare.
assert_eq!(to_linear(&[script(ScriptKind::SubSup, "10", "2.5")]), "V_10^2.5");
assert_eq!(to_linear(&[script(ScriptKind::Sub, "d", "")]), "V_d");
assert_eq!(to_linear(&[script(ScriptKind::Sub, "dd", "")]), "V_dd");
assert_eq!(to_linear(&[script(ScriptKind::Sup, "", "ab")]), "V^ab");
}
#[test]
fn multi_letter_script_operands_follow_word() {
// #1609, Word for Mac 16.113.4: `V_dd` and `V_(dd)` build up the same; `dd` is all subscript.
let sub_dd = MNode::Script { kind: ScriptKind::Sub, base: vec![run("V")], sub: vec![run("dd")], sup: vec![] };
assert_eq!(parse_linear("V_dd"), vec![sub_dd.clone()]);
assert_eq!(parse_linear("V_(dd)"), vec![sub_dd]);
let sup_ab = MNode::Script { kind: ScriptKind::Sup, base: vec![run("x")], sub: vec![], sup: vec![run("ab")] };
assert_eq!(parse_linear("x^ab"), vec![sup_ab.clone()]);
assert_eq!(parse_linear("x^(ab)"), vec![sup_ab]);
// Delimiters still end the operand.
let both = MNode::Script { kind: ScriptKind::SubSup, base: vec![run("x")], sub: vec![run("i")], sup: vec![run("2")] };
assert_eq!(parse_linear("x_i^2"), vec![both]);
let a_i = MNode::Script { kind: ScriptKind::Sub, base: vec![run("a")], sub: vec![run("i")], sup: vec![] };
assert_eq!(parse_linear("a_i b"), vec![a_i, run("b")]);
let both2 = MNode::Script { kind: ScriptKind::SubSup, base: vec![run("x")], sub: vec![run("ij")], sup: vec![run("kl")] };
assert_eq!(parse_linear("x_ij^kl"), vec![both2]);
// Round trip is Word's bare form.
assert_eq!(to_linear(&parse_linear("V_(dd)")), "V_dd");
}
#[test]
fn a_bare_script_does_not_absorb_what_follows() {
// #1609 review: `x_i` then `y` must not read back as x with subscript `iy`.
let sc =
|kind, base: &str, sub: &str, sup: &str| MNode::Script { kind, base: vec![run(base)], sub: parse_linear(sub), sup: parse_linear(sup) };
let sin_x = MNode::Func { name: vec![MNode::Run(MRun::plain("sin"))], e: vec![run("x")] };
let lim = MNode::Lim { upper: false, e: vec![run("a")], lim: vec![run("n")] };
for n in [
vec![sc(ScriptKind::Sub, "x", "i", ""), run("y")],
vec![sc(ScriptKind::Sub, "a", "i", ""), sc(ScriptKind::Sub, "b", "j", "")],
vec![sc(ScriptKind::Sup, "x", "", "2"), run("y")],
vec![sc(ScriptKind::Sup, "x", "", "2"), run("3")],
vec![sc(ScriptKind::Sub, "x", "i", ""), sin_x],
vec![lim.clone(), run("y")],
vec![sc(ScriptKind::Sub, "x", "ab", ""), run("y")],
vec![sc(ScriptKind::Sub, "x", "i", ""), run(".5")],
] {
let lin = to_linear(&n);
assert_eq!(parse_linear(&lin), n, "{lin}");
}
assert_eq!(to_linear(&[sc(ScriptKind::Sub, "x", "i", ""), run("y")]), "x_i y");
// Nothing extra when the script already ends in a bracket or a symbol follows.
assert_eq!(to_linear(&[sc(ScriptKind::Sub, "x", "i", ""), run("+")]), "x_i+");
}
#[test]
fn a_function_name_in_a_script_stays_upright() {
// Word for Mac (#1609): `x_max` and `x_(max)` both give an upright function name.
let up = vec![MNode::Script { kind: ScriptKind::Sub, base: vec![run("x")], sub: vec![MNode::Run(MRun::plain("max"))], sup: vec![] }];
assert_eq!(parse_linear("x_max"), up);
assert_eq!(parse_linear("x_(max)"), up);
assert_eq!(parse_linear(&to_linear(&up)), up, "{}", to_linear(&up));
// Other upright text is not a function name: it keeps its parentheses.
let other = vec![MNode::Script { kind: ScriptKind::Sub, base: vec![run("x")], sub: vec![MNode::Run(MRun::plain("ab"))], sup: vec![] }];
assert_eq!(to_linear(&other), "x_(ab)");
}
#[test]
fn digits_then_letters_continue_a_script() {
// Word for Mac (#1609): a script runs to a space or operator, so `x_2a` is subscript "2a".
let sub = |t: &str| vec![MNode::Script { kind: ScriptKind::Sub, base: vec![run("x")], sub: vec![run(t)], sup: vec![] }];
assert_eq!(parse_linear("x_2a"), sub("2a"));
assert_eq!(parse_linear("x_a2"), sub("a2"));
assert_eq!(parse_linear("x_αβ"), sub("αβ"));
assert_eq!(parse_linear("x_10"), sub("10"));
assert_eq!(parse_linear("x_2.5"), sub("2.5"));
assert_eq!(to_linear(&sub("2a")), "x_2a");
assert_eq!(parse_linear(&to_linear(&sub("2a"))), sub("2a"));
}
#[test]
+116 -13
View File
@@ -6,7 +6,7 @@
//! spacing and every named symbol in [`crate::math_symbols`]. Unknown commands stay as text.
use crate::math::{
Arg, FracKind, MAX_DEPTH, MNode, MRun, MScr, MSty, ScriptKind, WRAP_DEPTH, any_too_deep, is_nary_char, math_class, merge_runs, too_deep,
Arg, ColJc, FracKind, MAX_DEPTH, MNode, MRun, MScr, MSty, ScriptKind, WRAP_DEPTH, any_too_deep, is_nary_char, math_class, merge_runs, too_deep,
};
use crate::math_linear::{FUNCTION_NAMES, map_args};
use crate::math_symbols::{name_of, symbol};
@@ -380,9 +380,30 @@ impl Lp {
one(MNode::Run(MRun { text: t, nor: true, sty: Some(MSty::Italic), ..Default::default() }))
}
"operatorname" => {
let starred = self.peek() == Some(&Tok::Ch('*'));
if starred {
self.i += 1;
}
let t = self.raw_arg();
let name = vec![MNode::Run(MRun::plain(t))];
self.apply(name)
if starred && self.peek() == Some(&Tok::Sub) {
self.i += 1;
let lim = self.arg();
vec![MNode::Lim { upper: false, e: name, lim }]
} else {
self.apply(name)
}
}
"tag" => {
let starred = self.peek() == Some(&Tok::Ch('*'));
if starred {
self.i += 1;
}
let t = self.raw_arg();
let num_str = if starred { format!("#{t}") } else { format!("# ({t})") };
let mut r = MRun::new(num_str);
r.lit = false;
vec![MNode::Run(r)]
}
"mathrm" | "rm" => {
let a = self.arg();
@@ -424,13 +445,15 @@ impl Lp {
"underline" => one(MNode::Bar { top: false, e: self.arg() }),
"boxed" | "fbox" => one(MNode::BorderBox { hide: [false; 4], strike: [false; 4], e: self.arg() }),
"phantom" => one(MNode::Phant { show: false, zero_wid: false, zero_asc: false, zero_desc: false, e: self.arg() }),
"overbrace" | "underbrace" | "overparen" | "underparen" => {
"overbrace" | "underbrace" | "overparen" | "underparen" | "overbracket" | "underbracket" => {
let top = c.starts_with("over");
let chr = match c {
"overbrace" => '⏞',
"underbrace" => '⏟',
"overparen" => '⏜',
_ => '⏝',
"underparen" => '⏝',
"overbracket" => '⎴',
_ => '⎵',
};
let e = self.arg();
let g = MNode::GroupChr { chr, top, e };
@@ -540,6 +563,22 @@ impl Lp {
}
fn environment(&mut self, env: &str) -> Arg {
let base = env.trim_end_matches('*');
let mut col_jc = Vec::new();
if base == "array" || base == "tabular" {
let _ = self.opt();
let spec = self.raw_arg();
for ch in spec.chars() {
match ch {
'l' | 'L' => col_jc.push(ColJc::Left),
'r' | 'R' => col_jc.push(ColJc::Right),
'c' | 'C' => col_jc.push(ColJc::Center),
_ => {}
}
}
} else if base == "aligned" {
let _ = self.opt();
}
// Rows of `&`-separated cells up to `\end{…}`.
let mut rows: Vec<Vec<Arg>> = vec![vec![]];
loop {
@@ -567,7 +606,6 @@ impl Lp {
if rows.len() > 1 && rows.last().is_some_and(|r| r.iter().all(|c| c.is_empty())) {
rows.pop();
}
let base = env.trim_end_matches('*');
let delims = match base {
"pmatrix" => Some((Some('('), Some(')'))),
"bmatrix" => Some((Some('['), Some(']'))),
@@ -585,8 +623,8 @@ impl Lp {
elems: vec![vec![inner]],
};
match base {
"matrix" | "pmatrix" | "bmatrix" | "Bmatrix" | "vmatrix" | "Vmatrix" | "smallmatrix" | "array" => {
let m = MNode::Matrix { rows, col_jc: Vec::new() };
"matrix" | "pmatrix" | "bmatrix" | "Bmatrix" | "vmatrix" | "Vmatrix" | "smallmatrix" | "array" | "tabular" => {
let m = MNode::Matrix { rows, col_jc };
match delims {
Some((b, e)) => vec![wrap(m, b, e)],
None => vec![m],
@@ -774,6 +812,17 @@ fn delim_tex(c: Option<char>) -> String {
}
}
fn group_chr_tex(chr: char, top: bool) -> &'static str {
match chr {
'⏜' => "\\overparen",
'⏝' => "\\underparen",
'⎴' => "\\overbracket",
'⎵' => "\\underbracket",
_ if top => "\\overbrace",
_ => "\\underbrace",
}
}
fn node(n: &MNode, s: &mut String, depth: usize) {
match n {
MNode::Run(r) => {
@@ -877,12 +926,16 @@ fn node(n: &MNode, s: &mut String, depth: usize) {
s.push(' ');
braced(e, s, depth);
}
MNode::Delim { beg, end, elems, .. } => {
MNode::Delim { beg, end, sep, elems, .. } => {
s.push_str("\\left");
s.push_str(&delim_tex(*beg));
let mid = match sep {
Some(c) => format!("\\middle{}", delim_tex(Some(*c))),
None => "\\middle|".to_string(),
};
for (i, e) in elems.iter().enumerate() {
if i > 0 {
s.push_str("\\middle|");
s.push_str(&mid);
}
arg(e, s, depth);
}
@@ -901,11 +954,10 @@ fn node(n: &MNode, s: &mut String, depth: usize) {
s.push('_');
braced(lim, s, depth);
} else if let [MNode::GroupChr { chr, top, e: inner }] = e.as_slice() {
s.push_str(if *top { "\\overbrace" } else { "\\underbrace" });
s.push_str(group_chr_tex(*chr, *top));
braced(inner, s, depth);
s.push(if *upper { '^' } else { '_' });
braced(lim, s, depth);
let _ = chr;
} else {
s.push_str(if *upper { "\\overset" } else { "\\underset" });
braced(lim, s, depth);
@@ -941,8 +993,8 @@ fn node(n: &MNode, s: &mut String, depth: usize) {
braced(e, s, depth);
}
MNode::Boxed { e } => braced(e, s, depth),
MNode::GroupChr { top, e, .. } => {
s.push_str(if *top { "\\overbrace" } else { "\\underbrace" });
MNode::GroupChr { chr, top, e } => {
s.push_str(group_chr_tex(*chr, *top));
braced(e, s, depth);
}
MNode::EqArr { rows } => {
@@ -1034,6 +1086,15 @@ mod tests {
}
}
#[test]
fn group_character_latex_roundtrip() {
for cmd in ["\\overparen{a+b}", "\\underparen{a+b}", "\\overbracket{x}", "\\underbracket{x}"] {
let n = parse_latex(cmd);
let tex = to_latex(&n);
assert_eq!(tex, cmd);
}
}
#[test]
fn hostile_latex_never_panics() {
for s in ["", "\\", "{", "}", "^", "_", "\\frac", "\\sqrt[", "\\left(", "\\right)", "\\begin{matrix", "&&\\\\", "{{{{", "\\end{x}", "a^^b__c"]
@@ -1049,4 +1110,46 @@ mod tests {
let deep = "\\hat".repeat(5000) + "x";
let _ = parse_latex(&deep);
}
#[test]
fn array_column_specs_and_aligned_optional_args() {
let n = parse_latex(r"\begin{array}{lcr} a & b & c \\ d & e & f \end{array}");
let MNode::Matrix { rows, col_jc } = &n[0] else { panic!("{n:#?}") };
assert_eq!(col_jc, &[ColJc::Left, ColJc::Center, ColJc::Right]);
assert_eq!(rows.len(), 2);
assert_eq!(crate::math::to_linear(&rows[0][0]), "a");
assert_eq!(crate::math::to_linear(&rows[0][1]), "b");
assert_eq!(crate::math::to_linear(&rows[0][2]), "c");
let n = parse_latex(r"\begin{aligned}[t] x &= 1 \\ y &= 2 \end{aligned}");
assert!(matches!(&n[0], MNode::EqArr { rows } if rows.len() == 2));
}
#[test]
fn operatorname_starred_and_tag() {
let n = parse_latex(r"\operatorname*{argmin}_{x} f(x)");
assert!(matches!(&n[0], MNode::Lim { upper: false, .. }));
let n = parse_latex(r"E=mc^2 \tag{1}");
let MNode::EqArr { rows } = &n[0] else { panic!("{n:#?}") };
assert!(crate::math::has_number_mark(&rows[0]));
let n = parse_latex(r"E=mc^2 \tag*{(A)}");
let MNode::EqArr { rows } = &n[0] else { panic!("{n:#?}") };
assert!(crate::math::has_number_mark(&rows[0]));
}
#[test]
fn delim_preserves_custom_separator() {
let delim = MNode::Delim {
beg: Some('{'),
end: Some('}'),
sep: Some('‖'),
grow: true,
shp_match: false,
elems: vec![vec![MNode::Run(MRun::new("x"))], vec![MNode::Run(MRun::new("y"))]],
};
let tex = to_latex(&[delim]);
assert!(tex.contains(r"\middle\|"), "Expected \\middle\\|, got: {tex}");
}
}
+13 -7
View File
@@ -587,8 +587,8 @@ impl P {
}
}
/// A script or limit: a parenthesised group, a structure, a number, a signed operand or one
/// character.
/// A script or limit: a parenthesised group, a structure, a number, a signed operand, a run of
/// letters (and digits) or one other character.
fn script_operand(&mut self, stops: &[char]) -> Arg {
if self.tired() {
return self.rest();
@@ -612,7 +612,7 @@ impl P {
Some(c) if is_prefix(c) || closer(c).is_some() => self.atom(stops),
Some(c) if c.is_ascii_digit() => {
let start = self.i;
while self.peek().is_some_and(|k| k.is_ascii_digit() || k == '.') {
while self.peek().is_some_and(|k| k.is_alphanumeric() || k == '.') {
self.i += 1;
}
let t: String = self.c.get(start..self.i).unwrap_or(&[]).iter().collect();
@@ -624,10 +624,16 @@ impl P {
v.extend(self.script_operand(stops));
v
}
Some(c) if c.is_ascii_alphabetic() => {
// A function name as a script (`e^sin x` is rare): one letter.
self.i += 1;
vec![text(c)]
Some(c) if c.is_alphabetic() => {
// Word (#1609): the whole run of letters and digits up to a space or operator, so
// `V_dd` is V with a subscript `dd`, exactly like `V_(dd)`.
let start = self.i;
while self.peek().is_some_and(|k| k.is_alphanumeric()) {
self.i += 1;
}
let t: String = self.c.get(start..self.i).unwrap_or(&[]).iter().collect();
// A known function name stays upright, as Word builds it.
vec![MNode::Run(if FUNCTION_NAMES.contains(&t.as_str()) { MRun::plain(t) } else { MRun::new(t) })]
}
Some(c) if !stops.contains(&c) && c != ' ' && !is_closer(c) => {
self.i += 1;
+7 -2
View File
@@ -156,6 +156,11 @@ impl ListKind {
}
}
/// The bullet a level of a plain bulleted list shows: • ○ ▪ repeating (`level` is zero-based).
pub fn default_bullet(level: usize) -> char {
['•', '○', '▪'][level % 3]
}
fn bullet_level(i: usize, c: char) -> Level {
Level {
format: NumFormat::Bullet,
@@ -188,8 +193,8 @@ pub fn levels_for(kind: ListKind) -> Vec<Level> {
let lvl = |format: NumFormat, text: String| Level { format, text, indent: ind, hanging: 18.0, ..Level::default() };
let n = i + 1;
match kind {
ListKind::Bullet => bullet_level(i, ['•', '○', '▪'][i % 3]),
ListKind::BulletChar(c) => bullet_level(i, if i == 0 { c } else { ['•', '○', '▪'][i % 3] }),
ListKind::Bullet => bullet_level(i, default_bullet(i)),
ListKind::BulletChar(c) => bullet_level(i, if i == 0 { c } else { default_bullet(i) }),
ListKind::Numbered => lvl(cyc(NumFormat::Decimal, NumFormat::LowerLetter, NumFormat::LowerRoman), format!("%{n}.")),
ListKind::NumberedParen => lvl(cyc(NumFormat::Decimal, NumFormat::LowerLetter, NumFormat::LowerRoman), format!("%{n})")),
ListKind::UpperLetter => lvl(cyc(NumFormat::UpperLetter, NumFormat::LowerLetter, NumFormat::LowerRoman), format!("%{n}.")),
+96 -3
View File
@@ -216,6 +216,10 @@ pub struct Float {
/// laid out, as Word for Windows does (Word for Mac draws hidden VML shapes).
#[serde(skip_serializing_if = "is_false")]
pub hidden: bool,
/// Anchored in a table cell but placed on the page, not in the cell (`wp:anchor/@layoutInCell`
/// off, any VML shape in a header or footer): a picture watermark in a header that starts with a table.
#[serde(skip_serializing_if = "is_false")]
pub outside_cell: bool,
}
/// Longest alternative text kept (characters): hostile files and commands are capped.
@@ -316,6 +320,60 @@ pub enum NoteKind {
Endnote,
}
/// Where a positional tab ([`InlineObject::PositionalTab`]) lands on its line, and how the text
/// after it lines up there (ECMA-376 `ST_PTabAlignment`).
#[derive(Clone, Copy, Debug, PartialEq, Eq, Hash, Serialize, Deserialize, Default)]
#[serde(rename_all = "camelCase")]
pub enum PTabAlign {
#[default]
Left,
Center,
Right,
}
impl PTabAlign {
pub const ALL: [PTabAlign; 3] = [PTabAlign::Left, PTabAlign::Center, PTabAlign::Right];
/// The OOXML name (`left`, `center`, `right`).
pub fn ooxml(self) -> &'static str {
match self {
PTabAlign::Left => "left",
PTabAlign::Center => "center",
PTabAlign::Right => "right",
}
}
pub fn from_ooxml(v: &str) -> Option<PTabAlign> {
PTabAlign::ALL.into_iter().find(|a| a.ooxml() == v)
}
}
/// What a positional tab's place is measured against (ECMA-376 `ST_PTabRelativeTo`): the text
/// margins, or the paragraph's indents.
#[derive(Clone, Copy, Debug, PartialEq, Eq, Hash, Serialize, Deserialize, Default)]
#[serde(rename_all = "camelCase")]
pub enum PTabBase {
#[default]
Margin,
Indent,
}
impl PTabBase {
pub const ALL: [PTabBase; 2] = [PTabBase::Margin, PTabBase::Indent];
/// The OOXML name (`margin`, `indent`).
pub fn ooxml(self) -> &'static str {
match self {
PTabBase::Margin => "margin",
PTabBase::Indent => "indent",
}
}
pub fn from_ooxml(v: &str) -> Option<PTabBase> {
PTabBase::ALL.into_iter().find(|a| a.ooxml() == v)
}
}
/// How ruby (phonetic guide) text lines up over its base text (ECMA-376 `ST_RubyAlign`).
#[derive(Clone, Copy, Debug, PartialEq, Eq, Hash, Serialize, Deserialize, Default)]
#[serde(rename_all = "camelCase")]
@@ -549,6 +607,17 @@ pub enum InlineObject {
#[serde(default, skip_serializing_if = "CharProps::is_empty")]
props: Box<CharProps>,
},
/// A positional tab (Word's alignment tab, `w:ptab`): it goes to the left edge, centre or
/// right edge of the margins or of the indents, whatever the paragraph's tab stops, and the
/// text after it lines up there. Text already past that place sends it to the next line.
PositionalTab {
#[serde(default)]
align: PTabAlign,
#[serde(default)]
relative_to: PTabBase,
#[serde(default)]
leader: crate::props::TabLeader,
},
/// Something we don't model, kept for round-trip (raw XML of the source format).
Opaque {
format: String,
@@ -678,6 +747,14 @@ impl InlineObject {
})
.collect()
}
/// A legacy form field's definition (`FORMCHECKBOX`, `FORMTEXT`, `FORMDROPDOWN`), read from
/// the `w:ffData` it keeps; `None` for anything else.
pub fn legacy_form(&self) -> Option<crate::legacy_form::LegacyForm> {
match self {
InlineObject::Field { form, .. } if !form.is_empty() => crate::legacy_form::LegacyForm::parse(form),
_ => None,
}
}
/// Zero-width markers don't take part in layout.
pub fn is_marker(&self) -> bool {
matches!(
@@ -756,6 +833,7 @@ impl InlineObject {
InlineObject::Field { result, .. } => result,
InlineObject::Equation { linear, .. } => linear,
InlineObject::Ruby { base, .. } => base,
InlineObject::PositionalTab { .. } => "\t",
InlineObject::Opaque { text, .. } => text,
_ => "",
}
@@ -870,7 +948,7 @@ impl Paragraph {
k += 1;
continue;
}
let font = runs.peek().filter(|(r, _)| r.contains(&i)).and_then(|(_, p)| p.font.as_deref());
let font = runs.peek().filter(|(r, _)| r.contains(&i)).and_then(|(_, p)| p.shown_font());
out.push(font.and_then(|f| crate::symbol_fonts::symbol_char(f, c)).unwrap_or(c));
}
out
@@ -1322,7 +1400,7 @@ impl Paragraph {
let mut a = 0;
let mut prev_end = false;
for (i, c) in t.char_indices() {
if i >= off {
if i > off {
break;
}
if prev_end && c != ' ' {
@@ -1335,7 +1413,9 @@ impl Paragraph {
}
}
let mut b = t.len();
let mut ended = false;
// A caret on the space right after a sentence's end belongs to that sentence; a terminator
// directly followed by text ("Hello.World") doesn't end the sentence the caret is in.
let mut ended = t.get(..off).unwrap_or("").ends_with(['.', '!', '?']) && t.get(off..).unwrap_or("").starts_with(' ');
for (i, c) in t.get(off..).unwrap_or("").char_indices() {
if ended && c != ' ' {
b = off + i;
@@ -1353,6 +1433,19 @@ impl Paragraph {
mod tests {
use super::*;
#[test]
fn sentence_at_sentence_start() {
let p = Paragraph::with_text("Hello. World", CharProps::default());
assert_eq!(p.sentence_at(0), (0, 7));
assert_eq!(p.sentence_at(6), (0, 7));
assert_eq!(p.sentence_at(7), (7, 12));
assert_eq!(p.sentence_at(8), (7, 12));
assert_eq!(p.sentence_at(99), (7, 12));
let p = Paragraph::with_text("Hello.World", CharProps::default());
assert_eq!(p.sentence_at(6), (6, 11));
assert_eq!(p.sentence_at(8), (6, 11));
}
fn bold() -> CharProps {
CharProps { bold: Some(true), ..Default::default() }
}
+202 -6
View File
@@ -155,6 +155,46 @@ impl Underline {
}
}
/// The OpenType ligatures a run asks for ([MS-DOCX] `w14:ligatures`): standard (`liga`),
/// contextual (`clig`), historical (`hlig`) and discretional (`dlig`).
#[derive(Clone, Copy, Debug, PartialEq, Eq, Hash, Serialize, Deserialize, Default)]
#[serde(default, rename_all = "camelCase")]
pub struct Ligatures {
pub standard: bool,
pub contextual: bool,
pub historical: bool,
pub discretional: bool,
}
impl Ligatures {
/// Every combination of the four flags.
pub fn all() -> impl Iterator<Item = Ligatures> {
(0u8..16).map(|b| Ligatures { standard: b & 1 != 0, contextual: b & 2 != 0, historical: b & 4 != 0, discretional: b & 8 != 0 })
}
/// The `w14:ligatures/@w14:val` name: `none`, `all`, or the set flags joined in camelCase
/// (`standardContextual`).
pub fn ooxml(self) -> String {
let on: Vec<&str> =
[(self.standard, "standard"), (self.contextual, "contextual"), (self.historical, "historical"), (self.discretional, "discretional")]
.into_iter()
.filter_map(|(on, n)| on.then_some(n))
.collect();
match on.len() {
0 => "none".into(),
4 => "all".into(),
_ => on
.iter()
.enumerate()
.flat_map(|(i, n)| n.chars().enumerate().map(move |(j, c)| if i > 0 && j == 0 { c.to_ascii_uppercase() } else { c }))
.collect(),
}
}
/// `None` for a name [MS-DOCX] doesn't define.
pub fn from_ooxml(s: &str) -> Option<Ligatures> {
Ligatures::all().find(|l| l.ooxml() == s)
}
}
#[derive(Clone, Copy, Debug, PartialEq, Eq, Hash, Serialize, Deserialize, Default)]
#[serde(rename_all = "camelCase")]
pub enum VertAlign {
@@ -286,6 +326,11 @@ pub struct CharProps {
pub style: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub font: Option<String>,
/// The symbol's own font (OOXML `w:sym/@w:font`, ECMA-376 §17.3.3.30), set on a symbol
/// character from a `w:sym` or Insert › Symbol: it is shown in this font whatever `font` is,
/// so changing the text's font keeps the symbol. `None` = the run's font.
#[serde(skip_serializing_if = "Option::is_none")]
pub sym_font: Option<String>,
/// Points.
#[serde(skip_serializing_if = "Option::is_none")]
pub size: Option<f32>,
@@ -331,6 +376,10 @@ pub struct CharProps {
/// Kerning for fonts at or above this size (points); 0 = off.
#[serde(skip_serializing_if = "Option::is_none")]
pub kern: Option<f32>,
/// OpenType ligatures (`w14:ligatures`). `None` leaves WordCraft's default: standard and
/// contextual ligatures.
#[serde(skip_serializing_if = "Option::is_none")]
pub ligatures: Option<Ligatures>,
#[serde(skip_serializing_if = "Option::is_none")]
pub outline: Option<bool>,
#[serde(skip_serializing_if = "Option::is_none")]
@@ -372,6 +421,10 @@ pub struct CharProps {
/// BCP 47 language of complex-script text (`w:lang/@w:bidi`, e.g. `fa-IR`).
#[serde(skip_serializing_if = "Option::is_none")]
pub lang_bidi: Option<String>,
/// East Asian typography: two lines in one and horizontal in vertical (`w:eastAsianLayout`).
/// Kept for round trip; layout doesn't apply it yet.
#[serde(skip_serializing_if = "Option::is_none")]
pub east_asian_layout: Option<EastAsianLayout>,
/// Hyperlink target: a URL, or `#bookmark` for an internal link.
#[serde(skip_serializing_if = "Option::is_none")]
pub link: Option<String>,
@@ -386,6 +439,60 @@ pub struct CharProps {
pub fmt_change: Option<Box<PropChange<CharProps>>>,
}
/// East Asian typography settings of a run (ISO/IEC 29500-1 §17.3.2.10, `w:eastAsianLayout`):
/// *two lines in one* sets the run's text on two sub-lines inside one line (warichu), *horizontal
/// in vertical* turns it 90° to read across a vertical line (tate-chu-yoko). Each attribute is
/// `None` when the file doesn't say.
#[derive(Clone, Debug, PartialEq, Eq, Default, Serialize, Deserialize)]
#[serde(default, rename_all = "camelCase")]
pub struct EastAsianLayout {
/// East Asian typography run id (`w:id`).
#[serde(skip_serializing_if = "Option::is_none")]
pub id: Option<i32>,
/// Two lines in one (`w:combine`).
#[serde(skip_serializing_if = "Option::is_none")]
pub combine: Option<bool>,
/// Brackets around two lines in one (`w:combineBrackets`).
#[serde(skip_serializing_if = "Option::is_none")]
pub combine_brackets: Option<CombineBrackets>,
/// Horizontal in vertical (`w:vert`).
#[serde(skip_serializing_if = "Option::is_none")]
pub vert: Option<bool>,
/// Compress horizontal-in-vertical text to the line height (`w:vertCompress`).
#[serde(skip_serializing_if = "Option::is_none")]
pub vert_compress: Option<bool>,
}
/// Brackets around two lines in one (`ST_CombineBrackets`).
#[derive(Clone, Copy, Debug, PartialEq, Eq, Hash, Serialize, Deserialize)]
#[serde(rename_all = "camelCase")]
pub enum CombineBrackets {
None,
Round,
Square,
Angle,
Curly,
}
impl CombineBrackets {
pub const ALL: [CombineBrackets; 5] =
[CombineBrackets::None, CombineBrackets::Round, CombineBrackets::Square, CombineBrackets::Angle, CombineBrackets::Curly];
/// OOXML `w:combineBrackets` value.
pub fn ooxml(self) -> &'static str {
match self {
CombineBrackets::None => "none",
CombineBrackets::Round => "round",
CombineBrackets::Square => "square",
CombineBrackets::Angle => "angle",
CombineBrackets::Curly => "curly",
}
}
/// The value for `s`, `None` for one the schema doesn't have.
pub fn from_ooxml(s: &str) -> Option<CombineBrackets> {
CombineBrackets::ALL.into_iter().find(|b| b.ooxml() == s)
}
}
/// A tracked formatting change (ECMA-376 §17.13.5: `w:rPrChange`, `w:pPrChange`,
/// `w:tblPrChange`, `w:trPrChange`, `w:tcPrChange`, `w:sectPrChange`): the properties as they
/// were before the change, and the revision (author, date) that made it.
@@ -440,9 +547,9 @@ impl CharProps {
/// The fields of `self` that differ from `base` (equal ones unset): what `self` adds when
/// overlaid on `base`.
pub fn minus(mut self, base: &CharProps) -> CharProps {
drop_same_fields!(self, base; style, font, size, bold, italic, underline, underline_color, strike, double_strike, color, highlight,
drop_same_fields!(self, base; style, font, sym_font, size, bold, italic, underline, underline_color, strike, double_strike, color, highlight,
shading, border, vert_align, caps, small_caps, hidden, spacing, scale, position, kern, outline, shadow, emboss, engrave, lang, no_proof, rtl,
cs, font_cs, size_cs, bold_cs, italic_cs, lang_bidi, link, ins, del);
cs, font_cs, size_cs, bold_cs, italic_cs, lang_bidi, east_asian_layout, link, ins, del, ligatures);
self
}
@@ -451,9 +558,9 @@ impl CharProps {
/// Plain and complex-script properties inherit independently. Commands that intend to
/// format both scripts must set both values explicitly.
pub fn overlay(&mut self, patch: &CharProps) {
overlay_fields!(self, patch; style, font, size, bold, italic, underline, underline_color, strike, double_strike, color, highlight,
overlay_fields!(self, patch; style, font, sym_font, size, bold, italic, underline, underline_color, strike, double_strike, color, highlight,
shading, border, vert_align, caps, small_caps, hidden, spacing, scale, position, kern, outline, shadow, emboss, engrave, lang, no_proof, rtl,
cs, font_cs, font_ea, size_cs, bold_cs, italic_cs, lang_bidi, link, ins, del, fmt_change);
cs, font_cs, font_ea, size_cs, bold_cs, italic_cs, lang_bidi, east_asian_layout, link, ins, del, fmt_change, ligatures);
if let Some(p) = &patch.text_effects {
self.text_effects.get_or_insert_with(Default::default).overlay(p);
}
@@ -465,10 +572,29 @@ impl CharProps {
pub fn is_empty(&self) -> bool {
*self == CharProps::default()
}
/// The font a character of this run is shown in: the symbol's own font when it has one
/// ([`Self::sym_font`]), else the run's font.
pub fn shown_font(&self) -> Option<&str> {
self.sym_font.as_deref().filter(|f| !f.trim().is_empty()).or(self.font.as_deref())
}
/// The same formatting for text typed or painted next to a symbol: without the symbol's own
/// font.
pub fn without_symbol(mut self) -> CharProps {
self.sym_font = None;
self
}
/// Clear direct formatting but keep what isn't "formatting" (links, revisions, character style
/// is cleared too, as Word's Clear Formatting does).
pub fn cleared(&self) -> CharProps {
CharProps { link: self.link.clone(), ins: self.ins, del: self.del, fmt_change: self.fmt_change.clone(), ..Default::default() }
CharProps {
link: self.link.clone(),
ins: self.ins,
del: self.del,
fmt_change: self.fmt_change.clone(),
// A symbol stays the symbol it is.
sym_font: self.sym_font.clone(),
..Default::default()
}
}
/// Just the formatting: without the link, revision marks and tracked formatting change.
pub fn formatting(&self) -> CharProps {
@@ -793,6 +919,16 @@ pub struct ParaProps {
/// Drop cap: lines to drop (0 = none) — applies to the first character(s).
#[serde(skip_serializing_if = "Option::is_none")]
pub drop_cap: Option<u8>,
/// Drop cap "In margin" (`w:dropCap="margin"`): the letter hangs in the left margin and the
/// text keeps its indent. Unset = dropped into the text.
#[serde(skip_serializing_if = "Option::is_none")]
pub drop_cap_margin: Option<bool>,
/// The drop cap's distance from the text, in points (`w:hSpace`). Unset = 0.
#[serde(skip_serializing_if = "Option::is_none")]
pub drop_cap_distance: Option<f32>,
/// The drop cap letter's font family (the cap run's `w:rFonts`). Unset = the text's font.
#[serde(skip_serializing_if = "Option::is_none")]
pub drop_cap_font: Option<String>,
/// A text frame (`w:framePr` other than a drop cap): the paragraph is placed apart from the
/// text, which may wrap around it — a footer's page number in a frame, for instance.
#[serde(skip_serializing_if = "Option::is_none")]
@@ -833,7 +969,7 @@ impl ParaProps {
pub fn overlay(&mut self, patch: &ParaProps) {
overlay_fields!(self, patch; style, align, indent_left, indent_right, indent_first, indent_left_chars, indent_right_chars, indent_first_chars, space_before, space_after, space_before_auto, space_after_auto, line_spacing,
contextual_spacing, mirror_indents, keep_next, keep_lines, page_break_before, widow_control, outline_level, numbering, tabs, shading, borders,
suppress_hyphens, suppress_line_numbers, bidi, drop_cap, frame, kinsoku, word_wrap, overflow_punct, top_line_punct, auto_space_de, auto_space_dn,
suppress_hyphens, suppress_line_numbers, bidi, drop_cap, drop_cap_margin, drop_cap_distance, drop_cap_font, frame, kinsoku, word_wrap, overflow_punct, top_line_punct, auto_space_de, auto_space_dn,
snap_to_grid);
}
pub fn overlaid(mut self, patch: &ParaProps) -> ParaProps {
@@ -1133,6 +1269,9 @@ pub struct CellProps {
pub valign: Option<VAlign>,
/// Cell margins override (top, left, bottom, right), points.
pub margins: Option<[f32; 4]>,
/// Per-side overrides when `margins` is absent; omitted sides inherit table defaults.
#[serde(skip_serializing_if = "Option::is_none")]
pub partial_margins: Option<[Option<f32>; 4]>,
/// Which way the cell's text runs.
pub text_direction: TextDirection,
/// The file states the horizontal direction (`lrTb`) outright; saving writes it back.
@@ -1148,6 +1287,17 @@ pub struct CellProps {
pub fmt_change: Option<Box<PropChange<CellProps>>>,
}
impl CellProps {
/// Resolve top, left, bottom and right margins against the current table defaults.
/// A complete override takes precedence over per-side overrides.
pub fn resolved_margins(&self, defaults: [f32; 4]) -> [f32; 4] {
if let Some(margins) = self.margins {
return margins;
}
self.partial_margins.map_or(defaults, |sides| std::array::from_fn(|i| sides[i].unwrap_or(defaults[i])))
}
}
#[cfg(test)]
mod tests {
use super::*;
@@ -1196,6 +1346,37 @@ mod tests {
}
}
/// The 16 `w14:ligatures` values of [MS-DOCX], each to its flags and back.
#[test]
fn ligature_names() {
let l = |standard, contextual, historical, discretional| Ligatures { standard, contextual, historical, discretional };
for (name, flags) in [
("none", l(false, false, false, false)),
("standard", l(true, false, false, false)),
("contextual", l(false, true, false, false)),
("historical", l(false, false, true, false)),
("discretional", l(false, false, false, true)),
("standardContextual", l(true, true, false, false)),
("standardHistorical", l(true, false, true, false)),
("contextualHistorical", l(false, true, true, false)),
("standardDiscretional", l(true, false, false, true)),
("contextualDiscretional", l(false, true, false, true)),
("historicalDiscretional", l(false, false, true, true)),
("standardContextualHistorical", l(true, true, true, false)),
("standardContextualDiscretional", l(true, true, false, true)),
("standardHistoricalDiscretional", l(true, false, true, true)),
("contextualHistoricalDiscretional", l(false, true, true, true)),
("all", l(true, true, true, true)),
] {
assert_eq!(Ligatures::from_ooxml(name), Some(flags), "{name}");
assert_eq!(flags.ooxml(), name);
}
assert_eq!(Ligatures::all().count(), 16);
for bad in ["", "Standard", "contextualStandard", "standardstandard", "allx"] {
assert_eq!(Ligatures::from_ooxml(bad), None, "{bad:?}");
}
}
#[test]
fn text_direction_names() {
let mut d = TextDirection::Horizontal;
@@ -1257,4 +1438,19 @@ mod tests {
let back: CharProps = serde_json::from_str(&s).unwrap();
assert_eq!(back, p);
}
#[test]
fn symbol_font_is_optional_in_json_and_survives_font_changes() {
// Older documents (no `symFont`) still load.
let old: CharProps = serde_json::from_str(r#"{"font":"Arial"}"#).unwrap();
assert_eq!(old.sym_font, None);
let p = CharProps { font: Some("Arial".into()), sym_font: Some("Wingdings".into()), ..Default::default() };
let s = serde_json::to_string(&p).unwrap();
assert!(s.contains(r#""symFont":"Wingdings""#), "{s}");
assert_eq!(serde_json::from_str::<CharProps>(&s).unwrap(), p);
// A font change keeps the symbol's own font; Clear Formatting too.
let changed = p.clone().overlaid(&CharProps { font: Some("Times New Roman".into()), ..Default::default() });
assert_eq!(changed.sym_font.as_deref(), Some("Wingdings"));
assert_eq!(p.cleared().sym_font.as_deref(), Some("Wingdings"));
}
}
+36 -1
View File
@@ -11,6 +11,8 @@ use crate::styles::{StyleKind, StyleSheet};
/// Automatic (HTML-style) paragraph spacing, points: what Word for Windows gives `w:beforeAutospacing`
/// and `w:afterAutospacing` at any font size (measured).
pub const AUTO_SPACING: f32 = 14.0;
/// The largest drop cap distance from the text, in points (22", as Word's dialog allows).
pub const MAX_DROP_CAP_DISTANCE: f32 = 1584.0;
/// Concrete character formatting.
#[derive(Clone, Debug, PartialEq, Serialize, Deserialize)]
@@ -39,6 +41,10 @@ pub struct ResolvedChar {
/// Kerning: the smallest font size, in points, that is kerned (`w:kern`); 0 = never kerned.
#[serde(default)]
pub kern: f32,
/// OpenType ligatures (`w14:ligatures`); `None` = the shaper's default (standard and
/// contextual).
#[serde(default)]
pub ligatures: Option<crate::props::Ligatures>,
pub outline: bool,
pub shadow: bool,
pub emboss: bool,
@@ -70,7 +76,8 @@ pub struct ResolvedChar {
impl ResolvedChar {
fn from(c: &CharProps) -> ResolvedChar {
ResolvedChar {
font: c.font.clone().unwrap_or_else(|| crate::styles::BODY_FONT.to_string()),
// A symbol character is shown in its own font (`w:sym`), whatever the run's font.
font: c.shown_font().map(str::to_string).unwrap_or_else(|| crate::styles::BODY_FONT.to_string()),
size: c.size.unwrap_or(11.0).clamp(1.0, 1638.0),
bold: c.bold.unwrap_or(false),
italic: c.italic.unwrap_or(false),
@@ -90,6 +97,7 @@ impl ResolvedChar {
scale: c.scale.unwrap_or(100.0).clamp(1.0, 600.0),
position: c.position.unwrap_or(0.0).clamp(-800.0, 800.0),
kern: c.kern.filter(|k| k.is_finite()).unwrap_or(0.0).clamp(0.0, 1638.0),
ligatures: c.ligatures,
outline: c.outline.unwrap_or(false),
shadow: c.shadow.unwrap_or(false),
emboss: c.emboss.unwrap_or(false),
@@ -172,6 +180,15 @@ pub struct ResolvedPara {
pub borders: Option<Borders>,
pub bidi: bool,
pub drop_cap: u8,
/// The drop cap hangs in the margin (see `ParaProps::drop_cap_margin`).
#[serde(default)]
pub drop_cap_margin: bool,
/// The drop cap's distance from the text, in points.
#[serde(default)]
pub drop_cap_distance: f32,
/// The drop cap letter's font family; `None` = the text's.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub drop_cap_font: Option<String>,
/// A text frame the paragraph sits in (see `ParaProps::frame`).
#[serde(default, skip_serializing_if = "Option::is_none")]
pub frame: Option<crate::props::TextFrame>,
@@ -250,6 +267,9 @@ impl ResolvedPara {
borders: p.borders.filter(Borders::any_visible),
bidi: p.bidi.unwrap_or(false),
drop_cap: p.drop_cap.unwrap_or(0).min(10),
drop_cap_margin: p.drop_cap_margin.unwrap_or(false),
drop_cap_distance: p.drop_cap_distance.filter(|d| d.is_finite()).unwrap_or(0.0).clamp(0.0, MAX_DROP_CAP_DISTANCE),
drop_cap_font: p.drop_cap_font.clone().filter(|f| !f.trim().is_empty()),
frame: p.frame.clone().filter(|_| p.drop_cap.unwrap_or(0) == 0),
suppress_hyphens: p.suppress_hyphens.unwrap_or(false),
suppress_line_numbers: p.suppress_line_numbers.unwrap_or(false),
@@ -349,6 +369,8 @@ impl StyleSheet {
}
}
}
// Only a character itself has a symbol font of its own (`w:sym`); a style never gives one.
cp.sym_font = None;
let mut direct = c.clone();
direct.style = None;
cp.overlay(&direct);
@@ -360,6 +382,19 @@ impl StyleSheet {
mod tests {
use super::*;
/// #2036: only a character itself (direct formatting) has a symbol font of its own; a
/// style never gives one to its text.
#[test]
fn a_style_never_gives_text_a_symbol_font() {
let mut s = StyleSheet::builtin();
if let Some(n) = s.get_mut("Normal") {
n.chr.sym_font = Some("Wingdings".into());
}
assert_ne!(s.resolve_char(None, &CharProps::default()).font, "Wingdings");
let direct = CharProps { sym_font: Some("Wingdings".into()), ..Default::default() };
assert_eq!(s.resolve_char(None, &direct).font, "Wingdings");
}
#[test]
fn heading_resolves() {
let s = StyleSheet::builtin();
+10
View File
@@ -795,6 +795,16 @@ fn is_horizontal(d: &crate::props::TextDirection) -> bool {
}
impl SectionProps {
/// Join the section before this one into it, as when the break between them is deleted:
/// this section's setup stays, as in Word, and the headers and footers it leaves out (linked
/// to the previous section) are taken from `before`, so its text keeps showing them.
pub fn absorb(&mut self, before: &SectionProps) {
for (own, prev) in [(&mut self.headers, &before.headers), (&mut self.footers, &before.footers)] {
own.default = own.default.or(prev.default);
own.first = own.first.or(prev.first);
own.even = own.even.or(prev.even);
}
}
/// Width of the text area, points.
pub fn text_width(&self) -> f32 {
(self.page_w - self.margin_left - self.margin_right - self.gutter).max(18.0)
+17 -7
View File
@@ -68,13 +68,14 @@ impl SmartArtLayout {
}
}
/// The layout an id names (any ASCII case; `list` is the basic list).
/// The layout an id names (any ASCII case, ignoring hyphens, underscores and spaces;
/// `list` is the basic list; `basic process` or `process` is process, etc.).
pub fn from_id(id: &str) -> Option<SmartArtLayout> {
let id = id.trim();
if id.eq_ignore_ascii_case("list") {
let norm = id.chars().filter(|c| !c.is_whitespace() && *c != '-' && *c != '_').collect::<String>();
if norm.eq_ignore_ascii_case("list") {
return Some(SmartArtLayout::BasicList);
}
SmartArtLayout::ALL.into_iter().find(|l| l.id().eq_ignore_ascii_case(id))
SmartArtLayout::ALL.into_iter().find(|l| l.id().eq_ignore_ascii_case(&norm) || l.label().replace(' ', "").eq_ignore_ascii_case(&norm))
}
/// The name shown for it.
@@ -159,7 +160,8 @@ impl SmartArtColors {
}
pub fn from_id(id: &str) -> Option<SmartArtColors> {
SmartArtColors::ALL.into_iter().find(|c| c.id().eq_ignore_ascii_case(id.trim()))
let norm = id.chars().filter(|c| !c.is_whitespace() && *c != '-' && *c != '_').collect::<String>();
SmartArtColors::ALL.into_iter().find(|c| c.id().eq_ignore_ascii_case(&norm) || c.label().replace(' ', "").eq_ignore_ascii_case(&norm))
}
/// The name shown for it.
@@ -287,8 +289,9 @@ impl SmartArtSpec {
/// `s` as item text: control characters as spaces, ends trimmed, at most [`MAX_TEXT`] characters.
pub fn clean_text(s: &str) -> String {
let s: String = s.chars().take(MAX_TEXT * 4).map(|c| if c.is_control() { ' ' } else { c }).collect();
s.trim().chars().take(MAX_TEXT).collect::<String>().trim_end().to_string()
let trimmed = s.trim_start_matches(|c: char| c.is_whitespace() || c.is_control());
let s: String = trimmed.chars().take(MAX_TEXT).map(|c| if c.is_control() { ' ' } else { c }).collect();
s.trim_end().to_string()
}
#[cfg(test)]
@@ -319,6 +322,13 @@ mod tests {
assert_eq!(s.descendants(5), 6..6);
assert_eq!(s.descendants(99), 0..0);
assert_eq!(SmartArtLayout::from_id("VENN"), Some(SmartArtLayout::Venn));
assert_eq!(SmartArtLayout::from_id("basic-list"), Some(SmartArtLayout::BasicList));
assert_eq!(SmartArtLayout::from_id("Basic Process"), Some(SmartArtLayout::Process));
assert_eq!(SmartArtLayout::from_id("basic_cycle"), Some(SmartArtLayout::Cycle));
assert_eq!(SmartArtLayout::from_id("list"), Some(SmartArtLayout::BasicList));
assert_eq!(SmartArtColors::from_id("colorful"), Some(SmartArtColors::Colorful));
assert_eq!(SmartArtColors::from_id("Accent 2"), Some(SmartArtColors::Accent2));
assert_eq!(SmartArtColors::from_id("accent-4"), Some(SmartArtColors::Accent4));
assert_eq!(clean_text(&format!("{}Valid Text", " ".repeat(1500))), "Valid Text");
}
}
+16 -1
View File
@@ -22,7 +22,7 @@
/// The symbol fonts this module knows, by lowercase family name.
fn table(family: &str) -> Option<&'static [u32; 224]> {
let f = family.trim();
if f.eq_ignore_ascii_case("symbol") {
if f.eq_ignore_ascii_case("symbol") || f.eq_ignore_ascii_case("symbol (pcl6)") {
Some(&SYMBOL)
} else if f.eq_ignore_ascii_case("wingdings") {
Some(&WINGDINGS)
@@ -243,6 +243,21 @@ mod tests {
}
}
#[test]
fn pcl6_symbol_encoding() {
for family in ["Symbol (PCL6)", " symbol (pcl6) "] {
assert!(is_symbol_font(family));
for code in 0x20..=0xFF {
let c = char::from_u32(code).unwrap();
assert_eq!(symbol_char(family, c), symbol_char("Symbol", c));
let pua = char::from_u32(0xF000 + code).unwrap();
assert_eq!(symbol_char(family, pua), symbol_char("Symbol", pua));
}
assert_eq!(symbol_char(family, '\u{F0B3}'), Some('≥'));
}
assert!(!is_symbol_font("Symbol (unknown)"));
}
#[test]
fn symbol_uses_the_adobe_symbol_encoding() {
assert_eq!(symbol_char("Symbol", '\u{F061}'), Some('α'));
+144 -23
View File
@@ -127,9 +127,27 @@ impl Table {
}
pub fn delete_row(&mut self, r: usize) {
if r < self.rows.len() {
self.rows.remove(r);
if r >= self.rows.len() {
return;
}
// A vertical merge that starts in this row now starts in the row below.
let starts: Vec<usize> = (0..self.rows.get(r).map_or(0, |row| row.cells.len()))
.filter(|&c| self.rows.get(r).and_then(|row| row.cells.get(c)).is_some_and(|x| x.props.vmerge == VMerge::Restart))
.map(|c| self.grid_col(r, c))
.collect();
for g in starts {
let continues = |t: &Self, row: usize| {
t.cell_at_grid(row, g).and_then(|i| t.rows.get(row)?.cells.get(i)).is_some_and(|x| x.props.vmerge == VMerge::Continue)
};
if !continues(self, r + 1) {
continue;
}
let more = continues(self, r + 2);
if let Some(c) = self.cell_at_grid(r + 1, g).and_then(|i| self.rows.get_mut(r + 1)?.cells.get_mut(i)) {
c.props.vmerge = if more { VMerge::Restart } else { VMerge::None };
}
}
self.rows.remove(r);
}
/// Insert a grid column at grid index `g` (cells spanning it widen instead).
@@ -256,28 +274,15 @@ impl Table {
}
}
/// Split cell `c` of row `r` into `n` cells horizontally. When the cell spans fewer grid
/// columns than that, the grid is refined inside the cell only: the new borders divide it
/// Split cell `c` of row `r` into `n` cells horizontally. Where the cell's grid lines don't
/// fall on the equal parts, the grid is refined inside the cell only: the new borders divide it
/// into equal parts, and the cells of other rows widen over the new grid columns, so their
/// extent and the table's width stay as they were.
pub fn split_cell(&mut self, r: usize, c: usize, n: usize) {
/// extent and the table's width stay as they were. Returns the parts' spans, left to right.
pub fn split_cell(&mut self, r: usize, c: usize, n: usize) -> Vec<usize> {
let n = n.clamp(1, MAX_COLS);
let Some(span) = self.rows.get(r).and_then(|row| row.cells.get(c)).map(Cell::span) else { return };
let Some(span) = self.rows.get(r).and_then(|row| row.cells.get(c)).map(Cell::span) else { return Vec::new() };
if n <= 1 {
return;
}
if span >= n {
let Some(row) = self.rows.get_mut(r) else { return };
let Some(cell) = row.cells.get_mut(c) else { return };
let w = cell.props.width.map(|w| w / n as f32);
cell.props.width = w;
cell.props.span = (span - (n - 1)) as u32;
for k in 1..n {
let mut nc = Cell::empty();
nc.props.width = w;
row.cells.insert(c + k, nc);
}
return;
return vec![span];
}
let g = self.grid_col(r, c);
let mut end = g.saturating_add(span);
@@ -286,7 +291,7 @@ impl Table {
self.grid.push(72.0);
}
if self.grid.len() < end {
return;
return vec![span];
}
for w in self.grid.get_mut(g..end).into_iter().flatten() {
if !(w.is_finite() && *w > 0.0) {
@@ -341,7 +346,7 @@ impl Table {
// The cell now spans `g..end`: cut it at the new lines.
let grid = self.grid.clone();
let width = |a: usize, b: usize| grid.get(a..b).map(|x| x.iter().sum::<f32>());
let Some(row) = self.rows.get_mut(r) else { return };
let Some(row) = self.rows.get_mut(r) else { return Vec::new() };
let mut a = g;
let mut parts = Vec::with_capacity(lines.len() + 1);
for &b in lines.iter().chain(std::iter::once(&end)) {
@@ -363,6 +368,122 @@ impl Table {
row.cells.insert((c + k).min(row.cells.len()), nc);
}
}
parts.iter().map(|&(a, b)| b - a).collect()
}
/// The rows cell `c` of row `r` covers: (first row, count). A vertically merged cell runs from
/// its `Restart` cell down through the `Continue` cells that start at the same grid column.
pub fn vertical_extent(&self, r: usize, c: usize) -> Option<(usize, usize)> {
let g = self.grid_col(r, c);
let at = |rr: usize| -> Option<VMerge> {
let ci = self.cell_at_grid(rr, g)?;
if self.grid_col(rr, ci) != g {
return None;
}
Some(self.rows.get(rr)?.cells.get(ci)?.props.vmerge)
};
let mut top = r;
if self.rows.get(r)?.cells.get(c)?.props.vmerge == VMerge::Continue {
while let Some(up) = top.checked_sub(1).and_then(at) {
if up == VMerge::None {
break;
}
top -= 1;
if up == VMerge::Restart {
break;
}
}
}
let mut k = 1;
if at(top) != Some(VMerge::None) {
while at(top + k) == Some(VMerge::Continue) {
k += 1;
}
}
Some((top, k))
}
/// Split Cells: cut cell `c` of row `r` into `cols` × `rows` cells. A cell one row high gets
/// `rows - 1` new rows under it, over which the row's other cells merge down; a vertically
/// merged cell is regrouped instead, so `rows` must divide its height. The content stays in
/// the first part. Returns false (with the table unchanged) when the split can't be made.
pub fn split_cell_grid(&mut self, r: usize, c: usize, cols: usize, rows: usize) -> bool {
let (cols, rows) = (cols.clamp(1, MAX_COLS), rows.clamp(1, MAX_ROWS));
let Some((top, k)) = self.vertical_extent(r, c) else { return false };
let g = self.grid_col(r, c);
let Some(ci) = self.cell_at_grid(top, g) else { return false };
if (k > 1 && k % rows != 0) || (k == 1 && self.rows.len().saturating_add(rows - 1) > MAX_ROWS) {
return false;
}
let height = if k > 1 { k } else { rows };
if k > 1 && rows > 1 {
let q = k / rows;
for i in 0..k {
if let Some(cj) = self.cell_at_grid(top + i, g)
&& let Some(cell) = self.rows.get_mut(top + i).and_then(|row| row.cells.get_mut(cj))
{
cell.props.vmerge = if q == 1 {
VMerge::None
} else if i % q == 0 {
VMerge::Restart
} else {
VMerge::Continue
};
}
}
} else if k == 1 && rows > 1 {
let Some(src) = self.rows.get_mut(top) else { return false };
let mut like = Row { props: src.props.clone(), cells: Vec::with_capacity(src.cells.len()), controls: ControlWrap::default() };
like.props.header = false;
for (i, cell) in src.cells.iter_mut().enumerate() {
let mut nc = Cell::empty();
nc.props = cell.props.clone();
if i == ci {
cell.props.vmerge = VMerge::None;
nc.props.vmerge = VMerge::None;
} else {
// The row's other cells merge down over the new rows.
if cell.props.vmerge == VMerge::None {
cell.props.vmerge = VMerge::Restart;
}
nc.props.vmerge = VMerge::Continue;
}
like.cells.push(nc);
}
for _ in 1..rows {
self.rows.insert(top + 1, like.clone());
}
}
if cols > 1 {
let spans = self.split_cell(top, ci, cols);
let total: usize = spans.iter().sum();
let vmerge = self.rows.get(top).and_then(|row| row.cells.get(ci)).map(|cell| cell.props.vmerge).unwrap_or_default();
if let Some(row) = self.rows.get_mut(top) {
for cell in row.cells.iter_mut().skip(ci + 1).take(spans.len().saturating_sub(1)) {
cell.props.vmerge = vmerge;
}
}
// The cells under it, which the new grid lines widened, are cut the same way.
let grid = self.grid.clone();
let width = |a: usize, b: usize| grid.get(a..b).map(|x| x.iter().sum::<f32>());
for rr in top + 1..top + height {
let Some(cj) = self.cell_at_grid(rr, g).filter(|&cj| self.grid_col(rr, cj) == g) else { continue };
let Some(row) = self.rows.get_mut(rr) else { continue };
let Some(cell) = row.cells.get_mut(cj).filter(|cell| cell.span() == total) else { continue };
let mut a = g;
let mut parts = Vec::with_capacity(spans.len());
for (n, &span) in spans.iter().enumerate() {
let mut part = if n == 0 { cell.clone() } else { Cell::empty() };
part.props.vmerge = cell.props.vmerge;
part.props.span = span as u32;
part.props.width = width(a, a + span);
a += span;
parts.push(part);
}
row.cells.splice(cj..=cj, parts);
}
}
true
}
pub fn total_width(&self) -> f32 {
+45 -6
View File
@@ -27,7 +27,7 @@ fn u32_at(bytes: &[u8], offset: usize) -> Option<u32> {
}
fn parse(bytes: &[u8]) -> Option<CoreProps> {
if bytes.get(..2)? != [0xfe, 0xff] || bytes.get(28..44)? != SUMMARY_FMTID || u32_at(bytes, 24)? != 1 {
if bytes.get(..2)? != [0xfe, 0xff] || bytes.get(28..44)? != SUMMARY_FMTID || u32_at(bytes, 24)? < 1 {
return None;
}
if u16::from_le_bytes(bytes.get(2..4)?.try_into().ok()?) > 1 {
@@ -65,12 +65,14 @@ fn parse(bytes: &[u8]) -> Option<CoreProps> {
.min(set.len());
set.get(start..end)
};
let cp = property(1)?;
if u32_at(cp, 0)? != 2 {
return None;
} // VT_I2
let codepage = u16::from_le_bytes(cp.get(4..6)?.try_into().ok()?);
let codepage = property(1)
.and_then(|cp| {
let vt = u32_at(cp, 0)?;
if vt == 2 || vt == 3 { Some(u16::from_le_bytes(cp.get(4..6)?.try_into().ok()?)) } else { None }
})
.unwrap_or(1252);
let get = |id| property(id).and_then(|value| text(value, codepage)).unwrap_or_default();
let get_filetime = |id| property(id).and_then(filetime_iso).unwrap_or_default();
Some(CoreProps {
title: get(2),
subject: get(3),
@@ -79,10 +81,47 @@ fn parse(bytes: &[u8]) -> Option<CoreProps> {
description: get(6),
last_modified_by: get(8),
revision: get(9).parse().unwrap_or_default(),
created: get_filetime(12),
modified: get_filetime(13),
..Default::default()
})
}
fn civil(days: i64) -> (i64, u32, u32) {
let z = days + 719_468;
let era = (if z >= 0 { z } else { z - 146_096 }) / 146_097;
let doe = (z - era * 146_097) as u32;
let yoe = (doe - doe / 1460 + doe / 36524 - doe / 146096) / 365;
let y = yoe as i64 + era * 400;
let doy = doe - (365 * yoe + yoe / 4 - yoe / 100);
let mp = (5 * doy + 2) / 153;
let d = doy - (153 * mp + 2) / 5 + 1;
let m = if mp < 10 { mp + 3 } else { mp - 9 };
let y = y + i64::from(m <= 2);
(y, m, d)
}
fn filetime_iso(value: &[u8]) -> Option<String> {
let kind = u32_at(value, 0)?;
if kind != 64 {
return None;
}
let ft_low = u32_at(value, 4)? as u64;
let ft_high = u32_at(value, 8)? as u64;
let ft = (ft_high << 32) | ft_low;
if ft == 0 {
return None;
}
const EPOCH_DIFFERENCE: u64 = 116_444_736_000_000_000;
let unix_100ns = ft.checked_sub(EPOCH_DIFFERENCE)?;
let unix_secs = unix_100ns / 10_000_000;
let days = (unix_secs / 86_400) as i64;
let secs_of_day = (unix_secs % 86_400) as u32;
let (h, mi, s) = (secs_of_day / 3600, (secs_of_day / 60) % 60, secs_of_day % 60);
let (y, m, d) = civil(days);
Some(format!("{y:04}-{m:02}-{d:02}T{h:02}:{mi:02}:{s:02}Z"))
}
fn text(value: &[u8], codepage: u16) -> Option<String> {
let kind = u32_at(value, 0)?;
let count = u32_at(value, 4)? as usize;
+32 -9
View File
@@ -77,6 +77,7 @@ pub(crate) struct RowInfo {
/// automatic `Shd` (`None`) clears it.
shading: Vec<(u8, Option<Rgb>)>,
widths: Vec<(u8, f32)>,
width_pcts: Vec<(u8, f32)>,
valign: Vec<(u8, VAlign)>,
}
@@ -93,6 +94,18 @@ impl RowInfo {
pub(crate) fn shading_of(&self, i: u8) -> Option<Rgb> {
self.shading.iter().rfind(|(k, _)| *k == i).and_then(|(_, s)| *s)
}
/// The width cell `i` ends up with, for tests.
#[cfg(test)]
pub(crate) fn width_of(&self, i: u8) -> Option<f32> {
self.widths.iter().rfind(|(k, _)| *k == i).map(|(_, w)| *w)
}
/// The width percentage cell `i` ends up with, for tests.
#[cfg(test)]
pub(crate) fn width_pct_of(&self, i: u8) -> Option<f32> {
self.width_pcts.iter().rfind(|(k, _)| *k == i).map(|(_, w)| *w)
}
}
pub(crate) fn decode(papx: &[u8]) -> RowInfo {
@@ -187,13 +200,20 @@ pub(crate) fn decode(papx: &[u8]) -> RowInfo {
}
}
T_CELL_WIDTH => {
// cb, itcFirst, itcLim, ftsWidth (3 = twips), wWidth.
if let [a, b, fts, w0, w1, ..] = prl.operand
&& *fts == 3
{
let w = (i16::from_le_bytes([*w0, *w1]) as f32 / 20.0).clamp(0.0, 1584.0);
for c in *a..(*b).min(MAX_COLS as u8) {
r.widths.push((c, w));
// Operand: cb (1 byte), itcFirst (1 byte), itcLim (1 byte), ftsWidth (1 byte), wWidth (2 bytes).
// [MS-DOC] §2.9.290 sprmTCellWidth (0xD635).
// ftsWidth: 3 = twips (dxa), 2 = percentage (50ths of a percent).
if let [_, a, b, fts, w0, w1, ..] = prl.operand {
if *fts == 3 {
let w = (i16::from_le_bytes([*w0, *w1]) as f32 / 20.0).clamp(0.0, 1584.0);
for c in *a..(*b).min(MAX_COLS as u8) {
r.widths.push((c, w));
}
} else if *fts == 2 {
let pct = (u16::from_le_bytes([*w0, *w1]) as f32 / 50.0).clamp(0.0, 100.0);
for c in *a..(*b).min(MAX_COLS as u8) {
r.width_pcts.push((c, pct));
}
}
}
}
@@ -256,13 +276,16 @@ pub(crate) fn assemble(paras: Vec<ParaOut>) -> Blocks {
if let Some((_, m)) = info.vmerge.iter().find(|(k, _)| *k == i) {
c.props.vmerge = *m;
}
if let Some((_, w)) = info.widths.iter().find(|(k, _)| *k == i) {
if let Some((_, w)) = info.widths.iter().rfind(|(k, _)| *k == i) {
c.props.width = Some(*w);
}
if let Some((_, pct)) = info.width_pcts.iter().rfind(|(k, _)| *k == i) {
c.props.width_pct = Some(*pct);
}
if let Some((_, s)) = info.shading.iter().rfind(|(k, _)| *k == i) {
c.props.shading = *s;
}
if let Some((_, v)) = info.valign.iter().find(|(k, _)| *k == i) {
if let Some((_, v)) = info.valign.iter().rfind(|(k, _)| *k == i) {
c.props.valign = Some(*v);
}
}
+45
View File
@@ -2285,3 +2285,48 @@ fn second_table_gets_its_own_grid() {
assert_eq!(tables[1].props.align, None, "alignment reset");
assert_eq!(tables[1].props.indent, None, "indent reset");
}
#[test]
fn sprm_t_cell_width_twips_and_percentage() {
// [MS-DOC] §2.9.290 sprmTCellWidth (0xD635):
// cb (1 byte), itcFirst (1 byte), itcLim (1 byte), ftsWidth (1 byte), wWidth (2 bytes).
// ftsWidth = 3 (twips): 1440 twips = 72.0 pt.
let w_twips = [0u8, 2, 3, 0xA0, 0x05]; // cell 0..2, fts=3, 1440 dxa
// ftsWidth = 2 (percentage in 50ths of a percent): 2500 = 50.0%.
let w_pct = [2u8, 3, 2, 0xC4, 0x09]; // cell 2..3, fts=2, 2500 (50%)
let papx = [sprm_cb(0xD635, &w_twips), sprm_cb(0xD635, &w_pct)].concat();
let row = crate::table::decode(&papx);
assert_eq!(row.width_of(0), Some(72.0));
assert_eq!(row.width_of(1), Some(72.0));
assert_eq!(row.width_pct_of(2), Some(50.0));
// End-to-end document test: row cells receive the preferred widths.
let text = "a\u{7}b\u{7}c\u{7}\u{7}\r";
let in_tbl = grpprl(&[(0x2416, &[0x01]), (0x6649, &[1, 0, 0, 0])]);
let ttp = [
grpprl(&[(0x2416, &[0x01]), (0x2417, &[0x01]), (0x6649, &[1, 0, 0, 0])]),
t_def_edges(&[0, 1440, 2880, 4320]),
sprm_cb(0xD635, &w_twips),
sprm_cb(0xD635, &w_pct),
]
.concat();
let spec = StructSpec {
text,
header_stories: vec![],
sections: vec![(text.chars().count() as u32, Vec::new())],
chpx_runs: vec![(0, Vec::new())],
papx_entries: vec![(0, 0, in_tbl), (6, 0, ttp), (7, 0, Vec::new())],
plf_lst: Vec::new(),
plf_lfo: Vec::new(),
note_stories: vec![],
endnote_stories: vec![],
extra_pairs: Vec::new(),
data_stream: Vec::new(),
};
let doc = read(&struct_doc(&spec)).expect("opens");
let table = doc.body[0].as_table().expect("table block");
assert_eq!(table.rows[0].cells[0].props.width, Some(72.0));
assert_eq!(table.rows[0].cells[1].props.width, Some(72.0));
assert_eq!(table.rows[0].cells[2].props.width_pct, Some(50.0));
}
+66
View File
@@ -96,3 +96,69 @@ fn malformed_or_oversized_summary_does_not_break_doc_open() {
assert!(open_with_metadata(&huge, false).core.title.is_empty());
assert!(open_with_metadata(&vec![0; (2 << 20) + 1], false).core.title.is_empty());
}
fn custom_summary(props: &[(u32, Vec<u8>)], num_property_sets: u32) -> Vec<u8> {
let mut section = vec![0; 8 + 8 * props.len()];
section[4..8].copy_from_slice(&(props.len() as u32).to_le_bytes());
for (i, (id, value)) in props.iter().enumerate() {
section[8 + i * 8..12 + i * 8].copy_from_slice(&id.to_le_bytes());
let offset = section.len() as u32;
section[12 + i * 8..16 + i * 8].copy_from_slice(&offset.to_le_bytes());
section.extend_from_slice(value);
}
let size = section.len() as u32;
section[..4].copy_from_slice(&size.to_le_bytes());
let mut header = vec![0; 28 + 20 * num_property_sets as usize];
header[..2].copy_from_slice(&[0xfe, 0xff]);
header[24..28].copy_from_slice(&num_property_sets.to_le_bytes());
// First property set FMTID and offset
header[28..44].copy_from_slice(&[0xe0, 0x85, 0x9f, 0xf2, 0xf9, 0x4f, 0x68, 0x10, 0xab, 0x91, 0x08, 0, 0x2b, 0x27, 0xb3, 0xd9]);
let first_offset = header.len() as u32;
header[44..48].copy_from_slice(&first_offset.to_le_bytes());
if num_property_sets > 1 {
// Dummy second property set header
header[48..64].copy_from_slice(&[1; 16]);
let second_offset = (first_offset + size) as u32;
header[64..68].copy_from_slice(&second_offset.to_le_bytes());
}
header.extend(section);
header
}
#[test]
fn summary_defaults_codepage_and_accepts_vti4_and_num_property_sets() {
let title_prop = [30u32.to_le_bytes().as_slice(), &7u32.to_le_bytes(), b"Default\0"].concat();
// 1. Missing PID_CODEPAGE (property 1) should default to 1252 per MS-OLEPS.
let no_cp = custom_summary(&[(2, title_prop.clone())], 1);
assert_eq!(open_with_metadata(&no_cp, false).core.title, "Default");
// 2. PID_CODEPAGE as VT_I4 (type 3).
let cp_vti4 = [3u32.to_le_bytes().as_slice(), &1252u32.to_le_bytes()].concat();
let with_vti4 = custom_summary(&[(1, cp_vti4), (2, title_prop.clone())], 1);
assert_eq!(open_with_metadata(&with_vti4, false).core.title, "Default");
// 3. Stream with NumPropertySets = 2.
let multi_sets = custom_summary(&[(2, title_prop)], 2);
assert_eq!(open_with_metadata(&multi_sets, false).core.title, "Default");
}
#[test]
fn summary_parses_filetime_created_and_modified_dates() {
// 2024-03-15T12:30:45Z = 133549794450000000 in 100ns since 1601-01-01
let ft_val: u64 = 133_549_794_450_000_000;
let ft_low = (ft_val & 0xFFFFFFFF) as u32;
let ft_high = (ft_val >> 32) as u32;
let filetime_bytes = [64u32.to_le_bytes().as_slice(), &ft_low.to_le_bytes(), &ft_high.to_le_bytes()].concat();
let stream = custom_summary(
&[
(12, filetime_bytes.clone()), // PIDSI_CREATE_DTM
(13, filetime_bytes), // PIDSI_LASTSAVE_DTM
],
1,
);
let core = open_with_metadata(&stream, false).core;
assert_eq!(core.created, "2024-03-15T12:30:45Z");
assert_eq!(core.modified, "2024-03-15T12:30:45Z");
}
+1
View File
@@ -15,6 +15,7 @@ quick-xml = { workspace = true }
zip = { workspace = true }
thiserror = { workspace = true }
log = { workspace = true }
unicode-bidi = { workspace = true }
cfb = { workspace = true }
aes = { workspace = true }
sha1 = { workspace = true }
+280
View File
@@ -0,0 +1,280 @@
//! Word bibliography sources: a `b:Sources` document — the custom XML part of a document, or
//! Word's own `Sources.xml` master list, imported through Manage Sources (#2640).
use std::collections::HashSet;
use wordcraft_doc::Source;
use crate::DocxError;
use crate::xml::El;
/// Bibliography sources kept from one file (Word's own source manager holds far fewer).
pub const MAX_SOURCES: usize = 10_000;
/// Characters kept of a bibliography's citation style name.
pub const MAX_STYLE_NAME: usize = 64;
/// A `b:Sources` document read into the document model: its sources in file order, and the
/// citation style it names (`b:Sources/@StyleName`), when it has one.
#[derive(Clone, Debug, Default, PartialEq)]
pub struct Sources {
pub sources: Vec<Source>,
pub style: Option<String>,
}
/// The bibliography namespace Word's own `Sources.xml` files may carry instead of OOXML's (the
/// VBA `Source.XML` format, `schemas.microsoft.com/office/word/2004/10/bibliography`).
const WORD_2004_BIBLIOGRAPHY: &str = "http://schemas.microsoft.com/office/word/2004/10/bibliography";
/// Read a Word bibliography sources file — Word's `Sources.xml`, or any `b:Sources` document in
/// one of the bibliography namespaces (ECMA-376's, or Word's own 2004/10) — into [`Sources`].
pub fn read_sources(bytes: &[u8]) -> Result<Sources, DocxError> {
// `parse_ns` maps Word's 2004/10 namespace onto the canonical `b` prefix as well, so both
// formats read; the root check then refuses an un-namespaced or unrelated `<Sources>`.
let root = crate::xml::parse_ns(bytes, &[("b", WORD_2004_BIBLIOGRAPHY)])?;
if root.name != "b:Sources" {
return Err(DocxError::NotWord(format!("root element is {}", root.name)));
}
Ok(parse(&root, &mut Seen::default(), MAX_SOURCES))
}
/// Bibliography sources already read: tags, and (title, author) of untagged ones.
#[derive(Default)]
pub(crate) struct Seen {
tags: HashSet<String>,
untagged: HashSet<(String, String)>,
}
/// The fresh sources and style of a parsed `b:Sources` root element: `seen` carries the tags and
/// untagged (title, author) keys already taken, so repeated parts collapse into one source, and
/// at most `room` fresh sources are built (later ones are left unread, bounding hostile files).
pub(crate) fn parse(root: &El, seen: &mut Seen, room: usize) -> Sources {
let style = root
.attrs
.iter()
.find(|(k, _)| k.rsplit(':').next() == Some("StyleName"))
.map(|(_, v)| v.trim())
.filter(|name| !name.is_empty())
.map(|name| name.chars().take(MAX_STYLE_NAME).collect());
let child_text = |el: &El, local: &str| el.els().find(|e| e.local() == local).map(|e| e.text()).unwrap_or_default();
let parse_persons = |parent: &El| -> String {
let mut names = Vec::new();
let person_els: Vec<&El> = {
let in_lists: Vec<&El> =
parent.els().filter(|e| e.local() == "NameList").flat_map(|n| n.els().filter(|e| e.local() == "Person")).collect();
if in_lists.is_empty() { parent.els().filter(|e| e.local() == "Person").collect() } else { in_lists }
};
for name_el in person_els {
let last = child_text(name_el, "Last");
let first = child_text(name_el, "First");
let middle = child_text(name_el, "Middle");
let name = if !first.is_empty() && !middle.is_empty() {
format!("{last}, {first} {middle}")
} else if !first.is_empty() {
format!("{last}, {first}")
} else {
last
};
if !name.is_empty() {
names.push(name);
}
}
names.join("; ")
};
let mut out = Sources { sources: Vec::new(), style };
for src_el in root.els().filter(|e| e.local() == "Source") {
if out.sources.len() >= room {
log::warn!("docx: more than {MAX_SOURCES} bibliography sources; the rest are dropped");
break;
}
let tag = child_text(src_el, "Tag");
let raw_kind = child_text(src_el, "SourceType");
let kind = crate::write::from_source_type(&raw_kind);
let (author, editor) = match src_el.els().find(|e| e.local() == "Author") {
Some(a) => {
let auth = a.els().find(|e| e.local() == "Author").map(parse_persons).filter(|s| !s.is_empty()).unwrap_or_else(|| parse_persons(a));
let ed = a.els().find(|e| e.local() == "Editor").map(parse_persons).unwrap_or_default();
(auth, ed)
}
None => (String::new(), String::new()),
};
let editor = if editor.is_empty() { src_el.els().find(|e| e.local() == "Editor").map(parse_persons).unwrap_or_default() } else { editor };
let title = child_text(src_el, "Title");
let year = child_text(src_el, "Year");
let month = child_text(src_el, "Month");
let day = child_text(src_el, "Day");
let city = child_text(src_el, "City");
let publisher = child_text(src_el, "Publisher");
let journal = child_text(src_el, "JournalName");
let edition = child_text(src_el, "Edition");
let volume = child_text(src_el, "Volume");
let issue = child_text(src_el, "Issue");
let pages = child_text(src_el, "Pages");
let url = child_text(src_el, "URL");
let accessed = match child_text(src_el, "Accessed") {
a if !a.is_empty() => a,
_ => {
let y = child_text(src_el, "YearAccessed");
let m = child_text(src_el, "MonthAccessed");
let d = child_text(src_el, "DayAccessed");
if !y.is_empty() && !m.is_empty() && !d.is_empty() {
format!("{y}-{m}-{d}")
} else if !y.is_empty() && !m.is_empty() {
format!("{y}-{m}")
} else {
y
}
}
};
let doi = match child_text(src_el, "DOI") {
d if !d.is_empty() => d,
_ => child_text(src_el, "StandardNumber"),
};
// Keep sources with some content, once each (by tag, or title and author untagged).
if tag.is_empty() && title.is_empty() && author.is_empty() && editor.is_empty() {
continue;
}
let fresh = if tag.is_empty() {
seen.untagged.insert((title.clone(), if author.is_empty() { editor.clone() } else { author.clone() }))
} else {
seen.tags.insert(tag.clone())
};
if !fresh {
continue;
}
out.sources.push(Source {
tag,
kind,
author,
title,
year,
city,
publisher,
journal,
volume,
pages,
url,
editor,
edition,
issue,
month,
day,
accessed,
doi,
});
}
out
}
#[cfg(test)]
mod tests {
use super::*;
const WORD_SOURCES: &str = r#"<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<b:Sources xmlns:b="http://schemas.openxmlformats.org/officeDocument/2006/bibliography"
xmlns="http://schemas.openxmlformats.org/officeDocument/2006/bibliography"
SelectedStyle="\MLA.XSL" StyleName="MLA">
<b:Source>
<b:Tag>Riv21</b:Tag>
<b:SourceType>Book</b:SourceType>
<b:Author>
<b:Author>
<b:NameList>
<b:Person><b:Last>Rivera</b:Last><b:First>Alex</b:First></b:Person>
<b:Person><b:Last>Chen</b:Last><b:First>Mei</b:First><b:Middle>L.</b:Middle></b:Person>
</b:NameList>
</b:Author>
</b:Author>
<b:Title>Shared Spaces</b:Title>
<b:Year>2021</b:Year>
<b:City>Portland</b:City>
<b:Publisher>Harbor Books</b:Publisher>
</b:Source>
<b:Source>
<b:Tag>Web19</b:Tag>
<b:SourceType>InternetSite</b:SourceType>
<b:Author>
<b:Author>
<b:NameList>
<b:Person><b:Last>Okafor</b:Last><b:First>Ada</b:First></b:Person>
</b:NameList>
</b:Author>
</b:Author>
<b:Title>Dockyards of the north</b:Title>
<b:Year>2019</b:Year>
<b:URL>https://example.org/docks</b:URL>
<b:YearAccessed>2024</b:YearAccessed>
<b:MonthAccessed>May</b:MonthAccessed>
<b:DayAccessed>17</b:DayAccessed>
<b:StandardNumber>10.1234/x</b:StandardNumber>
</b:Source>
<b:Source>
<b:Tag>NoDate</b:Tag>
<b:SourceType>JournalArticle</b:SourceType>
<b:Title>Untitled notes</b:Title>
<b:JournalName>Field Notes</b:JournalName>
<b:Volume>4</b:Volume>
</b:Source>
<b:Source>
<b:Tag>Riv21</b:Tag>
<b:SourceType>Book</b:SourceType>
<b:Title>A duplicate of Riv21</b:Title>
<b:Year>2021</b:Year>
</b:Source>
<b:Source>
<b:SourceType>Book</b:SourceType>
</b:Source>
</b:Sources>"#;
/// A Word `Sources.xml` (`b:Sources`): its style, the `b:Source` elements with their names,
/// kinds and fallback fields; a repeated tag and an entry with no content at all are dropped.
#[test]
fn a_word_sources_file_reads_into_sources() {
let f = read_sources(WORD_SOURCES.as_bytes()).unwrap();
assert_eq!(f.style.as_deref(), Some("MLA"), "the citation style comes from StyleName");
assert_eq!(
f.sources.len(),
3,
"the repeated tag and the empty b:Source are dropped: {:?}",
f.sources.iter().map(|s| &s.tag).collect::<Vec<_>>()
);
let book = &f.sources[0];
assert_eq!((book.tag.as_str(), book.kind.as_str()), ("Riv21", "book"));
assert_eq!(book.author, "Rivera, Alex; Chen, Mei L.");
assert_eq!(book.title, "Shared Spaces", "the first source of a repeated tag wins");
assert_eq!(
(book.title.as_str(), book.year.as_str(), book.city.as_str(), book.publisher.as_str()),
("Shared Spaces", "2021", "Portland", "Harbor Books")
);
let web = &f.sources[1];
assert_eq!(web.kind, "website");
assert_eq!(web.author, "Okafor, Ada");
assert_eq!(web.accessed, "2024-May-17", "assemble the accessed date from its parts");
assert_eq!(web.doi, "10.1234/x", "StandardNumber stands in for a missing DOI");
let notes = &f.sources[2];
assert_eq!((notes.kind.as_str(), notes.journal.as_str(), notes.volume.as_str()), ("article", "Field Notes", "4"));
}
/// Anything whose root isn't `b:Sources` is refused, not read as an empty bibliography.
#[test]
fn a_file_that_is_not_b_sources_is_an_error() {
assert!(read_sources(b"<html><body>hi</body></html>").is_err());
assert!(matches!(read_sources(b"not xml at all"), Err(DocxError::Xml(_))));
assert!(read_sources(b"<Sources><Source><Tag>A</Tag></Source></Sources>").is_err(), "no namespace");
assert!(read_sources(br#"<o:Sources xmlns:o="urn:example:other"><o:Source/></o:Sources>"#).is_err(), "an unrelated namespace");
}
/// Word's own 2004/10 bibliography namespace (`Sources.xml` written by scripts and older
/// builds, the VBA `Source.XML` format) reads as well as OOXML's.
#[test]
fn word_2004_bibliography_namespace_is_accepted() {
let xml = br#"<b:Sources xmlns:b="http://schemas.microsoft.com/office/word/2004/10/bibliography" StyleName="MLA">
<b:Source><b:Tag>Riv21</b:Tag><b:SourceType>Book</b:SourceType><b:Title>Shared Spaces</b:Title><b:Year>2021</b:Year></b:Source>
</b:Sources>"#;
let f = read_sources(xml).unwrap();
assert_eq!(f.style.as_deref(), Some("MLA"));
assert_eq!(f.sources.len(), 1);
assert_eq!(f.sources[0].tag, "Riv21");
}
}
+16 -3
View File
@@ -14,10 +14,20 @@ const MAX_FONTS: usize = 64;
/// The font table's references to embedded font files, one per style.
const EMBEDS: [&str; 4] = ["w:embedRegular", "w:embedBold", "w:embedItalic", "w:embedBoldItalic"];
/// A font file a document embeds, and the name its font table gives the font: the name the text
/// asks for, which a subset Word embeds doesn't carry inside (#2240).
#[derive(Clone, Debug, PartialEq)]
pub struct EmbeddedFont {
/// `w:font/@w:name`.
pub name: String,
/// The font file, de-obfuscated.
pub data: Vec<u8>,
}
/// The font files `doc`'s font table embeds, de-obfuscated, in table order. Not checked to be
/// fonts: that is up to the font database. Empty for a document read from anything but a package
/// with a font table.
pub fn embedded_fonts(doc: &Document) -> Vec<Vec<u8>> {
pub fn embedded_fonts(doc: &Document) -> Vec<EmbeddedFont> {
let Some(table) = doc.passthrough.get(FONT_TABLE_PART).and_then(|b| crate::xml::parse(b).ok()) else {
return Vec::new();
};
@@ -26,7 +36,10 @@ pub fn embedded_fonts(doc: &Document) -> Vec<Vec<u8>> {
return Vec::new();
};
let mut out = Vec::new();
for embed in table.children("w:font").flat_map(|f| f.els().filter(|e| EMBEDS.contains(&e.name.as_str()))) {
let embeds = table
.children("w:font")
.flat_map(|f| f.els().filter(|e| EMBEDS.contains(&e.name.as_str())).map(move |e| (f.attr("w:name").unwrap_or(""), e)));
for (name, embed) in embeds {
if out.len() >= MAX_FONTS {
log::warn!("docx: more than {MAX_FONTS} embedded fonts; the rest are not loaded");
break;
@@ -52,7 +65,7 @@ pub fn embedded_fonts(doc: &Document) -> Vec<Vec<u8>> {
for (i, b) in data.iter_mut().take(32).enumerate() {
*b ^= key.get(i % 16).copied().unwrap_or(0);
}
out.push(data);
out.push(EmbeddedFont { name: name.to_string(), data });
}
out
}
+3 -1
View File
@@ -7,6 +7,7 @@
//! error or a best-effort document, never a panic.
#![deny(clippy::unwrap_used, clippy::expect_used, clippy::panic, clippy::unimplemented, clippy::todo, clippy::unreachable)]
mod bibliography;
mod chart_spec;
mod crypt;
mod custom;
@@ -21,13 +22,14 @@ mod wordml2003_write;
mod write;
mod xml;
pub use bibliography::{MAX_SOURCES, Sources, read_sources};
pub use chart_spec::{chart_items, chart_xml};
pub use crypt::{
DEFAULT_SPIN_COUNT, MAX_PASSWORD_CHARS, PasswordCheck, check_password, check_protection_password, decrypt, encrypt, encrypt_with_spin_count,
is_encrypted, protection_hash,
};
pub use flat::{XmlKind, sniff_xml, write as write_flat};
pub use fonts::embedded_fonts;
pub use fonts::{EmbeddedFont, embedded_fonts};
pub use read::read;
pub use smart_art::smart_art_items;
pub use wordml2003_write::write as write_wordml2003;
+66 -5
View File
@@ -2,6 +2,7 @@
use std::collections::{BTreeMap, HashMap};
use std::io::{Cursor, Read, Write};
use std::sync::Arc;
use crate::DocxError;
use crate::xml::{self, El};
@@ -44,6 +45,38 @@ pub mod rt {
pub const DIAGRAM_QUICK_STYLE: &str = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/diagramQuickStyle";
pub const DIAGRAM_COLORS: &str = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/diagramColors";
pub const CUSTOM_XML: &str = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/customXml";
pub const GLOSSARY: &str = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/glossaryDocument";
}
/// Content type of a glossary document part (ECMA-376 Part 1 §11.3.8).
pub const GLOSSARY_CONTENT_TYPE: &str = "application/vnd.openxmlformats-officedocument.wordprocessingml.document.glossary+xml";
/// `Document::passthrough` key: a fingerprint of `Document::glossary` as read, so a save can tell
/// an untouched glossary (written back verbatim) from an edited one (written from the model).
/// Not a package part: WordCraft's own bookkeeping.
pub const GLOSSARY_READ: &str = "wordcraft:glossary.read";
/// Most building blocks read from one glossary.
pub const MAX_GLOSSARY_ENTRIES: usize = 10_000;
/// The glossary entry types `w:type` takes (ST_DocPartType); others aren't kept.
pub const GLOSSARY_TYPES: [&str; 7] = ["none", "normal", "autoExp", "toolbar", "speller", "formFld", "bbPlcHdr"];
/// The insertion behaviours `w:behavior` takes (ST_DocPartBehavior); others aren't kept.
pub const GLOSSARY_BEHAVIORS: [&str; 3] = ["content", "p", "pg"];
/// A fingerprint of `glossary` (FNV-1a over its debug form): stable across runs and builds that
/// keep the model's shape, and cheap (nothing is allocated).
pub fn glossary_fingerprint(glossary: &[wordcraft_doc::BuildingBlock]) -> String {
struct Fnv(u64);
impl std::fmt::Write for Fnv {
fn write_str(&mut self, s: &str) -> std::fmt::Result {
for b in s.bytes() {
self.0 = (self.0 ^ u64::from(b)).wrapping_mul(0x0100_0000_01b3);
}
Ok(())
}
}
let mut h = Fnv(0xcbf2_9ce4_8422_2325);
// Writing into `Fnv` never fails.
let _ = std::fmt::Write::write_fmt(&mut h, format_args!("{glossary:?}"));
format!("{:016x}", h.0)
}
/// `Document::passthrough` key for a macro project, kept as opaque bytes (never parsed or run).
@@ -154,9 +187,10 @@ pub fn rel_is(t: &str, full: &str) -> bool {
t == full || tail(t) == tail(full)
}
/// The decompressed parts of a package (bounded).
/// The decompressed parts of a package (bounded). Shared, so a part the document keeps as it is (a
/// picture, an embedded object) is handed over rather than copied.
pub struct Package {
files: BTreeMap<String, Vec<u8>>,
files: BTreeMap<String, Arc<Vec<u8>>>,
}
impl Package {
@@ -198,7 +232,7 @@ impl Package {
return Err(DocxError::Limit(format!("part {name} decompresses past the size limit")));
}
total = total.saturating_add(buf.len() as u64);
files.insert(name, buf);
files.insert(name, Arc::new(buf));
}
Ok(Package { files })
}
@@ -206,16 +240,22 @@ impl Package {
/// A package from parts already in memory (a Flat OPC file, a converted document), by part
/// name without the leading `/`.
pub fn from_parts(files: BTreeMap<String, Vec<u8>>) -> Package {
Package { files }
Package { files: files.into_iter().map(|(name, bytes)| (name, Arc::new(bytes))).collect() }
}
/// Part bytes by name (leading `/` ignored; case-insensitive fallback).
pub fn get(&self, name: &str) -> Option<&[u8]> {
self.get_shared(name).map(|v| v.as_slice())
}
/// [`Package::get`], as the shared bytes themselves: for a part the document keeps whole, which
/// then costs no copy (a picture document's pictures used to be held twice while it opened).
pub fn get_shared(&self, name: &str) -> Option<&Arc<Vec<u8>>> {
let n = name.trim_start_matches('/');
if let Some(v) = self.files.get(n) {
return Some(v);
}
self.files.iter().find(|(k, _)| k.eq_ignore_ascii_case(n)).map(|(_, v)| v.as_slice())
self.files.iter().find(|(k, _)| k.eq_ignore_ascii_case(n)).map(|(_, v)| v)
}
pub fn xml(&self, name: &str) -> Result<Option<El>, DocxError> {
@@ -400,6 +440,27 @@ mod tests {
}
}
#[test]
fn the_document_shares_its_pictures_with_the_package() {
// Each picture was copied out of the package, so a picture document held its pictures
// twice while it opened, and left the package's copies behind as freed memory.
let ns = r#"xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships" xmlns:wp="http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing" xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main" xmlns:pic="http://schemas.openxmlformats.org/drawingml/2006/picture""#;
let document = format!(
r#"<w:document {ns}><w:body><w:p><w:r><w:drawing><wp:inline><wp:extent cx="635000" cy="635000"/><a:graphic><a:graphicData uri="http://schemas.openxmlformats.org/drawingml/2006/picture"><pic:pic><pic:blipFill><a:blip r:embed="rIdImg"/></pic:blipFill></pic:pic></a:graphicData></a:graphic></wp:inline></w:drawing></w:r></w:p></w:body></w:document>"#
);
let rels = r#"<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rIdImg" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/image" Target="media/pic.png"/></Relationships>"#;
let z = zip_of(&[
("word/document.xml", document.into_bytes()),
("word/_rels/document.xml.rels", rels.as_bytes().to_vec()),
("word/media/pic.png", b"\x89PNG picture bytes".to_vec()),
]);
let pkg = Package::open(&z).unwrap();
let doc = crate::read::read_opened(&pkg).unwrap();
let held = doc.media.values().next().expect("the picture is read");
assert!(Arc::ptr_eq(held, pkg.get_shared("word/media/pic.png").unwrap()), "the same bytes, not a copy");
assert_eq!(pkg.get("/WORD/media/PIC.png"), Some(&b"\x89PNG picture bytes"[..]), "lookups are unchanged");
}
#[test]
fn resolves_targets() {
assert_eq!(resolve("word", "media/image1.png"), "word/media/image1.png");
+28 -3
View File
@@ -79,7 +79,8 @@ pub(crate) fn color(e: &El, theme: &[Rgb]) -> Option<Rgb> {
/// A preset colour (`a:prstClr/@val`, ECMA-376 Part 1 §20.1.10.48): the common ones; others are
/// unknown (`None`).
pub(crate) fn preset_color(name: &str) -> Option<Rgb> {
Some(match name {
let lower = name.to_ascii_lowercase();
Some(match lower.as_str() {
"black" => Rgb(0, 0, 0),
"white" => Rgb(0xFF, 0xFF, 0xFF),
"red" => Rgb(0xFF, 0, 0),
@@ -91,8 +92,8 @@ pub(crate) fn preset_color(name: &str) -> Option<Rgb> {
"magenta" | "fuchsia" => Rgb(0xFF, 0, 0xFF),
"gray" | "grey" => Rgb(0x80, 0x80, 0x80),
"silver" => Rgb(0xC0, 0xC0, 0xC0),
"ltGray" | "ltGrey" => Rgb(0xD3, 0xD3, 0xD3),
"dkGray" | "dkGrey" => Rgb(0xA9, 0xA9, 0xA9),
"ltgray" | "ltgrey" | "lightgray" | "lightgrey" => Rgb(0xD3, 0xD3, 0xD3),
"dkgray" | "dkgrey" | "darkgray" | "darkgrey" => Rgb(0xA9, 0xA9, 0xA9),
"maroon" => Rgb(0x80, 0, 0),
"navy" => Rgb(0, 0, 0x80),
"olive" => Rgb(0x80, 0x80, 0),
@@ -101,6 +102,18 @@ pub(crate) fn preset_color(name: &str) -> Option<Rgb> {
"orange" => Rgb(0xFF, 0xA5, 0),
"brown" => Rgb(0xA5, 0x2A, 0x2A),
"pink" => Rgb(0xFF, 0xC0, 0xCB),
"gold" => Rgb(0xFF, 0xD7, 0),
"darkblue" | "dkblue" => Rgb(0, 0, 0x8B),
"darkred" | "dkred" => Rgb(0x8B, 0, 0),
"darkgreen" | "dkgreen" => Rgb(0, 0x64, 0),
"lightblue" | "ltblue" => Rgb(0xAD, 0xD8, 0xE6),
"coral" => Rgb(0xFF, 0x7F, 0x50),
"salmon" => Rgb(0xFA, 0x80, 0x72),
"tomato" => Rgb(0xFF, 0x63, 0x47),
"violet" => Rgb(0xEE, 0x82, 0xEE),
"indigo" => Rgb(0x4B, 0, 0x82),
"beige" => Rgb(0xF5, 0xF5, 0xDC),
"ivory" => Rgb(0xFF, 0xFF, 0xF0),
_ => return None,
})
}
@@ -241,4 +254,16 @@ mod tests {
let b = el("a:srgbClr", r#"val="808080""#, r#"<a:lumOff val="50000"/><a:lumMod val="50000"/>"#);
assert_ne!(color(&a, &theme), color(&b, &theme));
}
#[test]
fn preset_colors_are_case_insensitive_and_include_common_names() {
assert_eq!(preset_color("red"), Some(Rgb(255, 0, 0)));
assert_eq!(preset_color("Red"), Some(Rgb(255, 0, 0)));
assert_eq!(preset_color("darkBlue"), Some(Rgb(0, 0, 0x8B)));
assert_eq!(preset_color("dkBlue"), Some(Rgb(0, 0, 0x8B)));
assert_eq!(preset_color("lightGray"), Some(Rgb(0xD3, 0xD3, 0xD3)));
assert_eq!(preset_color("ltGray"), Some(Rgb(0xD3, 0xD3, 0xD3)));
assert_eq!(preset_color("gold"), Some(Rgb(0xFF, 0xD7, 0)));
assert_eq!(preset_color("unknownColor"), None);
}
}
+13 -3
View File
@@ -77,6 +77,12 @@ impl Reader<'_> {
if !self.doc.sources.is_empty() && self.xml(&rel.target).ok().flatten().is_some_and(|r| r.local() == "Sources") {
continue;
}
// Native ink shares the custom XML relationship type, but its part belongs to the
// drawing that refers to it: kept with it, or dropped with it (no orphan part).
let pkg = self.pkg;
if self.content_types.get_or_insert_with(|| ContentTypes::read(pkg)).of(&rel.target) == Some("application/inkml+xml") {
continue;
}
if self.keep_part_tree(&rel.target) {
roots.push_str(&format!("{owner}\t{}\t{}\n", rel.kind, rel.target));
}
@@ -94,17 +100,21 @@ impl Reader<'_> {
/// relationship the main document's markup refers to by id (its target goes with the
/// element, or not at all).
pub(super) fn keep_unmodelled(&mut self, package: &Rels, document: &Rels) {
let main = self.pkg.get(&self.main).map(|b| b.to_vec()).unwrap_or_default();
let pkg = self.pkg;
let main = pkg.get(&self.main).unwrap_or_default();
let referenced = |id: &str| {
let needle = format!("\"{id}\"");
main.windows(needle.len()).any(|w| w == needle.as_bytes())
};
let mut roots = String::new();
// One glossary per main document (its first relationship; see `read_glossary`).
let mut glossary = false;
for (owner, rels) in [("P", package), ("D", document)] {
for rel in rels.list.iter().filter(|r| !r.external).take(MAX_EMBEDDED_RELS) {
let tail = rel.kind.rsplit('/').next().unwrap_or("");
if HANDLED_RELS.contains(&tail)
|| (owner == "D" && referenced(&rel.id))
|| (owner == "D" && tail == "glossaryDocument" && std::mem::replace(&mut glossary, true))
|| ![&rel.kind, &rel.target].iter().all(|s| manifest_field_ok(s))
{
continue;
@@ -192,14 +202,14 @@ impl Reader<'_> {
log::warn!("docx: more than {MAX_EMBEDDED_PARTS} parts for embedded objects; the rest are dropped");
return false;
}
let Some(bytes) = self.pkg.get(part) else {
let Some(bytes) = self.pkg.get_shared(part) else {
log::warn!("docx: part {part} of an embedded object is missing");
return false;
};
let pkg = self.pkg;
let ct = self.content_types.get_or_insert_with(|| ContentTypes::read(pkg)).of(part).unwrap_or("application/octet-stream");
let content_type = if manifest_field_ok(ct) { ct.to_string() } else { "application/octet-stream".to_string() };
self.doc.passthrough.entry(part.to_string()).or_insert_with(|| Arc::new(bytes.to_vec()));
self.doc.passthrough.entry(part.to_string()).or_insert_with(|| bytes.clone());
self.embedded.parts.insert(part.to_string(), crate::package::EmbeddedPart { content_type, rels: Vec::new() });
true
}
+114
View File
@@ -0,0 +1,114 @@
//! The glossary document (ECMA-376 Part 1 §17.12): the document's own building blocks (Quick
//! Parts, AutoText…), read into `Document::glossary`.
//!
//! The glossary part and everything it reaches are also kept verbatim (`keep_unmodelled`), with a
//! fingerprint of the entries as read ([`GLOSSARY_READ`]): a save writes the kept part while the
//! entries are unchanged, so whatever the model doesn't hold survives, and writes the glossary from
//! the model once they change.
use std::sync::Arc;
use wordcraft_doc::{Blocks, BuildingBlock};
use super::Reader;
use super::story::StoryCtx;
use crate::package::{
ContentTypes, GLOSSARY_BEHAVIORS, GLOSSARY_CONTENT_TYPE, GLOSSARY_READ, GLOSSARY_TYPES, MAX_GLOSSARY_ENTRIES, Rels, glossary_fingerprint, rel_is,
rt,
};
use crate::xml::El;
/// Longest name, gallery, category, description, guid or behaviour value kept (characters).
const MAX_FIELD: usize = 1024;
/// The known values of `parent`'s `tag` children, each once, in file order.
fn listed(parent: Option<&El>, tag: &str, known: &[&str]) -> Vec<String> {
let mut out: Vec<String> = Vec::new();
for v in parent.into_iter().flat_map(|p| p.children(tag)).filter_map(|e| e.attr("w:val")) {
if known.contains(&v) && !out.iter().any(|o| o == v) {
out.push(v.to_string());
}
}
out
}
impl Reader<'_> {
/// Read the glossary the main document relates to (`document`: its relationships). The first
/// internal glossary relationship counts; one pointing at the main document, a part that isn't a
/// glossary, or a missing one gives no entries. Entries without a name or a body are skipped.
pub(super) fn read_glossary(&mut self, document: &Rels) {
// The first internal one, as `keep_unmodelled` keeps.
let Some(rel) = document.list.iter().find(|r| !r.external && rel_is(&r.kind, rt::GLOSSARY)) else { return };
let mut entries = Vec::new();
if !rel.target.eq_ignore_ascii_case(&self.main) {
let pkg = self.pkg;
let ct = self.content_types.get_or_insert_with(|| ContentTypes::read(pkg)).of(&rel.target).map(str::to_string);
let root = match self.xml(&rel.target) {
Ok(r) => r,
Err(e) => {
log::warn!("docx: ignoring unreadable glossary {}: {e}", rel.target);
None
}
};
// Without content types (a converted package) the root element alone decides.
if let Some(root) = root.filter(|r| r.name == "w:glossaryDocument" && ct.as_deref().is_none_or(|t| t == GLOSSARY_CONTENT_TYPE)) {
let rels = self.pkg.rels(&rel.target);
entries = self.doc_parts(&root, &rels);
}
}
self.doc.glossary = entries;
let print = glossary_fingerprint(&self.doc.glossary);
self.doc.passthrough.insert(GLOSSARY_READ.into(), Arc::new(print.into_bytes()));
}
/// The building blocks of a `w:glossaryDocument`, read with the glossary's relationships.
fn doc_parts(&mut self, root: &El, rels: &Rels) -> Vec<BuildingBlock> {
// The glossary's notes, comments, bookmarks, permissions and moves are its own: what the
// main document's ids mean doesn't apply to it, and its ids mustn't leak into the body's.
let footnotes = std::mem::take(&mut self.footnotes);
let endnotes = std::mem::take(&mut self.endnotes);
let comments = std::mem::take(&mut self.comment_map);
let ended = std::mem::take(&mut self.comments_ended);
let bookmarks = std::mem::take(&mut self.bookmarks);
let perms = std::mem::take(&mut self.perms);
let moves = std::mem::take(&mut self.moves);
let mut out = Vec::new();
for part in root.child("w:docParts").into_iter().flat_map(|d| d.children("w:docPart")) {
if out.len() >= MAX_GLOSSARY_ENTRIES {
log::warn!("docx: glossary has more than {MAX_GLOSSARY_ENTRIES} entries; the rest are dropped");
break;
}
let pr = part.child("w:docPartPr");
let val = |e: Option<&El>| e.and_then(|e| e.attr("w:val")).unwrap_or("").chars().take(MAX_FIELD).collect::<String>();
let name = val(pr.and_then(|p| p.child("w:name")));
let Some(body) = part.child("w:docPartBody").filter(|_| !name.is_empty()) else { continue };
let category = pr.and_then(|p| p.child("w:category"));
let mut sc = StoryCtx::default();
let mut blocks = Blocks::new();
self.read_blocks(&mut sc, body, rels, &mut blocks, 0);
self.flush_pending(&mut sc, &mut blocks);
if blocks.is_empty() {
continue;
}
out.push(BuildingBlock {
name,
gallery: val(category.and_then(|c| c.child("w:gallery"))),
category: val(category.and_then(|c| c.child("w:name"))),
description: val(pr.and_then(|p| p.child("w:description"))),
guid: val(pr.and_then(|p| p.child("w:guid"))),
style: val(pr.and_then(|p| p.child("w:style"))),
types: listed(pr.and_then(|p| p.child("w:types")), "w:type", &GLOSSARY_TYPES),
behaviors: listed(pr.and_then(|p| p.child("w:behaviors")), "w:behavior", &GLOSSARY_BEHAVIORS),
blocks,
});
}
self.footnotes = footnotes;
self.endnotes = endnotes;
self.comment_map = comments;
self.comments_ended = ended;
self.bookmarks = bookmarks;
self.perms = perms;
self.moves = moves;
out
}
}
+134 -5
View File
@@ -4,6 +4,7 @@ use wordcraft_doc::math::{
Arg, BrkBin, BrkBinSub, ColJc, FracKind, LimLoc, MAX_DEPTH, MNode, MRun, MScr, MSty, Math, MathJc, MathProps, ScriptKind, merge_runs,
};
use wordcraft_doc::props::Rgb;
use wordcraft_doc::symbol_fonts;
use crate::xml::El;
@@ -232,12 +233,21 @@ fn node(k: &El, depth: usize, out: &mut Arg) {
e: sub_arg(k, "m:e", d),
});
}
// Tracked insertions show; deletions don't.
"w:ins" | "w:moveTo" | "w:smartTag" | "w:customXml" | "m:oMath" if depth <= MAX_DEPTH => {
// Tracked insertions show; deletions don't. Hyperlinks, simple fields and direction
// wrappers (EG_PContentMath, §22.1.2.77) keep their runs (#1273).
"w:ins" | "w:moveTo" | "w:smartTag" | "w:customXml" | "w:hyperlink" | "w:fldSimple" | "w:dir" | "w:bdo" | "m:oMath" if depth <= MAX_DEPTH => {
for c in k.els() {
node(c, d, out);
}
}
// Markup compatibility: the math choice (we read math), else the fallback.
"mc:AlternateContent" if depth <= MAX_DEPTH => {
if let Some(c) = k.child("mc:Choice").or_else(|| k.child("mc:Fallback")) {
for c in c.els() {
node(c, d, out);
}
}
}
"w:sdt" => {
if let Some(c) = k.child("w:sdtContent") {
for c in c.els() {
@@ -249,13 +259,24 @@ fn node(k: &El, depth: usize, out: &mut Arg) {
}
}
/// `m:t` (and `w:t`) text of a run.
/// `m:t` (and `w:t`) text of a run, with symbol characters (`w:sym`) and special hyphens.
/// Text in a symbol font (Symbol, Wingdings…) becomes its Unicode equivalent: the math font has
/// no glyphs at those private-use codes (#1273).
fn run_text(r: &El) -> String {
let font = r.child("w:rPr").and_then(|p| p.child("w:rFonts")).and_then(|f| f.attr("w:ascii").or_else(|| f.attr("w:hAnsi"))).unwrap_or("");
let mut s = String::new();
for c in r.els() {
match c.name.as_str() {
"m:t" | "w:t" => s.push_str(&c.text()),
"m:t" | "w:t" => s.push_str(&symbol_fonts::map_text(font, &c.text())),
"w:tab" => s.push('\t'),
"w:noBreakHyphen" => s.push('\u{2011}'),
"w:sym" => {
let ch = c.attr("w:char").and_then(|h| u32::from_str_radix(h.trim(), 16).ok()).and_then(char::from_u32).filter(|c| *c as u32 >= 0x20);
if let Some(ch) = ch {
let f = c.attr("w:font").unwrap_or(font);
s.push(symbol_fonts::symbol_char(f, ch).unwrap_or(ch));
}
}
_ => {}
}
}
@@ -308,7 +329,10 @@ fn wrpr(pr: Option<&El>, out: &mut MRun) {
out.color = Some(c);
}
if out.nor {
if let Some(f) = pr.child("w:rFonts").and_then(|f| f.attr("w:ascii").or_else(|| f.attr("w:hAnsi"))) {
// Symbol-font text was mapped to Unicode: it takes the text font.
if let Some(f) =
pr.child("w:rFonts").and_then(|f| f.attr("w:ascii").or_else(|| f.attr("w:hAnsi"))).filter(|f| !symbol_fonts::is_symbol_font(f))
{
out.font = Some(f.to_string());
}
let bold = pr.child("w:b").is_some_and(crate::units::on_off);
@@ -364,4 +388,109 @@ mod tests {
assert!(matches!(m.nodes[2], MNode::Nary { chr: '∫', .. }));
assert!(matches!(&m.nodes[3], MNode::Matrix { rows, .. } if rows.len() == 1 && rows[0].len() == 2));
}
/// Every run's text in the tree, in reading order.
fn texts(m: &mut Math) -> Vec<String> {
let mut out = Vec::new();
wordcraft_doc::math_edit::each_run_mut(&mut m.nodes, &mut |r| out.push(r.text.clone()));
out
}
fn t(s: &str) -> String {
format!("<m:r><m:t>{s}</m:t></m:r>")
}
/// #1273: one equation using every Office Math element of ECMA-376 §22.1 (each argument
/// holding a distinct marker) keeps every argument's text: nothing is silently dropped.
#[test]
fn every_omml_element_keeps_its_content() {
let parts = [
format!("<m:acc><m:accPr><m:chr m:val=\"̃\"/></m:accPr><m:e>{}</m:e></m:acc>", t("A1")),
format!("<m:bar><m:barPr><m:pos m:val=\"top\"/></m:barPr><m:e>{}</m:e></m:bar>", t("B1")),
format!("<m:box><m:boxPr><m:opEmu m:val=\"1\"/></m:boxPr><m:e>{}</m:e></m:box>", t("C1")),
format!("<m:borderBox><m:borderBoxPr><m:strikeH/></m:borderBoxPr><m:e>{}</m:e></m:borderBox>", t("D1")),
format!("<m:d><m:dPr><m:begChr m:val=\"[\"/></m:dPr><m:e>{}</m:e><m:e>{}</m:e></m:d>", t("E1"), t("E2")),
format!("<m:eqArr><m:e>{}</m:e><m:e>{}</m:e></m:eqArr>", t("F1"), t("F2")),
format!("<m:f><m:fPr><m:type m:val=\"skw\"/></m:fPr><m:num>{}</m:num><m:den>{}</m:den></m:f>", t("G1"), t("G2")),
format!("<m:func><m:fName>{}</m:fName><m:e>{}</m:e></m:func>", t("H1"), t("H2")),
format!("<m:groupChr><m:groupChrPr><m:chr m:val=\"⏞\"/><m:pos m:val=\"top\"/></m:groupChrPr><m:e>{}</m:e></m:groupChr>", t("I1")),
format!("<m:limLow><m:e>{}</m:e><m:lim>{}</m:lim></m:limLow>", t("J1"), t("J2")),
format!("<m:limUpp><m:e>{}</m:e><m:lim>{}</m:lim></m:limUpp>", t("K1"), t("K2")),
format!("<m:m><m:mr><m:e>{}</m:e><m:e>{}</m:e></m:mr><m:mr><m:e>{}</m:e><m:e/></m:mr></m:m>", t("L1"), t("L2"), t("L3")),
format!(
"<m:nary><m:naryPr><m:chr m:val=\"∑\"/></m:naryPr><m:sub>{}</m:sub><m:sup>{}</m:sup><m:e>{}</m:e></m:nary>",
t("M1"),
t("M2"),
t("M3")
),
format!("<m:phant><m:phantPr><m:show m:val=\"0\"/></m:phantPr><m:e>{}</m:e></m:phant>", t("N1")),
format!("<m:rad><m:deg>{}</m:deg><m:e>{}</m:e></m:rad>", t("O1"), t("O2")),
format!("<m:sPre><m:sub>{}</m:sub><m:sup>{}</m:sup><m:e>{}</m:e></m:sPre>", t("P1"), t("P2"), t("P3")),
format!("<m:sSub><m:e>{}</m:e><m:sub>{}</m:sub></m:sSub>", t("Q1"), t("Q2")),
format!("<m:sSubSup><m:e>{}</m:e><m:sub>{}</m:sub><m:sup>{}</m:sup></m:sSubSup>", t("R1"), t("R2"), t("R3")),
format!("<m:sSup><m:e>{}</m:e><m:sup>{}</m:sup></m:sSup>", t("S1"), t("S2")),
// Run-level wrappers allowed in math (EG_PContentMath): hyperlinks, simple fields,
// tracked insertions, content controls, markup compatibility.
format!("<w:hyperlink w:anchor=\"x\">{}</w:hyperlink>", t("T1")),
format!("<w:fldSimple w:instr=\"QUOTE\">{}</w:fldSimple>", t("T2")),
format!("<w:ins w:id=\"1\" w:author=\"a\">{}</w:ins>", t("T3")),
format!("<w:sdt><w:sdtContent>{}</w:sdtContent></w:sdt>", t("T4")),
format!(
"<mc:AlternateContent><mc:Choice Requires=\"m\">{}</mc:Choice><mc:Fallback><w:r><w:t>picture</w:t></w:r></mc:Fallback></mc:AlternateContent>",
t("T5")
),
"<w:r><w:t>T6</w:t></w:r>".to_string(),
];
let x = format!(
r#"<m:oMath xmlns:m="http://schemas.openxmlformats.org/officeDocument/2006/math" xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006">{}</m:oMath>"#,
parts.concat()
);
let e = crate::xml::parse(x.as_bytes()).unwrap();
let mut m = read_omath(&e, MathJc::CenterGroup);
let got = texts(&mut m).concat();
let markers = [
"A1", "B1", "C1", "D1", "E1", "E2", "F1", "F2", "G1", "G2", "H1", "H2", "I1", "J1", "J2", "K1", "K2", "L1", "L2", "L3", "M1", "M2", "M3",
"N1", "O1", "O2", "P1", "P2", "P3", "Q1", "Q2", "R1", "R2", "R3", "S1", "S2", "T1", "T2", "T3", "T4", "T5", "T6",
];
for k in markers {
assert!(got.contains(k), "{k} dropped from {got}");
}
assert!(!got.contains("picture"), "the math choice wins over the fallback: {got}");
}
/// #1273: deleted text stays out; symbol characters and symbol-font text become Unicode.
#[test]
fn symbols_and_deletions_in_math() {
let e = omath(
r#"<m:r><m:t>x</m:t></m:r><w:del w:id="1" w:author="a"><m:r><m:t>gone</m:t></m:r></w:del><m:r><w:sym w:font="Symbol" w:char="F061"/></m:r><m:r><w:rPr><w:rFonts w:ascii="Symbol" w:hAnsi="Symbol"/></w:rPr><m:t>&#xF062;</m:t><w:noBreakHyphen/></m:r><w:r><w:rPr><w:rFonts w:ascii="Symbol"/></w:rPr><w:t>&#xF067;</w:t></w:r>"#,
);
let mut m = read_omath(&e, MathJc::CenterGroup);
let got = texts(&mut m).concat();
assert_eq!(got, "xαβ\u{2011}γ");
let mut font = None;
wordcraft_doc::math_edit::each_run_mut(&mut m.nodes, &mut |r| {
if r.nor {
font = Some(r.font.clone());
}
});
assert_eq!(font, Some(None), "mapped symbol text takes the text font");
}
/// Hostile nesting through the wrappers is bounded (no stack overflow, no panic).
#[test]
fn deeply_nested_wrappers_are_bounded() {
// Deeper than the math limit, within the XML parser's own depth limit.
let n = 120;
let inner = format!("{}{}{}", "<w:hyperlink>".repeat(n), t("deep"), "</w:hyperlink>".repeat(n));
let e = omath(&inner);
let _ = read_omath(&e, MathJc::CenterGroup);
let inner = format!(
"{}{}{}",
"<mc:AlternateContent xmlns:mc=\"http://schemas.openxmlformats.org/markup-compatibility/2006\"><mc:Choice Requires=\"m\">".repeat(n),
t("deep"),
"</mc:Choice></mc:AlternateContent>".repeat(n)
);
let e = omath(&inner);
let _ = read_omath(&e, MathJc::CenterGroup);
}
}
+24 -136
View File
@@ -5,6 +5,7 @@ pub(crate) mod diagram;
mod drawing_color;
mod embed;
mod freeform;
mod glossary;
pub(crate) mod math;
mod props;
mod sdt;
@@ -28,6 +29,7 @@ use wordcraft_doc::{
};
use crate::DocxError;
use crate::bibliography::{MAX_SOURCES, Seen};
use crate::package::{ContentTypes, EMBEDDED_PARTS, EmbeddedManifest, MAX_VBA_RELATED, Package, Rels, VBA_PROJECT_PART, VBA_RELATED, rel_is, rt};
use crate::units::{flag, int, on_off, tw, u32_of};
use crate::xml::El;
@@ -45,19 +47,8 @@ const MAX_HF_DEPTH: usize = 4;
/// Most graphic work (items plus path segments, see `story::graphic_work`) built for one file's
/// charts and diagrams; later ones are left empty.
const MAX_GRAPHIC_WORK: usize = 2_000_000;
/// Bibliography sources kept from a file (Word's own source manager holds far fewer).
const MAX_SOURCES: usize = 10_000;
/// Custom XML parts looked at for bibliography sources.
const MAX_BIBLIOGRAPHY_PARTS: usize = 64;
/// Characters kept of a bibliography's citation style name.
const MAX_STYLE_NAME: usize = 64;
/// Bibliography sources already read: tags, and (title, author) of untagged ones.
#[derive(Default)]
struct BibSeen {
tags: HashSet<String>,
untagged: HashSet<(String, String)>,
}
/// Longest attached-template path we keep (bytes); a longer one is not a real path.
const MAX_TEMPLATE_PATH: usize = 32 * 1024;
@@ -232,7 +223,7 @@ pub(crate) fn read_opened(pkg: &Package) -> Result<Document, DocxError> {
bib_parts.push(path);
}
}
let mut seen_sources = BibSeen::default();
let mut seen_sources = Seen::default();
for path in &bib_parts {
if r.doc.sources.len() >= MAX_SOURCES {
break;
@@ -281,11 +272,12 @@ pub(crate) fn read_opened(pkg: &Package) -> Result<Document, DocxError> {
// bytes, never parsed or run, so a .docm/.dotm saved again carries exactly what the file had.
// Only the relationships and content types are read.
if let Some(v) = part(rt::VBA_PROJECT)
&& let Some(b) = pkg.get(&v).filter(|b| !b.is_empty())
&& let Some(b) = pkg.get_shared(&v).filter(|b| !b.is_empty())
{
r.doc.passthrough.insert(VBA_PROJECT_PART.into(), Arc::new(b.to_vec()));
r.doc.passthrough.insert(VBA_PROJECT_PART.into(), b.clone());
r.read_vba_related(&v);
}
r.read_glossary(&rels);
r.keep_custom_xml(&root_rels, &rels);
r.keep_unmodelled(&root_rels, &rels);
if !r.embedded.parts.is_empty() {
@@ -577,7 +569,7 @@ impl Reader<'_> {
log::warn!("docx: not carrying VBA-related part {path:?}");
continue;
}
let Some(bytes) = self.pkg.get(path) else {
let Some(bytes) = self.pkg.get_shared(path) else {
log::warn!("docx: VBA-related part {path} is missing; keeping the project without it");
continue;
};
@@ -587,7 +579,7 @@ impl Reader<'_> {
}
kept += 1;
manifest.push_str(&format!("{}\t{path}\t{ct}\n", rel.kind));
self.doc.passthrough.entry(path.to_string()).or_insert_with(|| Arc::new(bytes.to_vec()));
self.doc.passthrough.entry(path.to_string()).or_insert_with(|| bytes.clone());
}
if !manifest.is_empty() {
self.doc.passthrough.insert(VBA_RELATED.into(), Arc::new(manifest.into_bytes()));
@@ -642,7 +634,7 @@ impl Reader<'_> {
if let Some(k) = self.media_by_path.get(&rel.target) {
return Some(k.clone());
}
let bytes = self.pkg.get(&rel.target)?;
let bytes = self.pkg.get_shared(&rel.target)?;
let base: String =
rel.target.rsplit('/').next().unwrap_or("image").chars().filter(|c| c.is_ascii_alphanumeric() || matches!(c, '.' | '_' | '-')).collect();
let base = if base.is_empty() || base.starts_with('.') { format!("image{base}") } else { base };
@@ -655,7 +647,7 @@ impl Reader<'_> {
None => format!("{base}_{n}"),
};
}
self.doc.media.insert(key.clone(), Arc::new(bytes.to_vec()));
self.doc.media.insert(key.clone(), bytes.clone());
self.media_by_path.insert(rel.target.clone(), key.clone());
Some(key)
}
@@ -816,9 +808,6 @@ impl Reader<'_> {
for s in root.children("w:style").take(10_000) {
let Some(raw_id) = s.attr("w:styleId").filter(|i| !i.is_empty()) else { continue };
let id = self.pc.style_id(raw_id);
if sheet.get(&id).is_some() {
continue;
}
let kind = match s.attr("w:type").unwrap_or("paragraph") {
"character" => StyleKind::Character,
"table" => StyleKind::Table,
@@ -860,7 +849,13 @@ impl Reader<'_> {
let parts = self.table_style_parts(s, &mut st);
st.table = (parts != TableStyleParts::default()).then_some(parts);
}
sheet.styles.push(st);
// A style defined twice (some generators write their own Heading 1 after a
// placeholder one) is the last definition, whole: Word drops the earlier one rather
// than merging the two (observed with synthetic documents).
match sheet.styles.iter_mut().find(|s| s.id == id) {
Some(earlier) => *earlier = st,
None => sheet.styles.push(st),
}
}
// Styles that give only a list level take the list from the style they are based on
// (Heading 3: level 3 of Heading 2's outline list).
@@ -1083,6 +1078,7 @@ impl Reader<'_> {
s.sum_paragraph_spacing = k.children("w:doNotUseHTMLParagraphAutoSpacing").any(on_off);
}
"w:trackRevisions" => s.track_changes = on_off(k),
"w:doNotTrackMoves" => s.track_moves = !on_off(k),
"w:defaultTabStop" => {
if let Some(v) = tw(k, "w:val").filter(|v| *v > 0.0) {
s.default_tab = v.clamp(1.0, 1584.0);
@@ -1318,125 +1314,17 @@ impl Reader<'_> {
/// Sources from a `b:Sources` custom XML part (other custom XML is left alone); at most
/// [`MAX_SOURCES`] in all, and the document's citation style from `StyleName`.
fn read_bibliography(&mut self, path: &str, seen: &mut BibSeen) -> Result<(), DocxError> {
fn read_bibliography(&mut self, path: &str, seen: &mut Seen) -> Result<(), DocxError> {
let Some(root) = self.xml(path)? else { return Ok(()) };
if root.local() != "Sources" {
return Ok(());
}
if self.doc.settings.bib_style.is_none()
&& let Some(name) = root.attrs.iter().find(|(k, _)| k.rsplit(':').next() == Some("StyleName")).map(|(_, v)| v.trim())
&& !name.is_empty()
{
self.doc.settings.bib_style = Some(name.chars().take(MAX_STYLE_NAME).collect());
}
let child_text = |el: &El, local: &str| el.els().find(|e| e.local() == local).map(|e| e.text()).unwrap_or_default();
let parse_persons = |parent: &El| -> String {
let mut names = Vec::new();
let person_els: Vec<&El> = {
let in_lists: Vec<&El> =
parent.els().filter(|e| e.local() == "NameList").flat_map(|n| n.els().filter(|e| e.local() == "Person")).collect();
if in_lists.is_empty() { parent.els().filter(|e| e.local() == "Person").collect() } else { in_lists }
};
for name_el in person_els {
let last = child_text(name_el, "Last");
let first = child_text(name_el, "First");
let middle = child_text(name_el, "Middle");
let name = if !first.is_empty() && !middle.is_empty() {
format!("{last}, {first} {middle}")
} else if !first.is_empty() {
format!("{last}, {first}")
} else {
last
};
if !name.is_empty() {
names.push(name);
}
}
names.join("; ")
};
for src_el in root.els().filter(|e| e.local() == "Source") {
if self.doc.sources.len() >= MAX_SOURCES {
log::warn!("docx: more than {MAX_SOURCES} bibliography sources; the rest are dropped");
break;
}
let tag = child_text(src_el, "Tag");
let raw_kind = child_text(src_el, "SourceType");
let kind = crate::write::from_source_type(&raw_kind);
let (author, editor) = match src_el.els().find(|e| e.local() == "Author") {
Some(a) => {
let auth =
a.els().find(|e| e.local() == "Author").map(parse_persons).filter(|s| !s.is_empty()).unwrap_or_else(|| parse_persons(a));
let ed = a.els().find(|e| e.local() == "Editor").map(parse_persons).unwrap_or_default();
(auth, ed)
}
None => (String::new(), String::new()),
};
let editor = if editor.is_empty() { src_el.els().find(|e| e.local() == "Editor").map(parse_persons).unwrap_or_default() } else { editor };
let title = child_text(src_el, "Title");
let year = child_text(src_el, "Year");
let month = child_text(src_el, "Month");
let day = child_text(src_el, "Day");
let city = child_text(src_el, "City");
let publisher = child_text(src_el, "Publisher");
let journal = child_text(src_el, "JournalName");
let edition = child_text(src_el, "Edition");
let volume = child_text(src_el, "Volume");
let issue = child_text(src_el, "Issue");
let pages = child_text(src_el, "Pages");
let url = child_text(src_el, "URL");
let accessed = match child_text(src_el, "Accessed") {
a if !a.is_empty() => a,
_ => {
let y = child_text(src_el, "YearAccessed");
let m = child_text(src_el, "MonthAccessed");
let d = child_text(src_el, "DayAccessed");
if !y.is_empty() && !m.is_empty() && !d.is_empty() {
format!("{y}-{m}-{d}")
} else if !y.is_empty() && !m.is_empty() {
format!("{y}-{m}")
} else {
y
}
}
};
let doi = match child_text(src_el, "DOI") {
d if !d.is_empty() => d,
_ => child_text(src_el, "StandardNumber"),
};
// Keep sources with some content, once each (by tag, or title and author untagged).
if tag.is_empty() && title.is_empty() && author.is_empty() && editor.is_empty() {
continue;
}
let fresh = if tag.is_empty() {
seen.untagged.insert((title.clone(), if author.is_empty() { editor.clone() } else { author.clone() }))
} else {
seen.tags.insert(tag.clone())
};
if fresh {
self.doc.sources.push(wordcraft_doc::Source {
tag,
kind,
author,
title,
year,
city,
publisher,
journal,
volume,
pages,
url,
editor,
edition,
issue,
month,
day,
accessed,
doi,
});
}
let room = MAX_SOURCES.saturating_sub(self.doc.sources.len());
let bib = crate::bibliography::parse(&root, seen, room);
if self.doc.settings.bib_style.is_none() {
self.doc.settings.bib_style = bib.style;
}
self.doc.sources.extend(bib.sources);
Ok(())
}
+56 -15
View File
@@ -4,8 +4,9 @@ use std::collections::HashMap;
use wordcraft_doc::para::cap_alt;
use wordcraft_doc::props::{
Align, Border, BorderStyle, Borders, CellProps, CharProps, HeightRule, Highlight, LineSpacing, NumRef, ParaProps, Rgb, RowProps, TabAlign,
TabLeader, TabStop, TableFloat, TableLook, TableProps, TextColor, TextDirection, Underline, VAlign, VMerge, VertAlign,
Align, Border, BorderStyle, Borders, CellProps, CharProps, CombineBrackets, EastAsianLayout, HeightRule, Highlight, LineSpacing, NumRef,
ParaProps, Rgb, RowProps, TabAlign, TabLeader, TabStop, TableFloat, TableLook, TableProps, TextColor, TextDirection, Underline, VAlign, VMerge,
VertAlign,
};
use wordcraft_doc::section::{
Columns, DocGrid, DocGridType, LineNumberRestart, LineNumbering, MAX_CHAR_SPACE, MAX_GRID_PITCH, NotePos, NoteProps, NoteRestart, NumFormat,
@@ -13,7 +14,7 @@ use wordcraft_doc::section::{
};
use super::drawing_color::adjust_lum;
use crate::units::{int, measure, on_off, tw, u32_of};
use crate::units::{int, line_multiple, measure, on_off, tw, u32_of};
/// A text frame (`w:framePr` without a drop cap, ECMA-376 §17.3.1.11), sizes and offsets in
/// points (clamped), names kept as OOXML writes them.
@@ -207,11 +208,8 @@ impl PropCtx {
"w:szCs" => c.size_cs = k.attr("w:val").and_then(|v| measure(v, 2.0)).filter(|v| *v > 0.0).map(|v| v.clamp(1.0, 1638.0)),
"w:highlight" => c.highlight = k.attr("w:val").map(Highlight::from_ooxml),
"w:u" => {
if let Some(v) = k.attr("w:val") {
c.underline = Some(Underline::from_ooxml(v));
} else {
c.underline = Some(Underline::Single);
}
// w:val has no default in the schema: without it Word draws no underline
c.underline = k.attr("w:val").map(Underline::from_ooxml);
c.underline_color =
self.theme_color(k, "w:themeColor", "w:themeTint", "w:themeShade").or_else(|| k.attr("w:color").and_then(Rgb::parse));
}
@@ -224,10 +222,15 @@ impl PropCtx {
_ => VertAlign::Baseline,
})
}
"w:eastAsianLayout" => {
let l = east_asian_layout(k);
c.east_asian_layout = (l != EastAsianLayout::default()).then_some(l);
}
"w:lang" => {
c.lang = k.attr("w:val").filter(|v| !v.is_empty() && v.len() < 64).map(str::to_string);
c.lang_bidi = k.attr("w:bidi").filter(|v| !v.is_empty() && v.len() < 64).map(str::to_string);
}
"w14:ligatures" => c.ligatures = k.attr("w14:val").and_then(wordcraft_doc::props::Ligatures::from_ooxml),
n if n.starts_with("w14:") => {
let fx = c.text_effects.get_or_insert_with(Default::default);
super::wordart::read(k, &self.theme, fx);
@@ -266,8 +269,11 @@ impl PropCtx {
"w:autoSpaceDN" => p.auto_space_dn = Some(on_off(k)),
"w:snapToGrid" => p.snap_to_grid = Some(on_off(k)),
"w:framePr" => {
if matches!(k.attr("w:dropCap"), Some("drop") | Some("margin")) {
if let Some(kind @ ("drop" | "margin")) = k.attr("w:dropCap") {
p.drop_cap = Some(k.attr("w:lines").and_then(u32_of).unwrap_or(3).clamp(1, 10) as u8);
p.drop_cap_margin = (kind == "margin").then_some(true);
p.drop_cap_distance =
tw(k, "w:hSpace").filter(|d| d.is_finite() && *d > 0.0).map(|d| d.min(wordcraft_doc::resolve::MAX_DROP_CAP_DISTANCE));
} else {
p.frame = Some(text_frame(k));
}
@@ -312,7 +318,8 @@ impl PropCtx {
p.line_spacing = match rule.as_str() {
"exact" => measure(line, 20.0).map(|v| LineSpacing::Exactly(v.abs())),
"atleast" => measure(line, 20.0).map(|v| LineSpacing::AtLeast(v.abs())),
_ => measure(line, 240.0).map(|v| LineSpacing::Multiple(v.abs().clamp(0.06, 132.0))),
// Strict OOXML writes the height with a unit (`13.90pt`, #2329).
_ => line_multiple(line).map(|v| LineSpacing::Multiple(v.abs().clamp(0.06, 132.0))),
};
}
}
@@ -372,14 +379,18 @@ fn tab_stop(t: &El) -> Option<TabStop> {
"clear" => TabAlign::Clear,
_ => TabAlign::Left,
};
let leader = match t.attr("w:leader").unwrap_or("none") {
Some(TabStop { pos, align, leader: tab_leader(t.attr("w:leader").unwrap_or("none")) })
}
/// A tab stop's or positional tab's `w:leader` (`ST_TabTlc`, `ST_PTabLeader`).
pub(crate) fn tab_leader(v: &str) -> TabLeader {
match v {
"dot" => TabLeader::Dot,
"hyphen" => TabLeader::Hyphen,
"underscore" | "heavy" => TabLeader::Underscore,
"middleDot" => TabLeader::MiddleDot,
_ => TabLeader::None,
};
Some(TabStop { pos, align, leader })
}
}
/// `w:shd/@w:fill` (a solid fill colour; `auto` = none), ignoring theme fills ([`PropCtx::shd_fill`]).
@@ -422,7 +433,11 @@ pub fn borders(e: &El) -> Borders {
/// Margins element (`w:tblCellMar`, `w:tcMar`) → [top, left, bottom, right].
pub fn margins(e: &El) -> [f32; 4] {
let get = |names: &[&str]| names.iter().find_map(|n| e.child(n)).and_then(|c| tw(c, "w:w")).unwrap_or(0.0).clamp(0.0, 1584.0);
margin_sides(e).map(|side| side.unwrap_or(0.0))
}
fn margin_sides(e: &El) -> [Option<f32>; 4] {
let get = |names: &[&str]| names.iter().find_map(|n| e.child(n)).and_then(|c| tw(c, "w:w")).map(|v| v.clamp(0.0, 1584.0));
[get(&["w:top"]), get(&["w:left", "w:start"]), get(&["w:bottom"]), get(&["w:right", "w:end"])]
}
@@ -557,7 +572,14 @@ pub fn tcpr(pc: &PropCtx, e: &El) -> (CellProps, bool) {
"w:shd" => c.shading = pc.shd_fill(k),
"w:noWrap" => c.no_wrap = on_off(k),
"w:hideMark" => c.hide_mark = on_off(k),
"w:tcMar" => c.margins = Some(margins(k)),
"w:tcMar" => {
let sides = margin_sides(k);
if sides.iter().all(Option::is_some) {
c.margins = Some(sides.map(|side| side.unwrap_or(0.0)));
} else {
c.partial_margins = Some(sides);
}
}
"w:textDirection" => {
c.text_direction = TextDirection::from_ooxml(k.attr("w:val").unwrap_or(""));
c.text_direction_set = k.attr("w:val") == Some("lrTb");
@@ -742,3 +764,22 @@ fn table_float(e: &El) -> TableFloat {
overlap: true,
}
}
/// `w:eastAsianLayout` (ISO/IEC 29500-1 §17.3.2.10). Attributes the schema doesn't allow (an id
/// out of `xsd:int`, an unknown bracket, an on/off that isn't one) are left unset.
fn east_asian_layout(e: &El) -> EastAsianLayout {
let flag = |name: &str| {
e.attr(name).and_then(|v| match v.trim() {
"1" | "true" | "on" => Some(true),
"0" | "false" | "off" => Some(false),
_ => None,
})
};
EastAsianLayout {
id: e.attr("w:id").and_then(int).and_then(|v| i32::try_from(v).ok()),
combine: flag("w:combine"),
combine_brackets: e.attr("w:combineBrackets").and_then(|v| CombineBrackets::from_ooxml(v.trim())),
vert: flag("w:vert"),
vert_compress: flag("w:vertCompress"),
}
}
+58 -8
View File
@@ -145,18 +145,63 @@ pub(crate) fn shape_paint(sppr: Option<&El>, style: Option<&El>, theme: &[Rgb])
ShapePaint { fill, stroke, stroke_width }
}
/// A VML colour (`strokecolor`, `fillcolor`): `#rrggbb` or `#rgb` (Word follows it with the
/// theme slot, as in `#4472c4 [3204]`), or a colour name.
/// Explicit DrawingML dash stops override the solid default.
pub(crate) fn line_dash(ln: Option<&El>) -> wordcraft_doc::line_dash::LineDash {
use wordcraft_doc::line_dash::{LineDash, MAX_DASH_STOPS};
let Some(ln) = ln else { return LineDash::Solid };
if let Some(custom) = ln.child("a:custDash") {
let stops: Vec<_> =
custom.children("a:ds").take(MAX_DASH_STOPS).filter_map(|e| Some((dash_ratio(e.attr("d")?)?, dash_ratio(e.attr("sp")?)?))).collect();
return if stops.is_empty() { LineDash::Solid } else { LineDash::Custom(stops) };
}
ln.child("a:prstDash").and_then(|e| e.attr("val")).map(LineDash::from_ooxml).unwrap_or_default()
}
/// Strict OOXML may spell percentages with a percent sign instead of integer units.
fn dash_ratio(value: &str) -> Option<u32> {
const PERCENT_UNITS: f64 = 1000.0;
let value = value.trim();
let Some(percent) = value.strip_suffix('%') else { return value.parse().ok() };
let units = percent.parse::<f64>().ok()? * PERCENT_UNITS;
if !units.is_finite() || !(0.0..=f64::from(u32::MAX)).contains(&units) {
return None;
}
Some(units.round() as u32)
}
/// A VML colour (`strokecolor`, `fillcolor`): `#rrggbb`, `rrggbb`, `#rgb` or `rgb` (Word follows it
/// with the theme slot, as in `#4472c4 [3204]` or `4472c4 [3204]`), an `rgb(...)` function, or a
/// colour name.
pub(crate) fn vml_color(v: &str) -> Option<Rgb> {
let v = v.split_whitespace().next()?;
if let Some(hex) = v.strip_prefix('#') {
if hex.len() == 3 && hex.is_ascii() {
let long: String = hex.chars().flat_map(|c| [c, c]).collect();
return Rgb::parse(&long);
let s = v.trim();
if let Some(inner) = s.strip_prefix("rgb(").or_else(|| s.strip_prefix("rgba(")) {
let nums: Vec<u8> = inner
.trim_end_matches(')')
.split(',')
.take(3)
.filter_map(|x| {
let x = x.trim();
if let Some(p) = x.strip_suffix('%') {
p.trim().parse::<f32>().ok().map(|f| (f.clamp(0.0, 100.0) * 2.55).round() as u8)
} else {
x.parse::<f32>().ok().map(|f| f.clamp(0.0, 255.0).round() as u8)
}
})
.collect();
if let [r, g, b] = nums[..] {
return Some(Rgb(r, g, b));
}
}
let tok = s.split_whitespace().next()?;
let hex = tok.strip_prefix('#').unwrap_or(tok);
if hex.len() == 3 && hex.chars().all(|c| c.is_ascii_hexdigit()) {
let long: String = hex.chars().flat_map(|c| [c, c]).collect();
return Rgb::parse(&long);
}
if hex.len() == 6 && hex.chars().all(|c| c.is_ascii_hexdigit()) {
return Rgb::parse(hex);
}
super::drawing_color::preset_color(v)
super::drawing_color::preset_color(tok)
}
#[cfg(test)]
@@ -255,8 +300,13 @@ mod tests {
#[test]
fn vml_colours() {
assert_eq!(vml_color("#ffc000 [3207]"), Some(Rgb(0xFF, 0xC0, 0)));
assert_eq!(vml_color("ffc000 [3207]"), Some(Rgb(0xFF, 0xC0, 0)));
assert_eq!(vml_color("#f00"), Some(Rgb(0xFF, 0, 0)));
assert_eq!(vml_color("f00"), Some(Rgb(0xFF, 0, 0)));
assert_eq!(vml_color("black"), Some(Rgb(0, 0, 0)));
assert_eq!(vml_color("Red"), Some(Rgb(0xFF, 0, 0)));
assert_eq!(vml_color("rgb(255, 128, 0)"), Some(Rgb(255, 128, 0)));
assert_eq!(vml_color("rgb(100%, 0%, 0%)"), Some(Rgb(255, 0, 0)));
assert_eq!(vml_color("#é"), None);
assert_eq!(vml_color(""), None);
}
+360 -35
View File
@@ -6,7 +6,7 @@ use wordcraft_doc::connector::{ConnEnd, ShapeExtra};
use wordcraft_doc::control::{ContentControl, ControlKind, ControlWrap};
use wordcraft_doc::effects::{Glow, Shadow, ShapeEffects};
use wordcraft_doc::graphic::{Embedded, Graphic, GraphicItem, GraphicKind};
use wordcraft_doc::para::{AltText, Anchor, Float, FloatAlign, NoteKind, ShapeKind, Wrap, WrapText, cap_alt};
use wordcraft_doc::para::{AltText, Anchor, Float, FloatAlign, NoteKind, PTabAlign, PTabBase, ShapeKind, Wrap, WrapText, cap_alt};
use wordcraft_doc::props::{CharProps, NumChange, PropChange, Rgb, TextColor};
use wordcraft_doc::table::{Cell, MAX_COLS, MAX_ROWS, Row, Table};
use wordcraft_doc::{Block, Blocks, InlineObject, Paragraph, PartKind, RevisionKind, Run, para_block};
@@ -30,6 +30,8 @@ const MAX_TABLE_DEPTH: usize = 24;
const MAX_FIELD_DEPTH: usize = 32;
/// Deepest text-box-in-text-box nesting.
const MAX_STORY_DEPTH: usize = 4;
/// Deepest `v:group` nesting followed (hostile files).
const MAX_VML_GROUP_DEPTH: usize = 8;
/// Longest field code we keep (Zotero's citation codes embed item data as JSON).
const MAX_INSTR: usize = 2 * 1024 * 1024;
/// Longest legacy form field definition (`w:ffData`) we keep.
@@ -176,7 +178,8 @@ fn clean_text(s: &str) -> String {
fn children_of_choice(ac: &El) -> Option<&El> {
// Prefer a Choice whose requirements we understand; else the Fallback; else any Choice.
const KNOWN: &[&str] = &["wps", "wpg", "wpc", "wpi", "w14", "w15", "wp14", "a14", "w16se", "w16cid", "w16", "w16cex", "w16sdtdh", "v"];
// `m`: Office Math (a producer may guard an equation with a picture fallback, #1273).
const KNOWN: &[&str] = &["wps", "wpg", "wpc", "wpi", "w14", "w15", "wp14", "a14", "w16se", "w16cid", "w16", "w16cex", "w16sdtdh", "v", "m"];
let ok = |c: &El| c.attr("Requires").unwrap_or("").split_whitespace().all(|r| KNOWN.contains(&r));
ac.children("mc:Choice").find(|c| ok(c)).or_else(|| ac.child("mc:Fallback")).or_else(|| ac.child("mc:Choice"))
}
@@ -199,10 +202,21 @@ impl Reader<'_> {
&& p.props.drop_cap.is_none()
&& let Some(Block::Para(mut first)) = out.pop().map(|b| (*b).clone())
{
let lines = first.props.drop_cap;
let mut props = p.props.clone();
props.drop_cap = first.props.drop_cap;
props.drop_cap_margin = first.props.drop_cap_margin;
props.drop_cap_distance = first.props.drop_cap_distance;
// A letter in a font of its own: the drop cap's font, the letter itself
// in the text's (so taking the drop cap away gives the letter its font back).
let cap_font = first.runs.first().and_then(|r| r.props.font.clone());
let text_font = p.runs.first().and_then(|r| r.props.font.clone());
if cap_font.is_some() && cap_font != text_font {
for r in &mut first.runs {
r.props.font = text_font.clone();
}
props.drop_cap_font = cap_font;
}
first.append(p);
props.drop_cap = lines;
first.props = props;
out.push(para_block(first));
continue;
@@ -218,6 +232,13 @@ impl Reader<'_> {
}
"w:sdt" => self.read_block_sdt(sc, e, rels, out, depth),
"w:customXml" | "w:smartTag" | "w:ins" | "w:moveTo" => self.read_blocks(sc, e, rels, out, depth + 1),
// Math is allowed straight in a story (EG_BlockLevelElts takes EG_RunLevelElements,
// which takes EG_MathContent): it gets a paragraph of its own (#1273).
"m:oMathPara" | "m:oMath" => {
let wrapped = El { name: "w:p".into(), attrs: Vec::new(), kids: vec![Node::El(e.clone())] };
let p = self.read_para(sc, &wrapped, rels, depth);
out.push(para_block(p));
}
"mc:AlternateContent" => {
if let Some(c) = children_of_choice(e) {
self.read_blocks(sc, c, rels, out, depth + 1);
@@ -241,6 +262,7 @@ impl Reader<'_> {
let Some(content) = e.child("w:sdtContent") else { return };
let mut ctl = super::sdt::sdt_pr(e.child("w:sdtPr"), true);
ctl.end_pr_xml = super::sdt::sdt_end_pr(e.child("w:sdtEndPr"));
ctl.run_props = e.child("w:sdtPr").and_then(|pr| pr.child("w:rPr")).map(|r| self.pc.rpr(r));
let mut inner = Blocks::new();
self.read_blocks(sc, content, rels, &mut inner, depth + 1);
// A table of contents' control belongs to the TOC: the writer puts it back around the
@@ -295,17 +317,25 @@ impl Reader<'_> {
}
fn marker(&mut self, e: &El) -> Option<InlineObject> {
let id = e.attr("w:id").unwrap_or("").to_string();
let id = e.attr("w:id").map(str::trim).unwrap_or("").to_string();
match e.name.as_str() {
"w:bookmarkStart" => {
let name = e.attr("w:name").unwrap_or("").to_string();
if self.bookmarks.len() >= 100_000 {
return None;
}
let name = e.attr("w:name").map(str::trim).unwrap_or("").to_string();
if name.is_empty() || name.len() > 256 {
return None;
}
self.bookmarks.insert(id, name.clone());
Some(InlineObject::BookmarkStart { name })
}
"w:bookmarkEnd" => self.bookmarks.get(&id).map(|n| InlineObject::BookmarkEnd { name: n.clone() }),
"w:bookmarkEnd" => self
.bookmarks
.get(&id)
.cloned()
.or_else(|| e.attr("w:name").map(str::trim).filter(|n| !n.is_empty() && n.len() <= 256).map(str::to_string))
.map(|name| InlineObject::BookmarkEnd { name }),
"w:commentRangeStart" => self.comment_id(&id).map(|id| InlineObject::CommentStart { id }),
"w:permStart" => {
// Bounded: a hostile file can't grow the map without end.
@@ -494,6 +524,7 @@ impl Reader<'_> {
let Some(c) = k.child("w:sdtContent") else { return };
let mut ctl = super::sdt::sdt_pr(k.child("w:sdtPr"), false);
ctl.end_pr_xml = super::sdt::sdt_end_pr(k.child("w:sdtEndPr"));
ctl.run_props = k.child("w:sdtPr").and_then(|pr| pr.child("w:rPr")).map(|r| self.pc.rpr(r));
// The markers take the formatting of the content's first run (typing at the
// start of the content continues it).
let props = c.find("w:rPr").map(|p| self.run_props(p, ctx)).unwrap_or_else(|| self.run_props_none(ctx));
@@ -618,7 +649,15 @@ impl Reader<'_> {
f.instr.push_str(&k.text());
}
}
"w:tab" | "w:ptab" => self.emit_text(sc, pb, "\t", props),
"w:tab" => self.emit_text(sc, pb, "\t", props),
// Alignment and base are required (ECMA-376 §17.3.3.23); without them it stays a plain tab.
"w:ptab" => match (k.attr("w:alignment").and_then(PTabAlign::from_ooxml), k.attr("w:relativeTo").and_then(PTabBase::from_ooxml)) {
(Some(align), Some(relative_to)) => {
let leader = super::props::tab_leader(k.attr("w:leader").unwrap_or("none"));
self.emit_obj(sc, pb, InlineObject::PositionalTab { align, relative_to, leader }, props)
}
_ => self.emit_text(sc, pb, "\t", props),
},
"w:br" => {
if sc.fields.iter().any(|f| f.phase == Phase::Instr) {
return;
@@ -641,8 +680,9 @@ impl Reader<'_> {
let ch = k.attr("w:char").and_then(|h| u32::from_str_radix(h.trim(), 16).ok()).and_then(char::from_u32).filter(|c| *c as u32 >= 0x20);
if let Some(ch) = ch {
let mut p = props.clone();
if let Some(f) = k.attr("w:font").filter(|f| !f.is_empty()) {
p.font = Some(f.to_string());
// The symbol's own font, kept apart from the run's (ECMA-376 §17.3.3.30).
if let Some(f) = k.attr("w:font").filter(|f| !f.trim().is_empty()) {
p.sym_font = Some(f.to_string());
}
self.emit_text(sc, pb, &ch.to_string(), &p);
}
@@ -662,7 +702,7 @@ impl Reader<'_> {
}
}
"w:pict" => {
if let Some(o) = self.read_vml(sc, k, rels) {
for o in self.read_vml_all(sc, k, rels) {
self.emit_obj(sc, pb, o, props);
}
}
@@ -879,6 +919,17 @@ impl Reader<'_> {
Some(e) => (e.instr(), if result.is_empty() { e.text } else { result }),
None => (f.instr.trim().to_string(), result),
};
// The visible result owns the field's formatting; empty results keep the begin's.
let props = f
.buf
.iter()
.find_map(|(it, p)| match it {
Item::Text(t) if !t.is_empty() => Some(p),
Item::Obj(o) if !o.plain_text().is_empty() => Some(p),
_ => None,
})
.unwrap_or(&f.props)
.clone();
let obj = InlineObject::Field { instr, result, locked: f.locked, dirty: f.dirty, code: false, form: f.form };
if let Some(parent) = sc.fields.last_mut()
&& !parent.spilled
@@ -887,7 +938,7 @@ impl Reader<'_> {
parent.instr.push_str(obj.plain_text());
return;
}
self.emit_obj(sc, pb, obj, &f.props);
self.emit_obj(sc, pb, obj, &props);
}
/// At a paragraph end, open fields can't keep buffering: emit them (empty result) and
@@ -968,7 +1019,9 @@ impl Reader<'_> {
float.set_spin(xfrm_spin(content.child("w14:xfrm")));
let (px, py) = float.spin_pad(w, h);
for (e, p) in float.effect.iter_mut().zip([px, py, px, py]) {
*e = (*e - p).clamp(0.0, 1584.0);
if *e != 0.0 {
*e = (*e - p).clamp(0.0, 1584.0);
}
}
// The renderer's existing missing-picture box marks ink without a usable preview.
// InkML stays opaque: no approximation of its strokes or conversion to freeforms.
@@ -1024,7 +1077,10 @@ impl Reader<'_> {
if let Some(f) = obj.float_mut() {
let (px, py) = f.spin_pad(w, h);
for (e, p) in f.effect.iter_mut().zip([px, py, px, py]) {
*e = (*e - p).clamp(0.0, 1584.0);
// Zero adds no effects room when rotation contracts this axis.
if *e != 0.0 {
*e = (*e - p).clamp(0.0, 1584.0);
}
}
}
Some(obj)
@@ -1081,6 +1137,29 @@ impl Reader<'_> {
Some(InlineObject::Graphic { w, h, alt, float, graphic })
}
/// A group or canvas member that is native ink (`content`, a `w14:contentPart`, inside
/// `alternate` when it has a picture fallback): what to write back (the whole alternate when
/// it can be kept, else the ink alone) and its fallback picture (a `Document::media` key).
/// `None` when the ink's part is missing or isn't InkML.
fn ink_member(&mut self, content: &El, alternate: Option<&El>, rels: &Rels) -> Option<(Arc<Embedded>, Option<String>)> {
let InlineObject::Graphic { graphic, .. } = self.read_ink_content_part(content, rels)? else { return None };
let mut source = graphic.source.clone()?;
let mut preview = None;
let choice = alternate.and_then(children_of_choice).filter(|c| c.name == "mc:Choice");
if let (Some(choice), Some(fallback)) = (choice, alternate.and_then(|a| a.child("mc:Fallback"))) {
let kept = El { name: "mc:AlternateContent".into(), attrs: Vec::new(), kids: vec![Node::El(choice.clone()), Node::El(fallback.clone())] };
// A malformed or over-budget fallback must not discard a valid native part.
if let Some(whole) = self.keep_embedded(&kept, rels, &[]) {
source = whole;
preview = match fallback.child("pic:pic").and_then(|p| self.read_pic(p, rels, 0.0, 0.0, String::new(), Float::default())) {
Some(InlineObject::Image { media, .. }) => Some(media),
_ => None,
};
}
}
Some((source, preview))
}
fn read_ink_content_part(&mut self, content: &El, rels: &Rels) -> Option<InlineObject> {
let part = super::part_of(rels, content.attr("r:id")?, INK_REL)?;
let pkg = self.pkg;
@@ -1212,16 +1291,32 @@ impl Reader<'_> {
if out.len() >= wordcraft_doc::para::MAX_GROUP_CHILDREN {
return;
}
let e = match e.name.as_str() {
"mc:AlternateContent" => match children_of_choice(e).and_then(|c| c.els().next()) {
Some(c) => c,
None => continue,
},
_ => e,
// Native ink (`w14:contentPart`), alone or with a picture fallback: kept whole, in its
// place among the members. Ink that can't be kept gives way to its fallback.
let (e, ink) = match e.name.as_str() {
"w14:contentPart" => (e, self.ink_member(e, None, rels)),
"mc:AlternateContent" => {
let choice = children_of_choice(e);
match choice.filter(|c| c.name == "mc:Choice").and_then(|c| c.els().next()).filter(|c| c.name == "w14:contentPart") {
Some(content) => match self.ink_member(content, Some(e), rels) {
Some(ink) => (content, Some(ink)),
None => match e.child("mc:Fallback").and_then(|f| f.els().next()) {
Some(c) => (c, None),
None => continue,
},
},
None => match choice.and_then(|c| c.els().next()) {
Some(c) => (c, None),
None => continue,
},
}
}
_ => (e, None),
};
let sppr = match e.name.as_str() {
"wps:wsp" => e.child("wps:spPr"),
"pic:pic" => e.child("pic:spPr"),
"w14:contentPart" if ink.is_some() => Some(e),
// A group inside a canvas is a whole `wpg:wgp`.
"wpg:grpSp" | "wpg:wgp" => e.child("wpg:grpSpPr"),
_ => continue,
@@ -1243,10 +1338,21 @@ impl Reader<'_> {
let max = crate::units::MAX_LEN_PT;
let fit = |v: f32, lo: f32| wordcraft_geom::finite(v).clamp(lo, max);
let (w, h) = (fit(ew * sx, 0.0), fit(eh * sy, 0.0));
let nv = if e.name == "pic:pic" { e.find("pic:cNvPr") } else { e.child("wps:cNvPr") };
let nv = match e.name.as_str() {
"pic:pic" => e.find("pic:cNvPr"),
"w14:contentPart" => e.child("w14:nvContentPartPr").and_then(|p| p.child("w14:cNvPr")),
_ => e.child("wps:cNvPr"),
};
let (alt, meta) = nv_alt(nv);
let float = Float { alt: meta, hidden: nv.is_some_and(|p| on_off_attr(p, "hidden")), ..Float::default() };
let obj = if e.name == "pic:pic" {
let obj = if let Some((source, preview)) = ink {
let mut float = float;
float.set_spin(xfrm_spin(e.child("w14:xfrm")));
// As drawing-level ink: its fallback picture, else the missing-picture box.
let items = vec![GraphicItem::Image { rect: [0.0, 0.0, w, h], media: preview.unwrap_or_default() }];
let graphic = Arc::new(Graphic { kind: GraphicKind::Ink, items, w, h, source: Some(source), ..Default::default() });
Some(InlineObject::Graphic { w, h, alt: alt.clone(), float, graphic })
} else if e.name == "pic:pic" {
self.read_pic(e, rels, w, h, alt.clone(), float)
} else {
let mut o = self.read_wsp(sc, e, rels, w, h, float);
@@ -1327,7 +1433,15 @@ impl Reader<'_> {
part.body.inset = (inset != wordcraft_doc::TEXT_BOX_INSET).then_some(inset);
part.body.auto_fit = bp.child("a:spAutoFit").is_some();
}
let mut extra = ShapeExtra::default();
// The shape style's text colour (`a:fontRef`, often `lt1`: white on the accent fill).
if let Some(id) = story
&& let Some(c) = wsp.child("wps:style").and_then(|s| s.child("a:fontRef"))
&& let Some(rgb) = super::drawing_color::color(c, &self.doc.settings.theme_colors)
&& let Some(part) = self.doc.parts.get_mut(&id)
{
part.body.auto_color = Some(rgb);
}
let mut extra = ShapeExtra { dash: super::shape_style::line_dash(ln), ..Default::default() };
// A rounded rectangle's own corner (`adj`, ECMA-376 §20.1.10.56); other presets drawn as
// rounded rectangles keep the default.
if prst == "roundRect" {
@@ -1380,20 +1494,125 @@ impl Reader<'_> {
self.doc.add_part(PartKind::TextBox, blocks)
}
/// Legacy VML pictures, text boxes, lines, rectangles, ovals and preset shapes (best effort).
/// Every legacy VML drawing in one `w:pict`, in document order: a `w:pict` may hold any
/// number of shapes (ECMA-376 Part 4, VML), and each one is its own object. A `v:group`
/// becomes an [`InlineObject::Group`].
fn read_vml_all(&mut self, sc: &mut StoryCtx, pict: &El, rels: &Rels) -> Vec<InlineObject> {
let mut out = Vec::new();
for e in pict.els().filter(|e| e.name.starts_with("v:") && e.local() != "shapetype") {
if out.len() >= wordcraft_doc::para::MAX_GROUP_CHILDREN {
break;
}
if sc.in_hf
&& let Some(wm) = vml_watermark(e)
{
// The document's watermark (the first one found); drawn on every page, not as
// a shape, so it is not one of the objects this picture contributes.
self.doc.settings.watermark.get_or_insert(wm);
continue;
}
let o = if e.local() == "group" { self.read_vml_group(sc, pict, e, rels) } else { self.read_vml_shape(sc, pict, e, rels, None) };
out.extend(o);
}
out
}
/// The first VML drawing in `pict`, for the places that show a single one: an OLE object's
/// picture and a `mc:Fallback` preview.
fn read_vml(&mut self, sc: &mut StoryCtx, pict: &El, rels: &Rels) -> Option<InlineObject> {
let shape = pict.els().find(|e| e.name.starts_with("v:") && e.local() != "shapetype")?;
if sc.in_hf
&& let Some(wm) = vml_watermark(shape)
{
// The document's watermark (the first one found); drawn on every page, not as a shape.
self.doc.settings.watermark.get_or_insert(wm);
self.read_vml_all(sc, pict, rels).into_iter().next()
}
/// A `v:group`: its own box from its `style`, its members in the space `coordsize` and
/// `coordorigin` give it, nested groups flattened into it as [`Self::group_members`] does for
/// DrawingML. `None` when it has no member we can show. `pict` is the picture the group sits
/// in, where a member's `v:shapetype` is written.
fn read_vml_group(&mut self, sc: &mut StoryCtx, pict: &El, grp: &El, rels: &Rels) -> Option<InlineObject> {
let style = grp.attr("style").unwrap_or("");
let len = |k: &str| vml_style(style, k).and_then(|v| measure(v, 1.0)).unwrap_or(0.0).clamp(0.0, crate::units::MAX_LEN_PT);
let (w, h) = (len("width"), len("height"));
let float = vml_float(grp, style, sc.in_hf);
// The members' space; without a usable `coordsize` it is the group's own box.
let (ch_w, ch_h) = vml_coordpair(grp, "coordsize").filter(|(cw, ch)| *cw > 0.0 && *ch > 0.0).unwrap_or((w.max(1.0), h.max(1.0)));
let (cx, cy) = vml_coordpair(grp, "coordorigin").unwrap_or((0.0, 0.0));
let mut children = Vec::new();
self.vml_group_members(sc, pict, grp, rels, (-cx, -cy, 1.0, 1.0), 0, &mut children);
if children.is_empty() {
return None;
}
Some(InlineObject::Group { w, h, float, ch_w, ch_h, children, canvas: None })
}
/// The members of VML group `g` into `out`, mapped into the outermost group's space by `t`
/// (x0, y0, sx, sy: outer = x0 + inner × sx). A member is placed by `left`/`top` in that
/// space, except a `v:line`, which carries its own offset in `from`.
fn vml_group_members(
&mut self,
sc: &mut StoryCtx,
pict: &El,
g: &El,
rels: &Rels,
t: (f32, f32, f32, f32),
depth: usize,
out: &mut Vec<wordcraft_doc::para::GroupChild>,
) {
let (x0, y0, sx, sy) = t;
let max = crate::units::MAX_LEN_PT;
let fit = |v: f32, lo: f32| wordcraft_geom::finite(v).clamp(lo, max);
for e in g.els() {
if out.len() >= wordcraft_doc::para::MAX_GROUP_CHILDREN {
return;
}
if !e.name.starts_with("v:") || e.local() == "shapetype" {
continue;
}
let style = e.attr("style").unwrap_or("");
let num = |k: &str| vml_style(style, k).and_then(|v| measure(v, 1.0)).map_or(0.0, wordcraft_geom::finite);
let ((ox, oy), (ew, eh)) = ((num("left"), num("top")), (num("width"), num("height")));
if e.local() == "group" {
if depth >= MAX_VML_GROUP_DEPTH {
continue;
}
// Its members' space maps onto its own box.
let (cw, ch) = vml_coordpair(e, "coordsize").filter(|(cw, ch)| *cw > 0.0 && *ch > 0.0).unwrap_or((ew.max(1.0), eh.max(1.0)));
let (cox, coy) = vml_coordpair(e, "coordorigin").unwrap_or((0.0, 0.0));
let (kx, ky) = (ew / cw, eh / ch);
let nt = (x0 + (ox - cox * kx) * sx, y0 + (oy - coy * ky) * sy, sx * kx, sy * ky);
self.vml_group_members(sc, pict, e, rels, nt, depth + 1, out);
continue;
}
let Some(mut obj) = self.read_vml_shape(sc, pict, e, rels, Some((sx, sy))) else { continue };
// A member is placed by `x`/`y`, so a line's own offset comes off the member itself.
let (dx, dy) = match obj.float_mut() {
Some(f) => {
let d = (f.x, f.y);
(f.x, f.y) = (0.0, 0.0);
d
}
None => (0.0, 0.0),
};
out.push(wordcraft_doc::para::GroupChild { x: fit(x0 + (ox + dx) * sx, -max), y: fit(y0 + (oy + dy) * sy, -max), obj });
}
}
/// One legacy VML picture, text box, line, rectangle, oval or preset shape (best effort).
/// `pict` is the picture it sits in, where its `v:shapetype` is written -- for a group's
/// member too, since Word writes each shapetype once, beside the group rather than inside
/// it. `grp` scales a member from its group's space, and leaves placing it to the caller.
fn read_vml_shape(&mut self, sc: &mut StoryCtx, pict: &El, shape: &El, rels: &Rels, grp: Option<(f32, f32)>) -> Option<InlineObject> {
let style = shape.attr("style").unwrap_or("");
let len = |k: &str| vml_style(style, k).and_then(|v| measure(v, 1.0)).unwrap_or(0.0).clamp(0.0, crate::units::MAX_LEN_PT);
let (mut w, mut h) = (len("width"), len("height"));
let mut float = vml_float(shape, style);
let (sx, sy) = grp.unwrap_or((1.0, 1.0));
let (mut w, mut h) = ((len("width") * sx).min(crate::units::MAX_LEN_PT), (len("height") * sy).min(crate::units::MAX_LEN_PT));
// A member of a group never floats: the group does, and places it.
let mut float = match grp {
Some(_) => {
let mut f = Float { hidden: vml_style(style, "visibility").is_some_and(|v| v.eq_ignore_ascii_case("hidden")), ..Float::default() };
f.set_spin(vml_spin(style));
f
}
None => vml_float(shape, style, sc.in_hf),
};
if let Some(img) = shape.find("v:imagedata") {
let id = img.attr("r:id").or_else(|| img.attr("r:pict"))?;
let media = self.media_for(rels, id)?;
@@ -1420,6 +1639,52 @@ impl Reader<'_> {
}
_ => None,
};
// Word 2007 WordArt: a `v:shape` whose `v:textpath` draws its `string` (the header watermark
// is taken before this). Shown as a text box holding that text, in the textpath's font and the
// shape's `fillcolor`; the box itself has neither fill nor outline, as the letters are the art.
if shape.local() == "shape"
&& let Some(tp) = shape.child("v:textpath")
&& !tp.attr("on").is_some_and(|v| matches!(v, "f" | "false" | "0"))
&& let Some(text) = tp.attr("string").map(|t| t.chars().take(200).collect::<String>()).filter(|t| !t.trim().is_empty())
&& sc.story_depth < MAX_STORY_DEPTH
{
let font = tp
.attr("style")
.and_then(|s| vml_style(s, "font-family"))
.map(|f| f.trim_matches(|c: char| c == '"' || c == '\'' || c.is_whitespace()).chars().take(100).collect::<String>())
.filter(|f| !f.is_empty());
// The shapetype's WordArt shape (`o:spt` 136-175) bends the text like DrawingML's warp.
let warp = vml_spt(shape, shapetype, type_id)
.and_then(|spt| spt.checked_sub(136))
.and_then(|i| VML_TEXT_WARPS.get(i as usize))
.copied()
.and_then(TextWarp::new);
// The letters fill the box (`fitshape`): sized to its height.
let props = CharProps {
font,
size: Some((h * 0.75).clamp(8.0, 200.0)),
color: attr("fillcolor").and_then(vml_color).map(TextColor::Rgb),
..CharProps::default()
};
let id = self.doc.add_part(PartKind::TextBox, vec![para_block(Paragraph::with_text(&text, props))]);
if let Some(part) = self.doc.parts.get_mut(&id) {
part.body.anchor = wordcraft_doc::props::VAlign::Center;
part.body.inset = Some([0; 4]);
}
return Some(InlineObject::Shape {
kind: ShapeKind::TextBox,
w,
h,
fill: None,
stroke: None,
stroke_width,
float,
story: Some(id),
freeform: None,
effects: Default::default(),
extra: ShapeExtra { warp, ..ShapeExtra::default() },
});
}
if let Some((tb, t)) = vml_textbox(shape, 0)
&& sc.story_depth < MAX_STORY_DEPTH
{
@@ -1513,7 +1778,7 @@ impl Reader<'_> {
};
let ((x0, y0), (x1, y1)) = (pt("from"), pt("to"));
let max = crate::units::MAX_LEN_PT;
(w, h) = ((x1 - x0).abs().min(max), (y1 - y0).abs().min(max));
(w, h) = (((x1 - x0).abs() * sx).min(max), ((y1 - y0).abs() * sy).min(max));
float.x = (float.x + x0.min(x1)).clamp(-max, max);
float.y = (float.y + y0.min(y1)).clamp(-max, max);
// Drawn top-left to bottom-right; flipped for the other directions.
@@ -1732,6 +1997,8 @@ fn anchor_float(c: &El) -> Float {
f.dist = dist("distL").max(dist("distR"));
f.dist_top = dist("distT");
f.dist_bottom = dist("distB");
// Required by the schema; a file leaving it out lays out in the cell, as before.
f.outside_cell = !on_off_attr(c, "layoutInCell") && c.attr("layoutInCell").is_some();
f
}
@@ -1759,6 +2026,51 @@ fn wrap_polygon(e: &El) -> Option<wordcraft_doc::wrap::WrapPolygon> {
wordcraft_doc::wrap::WrapPolygon::new(edited, pts)
}
/// The DrawingML text warps of VML's WordArt shapes, `o:spt` 136 (plain text) to 175 (can down), in
/// order: the same shapes under their DrawingML names (`ST_TextShapeType`).
const VML_TEXT_WARPS: [&str; 40] = [
"textPlain",
"textStop",
"textTriangle",
"textTriangleInverted",
"textChevron",
"textChevronInverted",
"textRingInside",
"textRingOutside",
"textArchUp",
"textArchDown",
"textCircle",
"textButton",
"textArchUpPour",
"textArchDownPour",
"textCirclePour",
"textButtonPour",
"textCurveUp",
"textCurveDown",
"textCascadeUp",
"textCascadeDown",
"textWave1",
"textWave2",
"textDoubleWave1",
"textWave4",
"textInflate",
"textDeflate",
"textInflateBottom",
"textDeflateBottom",
"textInflateTop",
"textDeflateTop",
"textDeflateInflate",
"textDeflateInflateDeflate",
"textFadeRight",
"textFadeLeft",
"textFadeUp",
"textFadeDown",
"textSlantUp",
"textSlantDown",
"textCanUp",
"textCanDown",
];
/// A VML shape's preset type (`o:spt`): its own, its shapetype's, or the number in Word's
/// shapetype id `_x0000_tNN` (`type_id`), Word writing each shapetype only at its first use.
fn vml_spt(shape: &El, shapetype: Option<&El>, type_id: Option<&str>) -> Option<u32> {
@@ -1773,6 +2085,15 @@ fn vml_spt(shape: &El, shapetype: Option<&El>, type_id: Option<&str>) -> Option<
/// The value of property `key` in a VML `style` attribute (CSS declarations; the last one wins).
/// The first `v:textbox` under `e` holding text, with its `w:txbxContent` (depth-first, bounded
/// like `El::find`): in a group, a member's margins go with that member's text.
/// A VML `coordsize="w,h"` or `coordorigin="x,y"` pair (ECMA-376 Part 4 §14.1.2): the member
/// space of a group, in its own unitless coordinates.
fn vml_coordpair(e: &El, name: &str) -> Option<(f32, f32)> {
let v = e.attr(name)?;
let (x, y) = v.split_once(',')?;
let n = |s: &str| measure(s.trim(), 1.0).map(wordcraft_geom::finite);
Some((n(x)?, n(y)?))
}
fn vml_textbox(e: &El, depth: usize) -> Option<(&El, &El)> {
if depth > crate::xml::MAX_DEPTH {
return None;
@@ -1793,7 +2114,7 @@ fn vml_style<'a>(style: &'a str, key: &str) -> Option<&'a str> {
/// `mso-position-horizontal` / `mso-position-vertical`), wrapped as its `w10:wrap` says. Without
/// wrapping (`none`, or no `w10:wrap`) it's in front of the text, or behind it at a negative
/// `z-index`. Anything else stays inline.
fn vml_float(shape: &El, style: &str) -> Float {
fn vml_float(shape: &El, style: &str, in_hf: bool) -> Float {
let get = |k: &str| vml_style(style, k);
let hidden = get("visibility").is_some_and(|v| v.eq_ignore_ascii_case("hidden"));
if !get("position").is_some_and(|p| p.eq_ignore_ascii_case("absolute")) {
@@ -1846,6 +2167,9 @@ fn vml_float(shape: &El, style: &str) -> Float {
dist_top: dist("mso-wrap-distance-top"),
dist_bottom: dist("mso-wrap-distance-bottom"),
hidden,
// Word for Mac draws a VML shape anchored in a header table cell on the page whatever
// `o:allowincell` says (user-checked), as for Word's picture and text watermarks.
outside_cell: in_hf,
..Float::default()
};
f.set_spin(vml_spin(style));
@@ -2000,7 +2324,8 @@ fn xfrm_spin(sppr: Option<&El>) -> wordcraft_geom::Spin {
/// The `a:xfrm` in shape properties `sppr` (zeros when missing).
fn group_xfrm(sppr: Option<&El>) -> Xfrm {
let x = sppr.and_then(|s| s.child("a:xfrm"));
// A content part (native ink) has its own `w14:xfrm`.
let x = sppr.and_then(|s| s.child("a:xfrm").or_else(|| s.child("w14:xfrm")));
let pair = |name: &str, a: &str, b: &str, lim: f32| {
let e = x.and_then(|x| x.child(name))?;
let v = |n: &str| e.attr(n).and_then(|v| measure(v, 12_700.0)).map(|v| v.clamp(-lim, lim)).unwrap_or(0.0);
+21
View File
@@ -34,6 +34,17 @@ pub fn measure(s: &str, per_pt: f32) -> Option<f32> {
pt.is_finite().then_some(pt.clamp(-MAX_LEN_PT, MAX_LEN_PT))
}
/// `w:line` under `lineRule="auto"`: the multiple of single line spacing, in 240ths (240 =
/// single). Without a unit the value already is that — the transitional form. A universal
/// measure is a height (strict OOXML writes `w:line="13.90pt"`, ISO 29500
/// `ST_SignedTwipsMeasure`), so it goes through twips as Word reads it: × 20 ÷ 240 (13.90 pt =
/// 278 twips, the 1.1583 multiple Word renders).
pub fn line_multiple(s: &str) -> Option<f32> {
let t = s.trim();
let measured = t.ends_with(|c: char| c.is_ascii_alphabetic());
measure(t, 240.0).map(|v| if measured { v * 20.0 / 240.0 } else { v })
}
/// Plain integer (accepts decimals, rounds; clamps into i64 range).
pub fn int(s: &str) -> Option<i64> {
let t = s.trim();
@@ -98,6 +109,16 @@ mod tests {
assert_eq!(measure("1e400", 20.0), None);
assert_eq!(measure("99999999999999", 20.0), Some(MAX_LEN_PT));
assert_eq!(measure("x", 20.0), None);
// `w:line` auto: the multiple in 240ths without a unit, through twips with one (#2329).
assert_eq!(line_multiple("240"), Some(1.0));
assert_eq!(line_multiple("276"), Some(1.15));
assert_eq!(line_multiple("0"), Some(0.0));
let strict = line_multiple("13.90pt").expect("strict line height");
assert!((strict - 278.0 / 240.0).abs() < 1e-5, "{strict}");
let inches = line_multiple("1.2in").expect("inches");
assert!((inches - 7.2).abs() < 1e-5, "{inches}");
assert_eq!(line_multiple("junk"), None);
assert_eq!(line_multiple(""), None);
assert_eq!(int("12.6"), Some(13));
assert_eq!(u32_of("-5"), Some(0));
}
+148
View File
@@ -0,0 +1,148 @@
//! Run direction for mixed right-to-left and left-to-right text (#1543).
//!
//! WordCraft lays a paragraph out with the Unicode Bidirectional Algorithm (UAX #9) from its
//! text and direction alone, so a run need not say its direction. Word orders a run's characters
//! by the run's `w:rtl` (ECMA-376 Part 1 §17.3.2.30): text in a run without it is laid out left
//! to right, so Hebrew words on either side of a Latin name, and the punctuation between them,
//! change places. A run that doesn't say its direction is therefore split where the resolved
//! level changes parity, and the right-to-left pieces are written `w:rtl`, as Word writes text
//! typed in a right-to-left paragraph. Runs that say their direction, and paragraphs without
//! right-to-left characters, are written as they are.
//!
//! Word gives right-to-left text without a bidi language (`w:lang w:bidi`, §17.3.2.20) its
//! default complex-script language, Arabic, so Hebrew would proof and show as Arabic. The pieces
//! marked `w:rtl` therefore also say their language, unless the run or its styles already do.
use std::borrow::Cow;
use std::ops::Range;
use wordcraft_doc::props::CharProps;
use wordcraft_doc::styles::StyleSheet;
/// The language Word expects for right-to-left text of `c`'s script, for the scripts that have
/// one usual language.
fn script_lang(c: char) -> Option<&'static str> {
match u32::from(c) {
0x0590..=0x05FF | 0xFB1D..=0xFB4F => Some("he-IL"),
0x0600..=0x06FF | 0x0750..=0x077F | 0x0870..=0x08FF | 0xFB50..=0xFDFF | 0xFE70..=0xFEFF => Some("ar-SA"),
0x0700..=0x074F => Some("syr-SY"),
0x0780..=0x07BF => Some("dv-MV"),
_ => None,
}
}
/// The language of the first right-to-left letter of `text` with a usual language.
fn text_lang(text: &str) -> Option<&'static str> {
text.chars().find_map(script_lang)
}
/// A language tag of a right-to-left language (`he-IL`, `yi`, `fa-IR`, `ar-EG`…)?
fn is_rtl_lang(tag: &str) -> bool {
let primary = tag.split(['-', '_']).next().unwrap_or("").to_ascii_lowercase();
matches!(primary.as_str(), "he" | "iw" | "yi" | "ar" | "fa" | "ur" | "ps" | "sd" | "ug" | "ckb" | "syr" | "dv" | "ku")
}
/// Does a paragraph of style `para_style` inherit a bidi language from the document defaults or
/// its style chain? (Resolved once per paragraph.)
pub(super) fn para_inherits_lang_bidi(styles: &StyleSheet, para_style: Option<&str>) -> bool {
styles.para_style_props(para_style.unwrap_or("Normal")).1.lang_bidi.is_some()
}
/// Does `c`'s character style chain give it a bidi language?
pub(super) fn char_style_has_lang_bidi(styles: &StyleSheet, c: &CharProps) -> bool {
c.style.as_deref().is_some_and(|s| styles.chain(s).iter().any(|st| st.chr.lang_bidi.is_some()))
}
/// Add `rtl` and, when given, the bidi language `lang` to `c` where it doesn't say them.
fn direct(c: &mut CharProps, lang: Option<&String>) {
c.rtl.get_or_insert(true);
if let Some(l) = lang
&& c.lang_bidi.is_none()
{
c.lang_bidi = Some(l.clone());
}
}
/// `runs` (byte ranges of `text` with their props), split where the bidi level of `text` in a
/// paragraph of direction `rtl_para` changes parity; right-to-left pieces get `rtl: Some(true)`
/// and, unless they have one or `inherits_lang` says it comes from a style, a bidi language.
pub(super) fn direction_spans<'a>(
text: &str,
rtl_para: bool,
runs: impl Iterator<Item = (Range<usize>, &'a CharProps)>,
inherits_lang: impl Fn(&CharProps) -> bool,
) -> Vec<(Range<usize>, Cow<'a, CharProps>)> {
if !wordcraft_doc::bidi::has_rtl(text) {
return runs.map(|(r, p)| (r, Cow::Borrowed(p))).collect();
}
let level = if rtl_para { unicode_bidi::Level::rtl() } else { unicode_bidi::Level::ltr() };
let info = unicode_bidi::ParagraphBidiInfo::new(text, Some(level));
let odd = |at: usize| info.levels.get(at).is_some_and(|l| l.is_rtl());
let para_lang = text_lang(text);
let mut out = Vec::new();
for (range, props) in runs {
let Some(run) = text.get(range.clone()).filter(|_| props.rtl.is_none()) else {
out.push((range, Cow::Borrowed(props)));
continue;
};
let named = props.lang_bidi.is_some() || inherits_lang(props);
let piece = |r: Range<usize>, rtl: bool| {
if !rtl {
return (r, Cow::Borrowed(props));
}
let lang = if named {
None
} else {
let own = props.lang.clone().filter(|l| is_rtl_lang(l));
own.or_else(|| text.get(r.clone()).and_then(text_lang).or(para_lang).map(str::to_string))
};
let mut p = props.clone();
direct(&mut p, lang.as_ref());
// A tracked formatting change's "before" copy gets the same, so the change lists
// nothing the author didn't do.
if let Some(ch) = p.fmt_change.as_mut() {
direct(&mut ch.old, lang.as_ref());
}
(r, Cow::Owned(p))
};
let mut start = range.start;
let mut cur: Option<bool> = None;
for (i, _) in run.char_indices() {
let at = range.start.saturating_add(i);
let rtl = odd(at);
match cur {
Some(c) if c != rtl => {
out.push(piece(start..at, c));
start = at;
}
_ => {}
}
cur = Some(rtl);
}
out.push(piece(start..range.end, cur.unwrap_or(false)));
}
out
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn hostile_ranges_never_panic() {
let p = CharProps::default();
let text = "עברית WordCraft";
let runs = vec![(0..3, &p), (3..1000, &p), (usize::MAX - 1..usize::MAX, &p), (2..text.len(), &p)];
let out = direction_spans(text, true, runs.into_iter(), |_| false);
assert_eq!(out.len(), 5, "bad ranges pass through unsplit: {out:?}");
}
#[test]
fn languages_by_script() {
assert_eq!(text_lang("(שלום"), Some("he-IL"));
assert_eq!(text_lang("سلام"), Some("ar-SA"));
assert_eq!(text_lang("ܫܠܡܐ"), Some("syr-SY"));
assert_eq!(text_lang(". ,"), None);
assert!(is_rtl_lang("he-IL") && is_rtl_lang("yi") && is_rtl_lang("FA_ir") && !is_rtl_lang("en-US") && !is_rtl_lang(""));
}
}
+46 -2
View File
@@ -46,6 +46,8 @@ pub(crate) struct EmbedWriter<'d> {
markup_bytes: u64,
/// Parts to write: (name, bytes, content type, relationships).
pub parts: Vec<(String, Vec<u8>, String, PartRels)>,
/// Leave out the main document's kept glossary (it's written from the model instead).
pub drop_glossary: bool,
}
impl<'d> EmbedWriter<'d> {
@@ -90,6 +92,7 @@ impl<'d> EmbedWriter<'d> {
copy_bytes: MAX_COPY_BYTES,
markup_bytes: MAX_MARKUP_BYTES,
parts: Vec::new(),
drop_glossary: false,
}
}
@@ -142,7 +145,7 @@ impl<'d> EmbedWriter<'d> {
}
/// A number `n` for which every `{stem}{n}{ext}` is a free part name, with those names taken.
fn fresh_set(&mut self, stems: &[&str], ext: &str) -> Vec<String> {
pub(super) fn fresh_set(&mut self, stems: &[&str], ext: &str) -> Vec<String> {
let mut n: u32 = 1;
loop {
let names: Vec<String> = stems.iter().map(|s| format!("{s}{n}{ext}")).collect();
@@ -205,7 +208,7 @@ impl<'d> EmbedWriter<'d> {
for line in String::from_utf8_lossy(&roots).lines().take(MAX_EMBEDDED_RELS * 2) {
let fields: Vec<&str> = line.split('\t').collect();
let [source, kind, target] = fields.as_slice() else { continue };
if *source != owner {
if *source != owner || (self.drop_glossary && owner == "D" && rel_is(kind, rt::GLOSSARY)) {
continue;
}
if let Some(name) = self.copy_tree(target, &mut memo, &mut written) {
@@ -216,6 +219,15 @@ impl<'d> EmbedWriter<'d> {
self.first = memo;
}
/// Whether `unmodelled_rels("D", …)` can write the main document's kept glossary back.
pub fn kept_glossary(&self) -> bool {
let Some(roots) = self.passthrough.get(crate::package::UNMODELLED_ROOTS) else { return false };
String::from_utf8_lossy(roots).lines().take(MAX_EMBEDDED_RELS * 2).any(|line| {
let fields: Vec<&str> = line.split('\t').collect();
matches!(fields.as_slice(), [source, kind, target] if *source == "D" && rel_is(kind, rt::GLOSSARY) && (self.first.contains_key(*target) || self.available(target)))
})
}
/// What copying the parts reached from `roots` costs against the copy budget.
fn tree_cost(&self, roots: &[&str]) -> u64 {
let mut seen: HashSet<&str> = HashSet::new();
@@ -519,6 +531,38 @@ impl super::Writer<'_> {
Self::ink_transform(c, from, to, float, depth + 1);
}
}
/// Put a group or canvas member kept from a file (native ink and its picture fallback) at
/// `frame` ([x, y, w, h] in points, in the group's space, as the group is written): its
/// `w14:xfrm` and the fallback's `a:xfrm` get that offset and extent and the member's turn.
/// It rewrites every `w14:xfrm` and `a:xfrm` under `e` and doesn't descend into them: right
/// for the forms Word writes (the content part's own transform and one `pic:pic` fallback's
/// `pic:spPr/a:xfrm`); a fallback holding several transforms (a group) would get one frame.
pub(super) fn place_member(e: &mut El, frame: [f32; 4], float: &Float, depth: usize) {
if depth > MAX_DEPTH {
return;
}
if matches!(e.name.as_str(), "w14:xfrm" | "a:xfrm") {
e.attrs.retain(|(k, _)| !matches!(k.as_str(), "rot" | "flipH" | "flipV"));
for (k, v) in super::story::spin_attrs(float.spin()) {
e.attrs.push((k.to_string(), v));
}
let [x, y, w, h] = frame.map(|v| v.clamp(-MAX_LEN_PT, MAX_LEN_PT));
e.kids.retain(|k| !matches!(k, xml::Node::El(c) if matches!(c.name.as_str(), "a:off" | "a:ext")));
let el = |name: &str, attrs: [(&str, String); 2]| {
xml::Node::El(El { name: name.to_string(), attrs: attrs.map(|(k, v)| (k.to_string(), v)).to_vec(), kids: Vec::new() })
};
e.kids.insert(0, el("a:ext", [("cx", emu(w.max(0.0))), ("cy", emu(h.max(0.0)))]));
e.kids.insert(0, el("a:off", [("x", emu(x)), ("y", emu(y))]));
return;
}
for c in e.kids.iter_mut().filter_map(|k| match k {
xml::Node::El(c) => Some(c),
_ => None,
}) {
Self::place_member(c, frame, float, depth + 1);
}
}
}
/// Set an extent's `cx` and `cy` (those it has) to `w` × `h` points.
+95
View File
@@ -0,0 +1,95 @@
//! Writing the glossary document (ECMA-376 Part 1 §17.12) from `Document::glossary`, once its
//! entries differ from the file's (an untouched glossary is written back verbatim: see
//! `read::glossary`).
use wordcraft_doc::{Document, QUICK_PARTS_GALLERY};
use super::{CT_WML, PartRels, Writer, numbering_xml, push_part, styles_xml};
use crate::package::{GLOSSARY_BEHAVIORS, GLOSSARY_CONTENT_TYPE, GLOSSARY_READ, GLOSSARY_TYPES, glossary_fingerprint, rt};
use crate::xml::{self, W};
/// `values` known to `known`, as `parent` with one `tag` child each (nothing when none are).
fn listed(w: &mut W, parent: &str, tag: &str, values: &[String], known: &[&str]) {
let mut values = values.iter().filter(|v| known.contains(&v.as_str())).peekable();
if values.peek().is_none() {
return;
}
w.open(parent, &[]);
for v in values {
w.empty(tag, &[("w:val", v)]);
}
w.close(parent);
}
/// Whether the glossary differs from the one read from the file (none read: an empty one).
pub(super) fn edited(doc: &Document) -> bool {
let print = glossary_fingerprint(&doc.glossary);
match doc.passthrough.get(GLOSSARY_READ) {
Some(read) => read.as_slice() != print.as_bytes(),
None => print != glossary_fingerprint(&[]),
}
}
impl Writer<'_> {
/// Write the glossary part from the model, with a styles part of its own (the document's
/// styles), related from the main document's `rels`. Its parts live in `word/`, beside the
/// main document, so the relationships its stories add resolve as theirs do.
pub(super) fn glossary(&mut self, entries: &mut Vec<(String, Vec<u8>)>, overrides: &mut Vec<(String, String)>, rels: &mut PartRels) {
let doc = self.doc;
if doc.glossary.is_empty() {
return;
}
let names = self.embeds.fresh_set(&["word/glossary", "word/glossaryStyles", "word/glossaryNumbering"], ".xml");
let [part, styles, numbering] = names.as_slice() else { return };
let ns = xml::body_ns();
let ns_refs: Vec<(&str, &str)> = ns.iter().map(|(k, v)| (k.as_str(), v.as_str())).collect();
let mut prels = PartRels::default();
let mut w = W::new();
w.open("w:glossaryDocument", &ns_refs);
w.open("w:docParts", &[]);
for e in &doc.glossary {
w.open("w:docPart", &[]);
w.open("w:docPartPr", &[]);
w.empty("w:name", &[("w:val", &e.name)]);
if !e.style.is_empty() {
w.empty("w:style", &[("w:val", &e.style)]);
}
w.open("w:category", &[]);
w.empty("w:name", &[("w:val", if e.category.is_empty() { "General" } else { &e.category })]);
w.empty("w:gallery", &[("w:val", if e.gallery.is_empty() { QUICK_PARTS_GALLERY } else { &e.gallery })]);
w.close("w:category");
listed(&mut w, "w:types", "w:type", &e.types, &GLOSSARY_TYPES);
listed(&mut w, "w:behaviors", "w:behavior", &e.behaviors, &GLOSSARY_BEHAVIORS);
if !e.description.is_empty() {
w.empty("w:description", &[("w:val", &e.description)]);
}
if !e.guid.is_empty() {
w.empty("w:guid", &[("w:val", &e.guid)]);
}
w.close("w:docPartPr");
w.open("w:docPartBody", &[]);
if e.blocks.is_empty() {
w.empty("w:p", &[]);
} else {
self.blocks(&mut w, &e.blocks, &mut prels, false, 0);
}
w.close("w:docPartBody");
w.close("w:docPart");
}
w.close("w:docParts");
w.close("w:glossaryDocument");
let file = |p: &str| p.strip_prefix("word/").unwrap_or(p).to_string();
prels.add(rt::STYLES, &file(styles), false);
// Lists in entries need their definitions within the glossary's reach.
let lists = !doc.numbering.nums.is_empty() || !doc.numbering.abstracts.is_empty();
if lists {
prels.add(rt::NUMBERING, &file(numbering), false);
}
rels.add(rt::GLOSSARY, &file(part), false);
push_part(entries, overrides, part, w.into_bytes(), GLOSSARY_CONTENT_TYPE, prels);
push_part(entries, overrides, styles, styles_xml(doc), &format!("{CT_WML}styles+xml"), PartRels::default());
if lists {
push_part(entries, overrides, numbering, numbering_xml(doc), &format!("{CT_WML}numbering+xml"), PartRels::default());
}
}
}
+12
View File
@@ -1,6 +1,8 @@
//! [`Document`] → DOCX.
mod bidi;
mod embed;
mod glossary;
mod settings_merge;
pub(crate) use settings_merge::has_unowned as settings_has_unowned;
mod math;
@@ -263,6 +265,13 @@ pub(crate) fn write_parts(doc: &Document, flavor: Flavor) -> Result<Vec<(String,
);
}
// The glossary: kept verbatim while its entries are as read and its parts were kept, else
// written from the model (before the notes: references its entries hold are collected too).
if glossary::edited(doc) || (!doc.glossary.is_empty() && !wr.embeds.kept_glossary()) {
wr.embeds.drop_glossary = true;
wr.glossary(&mut entries, &mut overrides, &mut rels);
}
// Notes (references were collected from the body and headers).
for foot in [true, false] {
let seps = if foot { &doc.footnote_separators } else { &doc.endnote_separators };
@@ -997,6 +1006,9 @@ fn settings_xml(doc: &Document, footnotes: bool, endnotes: bool, template_rel: O
if s.track_changes {
w.empty("w:trackRevisions", &[]);
}
if !s.track_moves {
w.empty("w:doNotTrackMoves", &[]);
}
if let Some(p) = &s.protection {
let mut a: Vec<(&str, &str)> = Vec::new();
if p != "none" {
+71 -22
View File
@@ -1,8 +1,8 @@
//! Property writers, in ECMA-376 schema element order.
use wordcraft_doc::props::{
Align, Border, BorderStyle, Borders, CellProps, CharProps, HeightRule, Highlight, LineSpacing, ParaProps, RowProps, TabAlign, TabLeader,
TableFloat, TableLook, TableProps, TextColor, VAlign, VMerge, VertAlign,
Align, Border, BorderStyle, Borders, CellProps, CharProps, EastAsianLayout, HeightRule, Highlight, LineSpacing, ParaProps, RowProps, TabAlign,
TabLeader, TableFloat, TableLook, TableProps, TextColor, VAlign, VMerge, VertAlign,
};
use crate::units::{n, twips};
@@ -81,11 +81,12 @@ pub fn rpr_inner(w: &mut W, c: &CharProps) {
if let Some(h) = c.highlight {
w.val("w:highlight", if h == Highlight::None { "none" } else { h.ooxml() });
}
if let Some(u) = c.underline {
match c.underline_color {
Some(col) => w.empty("w:u", &[("w:val", u.ooxml()), ("w:color", &col.hex())]),
None => w.val("w:u", u.ooxml()),
}
match (c.underline, c.underline_color) {
(Some(u), Some(col)) => w.empty("w:u", &[("w:val", u.ooxml()), ("w:color", &col.hex())]),
(Some(u), None) => w.val("w:u", u.ooxml()),
// A colour-only w:u, as read: keep it without inventing an underline
(None, Some(col)) => w.empty("w:u", &[("w:color", &col.hex())]),
(None, None) => {}
}
if let Some(b) = &c.border {
border_el(w, "w:bdr", b);
@@ -111,9 +112,39 @@ pub fn rpr_inner(w: &mut W, c: &CharProps) {
(None, Some(b)) => w.empty("w:lang", &[("w:bidi", b)]),
(None, None) => {}
}
// After w:lang (CT_RPr's sequence).
if let Some(l) = &c.east_asian_layout {
east_asian_layout(w, l);
}
if let Some(fx) = &c.text_effects {
super::wordart::rpr_effects(w, fx);
}
if let Some(l) = c.ligatures {
w.empty("w14:ligatures", &[("w14:val", &l.ooxml())]);
}
}
/// `w:eastAsianLayout` with the attributes that are set (ISO/IEC 29500-1 §17.3.2.10).
fn east_asian_layout(w: &mut W, l: &EastAsianLayout) {
let on = |v: bool| if v { "1" } else { "0" };
let id = l.id.map(|v| v.to_string());
let mut a: Vec<(&str, &str)> = Vec::new();
if let Some(id) = &id {
a.push(("w:id", id));
}
if let Some(v) = l.combine {
a.push(("w:combine", on(v)));
}
if let Some(b) = l.combine_brackets {
a.push(("w:combineBrackets", b.ooxml()));
}
if let Some(v) = l.vert {
a.push(("w:vert", on(v)));
}
if let Some(v) = l.vert_compress {
a.push(("w:vertCompress", on(v)));
}
w.empty("w:eastAsianLayout", &a);
}
/// Does `c` produce any `w:rPr` content (besides a tracked change)?
@@ -157,6 +188,17 @@ pub fn align_val(a: Align) -> &'static str {
}
}
/// A tab stop's or positional tab's `w:leader` value.
pub fn tab_leader(l: TabLeader) -> &'static str {
match l {
TabLeader::None => "none",
TabLeader::Dot => "dot",
TabLeader::Hyphen => "hyphen",
TabLeader::Underscore => "underscore",
TabLeader::MiddleDot => "middleDot",
}
}
fn border_el(w: &mut W, name: &str, b: &Border) {
let sz = n(round(b.width.clamp(0.0, 12.0) * 8.0));
let space = n(round(b.space.clamp(0.0, 31.0)));
@@ -205,10 +247,17 @@ pub fn ppr_inner(w: &mut W, p: &ParaProps, framed: bool, num_change: Option<&Cha
toggle(w, "w:keepLines", p.keep_lines);
toggle(w, "w:pageBreakBefore", p.page_break_before);
if framed && let Some(lines) = p.drop_cap.filter(|l| *l > 0) {
w.empty(
"w:framePr",
&[("w:dropCap", "drop"), ("w:lines", &n(lines.min(10) as i64)), ("w:wrap", "around"), ("w:vAnchor", "text"), ("w:hAnchor", "text")],
);
// ECMA-376 §17.3.1.11: the drop cap's frame — dropped or in the margin, the lines it
// spans and its distance from the text (`w:hSpace`, twips).
let kind = if p.drop_cap_margin == Some(true) { "margin" } else { "drop" };
let lines = n(lines.min(10) as i64);
let space = p.drop_cap_distance.filter(|d| d.is_finite() && *d > 0.0).map(|d| twips(d.min(wordcraft_doc::resolve::MAX_DROP_CAP_DISTANCE)));
let mut a: Vec<(&str, &str)> = vec![("w:dropCap", kind), ("w:lines", &lines)];
if let Some(h) = &space {
a.push(("w:hSpace", h));
}
a.extend([("w:wrap", "around"), ("w:vAnchor", "text"), ("w:hAnchor", "text")]);
w.empty("w:framePr", &a);
} else if let Some(f) = &p.frame {
let (fw, fh, fx, fy) = (f.w.map(twips), f.h.map(twips), twips(f.x), twips(f.y));
let (hs, vs) = (twips(f.h_space), twips(f.v_space));
@@ -273,15 +322,7 @@ pub fn ppr_inner(w: &mut W, p: &ParaProps, framed: bool, num_change: Option<&Cha
let pos = twips(t.pos);
match t.leader {
TabLeader::None => w.empty("w:tab", &[("w:val", val), ("w:pos", &pos)]),
l => {
let lv = match l {
TabLeader::Dot => "dot",
TabLeader::Hyphen => "hyphen",
TabLeader::Underscore => "underscore",
_ => "middleDot",
};
w.empty("w:tab", &[("w:val", val), ("w:leader", lv), ("w:pos", &pos)]);
}
l => w.empty("w:tab", &[("w:val", val), ("w:leader", tab_leader(l)), ("w:pos", &pos)]),
}
}
w.close("w:tabs");
@@ -367,9 +408,15 @@ pub fn ppr_inner(w: &mut W, p: &ParaProps, framed: bool, num_change: Option<&Cha
}
pub fn margins(w: &mut W, tag: &str, m: &[f32; 4]) {
margin_sides(w, tag, &m.map(Some));
}
fn margin_sides(w: &mut W, tag: &str, m: &[Option<f32>; 4]) {
w.open(tag, &[]);
for (name, v) in ["w:top", "w:left", "w:bottom", "w:right"].iter().zip(m.iter()) {
w.empty(name, &[("w:w", &twips(v.clamp(0.0, 1584.0))), ("w:type", "dxa")]);
for (name, value) in ["w:top", "w:left", "w:bottom", "w:right"].iter().zip(m.iter()) {
if let Some(v) = value {
w.empty(name, &[("w:w", &twips(v.clamp(0.0, 1584.0))), ("w:type", "dxa")]);
}
}
w.close(tag);
}
@@ -573,6 +620,8 @@ fn tcpr_inner(w: &mut W, c: &CellProps) {
}
if let Some(m) = &c.margins {
margins(w, "w:tcMar", m);
} else if let Some(m) = &c.partial_margins {
margin_sides(w, "w:tcMar", m);
}
if c.text_direction.is_turned() || c.text_direction_set {
w.val("w:textDirection", c.text_direction.ooxml());
+18 -3
View File
@@ -21,6 +21,7 @@ const OWNED: &[&str] = &[
"w:attachedTemplate",
"w:linkStyles",
"w:trackRevisions",
"w:doNotTrackMoves",
"w:documentProtection",
"w:defaultTabStop",
"w:autoHyphenation",
@@ -222,9 +223,15 @@ fn merge_compat(orig: &str, generated: &str) -> Option<String> {
match (mode(orig), mode(generated)) {
(Some((a, b)), Some((c, d))) => Some(format!("{}{}{}", orig.get(..a)?, generated.get(c..d)?, orig.get(b..)?)),
(None, Some((c, d))) => {
// No mode in the original: add the generated one before `</w:compat>`.
let close = orig.rfind("</w:compat>")?;
Some(format!("{}{}{}", orig.get(..close)?, generated.get(c..d)?, orig.get(close..)?))
// No mode in the original: add the generated one before `</w:compat>`, or expand a self-closing `<w:compat/>`.
if let Some(close) = orig.rfind("</w:compat>") {
Some(format!("{}{}{}", orig.get(..close)?, generated.get(c..d)?, orig.get(close..)?))
} else if let Some(slash) = orig.rfind("/>") {
let open = orig.get(..slash)?.trim_end();
Some(format!("{open}>{}</w:compat>", generated.get(c..d)?))
} else {
None
}
}
_ => Some(orig.to_string()),
}
@@ -344,4 +351,12 @@ mod tests {
assert!(has_unowned(orig.as_bytes()));
assert!(!has_unowned(GEN.as_bytes()));
}
#[test]
fn a_self_closing_compat_block_gets_the_generated_mode() {
let orig = r#"<w:settings xmlns:w="w" xmlns:m="m" xmlns:r="r"><w:compat/><w:rsids><w:rsidRoot w:val="00112233"/></w:rsids></w:settings>"#;
let out = String::from_utf8(merge(orig.as_bytes(), GEN.as_bytes()).unwrap()).unwrap();
assert!(out.contains("<w:compat><w:compatSetting") && out.contains("</w:compat>"), "{out}");
assert!(out.contains("<w:rsidRoot w:val=\"00112233\"/>"), "{out}");
}
}
+115 -3
View File
@@ -21,6 +21,8 @@ enum Piece<'a> {
Br(Option<&'static str>),
NbHyphen,
SoftHyphen,
/// A symbol character in its own font (`w:sym`, ECMA-376 §17.3.3.30).
Sym(&'a str, char),
/// An object and its byte offset in the paragraph.
Obj(&'a InlineObject, usize),
}
@@ -137,6 +139,12 @@ impl Writer<'_> {
let tail = head.split_off(first.len_utf8()).ok();
if let Some(mut tail) = tail {
head.props.drop_cap = Some(lines);
// The letter's own font goes on its run (`w:rFonts`).
if let Some(f) = p.props.drop_cap_font.as_ref().filter(|f| !f.trim().is_empty()) {
for r in &mut head.runs {
r.props.font = Some(f.clone());
}
}
head.section = None;
self.para_inner(w, &head, rels, false, depth, true);
tail.props.drop_cap = None;
@@ -247,9 +255,19 @@ impl Writer<'_> {
// Split into pieces with their props.
let mut pieces: Vec<(Piece, &CharProps)> = Vec::new();
let mut k = 0usize;
for (range, props) in p.run_ranges() {
// Runs of mixed direction are split so Word orders them as WordCraft does (#1543).
let rtl_para = self.doc.styles.resolve_para(&p.props).bidi;
let styles = &self.doc.styles;
let para_style = p.props.style.as_deref();
let para_lang = super::bidi::para_inherits_lang_bidi(styles, para_style);
let spans =
super::bidi::direction_spans(&p.text, rtl_para, p.run_ranges(), |c| para_lang || super::bidi::char_style_has_lang_bidi(styles, c));
for (range, props) in &spans {
let props: &CharProps = props;
let range = range.clone();
let base = range.start;
let Some(text) = p.text.get(range) else { continue };
let sym_font = props.sym_font.as_deref().filter(|f| !f.trim().is_empty());
let mut buf = String::new();
for (ci, c) in text.char_indices() {
let special = match c {
@@ -267,6 +285,9 @@ impl Writer<'_> {
None => Piece::Text(String::new()),
})
}
// `w:char` is a two-byte code (ST_ShortHexNumber) from U+0020, as the reader
// takes it; anything else stays text.
c if (0x20..=0xFFFF).contains(&u32::from(c)) => sym_font.map(|f| Piece::Sym(f, c)),
_ => None,
};
match special {
@@ -349,6 +370,7 @@ impl Writer<'_> {
Piece::Br(Some(t)) => w.empty("w:br", &[("w:type", t)]),
Piece::NbHyphen => w.empty("w:noBreakHyphen", &[]),
Piece::SoftHyphen => w.empty("w:softHyphen", &[]),
Piece::Sym(font, c) => w.empty("w:sym", &[("w:font", font), ("w:char", &format!("{:04X}", u32::from(*c)))]),
_ => {}
}
}
@@ -520,7 +542,12 @@ impl Writer<'_> {
self.rev_open(w, props);
w.open("w:r", &[]);
self.rpr(w, &p);
w.empty(mark, &[]);
// A custom mark is plain text; the placeholder would turn it into a number.
if custom.is_empty() {
w.empty(mark, &[]);
} else {
w.leaf("w:t", &[("xml:space", "preserve")], custom);
}
w.close("w:r");
self.rev_close(w, props);
}
@@ -551,6 +578,12 @@ impl Writer<'_> {
let root = match (graphic.chart.as_deref(), graphic.smart_art.as_deref(), graphic.source.as_deref()) {
(Some(spec), _, _) => crate::xml::parse(self.chart_graphic(spec, rels).as_bytes()).ok(),
(None, Some(spec), _) => crate::xml::parse(self.smart_art_graphic(spec, *gw, *gh, rels).as_bytes()).ok(),
// Ink in a group member's form (no drawing of its own) can't stand alone: left
// out before its parts are queued, so no orphan part or relationship is written.
(None, None, Some(src)) if ink && !standalone_ink(src) => {
log::warn!("docx: group-member ink outside its group can't be written; it is left out");
None
}
(None, None, Some(src)) => self.embedded_tree(src, rels, ink.then_some((*gw, *gh, float, alt.as_str()))),
(None, None, None) => None,
};
@@ -728,6 +761,15 @@ impl Writer<'_> {
let id = self.shape_docpr(extra.id);
self.wsp(w, &c.obj, Some(&id), at, rels, depth);
}
// Native ink kept from the file, with its fallback, at the member's place.
InlineObject::Graphic { w: iw, h: ih, float, graphic, .. } if graphic.kind == wordcraft_doc::graphic::GraphicKind::Ink => {
if let Some(src) = graphic.source.as_deref()
&& let Some(mut root) = self.embedded_tree(src, rels, None)
{
Self::place_member(&mut root, [c.x, c.y, *iw, *ih], float, 0);
w.raw(&root.to_xml());
}
}
_ => {}
}
}
@@ -780,6 +822,17 @@ impl Writer<'_> {
w.close("w:r");
self.rev_close(w, props);
}
InlineObject::PositionalTab { align, relative_to, leader } => {
self.rev_open(w, props);
w.open("w:r", &[]);
self.rpr(w, props);
w.empty(
"w:ptab",
&[("w:relativeTo", relative_to.ooxml()), ("w:alignment", align.ooxml()), ("w:leader", super::props::tab_leader(*leader))],
);
w.close("w:r");
self.rev_close(w, props);
}
InlineObject::Opaque { format, xml, text } => {
if format == "docx" && crate::xml::parse(xml.as_bytes()).is_ok() {
w.raw(xml);
@@ -1043,6 +1096,7 @@ impl Writer<'_> {
w.empty("a:srgbClr", &[("val", &c.hex())]);
}
w.close("a:solidFill");
line_dash(w, &extra.dash);
if ink {
w.empty("a:round", &[]);
}
@@ -1052,6 +1106,7 @@ impl Writer<'_> {
None => {
w.open("a:ln", &[]);
w.empty("a:noFill", &[]);
line_dash(w, &extra.dash);
w.close("a:ln");
}
}
@@ -1142,7 +1197,7 @@ impl Writer<'_> {
("relativeHeight", &self.z.to_string()),
("behindDoc", behind),
("locked", "0"),
("layoutInCell", "1"),
("layoutInCell", if float.outside_cell { "0" } else { "1" }),
("allowOverlap", "1"),
],
);
@@ -1488,6 +1543,28 @@ fn cust_geom(w: &mut W, f: &wordcraft_doc::freeform::Freeform, sw: f32, sh: f32)
w.close("a:custGeom");
}
/// Explicit DrawingML outline dashes, before joins and arrowheads.
fn line_dash(w: &mut W, dash: &wordcraft_doc::line_dash::LineDash) {
use wordcraft_doc::line_dash::{LineDash, MAX_DASH_STOPS};
if dash.is_solid() {
return;
}
if let Some(preset) = dash.ooxml() {
w.empty("a:prstDash", &[("val", preset)]);
return;
}
if let LineDash::Custom(stops) = dash {
if stops.is_empty() {
return;
}
w.open("a:custDash", &[]);
for (d, sp) in stops.iter().take(MAX_DASH_STOPS) {
w.empty("a:ds", &[("d", &d.to_string()), ("sp", &sp.to_string())]);
}
w.close("a:custDash");
}
}
/// A shape's `a:effectLst` (ECMA-376 Part 1 §20.1.8.26), children in schema order.
/// A line's arrowheads (`a:headEnd`, `a:tailEnd`, ECMA-376 §20.1.8.38 / §20.1.8.57).
fn arrow_ends(w: &mut W, extra: &wordcraft_doc::connector::ShapeExtra) {
@@ -1696,3 +1773,38 @@ fn comment_reference_style(doc: &wordcraft_doc::Document) -> &str {
.find(|s| s.kind == wordcraft_doc::styles::StyleKind::Character && s.name.eq_ignore_ascii_case("annotation reference"))
.map_or("CommentReference", |s| s.id.as_str())
}
/// Is kept ink `src` in a form the drawing-level writer writes: its `a:graphic`, a direct run-level
/// `w:contentPart`, or an `mc:AlternateContent` whose choice holds a drawing (not a group member's
/// bare `w14:contentPart`, nor one in `mc:AlternateContent` with a `pic:pic` fallback)?
fn standalone_ink(src: &wordcraft_doc::graphic::Embedded) -> bool {
match crate::xml::parse(src.xml.as_bytes()) {
Ok(root) => match root.name.as_str() {
"a:graphic" | "w:contentPart" => true,
"mc:AlternateContent" => root.child("mc:Choice").and_then(|c| c.child("w:drawing")).is_some_and(|d| d.find("a:graphic").is_some()),
_ => false,
},
Err(_) => false,
}
}
#[cfg(test)]
mod standalone_ink_tests {
use super::standalone_ink;
use wordcraft_doc::graphic::Embedded;
fn emb(xml: &str) -> Embedded {
Embedded { xml: xml.into(), ..Default::default() }
}
/// Review: an alternate whose choice has a drawing but no `a:graphic` is not writable as ink.
#[test]
fn alternate_content_needs_a_graphic() {
let ns = r#"xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006" xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main""#;
let with =
format!("<mc:AlternateContent {ns}><mc:Choice Requires=\"w14\"><w:drawing><a:graphic/></w:drawing></mc:Choice></mc:AlternateContent>");
let without = format!("<mc:AlternateContent {ns}><mc:Choice Requires=\"w14\"><w:drawing/></mc:Choice></mc:AlternateContent>");
assert!(standalone_ink(&emb(&with)));
assert!(!standalone_ink(&emb(&without)));
}
}
+194
View File
@@ -0,0 +1,194 @@
//! Runs of mixed right-to-left and left-to-right text (#1543). WordCraft lays a paragraph out with
//! the Unicode Bidirectional Algorithm (UAX #9) from its text and direction alone, while Word
//! orders a run's characters by the run's `w:rtl` (ECMA-376 Part 1 §17.3.2.30): text in a run
//! without it is laid out left to right, so Hebrew words around a Latin name, and the punctuation
//! between them, change places. The writer therefore splits such a run where the resolved
//! direction changes and marks the right-to-left pieces `w:rtl`, as Word does with text typed
//! in a right-to-left paragraph.
use std::io::Read;
use wordcraft_doc::props::CharProps;
use wordcraft_doc::{Document, Paragraph, para_block};
/// `word/document.xml` of a written package.
fn document_xml(bytes: &[u8]) -> String {
let mut z = zip::ZipArchive::new(std::io::Cursor::new(bytes)).unwrap();
let mut s = String::new();
z.by_name("word/document.xml").unwrap().read_to_string(&mut s).unwrap();
s
}
/// The runs of the first paragraph of `xml`: (has `w:rtl`, text with `w:br` as `\n`).
fn runs(xml: &str) -> Vec<(bool, String)> {
let body = xml.split("<w:body>").nth(1).unwrap();
let para = body.split("</w:p>").next().unwrap();
para.split("<w:r>")
.skip(1)
.map(|r| {
let r = r.split("</w:r>").next().unwrap();
let rtl = r.contains("<w:rtl/>");
let mut text = String::new();
for part in r.split('<').skip(1) {
if part.starts_with("w:br/>") {
text.push('\n');
} else if part.starts_with("w:t ") || part.starts_with("w:t>") {
text.push_str(part.split_once('>').unwrap().1);
}
}
(rtl, text)
})
.collect()
}
fn arial() -> CharProps {
CharProps { font: Some("Arial".into()), font_cs: Some("Arial".into()), size: Some(16.0), size_cs: Some(16.0), ..Default::default() }
}
fn doc_of(text: &str, rtl_para: bool, props: CharProps) -> Document {
let mut p = Paragraph::with_text(text, props);
if rtl_para {
p.props.bidi = Some(true);
}
let mut d = Document::new();
d.body = vec![para_block(p)];
d
}
fn written_runs(d: &Document) -> Vec<(bool, String)> {
runs(&document_xml(&wordcraft_docx::write(d).unwrap()))
}
fn owned(v: &[(bool, &str)]) -> Vec<(bool, String)> {
v.iter().map(|(r, t)| (*r, (*t).to_string())).collect()
}
#[test]
fn rtl_paragraph_splits_latin_out_of_rtl_runs() {
let d = doc_of("עברית WordCraft 2026 ואחריה עוד עברית.", true, arial());
assert_eq!(written_runs(&d), owned(&[(true, "עברית "), (false, "WordCraft 2026"), (true, " ואחריה עוד עברית.")]));
}
#[test]
fn rtl_paragraph_brackets_email_and_final_stop_take_the_paragraph_direction() {
let d = doc_of("לפני (ABC 123) אחרי; דוא״ל [email protected].", true, arial());
assert_eq!(written_runs(&d), owned(&[(true, "לפני ("), (false, "ABC 123"), (true, ") אחרי; דוא״ל "), (false, "[email protected]"), (true, ".")]));
}
#[test]
fn line_breaks_stay_in_the_split_runs() {
let d = doc_of("שלום עולם.\nעברית WordCraft 2026 ואחריה.", true, arial());
assert_eq!(written_runs(&d), owned(&[(true, "שלום עולם.\nעברית "), (false, "WordCraft 2026"), (true, " ואחריה.")]));
}
#[test]
fn split_runs_keep_their_formatting() {
let d = doc_of("עברית WordCraft עברית", true, arial());
let xml = document_xml(&wordcraft_docx::write(&d).unwrap());
let body = xml.split("<w:body>").nth(1).unwrap();
assert_eq!(body.matches("<w:r>").count(), 3);
assert_eq!(body.matches(r#"<w:rFonts w:ascii="Arial" w:hAnsi="Arial" w:cs="Arial"/>"#).count(), 4, "three runs and the mark: {body}");
assert_eq!(body.matches(r#"<w:szCs w:val="32"/>"#).count(), 4, "{body}");
}
#[test]
fn hebrew_inside_an_ltr_paragraph_is_marked_right_to_left() {
let d = doc_of("English with עברית מעורבת inside.", false, CharProps::default());
assert_eq!(written_runs(&d), owned(&[(false, "English with "), (true, "עברית מעורבת"), (false, " inside.")]));
}
#[test]
fn text_without_rtl_characters_is_one_run() {
let d = doc_of("Plain English, 2026.", true, arial());
assert_eq!(written_runs(&d), owned(&[(false, "Plain English, 2026.")]));
let d = doc_of("Plain English, 2026.", false, arial());
assert_eq!(written_runs(&d), owned(&[(false, "Plain English, 2026.")]));
}
#[test]
fn explicit_run_direction_is_kept() {
// A run that already says its direction is written as it is, unsplit.
let d = doc_of("עברית WordCraft", true, CharProps { rtl: Some(true), ..arial() });
assert_eq!(written_runs(&d), owned(&[(true, "עברית WordCraft")]));
let d = doc_of("עברית WordCraft", true, CharProps { rtl: Some(false), ..arial() });
let xml = document_xml(&wordcraft_docx::write(&d).unwrap());
assert!(xml.contains(r#"<w:rtl w:val="0"/>"#), "{xml}");
assert_eq!(runs(&xml).len(), 1);
}
#[test]
fn mixed_paragraph_round_trips_text_and_direction() {
let text = "לפני (ABC 123) אחרי; דוא״ל [email protected].\nעברית WordCraft 2026 ואחריה עוד עברית.";
let d = doc_of(text, true, arial());
let bytes = wordcraft_docx::write(&d).unwrap();
let r = wordcraft_docx::read(&bytes).unwrap();
let p = r.body.iter().find_map(|b| b.as_para()).unwrap();
assert_eq!(p.text, text);
assert_eq!(p.props.bidi, Some(true));
for run in &p.runs {
assert_eq!(run.props.font.as_deref(), Some("Arial"));
assert_eq!(run.props.size, Some(16.0));
}
assert!(p.runs.iter().any(|r| r.props.rtl == Some(true)));
// Saving again writes the same runs: the read-back direction is now explicit.
assert_eq!(written_runs(&r), written_runs(&d));
}
// Word gives right-to-left text without a `w:lang w:bidi` its default complex-script language
// (Arabic), so its status bar, proofing and hyphenation call Hebrew "Arabic". Right-to-left
// pieces the writer marks say their language (§17.3.2.20), unless the run or a style already does.
/// `w:bidi` attribute of each run's `w:lang` in the first paragraph: (has `w:rtl`, bidi language).
fn run_langs(xml: &str) -> Vec<(bool, Option<String>)> {
let body = xml.split("<w:body>").nth(1).unwrap();
let para = body.split("</w:p>").next().unwrap();
para.split("<w:r>")
.skip(1)
.map(|r| {
let r = r.split("</w:r>").next().unwrap();
let lang = r.split("<w:lang ").nth(1).and_then(|l| l.split("w:bidi=\"").nth(1)).map(|v| v.split('"').next().unwrap().to_string());
(r.contains("<w:rtl/>"), lang)
})
.collect()
}
fn langs(d: &Document) -> Vec<(bool, Option<String>)> {
run_langs(&document_xml(&wordcraft_docx::write(d).unwrap()))
}
fn he(rtl: bool) -> (bool, Option<String>) {
(rtl, rtl.then(|| "he-IL".to_string()))
}
#[test]
fn hebrew_pieces_say_they_are_hebrew() {
let d = doc_of("לפני (ABC 123) אחרי; דוא״ל [email protected].", true, arial());
// The final stop has no letters of its own; it takes the paragraph's language.
assert_eq!(langs(&d), vec![he(true), he(false), he(true), he(false), he(true)]);
let d = doc_of("English with עברית מעורבת inside.", false, CharProps::default());
assert_eq!(langs(&d), vec![he(false), he(true), he(false)]);
}
#[test]
fn arabic_pieces_say_they_are_arabic() {
let d = doc_of("مرحبا WordCraft بالعالم", true, arial());
let ar = Some("ar-SA".to_string());
assert_eq!(langs(&d), vec![(true, ar.clone()), (false, None), (true, ar)]);
}
#[test]
fn a_language_the_run_already_has_is_kept() {
let d = doc_of("سلام WordCraft دنیا", true, CharProps { lang_bidi: Some("fa-IR".into()), ..arial() });
let fa = Some("fa-IR".to_string());
assert_eq!(langs(&d), vec![(true, fa.clone()), (false, fa.clone()), (true, fa)]);
// A right-to-left proofing language on the run (w:val) names the right-to-left text too.
let d = doc_of("שלום WordCraft", true, CharProps { lang: Some("yi".into()), ..arial() });
assert_eq!(langs(&d), vec![(true, Some("yi".into())), (false, None)]);
}
#[test]
fn an_inherited_bidi_language_is_not_repeated() {
let mut d = doc_of("سلام WordCraft دنیا", true, arial());
d.styles.default_chr.lang_bidi = Some("fa-IR".into());
assert_eq!(langs(&d), vec![(true, None), (false, None), (true, None)]);
}
@@ -0,0 +1,30 @@
//! Original partial-cell-margin fixture (ECMA-376 §17.4.69).
use std::io::{Cursor, Read, Write};
fn fixture() -> Vec<u8> {
let document = r#"<w:document xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:body><w:tbl><w:tblPr><w:tblCellMar><w:top w:w="55" w:type="dxa"/><w:left w:w="45" w:type="dxa"/><w:bottom w:w="55" w:type="dxa"/><w:right w:w="55" w:type="dxa"/></w:tblCellMar></w:tblPr><w:tblGrid><w:gridCol w:w="4320"/></w:tblGrid><w:tr><w:tc><w:tcPr><w:tcMar><w:left w:w="0" w:type="dxa"/></w:tcMar></w:tcPr><w:p><w:r><w:t>Point</w:t></w:r></w:p></w:tc></w:tr></w:tbl></w:body></w:document>"#;
let mut zip = zip::ZipWriter::new(Cursor::new(Vec::new()));
zip.start_file("word/document.xml", zip::write::SimpleFileOptions::default()).unwrap();
zip.write_all(document.as_bytes()).unwrap();
zip.finish().unwrap().into_inner()
}
#[test]
fn partial_margin_stays_partial() {
let doc = wordcraft_docx::read(&fixture()).unwrap();
let table = doc.body[0].as_table().unwrap();
assert_eq!(table.rows[0].cells[0].props.margins, None, "a left-only margin is not a complete override");
assert_eq!(table.rows[0].cells[0].props.resolved_margins([6.0, 7.0, 8.0, 9.0]), [6.0, 0.0, 8.0, 9.0]);
let bytes = wordcraft_docx::write(&doc).unwrap();
let reopened = wordcraft_docx::read(&bytes).unwrap();
let cell = &reopened.body[0].as_table().unwrap().rows[0].cells[0];
assert_eq!(cell.props.resolved_margins([6.0, 7.0, 8.0, 9.0]), [6.0, 0.0, 8.0, 9.0]);
let mut zip = zip::ZipArchive::new(Cursor::new(bytes)).unwrap();
let mut xml = String::new();
zip.by_name("word/document.xml").unwrap().read_to_string(&mut xml).unwrap();
let margins = xml.split("<w:tcMar>").nth(1).unwrap().split("</w:tcMar>").next().unwrap();
assert!(margins.contains("<w:left w:w=\"0\""));
assert!(!margins.contains("<w:top"));
assert!(!margins.contains("<w:bottom"));
assert!(!margins.contains("<w:right"));
}
@@ -2,7 +2,8 @@
//! Synthetic packages, written here; no external documents or installed fonts are needed.
use std::io::Write;
use wordcraft_doc::{Document, Paragraph};
use wordcraft_doc::props::CharProps;
use wordcraft_doc::{Document, Paragraph, Run};
fn package(run: &str, defaults: &str, styles: &str, paragraph_style: &str) -> Vec<u8> {
let ns = r#"xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main""#;
@@ -29,13 +30,41 @@ fn paragraph(d: &Document) -> &Paragraph {
d.body.iter().find_map(|b| b.as_para()).unwrap()
}
/// `runs` without run direction and bidi language (in the formatting change too), adjacent equal
/// runs merged.
fn undirected(runs: &[Run]) -> Vec<Run> {
let strip = |c: &mut CharProps| {
c.rtl = None;
c.lang_bidi = None;
};
let mut out: Vec<Run> = Vec::new();
for r in runs {
let mut props = r.props.clone();
strip(&mut props);
if let Some(ch) = props.fmt_change.as_mut() {
strip(&mut ch.old);
}
match out.last_mut() {
Some(last) if last.props == props => last.len += r.len,
_ => out.push(Run { len: r.len, props }),
}
}
out
}
fn check_generations(bytes: &[u8], check: impl Fn(&Document)) {
let mut d = wordcraft_docx::read(bytes).unwrap();
for _ in 0..3 {
for generation in 0..3 {
check(&d);
let next = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
assert_eq!(paragraph(&next).text, paragraph(&d).text);
assert_eq!(paragraph(&next).runs, paragraph(&d).runs);
if generation == 0 {
// The first save splits the mixed Hebrew/Latin run by direction and names the
// Hebrew's language (#1543); apart from that nothing changes.
assert_eq!(undirected(&paragraph(&next).runs), undirected(&paragraph(&d).runs));
} else {
assert_eq!(paragraph(&next).runs, paragraph(&d).runs);
}
assert_eq!(paragraph(&next).props.bidi, Some(true));
d = next;
}
@@ -133,3 +162,39 @@ fn tracked_formatting_keeps_independent_complex_script_values() {
assert_eq!(overlaid.fmt_change, c.fmt_change);
});
}
/// `(w:rtl, w:lang)` elements of a run's current `w:rPr` and of its `w:rPrChange` "before" copy.
fn direction_markup(run: &str) -> (String, String) {
let pick = |rpr: &str| -> String {
rpr.split('<').filter(|t| t.starts_with("w:rtl") || t.starts_with("w:lang")).map(|t| format!("<{t}")).collect::<Vec<_>>().join("")
};
let (now, change) = run.split_once("<w:rPrChange").unwrap_or((run, ""));
(pick(now), pick(change))
}
#[test]
fn split_runs_add_no_phantom_tracked_formatting() {
// #1543: direction and language the writer adds to a piece of a run that has a tracked
// formatting change go into the change's "before" copy too, so Word lists no change the
// author never made ("Formatted: Right-to-left, Hebrew").
let run = r#"<w:sz w:val="24"/><w:rPrChange w:id="7" w:author="Synthetic Author" w:date="2026-10-11T00:00:00Z"><w:rPr><w:sz w:val="20"/></w:rPr></w:rPrChange>"#;
let d = wordcraft_docx::read(&package(run, "", "", "")).unwrap();
let bytes = wordcraft_docx::write(&d).unwrap();
let mut xml = String::new();
std::io::Read::read_to_string(&mut zip::ZipArchive::new(std::io::Cursor::new(&bytes)).unwrap().by_name("word/document.xml").unwrap(), &mut xml)
.unwrap();
let runs: Vec<&str> = xml.split("<w:r>").skip(1).map(|r| r.split("</w:r>").next().unwrap()).collect();
assert!(runs.len() >= 2, "the mixed run is split: {xml}");
assert!(runs.iter().any(|r| r.contains("<w:rtl/>")), "{xml}");
for r in runs {
assert!(r.contains("<w:rPrChange"), "every piece keeps the tracked change: {r}");
let (now, before) = direction_markup(r);
assert_eq!(now, before, "no direction or language difference to report: {r}");
}
// And WordCraft reads no difference back either.
let r = wordcraft_docx::read(&bytes).unwrap();
for run in &paragraph(&r).runs {
let old = &run.props.fmt_change.as_ref().unwrap().old;
assert_eq!((old.rtl, old.lang_bidi.as_deref()), (run.props.rtl, run.props.lang_bidi.as_deref()));
}
}
+17
View File
@@ -228,3 +228,20 @@ fn deeply_nested_controls_are_bounded() {
let back = read(&write(&d).unwrap()).unwrap();
assert_eq!(controls(&back).len(), 60);
}
/// A control's `w:rPr` (ECMA-376 §17.5.2.27) is read as the formatting entered text takes, and
/// still written back as it was.
#[test]
fn a_controls_run_properties_are_read_and_kept() {
let docx = package(&p(&format!(
"{}{}",
t("Name: "),
sdt(r#"<w:rPr><w:rStyle w:val="Eingabe"/><w:sz w:val="20"/></w:rPr><w:id w:val="11"/><w:showingPlcHdr/>"#, &t("Eingabe"))
)));
let d = read(&docx).unwrap();
let c = by_id(&d, 11);
let rp = c.run_props.expect("run properties read");
assert_eq!((rp.style.as_deref(), rp.size), (Some("Eingabe"), Some(10.0)));
let out = document_xml(&write(&d).unwrap());
assert!(out.contains(r#"<w:rPr><w:rStyle w:val="Eingabe"/><w:sz w:val="20"/></w:rPr>"#), "{out}");
}
+121
View File
@@ -0,0 +1,121 @@
//! East Asian typography settings of a run (`w:eastAsianLayout`, ISO/IEC 29500-1 §17.3.2.10):
//! two lines in one and horizontal in vertical are read and saved again. Fixtures are
//! hand-written from the spec.
use std::io::{Read, Write};
use wordcraft_doc::props::{CombineBrackets, EastAsianLayout};
fn docx(body: &str, styles: Option<&str>) -> Vec<u8> {
let root = r#"<?xml version="1.0" encoding="UTF-8"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="word/document.xml"/></Relationships>"#;
let doc_rels = r#"<?xml version="1.0" encoding="UTF-8"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/styles" Target="styles.xml"/></Relationships>"#;
let doc = format!(
r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:document xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:body>{body}</w:body></w:document>"#
);
let mut parts = vec![("_rels/.rels", root.to_string()), ("word/document.xml", doc)];
if let Some(s) = styles {
parts.push(("word/_rels/document.xml.rels", doc_rels.to_string()));
parts.push((
"word/styles.xml",
format!(r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:styles xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main">{s}</w:styles>"#),
));
}
let mut zw = zip::ZipWriter::new(std::io::Cursor::new(Vec::new()));
for (name, xml) in &parts {
zw.start_file(*name, zip::write::SimpleFileOptions::default()).unwrap();
zw.write_all(xml.as_bytes()).unwrap();
}
zw.finish().unwrap().into_inner()
}
fn part(bytes: &[u8], name: &str) -> String {
let mut z = zip::ZipArchive::new(std::io::Cursor::new(bytes)).unwrap();
let mut s = String::new();
z.by_name(name).unwrap().read_to_string(&mut s).unwrap();
s
}
fn run(text: &str, rpr: &str) -> String {
format!(r#"<w:r><w:rPr>{rpr}</w:rPr><w:t>{text}</w:t></w:r>"#)
}
/// The East Asian layout of the run holding `text` in the first paragraph.
fn layout_of(d: &wordcraft_doc::Document, text: &str) -> Option<EastAsianLayout> {
let p = d.body.iter().find_map(|b| b.as_para()).unwrap();
let start = p.text.find(text).unwrap();
let mut at = 0;
for r in &p.runs {
if at == start {
return r.props.east_asian_layout.clone();
}
at += r.len;
}
panic!("no run starts at {text:?}");
}
#[test]
fn two_lines_in_one_and_horizontal_in_vertical_survive_a_save() {
// Warichu with round brackets, tate-chu-yoko compressed to the line, and a plain run beside them.
let body = format!(
r#"<w:p>{}{}{}{}</w:p><w:sectPr><w:textDirection w:val="tbRl"/></w:sectPr>"#,
run("本文", ""),
run("二行の注", r#"<w:eastAsianLayout w:id="1" w:combine="on" w:combineBrackets="round"/>"#),
run("12", r#"<w:eastAsianLayout w:id="2" w:vert="on" w:vertCompress="true"/>"#),
run("月", "")
);
let d = wordcraft_docx::read(&docx(&body, None)).unwrap();
let warichu = EastAsianLayout { id: Some(1), combine: Some(true), combine_brackets: Some(CombineBrackets::Round), ..Default::default() };
let tcy = EastAsianLayout { id: Some(2), vert: Some(true), vert_compress: Some(true), ..Default::default() };
assert_eq!(layout_of(&d, "二行の注"), Some(warichu.clone()));
assert_eq!(layout_of(&d, "12"), Some(tcy.clone()));
assert_eq!(layout_of(&d, "本文"), None);
assert_eq!(layout_of(&d, "月"), None);
let out = wordcraft_docx::write(&d).unwrap();
let xml = part(&out, "word/document.xml");
assert_eq!(xml.matches("<w:eastAsianLayout ").count(), 2, "{xml}");
assert!(xml.contains(r#"<w:eastAsianLayout w:id="1" w:combine="1" w:combineBrackets="round"/>"#), "{xml}");
assert!(xml.contains(r#"<w:eastAsianLayout w:id="2" w:vert="1" w:vertCompress="1"/>"#), "{xml}");
let back = wordcraft_docx::read(&out).unwrap();
assert_eq!(layout_of(&back, "二行の注"), Some(warichu));
assert_eq!(layout_of(&back, "12"), Some(tcy));
}
#[test]
fn east_asian_layout_is_written_after_lang() {
// CT_RPr's sequence: … w:lang, w:eastAsianLayout, w:specVanish.
let body = format!(r#"<w:p>{}</w:p>"#, run("12", r#"<w:eastAsianLayout w:vert="1"/><w:lang w:val="ja-JP"/><w:b/>"#));
let d = wordcraft_docx::read(&docx(&body, None)).unwrap();
let xml = part(&wordcraft_docx::write(&d).unwrap(), "word/document.xml");
let lang = xml.find("<w:lang ").unwrap();
let eal = xml.find("<w:eastAsianLayout ").unwrap();
assert!(lang < eal, "{xml}");
assert!(xml.contains(r#"<w:eastAsianLayout w:vert="1"/>"#), "{xml}");
}
#[test]
fn a_character_style_keeps_its_east_asian_layout() {
let styles = r#"<w:style w:type="character" w:styleId="TCY"><w:name w:val="Tate Chu Yoko"/><w:rPr><w:eastAsianLayout w:id="7" w:vert="on"/></w:rPr></w:style>"#;
let body = format!(r#"<w:p>{}</w:p>"#, run("12", r#"<w:rStyle w:val="TCY"/>"#));
let d = wordcraft_docx::read(&docx(&body, Some(styles))).unwrap();
let out = wordcraft_docx::write(&d).unwrap();
let xml = part(&out, "word/styles.xml");
assert!(xml.contains(r#"<w:eastAsianLayout w:id="7" w:vert="1"/>"#), "{xml}");
}
#[test]
fn values_outside_the_schema_are_dropped_not_guessed() {
// An id past xsd:int, an unknown bracket and an on/off that isn't one: those attributes are
// unset; the valid one beside them is kept. An element with nothing valid writes nothing.
let body = format!(
r#"<w:p>{}{}</w:p>"#,
run("注", r#"<w:eastAsianLayout w:id="99999999999" w:combine="yes" w:combineBrackets="wavy" w:vert="off"/>"#),
run("x", r#"<w:eastAsianLayout w:combineBrackets="wavy"/>"#)
);
let d = wordcraft_docx::read(&docx(&body, None)).unwrap();
assert_eq!(layout_of(&d, "注"), Some(EastAsianLayout { vert: Some(false), ..Default::default() }));
assert_eq!(layout_of(&d, "x"), None);
let xml = part(&wordcraft_docx::write(&d).unwrap(), "word/document.xml");
assert_eq!(xml.matches("<w:eastAsianLayout ").count(), 1, "{xml}");
assert!(xml.contains(r#"<w:eastAsianLayout w:vert="0"/>"#), "{xml}");
}
+712 -2
View File
@@ -3,7 +3,7 @@
use std::io::Write;
use wordcraft_doc::para::{Anchor, FloatAlign, NoteKind, ShapeKind, Wrap};
use wordcraft_doc::props::{Align, Border, BorderStyle, Rgb, TextColor, VMerge};
use wordcraft_doc::props::{Align, Border, BorderStyle, Rgb, TextColor, Underline, VMerge};
use wordcraft_doc::{Block, Document, InlineObject, Paragraph, RevisionKind};
const W_NS: &str = r#"xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships" xmlns:wp="http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing" xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main" xmlns:pic="http://schemas.openxmlformats.org/drawingml/2006/picture" xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006" xmlns:wps="http://schemas.microsoft.com/office/word/2010/wordprocessingShape" xmlns:wpg="http://schemas.microsoft.com/office/word/2010/wordprocessingGroup" xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office""#;
@@ -97,6 +97,28 @@ fn paras(d: &Document) -> Vec<&Paragraph> {
d.body.iter().filter_map(|b| b.as_para()).collect()
}
#[test]
fn ref_links_keep_cached_results() {
let d = read_body(
r#"
<w:p><w:fldSimple w:instr="REF destination \h" w:fldLock="1"><w:r><w:rPr><w:b/></w:rPr><w:t>Cached simple</w:t></w:r></w:fldSimple></w:p>
<w:p><w:r><w:fldChar w:fldCharType="begin"/></w:r><w:r><w:instrText xml:space="preserve">REF destination </w:instrText></w:r><w:r><w:instrText>\w \h</w:instrText></w:r><w:r><w:fldChar w:fldCharType="separate"/></w:r><w:r><w:t>Cached complex</w:t></w:r><w:r><w:fldChar w:fldCharType="end"/></w:r></w:p>
<w:p><w:bookmarkStart w:id="0" w:name="destination"/><w:r><w:t>Changed destination</w:t></w:r><w:bookmarkEnd w:id="0"/></w:p>"#,
);
let again = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
for doc in [&d, &again] {
let p = paras(doc);
for (index, instr, cached, locked) in
[(0, r"REF destination \h", "Cached simple", true), (1, r"REF destination \w \h", "Cached complex", false)]
{
assert!(matches!(&p[index].objects[0], InlineObject::Field { instr: actual, result, locked: lock, .. }
if actual == instr && result == cached && *lock == locked));
}
assert_eq!(p[0].props_of_char(0).bold, Some(true));
assert!(doc.bookmarks().iter().any(|(name, _)| name == "destination"));
}
}
#[test]
fn single_entry_toc_keeps_formatted_result_and_nested_page_field() {
let d = read_body(
@@ -205,7 +227,7 @@ fn hyperlink_runs_and_wrappers() {
assert_eq!(p[0].props_of_char(0).style.as_deref(), Some("Hyperlink"));
assert_eq!(p[0].props_of_char(4).link.as_deref(), Some("https://example.org/#frag"));
let sym_off = p[0].text.find('\u{F04A}').unwrap();
assert_eq!(p[0].props_of_char(sym_off).font.as_deref(), Some("Wingdings"));
assert_eq!(p[0].props_of_char(sym_off).sym_font.as_deref(), Some("Wingdings"), "the symbol keeps its own font");
let t = p[0].props_of_char(p[0].text.len() - 1);
assert_eq!((t.bold, t.italic, t.caps, t.strike, t.color), (Some(false), Some(false), Some(false), Some(true), Some(TextColor::Auto)));
assert_eq!(p[1].plain_text(), "in block sdt");
@@ -236,6 +258,86 @@ fn symbol_font_text_reads_as_unicode_and_round_trips_unchanged() {
assert_eq!(q[0].shown_text(), p[0].shown_text());
}
/// The `w:sym` elements of a document's main part, as written.
fn sym_elements(bytes: &[u8]) -> Vec<String> {
let parts = unzip(bytes);
let xml = String::from_utf8_lossy(parts.get("word/document.xml").unwrap()).to_string();
xml.match_indices("<w:sym ").map(|(i, _)| xml[i..].split("/>").next().unwrap().to_string() + "/>").collect()
}
#[test]
fn a_sym_character_keeps_its_symbol_font_when_the_text_font_changes() {
// #2036: ECMA-376 §17.3.3.30 — a `w:sym` character's font is its own `w:font`, whatever the
// run's font. Setting the whole paragraph in Times New Roman keeps the check mark.
let body = r#"<w:p><w:r><w:t>a</w:t></w:r><w:r><w:sym w:font="Wingdings" w:char="F0FC"/></w:r><w:r><w:t>b</w:t></w:r></w:p>"#;
let mut d = wordcraft_docx::read(&docx(body, &[], &[])).unwrap();
let len = paras(&d)[0].text.len();
d.format_range(&wordcraft_doc::Pos::body(0, 0), &wordcraft_doc::Pos::body(0, len), &|c| c.font = Some("Times New Roman".into())).unwrap();
let p = paras(&d)[0];
assert_eq!(p.plain_text(), "a\u{F0FC}b", "the stored character is unchanged");
assert_eq!(p.props_of_char(1).font.as_deref(), Some("Times New Roman"), "the run font did change");
assert_eq!(p.shown_text(), "a\u{2713}b");
assert_eq!(d.shown_text(wordcraft_doc::StoryRef::Body), "a\u{2713}b");
// Written back as a `w:sym` in its symbol font, so Word shows the symbol too.
let bytes = wordcraft_docx::write(&d).unwrap();
assert_eq!(sym_elements(&bytes), vec![r#"<w:sym w:font="Wingdings" w:char="F0FC"/>"#.to_string()]);
let again = wordcraft_docx::read(&bytes).unwrap();
let q = paras(&again)[0];
assert_eq!(q.plain_text(), "a\u{F0FC}b");
assert_eq!(q.shown_text(), "a\u{2713}b");
assert_eq!(q.props_of_char(1).font.as_deref(), Some("Times New Roman"));
}
#[test]
fn a_sym_character_round_trips_as_w_sym() {
let body = r#"<w:p><w:r><w:t xml:space="preserve">Done </w:t></w:r><w:r><w:sym w:font="Wingdings" w:char="F0FC"/><w:sym w:font="Symbol" w:char="F061"/></w:r></w:p>"#;
let d = wordcraft_docx::read(&docx(body, &[], &[])).unwrap();
let bytes = wordcraft_docx::write(&d).unwrap();
assert_eq!(
sym_elements(&bytes),
vec![r#"<w:sym w:font="Wingdings" w:char="F0FC"/>"#.to_string(), r#"<w:sym w:font="Symbol" w:char="F061"/>"#.to_string()]
);
let again = wordcraft_docx::read(&bytes).unwrap();
assert_eq!(paras(&again)[0].shown_text(), "Done \u{2713}\u{3B1}");
assert_eq!(sym_elements(&wordcraft_docx::write(&again).unwrap()), sym_elements(&bytes));
}
#[test]
fn hostile_sym_fonts_never_panic() {
// Built in the model (an XML file couldn't even hold some of these names).
let huge = "W".repeat(100_000);
for font in ["", " ", "\u{1}", huge.as_str(), "Wingdings\u{FFFF}", "<&\"'>"] {
let mut d = Document::from_text("a\u{F0FC}\u{1}\u{1F600}b");
let len = paras(&d)[0].text.len();
let f = font.to_string();
d.format_range(&wordcraft_doc::Pos::body(0, 0), &wordcraft_doc::Pos::body(0, len), &|c| c.sym_font = Some(f.clone())).unwrap();
let _ = d.shown_text(wordcraft_doc::StoryRef::Body);
let bytes = wordcraft_docx::write(&d).unwrap();
let _ = wordcraft_docx::read(&bytes).unwrap();
}
}
#[test]
fn only_printable_characters_are_written_as_w_sym() {
// `w:sym` holds a character code from U+0020 (the reader drops lower ones), like `w:t`.
let mut d = Document::from_text("a\u{7}");
d.format_range(&wordcraft_doc::Pos::body(0, 0), &wordcraft_doc::Pos::body(0, 2), &|c| c.sym_font = Some("Wingdings".into())).unwrap();
assert_eq!(sym_elements(&wordcraft_docx::write(&d).unwrap()), vec![r#"<w:sym w:font="Wingdings" w:char="0061"/>"#.to_string()]);
}
#[test]
fn pcl6_symbol_roundtrip() {
let body = r#"<w:p><w:r><w:sym w:font="Symbol (PCL6)" w:char="F0B3"/></w:r></w:p>"#;
let d = wordcraft_docx::read(&docx(body, &[], &[])).unwrap();
let again = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
for doc in [&d, &again] {
let p = paras(doc)[0];
assert_eq!(p.text, "\u{F0B3}");
assert_eq!(p.props_of_char(0).sym_font.as_deref(), Some("Symbol (PCL6)"));
assert_eq!(p.shown_text(), "≥");
}
}
#[test]
fn alternate_content_choice_and_fallback() {
let body = r#"
@@ -560,6 +662,26 @@ fn empty_tbl_borders_elements_leave_sides_unset() {
assert!(b.top.is_some() && b.between.is_some() && b.inside_v.is_some());
}
/// A style ID defined twice is its last definition, whole (Word drops the earlier one rather
/// than merging: observed with synthetic documents). Generators that write a placeholder
/// Heading 1 and then their own made headings come out in the placeholder's font and size.
#[test]
fn a_style_defined_twice_is_its_last_definition() {
let styles = format!(
r#"<w:styles {W_NS}>
<w:style w:type="paragraph" w:styleId="Heading1"><w:name w:val="heading 1"/><w:pPr><w:jc w:val="center"/></w:pPr><w:rPr><w:i/><w:color w:val="FF0000"/><w:sz w:val="40"/></w:rPr></w:style>
<w:style w:type="paragraph" w:styleId="Heading1"><w:name w:val="heading 1"/><w:rPr><w:b/><w:color w:val="0000FF"/><w:sz w:val="24"/></w:rPr></w:style>
</w:styles>"#
);
let body = r#"<w:p><w:pPr><w:pStyle w:val="Heading1"/></w:pPr><w:r><w:t>Probe</w:t></w:r></w:p>"#;
let d = wordcraft_docx::read(&docx(body, &[("rId1", "styles", "styles.xml")], &[("word/styles.xml", &styles)])).unwrap();
assert_eq!(d.styles.styles.iter().filter(|s| s.id == "Heading1").count(), 1);
let h = d.styles.get("Heading1").unwrap();
assert_eq!((h.chr.bold, h.chr.italic, h.chr.size), (Some(true), None, Some(12.0)));
assert_eq!(h.chr.color, Some(TextColor::Rgb(Rgb(0, 0, 0xFF))));
assert_eq!(h.para.align, None, "nothing kept from the first definition");
}
#[test]
fn vml_image_and_inline_drawing() {
let png: &[u8] = b"\x89PNG\r\n\x1a\nnot really a png";
@@ -616,6 +738,24 @@ fn chart_and_diagram_drawings_are_graphic_objects() {
}
}
/// A text box shape's style text colour (`wps:style/a:fontRef`) is kept on its story, as what
/// automatic text colour means inside it; a box without one has none.
#[test]
fn a_text_box_s_style_font_colour_is_read() {
let shape = |style: &str| {
format!(
r#"<w:r><w:drawing><wp:anchor behindDoc="0"><wp:positionH relativeFrom="column"><wp:posOffset>0</wp:posOffset></wp:positionH><wp:positionV relativeFrom="paragraph"><wp:posOffset>0</wp:posOffset></wp:positionV><wp:extent cx="1270000" cy="635000"/><wp:wrapNone/><wp:docPr id="1" name="S"/><a:graphic><a:graphicData uri="http://schemas.microsoft.com/office/word/2010/wordprocessingShape"><wps:wsp><wps:spPr><a:prstGeom prst="rect"/></wps:spPr>{style}<wps:txbx><w:txbxContent><w:p><w:r><w:t>In the box</w:t></w:r></w:p></w:txbxContent></wps:txbx><wps:bodyPr/></wps:wsp></a:graphicData></a:graphic></wp:anchor></w:drawing></w:r>"#
)
};
let style = r#"<wps:style><a:lnRef idx="2"><a:srgbClr val="000000"/></a:lnRef><a:fillRef idx="1"><a:srgbClr val="203040"/></a:fillRef><a:effectRef idx="0"><a:srgbClr val="000000"/></a:effectRef><a:fontRef idx="minor"><a:srgbClr val="FFFFFF"/></a:fontRef></wps:style>"#;
let d = read_body(&format!("<w:p>{}{}</w:p>", shape(style), shape("")));
let boxes: Vec<_> =
paras(&d)[0].objects.iter().filter_map(|o| if let InlineObject::Shape { story: Some(id), .. } = o { Some(*id) } else { None }).collect();
let [styled, plain] = boxes.as_slice() else { panic!("{boxes:?}") };
assert_eq!(d.parts[styled].body.auto_color, Some(Rgb(0xFF, 0xFF, 0xFF)));
assert_eq!(d.parts[plain].body.auto_color, None);
}
#[test]
fn a_chart_part_is_built_once_and_only_through_a_chart_relationship() {
let chart = r#"<c:chartSpace xmlns:c="http://schemas.openxmlformats.org/drawingml/2006/chart"><c:chart><c:plotArea><c:barChart><c:ser><c:val><c:numLit><c:ptCount val="2"/><c:pt idx="0"><c:v>1</c:v></c:pt><c:pt idx="1"><c:v>2</c:v></c:pt></c:numLit></c:val></c:ser></c:barChart></c:plotArea></c:chart></c:chartSpace>"#;
@@ -809,6 +949,28 @@ fn line_rule_is_read_case_insensitively() {
assert_eq!(p[1].props.line_spacing, Some(wordcraft_doc::props::LineSpacing::Exactly(15.0)));
}
/// #2329: Word 2024 saves strict OOXML with the line height written out with a unit
/// (`w:line="13.90pt"`). Under `auto` it reads through twips, as Word renders it (13.90 × 20 =
/// 278; the 1.1583 multiple), not as 240ths of a line like the bare number — a Word file with
/// a `line="13.90pt"` default came in at 13.9 lines per line of text.
#[test]
fn strict_line_heights_with_a_unit_are_read_as_point_multiples() {
let d = read_body(
r#"<w:p><w:pPr><w:spacing w:after="8pt" w:line="13.90pt" w:lineRule="auto"/></w:pPr><w:r><w:t>x</w:t></w:r></w:p><w:p><w:pPr><w:spacing w:line="13.90pt" w:lineRule="exact"/></w:pPr></w:p>"#,
);
let p = paras(&d);
let Some(wordcraft_doc::props::LineSpacing::Multiple(m)) = p[0].props.line_spacing else {
panic!("auto line spacing: {:?}", p[0].props.line_spacing)
};
assert!((m - 278.0 / 240.0).abs() < 1e-5, "{m}");
assert_eq!(p[0].props.space_after, Some(8.0), "the unit form of w:after too");
assert_eq!(p[1].props.line_spacing, Some(wordcraft_doc::props::LineSpacing::Exactly(13.9)), "exact keeps reading it as points");
// Saves as the transitional 240ths form and reads back the same multiple.
let again = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
let Some(wordcraft_doc::props::LineSpacing::Multiple(m2)) = paras(&again)[0].props.line_spacing else { panic!("multiple after a save") };
assert!((m2 - m).abs() < 1e-5, "{m2} vs {m}");
}
#[test]
fn char_border_reads_and_nil_resolves_off() {
let d = read_body(
@@ -960,6 +1122,39 @@ fn vml_textpath_in_header_is_the_watermark() {
assert_eq!(read(r#"<v:shape id="PowerPlusWaterMarkObject2"><v:textpath string=" "/></v:shape>"#, true).0, None);
}
/// Issue #2875: Word 2007 WordArt, a body `v:shape` with a `v:textpath`, is shown as a text shape
/// holding its string in the textpath's font and the shape's colour, bent by the warp of its WordArt
/// shape, not dropped; a textpath turned off or with an empty string draws nothing.
#[test]
fn vml_textpath_in_body_is_shown() {
let read = |shape: &str| {
let body = format!("<w:p><w:r><w:t>Before</w:t></w:r><w:r><w:pict>{shape}</w:pict></w:r></w:p>");
wordcraft_docx::read(&docx(&body, &[], &[])).unwrap()
};
let art = r##"<v:shape type="#_x0000_t136" style="position:absolute;width:300pt;height:60pt" fillcolor="#336699" stroked="f"><v:textpath style="font-family:&quot;Impact&quot;;font-size:1pt" string="Sale"/></v:shape>"##;
let d = read(art);
let objs = &paras(&d)[0].objects;
assert_eq!(objs.len(), 1);
let InlineObject::Shape { story: Some(id), w, h, .. } = &objs[0] else { panic!("not a text shape: {:?}", objs[0]) };
assert_eq!((*w, *h), (300.0, 60.0));
let p = d.parts.get(id).unwrap().blocks[0].as_para().unwrap();
assert_eq!(p.text, "Sale");
assert_eq!(p.runs[0].props.font.as_deref(), Some("Impact"));
assert_eq!(p.runs[0].props.color, Some(TextColor::Rgb(Rgb(0x33, 0x66, 0x99))));
// Its WordArt shape bends the text: shapetype 144 (Arch Up) is DrawingML's `textArchUp`.
let warp = |shape: &str| match &paras(&read(shape))[0].objects[..] {
[InlineObject::Shape { extra, .. }] => extra.warp.as_ref().map(|w| w.preset.clone()),
other => panic!("not one shape: {other:?}"),
};
assert_eq!(warp(&art.replace("_x0000_t136", "_x0000_t144")).as_deref(), Some("textArchUp"));
assert_eq!(warp(&art.replace("_x0000_t136", "_x0000_t175")).as_deref(), Some("textCanDown"));
assert_eq!(warp(art).as_deref(), Some("textPlain"));
assert!(paras(&read(&art.replace("<v:textpath ", r#"<v:textpath on="f" "#)))[0].objects.is_empty());
assert!(paras(&read(&art.replace(r#"string="Sale""#, r#"string=" ""#)))[0].objects.is_empty());
}
/// Issue #1290: a file with Word's watermark keeps it through save and reopen: the same
/// watermark, the same headers with the same text and paragraphs, one shape per header written;
/// header WordArt that isn't a watermark doesn't become one on the way.
@@ -1007,6 +1202,79 @@ fn word_watermark_survives_save() {
assert_eq!(d2.settings.watermark, None);
}
/// Issue #1775: a picture watermark in a header that starts with a table is anchored in a cell
/// but placed on the page: any VML shape in a header (Word's picture watermark, `o:allowincell="f"`; user-checked
/// in Word for Mac, the attribute makes no difference) or DrawingML `layoutInCell="0"`. That survives save and reopen; a shape laid out in its cell stays so.
#[test]
fn header_cell_picture_placed_on_the_page() {
let vml = |allow: &str| {
format!(
r##"<w:r><w:pict><v:shape id="WordPictureWatermark1" o:spid="_x0000_s2049" type="#_x0000_t75" style="position:absolute;margin-left:0;margin-top:0;width:300pt;height:200pt;z-index:-251657216;mso-position-horizontal:center;mso-position-horizontal-relative:margin;mso-position-vertical:center;mso-position-vertical-relative:margin" {allow}><v:imagedata r:id="rIdP" o:title="" gain="19661f" blacklevel="22938f"/></v:shape></w:pict></w:r>"##
)
};
let anchor = |in_cell: &str| {
format!(
r#"<w:r><w:drawing><wp:anchor distT="0" distB="0" distL="0" distR="0" simplePos="0" relativeHeight="1" behindDoc="1" locked="0" layoutInCell="{in_cell}" allowOverlap="1"><wp:simplePos x="0" y="0"/><wp:positionH relativeFrom="margin"><wp:align>center</wp:align></wp:positionH><wp:positionV relativeFrom="margin"><wp:align>center</wp:align></wp:positionV><wp:extent cx="3810000" cy="2540000"/><wp:wrapNone/><wp:docPr id="1" name="Picture 1"/><a:graphic><a:graphicData uri="http://schemas.openxmlformats.org/drawingml/2006/picture"><pic:pic><pic:nvPicPr><pic:cNvPr id="1" name="p.png"/><pic:cNvPicPr/></pic:nvPicPr><pic:blipFill><a:blip r:embed="rIdP"/></pic:blipFill><pic:spPr><a:xfrm><a:off x="0" y="0"/><a:ext cx="3810000" cy="2540000"/></a:xfrm><a:prstGeom prst="rect"/></pic:spPr></pic:pic></a:graphicData></a:graphic></wp:anchor></w:drawing></w:r>"#
)
};
let read = |run: &str| {
let hdr = format!(
r#"<w:hdr {W_NS}><w:tbl><w:tblGrid><w:gridCol w:w="4680"/><w:gridCol w:w="4680"/></w:tblGrid><w:tr><w:tc><w:p><w:r><w:t>Left</w:t></w:r></w:p></w:tc><w:tc><w:p>{run}<w:r><w:t>Right</w:t></w:r></w:p></w:tc></w:tr></w:tbl><w:p/></w:hdr>"#
);
let hrels = rels(&[("rIdP", "image", "media/p.png")]);
let body = r#"<w:p><w:r><w:t>Body</w:t></w:r></w:p><w:sectPr><w:headerReference w:type="default" r:id="rIdH"/></w:sectPr>"#;
let parts = [("word/header1.xml", hdr.as_str()), ("word/_rels/header1.xml.rels", hrels.as_str()), ("word/media/p.png", "\u{89}PNG stand-in")];
wordcraft_docx::read(&docx(body, &[("rIdH", "header", "header1.xml")], &parts)).unwrap()
};
let floats = |d: &Document| -> Vec<(bool, Wrap)> {
let hid = d.last_section.headers.default.unwrap();
let mut out = Vec::new();
for b in &d.parts[&hid].blocks {
if let Block::Table(t) = &**b {
for c in &t.rows[0].cells {
for p in c.blocks.iter().filter_map(|b| b.as_para()) {
out.extend(p.objects.iter().filter_map(|o| o.frame()).map(|(_, _, f)| (f.outside_cell, f.wrap)));
}
}
}
}
out
};
for (run, outside) in [
(vml(r#"o:allowincell="f""#), true),
(vml(r#"o:allowincell="false""#), true),
// Word for Mac draws every VML picture in a header cell on the page, ticked or not.
(vml(""), true),
(vml(r#"o:allowincell="t""#), true),
(anchor("0"), true),
(anchor("false"), true),
(anchor("1"), false),
] {
let d = read(&run);
assert_eq!(floats(&d), vec![(outside, Wrap::BehindText)], "{run}");
let d2 = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
assert_eq!(floats(&d2), vec![(outside, Wrap::BehindText)], "saved: {run}");
}
}
/// Issue #1775: only header and footer shapes are placed on the page from a cell (what was checked
/// in Word); a VML picture in a body table cell stays laid out in its cell.
#[test]
fn body_cell_vml_picture_stays_in_its_cell() {
let pic = r##"<w:r><w:pict><v:shape type="#_x0000_t75" style="position:absolute;margin-left:0;margin-top:0;width:100pt;height:80pt;z-index:-1;mso-position-horizontal:center;mso-position-horizontal-relative:margin;mso-position-vertical:center;mso-position-vertical-relative:margin"><v:imagedata r:id="rIdP"/></v:shape></w:pict></w:r>"##;
let body = format!(r#"<w:tbl><w:tblGrid><w:gridCol w:w="4680"/></w:tblGrid><w:tr><w:tc><w:p>{pic}</w:p></w:tc></w:tr></w:tbl><w:p/>"#);
let parts = [("word/media/p.png", "\u{89}PNG stand-in")];
let d = wordcraft_docx::read(&docx(&body, &[("rIdP", "image", "media/p.png")], &parts)).unwrap();
let Block::Table(t) = &*d.body[0] else { panic!("table first") };
let floats: Vec<bool> = t.rows[0].cells[0]
.blocks
.iter()
.filter_map(|b| b.as_para())
.flat_map(|p| p.objects.iter().filter_map(|o| o.frame()).map(|(_, _, f)| f.outside_cell))
.collect();
assert_eq!(floats, vec![false]);
}
/// Issue #149: an absolutely positioned VML text box in a header floats where its style puts it
/// (relative to the page here, behind the text at a negative z-index, no wrapping) instead of
/// sitting inline in the header's flow, and it keeps that placement when saved and read back.
@@ -1593,6 +1861,23 @@ fn theme_colour_without_a_theme_keeps_the_fixed_colour() {
assert_eq!(paras(&d)[0].props_of_char(0).color, Some(TextColor::Rgb(Rgb(0, 0xFF, 0))));
}
#[test]
fn underline_without_a_val_draws_no_underline() {
// w:val has no schema default (ECMA-376 CT_Underline); Word leaves a bare w:u with a colour plain.
let d = read_body(
r#"<w:p><w:r><w:rPr><w:u w:val="single"/></w:rPr><w:t>a</w:t></w:r><w:r><w:rPr><w:u w:color="FF0000"/></w:rPr><w:t>b</w:t></w:r><w:r><w:t>c</w:t></w:r></w:p>"#,
);
let p = paras(&d);
assert_eq!(p[0].props_of_char(0).underline, Some(Underline::Single));
let b = p[0].props_of_char(1);
assert_eq!(b.underline, None);
assert_eq!(b.underline_color, Some(Rgb(0xFF, 0, 0)), "the colour is still read");
assert_eq!(p[0].props_of_char(2).underline, None);
let back = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
let b = paras(&back)[0].props_of_char(1);
assert_eq!((b.underline, b.underline_color), (None, Some(Rgb(0xFF, 0, 0))), "saving keeps the colour, not an underline");
}
#[test]
fn theme_font_wins_over_the_fixed_font() {
let styles = r#"<w:style w:type="paragraph" w:styleId="Heading1"><w:name w:val="heading 1"/><w:rPr><w:rFonts w:ascii="Calibri Light" w:hAnsi="Calibri Light" w:asciiTheme="majorHAnsi" w:hAnsiTheme="majorHAnsi"/></w:rPr></w:style>"#;
@@ -2518,6 +2803,277 @@ fn native_ink_preserves_inner_transform_and_rotation_padding_on_unrelated_saves(
}
}
// ---- native Word ink as a member of a group or a drawing canvas (gap 2) ----
const GROUP_URI: &str = "http://schemas.microsoft.com/office/word/2010/wordprocessingGroup";
const CANVAS_URI: &str = "http://schemas.microsoft.com/office/word/2010/wordprocessingCanvas";
const W14_NS: &str = "http://schemas.microsoft.com/office/word/2010/wordml";
/// A rectangle member `name` at `x` (EMU) in its group's space.
fn rect_member(name: &str, x: u32) -> String {
format!(
r#"<wps:wsp><wps:cNvPr id="{x}" name="{name}"/><wps:cNvSpPr/><wps:spPr><a:xfrm><a:off x="{x}" y="0"/><a:ext cx="914400" cy="914400"/></a:xfrm><a:prstGeom prst="rect"><a:avLst/></a:prstGeom></wps:spPr><wps:bodyPr/></wps:wsp>"#
)
}
/// A native-ink member (`w14:contentPart`) at `x` (EMU), optionally inside `mc:AlternateContent`
/// with a picture fallback.
fn ink_member(x: u32, fallback: bool) -> String {
let content = format!(
r#"<w14:contentPart xmlns:w14="{W14_NS}" bwMode="auto" r:id="rInk"><w14:nvContentPartPr><w14:cNvPr id="9" name="Ink 9"/><w14:cNvContentPartPr/></w14:nvContentPartPr><w14:xfrm><a:off x="{x}" y="0"/><a:ext cx="914400" cy="457200"/></w14:xfrm></w14:contentPart>"#
);
if !fallback {
return content;
}
format!(
r#"<mc:AlternateContent><mc:Choice xmlns:wpi="{INK_NS}" Requires="wpi">{content}</mc:Choice><mc:Fallback><pic:pic><pic:nvPicPr><pic:cNvPr id="9" name="Ink 9"/><pic:cNvPicPr/></pic:nvPicPr><pic:blipFill><a:blip r:embed="rPreview"/><a:stretch><a:fillRect/></a:stretch></pic:blipFill><pic:spPr><a:xfrm><a:off x="{x}" y="0"/><a:ext cx="914400" cy="457200"/></a:xfrm><a:prstGeom prst="rect"><a:avLst/></a:prstGeom></pic:spPr></pic:pic></mc:Fallback></mc:AlternateContent>"#
)
}
/// A document whose second paragraph holds a group (or canvas) of a rectangle, `ink` and another
/// rectangle, in that order.
fn nested_ink_docx(canvas: bool, ink: &str, fallback: bool) -> Vec<u8> {
let members = format!("{}{ink}{}", rect_member("Before", 0), rect_member("After", 1828800));
let graphic = if canvas {
format!(r#"<a:graphicData uri="{CANVAS_URI}"><wpc:wpc xmlns:wpc="{CANVAS_URI}"><wpc:bg/><wpc:whole/>{members}</wpc:wpc></a:graphicData>"#)
} else {
format!(
r#"<a:graphicData uri="{GROUP_URI}"><wpg:wgp><wpg:cNvGrpSpPr/><wpg:grpSpPr><a:xfrm><a:off x="0" y="0"/><a:ext cx="2743200" cy="914400"/><a:chOff x="0" y="0"/><a:chExt cx="2743200" cy="914400"/></a:xfrm></wpg:grpSpPr>{members}</wpg:wgp></a:graphicData>"#
)
};
let body = format!(
r#"<w:p><w:r><w:t>Original</w:t></w:r></w:p><w:p><w:r><w:drawing><wp:inline><wp:extent cx="2743200" cy="914400"/><wp:docPr id="1" name="Group 1"/><a:graphic>{graphic}</a:graphic></wp:inline></w:drawing></w:r></w:p>"#
);
let mut relationships = vec![("rInk", "customXml", "ink/ink1.xml")];
let mut extra = vec![("[Content_Types].xml", INK_TYPES), ("word/ink/ink1.xml", INK_XML)];
if fallback {
relationships.push(("rPreview", "image", "media/preview.png"));
extra.push(("word/media/preview.png", "original preview bytes"));
}
docx(&body, &relationships, &extra)
}
/// The members of the one group or canvas in `d`.
fn group_children(d: &Document) -> Vec<wordcraft_doc::para::GroupChild> {
let groups: Vec<_> = paras(d)
.iter()
.flat_map(|p| p.objects.iter())
.filter_map(|o| match o {
InlineObject::Group { children, .. } => Some(children.clone()),
_ => None,
})
.collect();
let [children] = groups.try_into().unwrap_or_else(|g: Vec<_>| panic!("expected one group, found {}", g.len()));
children
}
fn assert_ink_between_rects(d: &Document) {
let c = group_children(d);
let kinds: Vec<_> = c
.iter()
.map(|c| match &c.obj {
InlineObject::Shape { .. } => "shape",
InlineObject::Graphic { graphic, .. } if graphic.kind.noun() == "ink" => "ink",
_ => "other",
})
.collect();
assert_eq!(kinds, ["shape", "ink", "shape"], "the ink member keeps its place among the members");
let ink = &c[1];
assert_eq!((ink.x, ink.y), (72.0, 0.0));
let InlineObject::Graphic { w, h, .. } = &ink.obj else { panic!() };
assert_eq!((*w, *h), (72.0, 36.0));
}
fn edit_first_para(d: &mut Document) {
let p = std::sync::Arc::make_mut(&mut d.body[0]);
let Block::Para(p) = p else { panic!() };
p.delete(0, p.len()).unwrap();
p.insert_text(0, "Edited", &Default::default()).unwrap();
}
#[test]
fn native_ink_in_a_group_or_canvas_survives_unrelated_edit_and_two_saves() {
for canvas in [false, true] {
for fallback in [false, true] {
let what = format!("canvas {canvas}, fallback {fallback}");
let mut d = wordcraft_docx::read(&nested_ink_docx(canvas, &ink_member(914400, fallback), fallback)).unwrap();
assert_ink_between_rects(&d);
edit_first_para(&mut d);
for _ in 0..2 {
let out = wordcraft_docx::write(&d).unwrap();
let files = unzip(&out);
check_ink(&files);
let body = text(&files, "word/document.xml");
assert!(body.contains("Edited"));
let at = |s: &str| body.find(s).unwrap_or_else(|| panic!("{s} missing ({what}): {body}"));
let ink = at("<w14:contentPart");
assert!(at("<wps:wsp") < ink && body.rfind("<wps:wsp").unwrap() > ink, "member order lost ({what}): {body}");
assert!(body.contains("<w14:nvContentPartPr>") && body.contains(r#"name="Ink 9""#), "{what}: {body}");
let xfrm = &body[ink..];
let xfrm = &xfrm[xfrm.find("<w14:xfrm").unwrap()..];
assert_eq!(attr_values(xfrm, "a:off", "x").first().map(String::as_str), Some("914400"), "{what}");
assert_eq!(attr_values(xfrm, "a:ext", "cx").first().map(String::as_str), Some("914400"), "{what}");
if fallback {
assert!(body.contains("<mc:Fallback>"), "{what}: {body}");
let [id] = attr_values(&body, "a:blip", "r:embed").try_into().unwrap();
let (_, target) = rel(&text(&files, "word/_rels/document.xml.rels"), &id);
assert_eq!(text(&files, &part_at("word", &target)), "original preview bytes");
}
d = wordcraft_docx::read(&out).unwrap();
assert_ink_between_rects(&d);
}
}
}
}
#[test]
fn native_ink_in_a_group_shows_its_fallback_picture() {
let d = wordcraft_docx::read(&nested_ink_docx(false, &ink_member(914400, true), true)).unwrap();
let c = group_children(&d);
let InlineObject::Graphic { graphic, .. } = &c[1].obj else { panic!() };
assert!(
graphic.items.iter().any(|item| matches!(item, wordcraft_doc::graphic::GraphicItem::Image { media, .. } if d.media.contains_key(media))),
"{:?}",
graphic.items
);
}
#[test]
fn native_ink_in_a_nested_group_is_written_at_its_place_in_the_outer_group() {
// The inner group doubles its members: 1 inch of its space is 2 inches of the outer group's.
let inner = format!(
r#"<wpg:grpSp><wpg:cNvGrpSpPr/><wpg:grpSpPr><a:xfrm><a:off x="914400" y="0"/><a:ext cx="1828800" cy="1828800"/><a:chOff x="0" y="0"/><a:chExt cx="914400" cy="914400"/></a:xfrm></wpg:grpSpPr>{}</wpg:grpSp>"#,
ink_member(0, false)
);
let d = wordcraft_docx::read(&nested_ink_docx(false, &inner, false)).unwrap();
let c = group_children(&d);
let ink = c.iter().find(|c| matches!(c.obj, InlineObject::Graphic { .. })).expect("ink member kept");
let InlineObject::Graphic { w, h, .. } = &ink.obj else { panic!() };
assert_eq!((ink.x, ink.y, *w, *h), (72.0, 0.0, 144.0, 72.0));
let files = unzip(&wordcraft_docx::write(&d).unwrap());
check_ink(&files);
let body = text(&files, "word/document.xml");
let xfrm = &body[body.find("<w14:xfrm").unwrap()..];
assert_eq!(attr_values(xfrm, "a:off", "x").first().map(String::as_str), Some("914400"));
assert_eq!(attr_values(xfrm, "a:ext", "cx").first().map(String::as_str), Some("1828800"));
assert_eq!(attr_values(xfrm, "a:ext", "cy").first().map(String::as_str), Some("914400"));
}
/// Ink parts share the custom XML relationship type with real custom XML stores: the store is
/// carried as before, the ink part only with the ink that refers to it.
#[test]
fn custom_xml_store_beside_native_ink_is_kept_and_ink_leaves_no_orphan() {
const STORE: &str = r#"<?xml version="1.0" encoding="UTF-8"?><data xmlns="urn:example:store"><item>kept</item></data>"#;
let types = INK_TYPES.replace("<Default ", r#"<Override PartName="/customXml/item1.xml" ContentType="application/xml"/><Default "#);
let mut files = unzip(&nested_ink_docx(false, &ink_member(914400, false), false));
let rels = text(&files, "word/_rels/document.xml.rels").replace(
"</Relationships>",
r#"<Relationship Id="rStore" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/customXml" Target="../customXml/item1.xml"/></Relationships>"#,
);
files.insert("word/_rels/document.xml.rels".into(), rels.into_bytes());
files.insert("[Content_Types].xml".into(), types.into_bytes());
files.insert("customXml/item1.xml".into(), STORE.as_bytes().to_vec());
let entries: Vec<_> = files.iter().map(|(name, bytes)| (name.as_str(), bytes.as_slice())).collect();
let mut d = wordcraft_docx::read(&zip(&entries)).unwrap();
let out = unzip(&wordcraft_docx::write(&d).unwrap());
check_ink(&out);
assert!(out.values().any(|b| b == STORE.as_bytes()), "the store is kept");
assert_eq!(out.values().filter(|b| *b == INK_XML.as_bytes()).count(), 1, "one ink part, not a second copy as a store");
// Without the ink the store stays and the ink part goes.
d.body.truncate(1);
let out = unzip(&wordcraft_docx::write(&d).unwrap());
assert!(out.values().any(|b| b == STORE.as_bytes()), "the store is kept");
assert!(!out.values().any(|b| b == INK_XML.as_bytes()), "orphan ink part: {:?}", out.keys().collect::<Vec<_>>());
assert!(!text(&out, "[Content_Types].xml").contains("inkml"));
assert_eq!(text(&out, "word/_rels/document.xml.rels").matches("/customXml\"").count(), 1, "only the store's relationship");
}
#[test]
fn deleting_drawing_level_native_ink_writes_no_orphan_ink_part() {
for fallback in [false, true] {
let mut d = wordcraft_docx::read(&ink_docx(fallback, false)).unwrap();
assert_eq!(graphics(&d).len(), 1);
d.body.truncate(1);
let files = unzip(&wordcraft_docx::write(&d).unwrap());
assert!(!files.values().any(|b| b == INK_XML.as_bytes()), "orphan ink part: {:?}", files.keys().collect::<Vec<_>>());
assert!(!text(&files, "[Content_Types].xml").contains("inkml"));
assert!(!text(&files, "word/_rels/document.xml.rels").contains("customXml"));
}
}
/// A group member's ink taken out of its group (not something WordCraft's commands do: Ungroup
/// refuses such a group) can't be written as a stand-alone drawing: it is left out whole, never
/// written in the wrong place, and its parts aren't written either.
#[test]
fn group_member_ink_on_its_own_is_left_out_without_orphans() {
for fallback in [false, true] {
let d = wordcraft_docx::read(&nested_ink_docx(false, &ink_member(914400, fallback), fallback)).unwrap();
let ink = group_children(&d).into_iter().find(|c| matches!(c.obj, InlineObject::Graphic { .. })).expect("ink member").obj;
let mut d = d;
d.body.truncate(1);
let mut p = Paragraph::new();
p.insert_object(0, ink, &Default::default()).unwrap();
d.body.push(wordcraft_doc::para_block(p));
let files = unzip(&wordcraft_docx::write(&d).unwrap());
let body = text(&files, "word/document.xml");
assert!(!body.contains("contentPart"), "fallback {fallback}: {body}");
assert!(!files.values().any(|b| b == INK_XML.as_bytes()), "orphan ink part: {:?}", files.keys().collect::<Vec<_>>());
assert!(!text(&files, "word/_rels/document.xml.rels").contains("customXml"));
}
}
#[test]
fn deleting_a_group_with_native_ink_writes_no_orphan_ink_part() {
let mut d = wordcraft_docx::read(&nested_ink_docx(false, &ink_member(914400, true), true)).unwrap();
d.body.truncate(1);
let files = unzip(&wordcraft_docx::write(&d).unwrap());
assert!(!text(&files, "word/document.xml").contains("contentPart"));
assert!(!files.values().any(|b| b == INK_XML.as_bytes()), "orphan ink part: {:?}", files.keys().collect::<Vec<_>>());
assert!(!text(&files, "[Content_Types].xml").contains("inkml"));
assert!(!text(&files, "word/_rels/document.xml.rels").contains("customXml"));
}
#[test]
fn hostile_native_ink_in_a_group_never_panics() {
// A dangling relationship, a missing part, or a part that isn't InkML: the ink member is
// dropped (as drawing-level ink is), the other members stay.
for fallback in [false, true] {
for case in ["dangling", "missing", "wrong type"] {
let mut files = unzip(&nested_ink_docx(false, &ink_member(914400, fallback), fallback));
match case {
"dangling" => {
let body = text(&files, "word/document.xml").replace(r#"r:id="rInk""#, r#"r:id="rNowhere""#);
files.insert("word/document.xml".into(), body.into_bytes());
}
"missing" => {
files.remove("word/ink/ink1.xml");
}
_ => {
let types = text(&files, "[Content_Types].xml").replace("application/inkml+xml", "application/xml");
files.insert("[Content_Types].xml".into(), types.into_bytes());
}
}
let entries: Vec<_> = files.iter().map(|(name, bytes)| (name.as_str(), bytes.as_slice())).collect();
let d = wordcraft_docx::read(&zip(&entries)).unwrap();
let c = group_children(&d);
assert!(!c.iter().any(|c| matches!(c.obj, InlineObject::Graphic { .. })), "{case}");
// With a picture fallback the picture takes the ink's place.
let pictures = c.iter().filter(|c| matches!(c.obj, InlineObject::Image { .. })).count();
assert_eq!(c.len(), if fallback { 3 } else { 2 }, "{case}, fallback {fallback}");
assert_eq!(pictures, usize::from(fallback), "{case}");
let out = unzip(&wordcraft_docx::write(&d).unwrap());
assert!(!text(&out, "word/document.xml").contains("contentPart"), "{case}");
}
}
// A garbled ink part is kept as it is: InkML stays opaque.
let mut files = unzip(&nested_ink_docx(true, &ink_member(914400, false), false));
files.insert("word/ink/ink1.xml".into(), vec![0xff, 0x00, b'<', 0xfe]);
let entries: Vec<_> = files.iter().map(|(name, bytes)| (name.as_str(), bytes.as_slice())).collect();
let d = wordcraft_docx::read(&zip(&entries)).unwrap();
let out = unzip(&wordcraft_docx::write(&d).unwrap());
assert!(out.values().any(|b| b == &[0xff, 0x00, b'<', 0xfe]));
}
/// Word draws a table of contents' entry links without the Hyperlink character style (measured in
/// Word for Windows); the import drops the style from them, keeping the link. A Hyperlink-styled run
/// outside a hyperlink, and a hyperlink in another field, keep it.
@@ -2830,3 +3386,157 @@ fn row_table_borders_round_trip() {
let again = wordcraft_docx::read(&saved).unwrap();
assert_eq!(d.body, again.body);
}
#[test]
fn field_result_keeps_run_props() {
let styles = format!(
r#"<w:styles {W_NS}><w:style w:type="paragraph" w:default="1" w:styleId="Normal"><w:name w:val="Normal"/><w:rPr><w:sz w:val="8"/></w:rPr></w:style><w:style w:type="character" w:styleId="Result"><w:name w:val="Result"/><w:rPr><w:sz w:val="16"/><w:b/></w:rPr></w:style></w:styles>"#
);
for result_pr in [r#"<w:sz w:val="16"/><w:b/>"#, r#"<w:rStyle w:val="Result"/>"#, ""] {
for simple in [false, true] {
let result = format!(r#"<w:r><w:rPr>{result_pr}</w:rPr><w:t>7</w:t></w:r>"#);
let field = if simple {
format!(r#"<w:fldSimple w:instr="PAGE" w:fldLock="1" w:dirty="1">{result}</w:fldSimple>"#)
} else {
format!(
r#"<w:r><w:rPr><w:sz w:val="6"/><w:i/></w:rPr><w:fldChar w:fldCharType="begin" w:fldLock="1" w:dirty="1"/></w:r><w:r><w:instrText>PAGE</w:instrText></w:r><w:r><w:fldChar w:fldCharType="separate"/></w:r><w:r><w:rPr><w:sz w:val="40"/></w:rPr><w:t/></w:r>{result}<w:r><w:rPr><w:sz w:val="50"/></w:rPr><w:fldChar w:fldCharType="end"/></w:r>"#
)
};
let body = format!(r#"<w:p><w:r><w:rPr><w:sz w:val="16"/></w:rPr><w:t>Counter </w:t></w:r>{field}</w:p>"#);
let mut d = wordcraft_docx::read(&docx(&body, &[("styles", "styles", "styles.xml")], &[("word/styles.xml", &styles)])).unwrap();
for _ in 0..2 {
let p = paras(&d)[0];
let props = p.props_of_char("Counter ".len());
assert_eq!(p.props_at(0).size, Some(8.0));
assert_eq!(props.size, result_pr.starts_with("<w:sz").then_some(8.0), "simple={simple}, result={result_pr}");
assert_eq!(props.style.as_deref(), result_pr.starts_with("<w:rStyle").then_some("Result"));
assert_eq!(props.italic, None);
assert!(
matches!(&p.objects[0], InlineObject::Field { instr, result, locked: true, dirty: true, .. } if instr == "PAGE" && result == "7")
);
d = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
}
}
}
}
#[test]
fn empty_field_keeps_begin_props() {
let d = read_body(
r#"<w:p><w:r><w:rPr><w:sz w:val="12"/><w:i/></w:rPr><w:fldChar w:fldCharType="begin"/></w:r><w:r><w:instrText>PAGE</w:instrText></w:r><w:r><w:fldChar w:fldCharType="separate"/></w:r><w:r><w:rPr><w:sz w:val="40"/></w:rPr><w:t/></w:r><w:r><w:fldChar w:fldCharType="end"/></w:r></w:p>"#,
);
let p = paras(&d)[0];
assert_eq!(p.props_at(0).size, Some(6.0));
assert_eq!(p.props_at(0).italic, Some(true));
assert!(matches!(&p.objects[0], InlineObject::Field { result, .. } if result.is_empty()));
}
#[test]
fn shape_dashes_import_and_save() {
use wordcraft_doc::line_dash::LineDash;
for (xml, expected) in [
(r#"<a:prstDash val="dash"/>"#, LineDash::Dash),
(r#"<a:prstDash val="sysDashDot"/>"#, LineDash::SystemDashDot),
(
r#"<a:custDash><a:ds d="400000" sp="300000"/><a:ds d="100000" sp="300000"/></a:custDash>"#,
LineDash::Custom(vec![(400_000, 300_000), (100_000, 300_000)]),
),
(r#"<a:custDash><a:ds d="400%" sp="300%"/></a:custDash>"#, LineDash::Custom(vec![(400_000, 300_000)])),
(r#"<a:custDash><a:ds d="NaN%" sp="-1"/></a:custDash>"#, LineDash::Solid),
(r#"<a:prstDash val="solid"/>"#, LineDash::Solid),
("", LineDash::Solid),
] {
let sppr = format!(r#"<a:prstGeom prst="rect"/><a:noFill/><a:ln w="12700"><a:solidFill><a:srgbClr val="C00000"/></a:solidFill>{xml}</a:ln>"#);
let body = anchored_wsp("Annotation", 889000, 355600, &sppr, "", "<wps:cNvSpPr/>");
let d = read_body(&format!("<w:p>{body}</w:p>"));
let saved = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
for doc in [&d, &saved] {
let InlineObject::Shape { w, h, stroke, stroke_width, extra, .. } = &paras(doc)[0].objects[0] else { panic!("shape") };
assert_eq!(extra.dash, expected);
assert_eq!((*w, *h, *stroke, *stroke_width), (70.0, 28.0, Some(Rgb(192, 0, 0)), 1.0));
}
}
}
#[test]
fn bookmark_end_name_and_id_trim() {
let body = r#"<w:p>
<w:bookmarkStart w:id=" 42 " w:name="section_intro"/>
<w:r><w:t>Hello</w:t></w:r>
<w:bookmarkEnd w:id="42"/>
<w:bookmarkEnd w:id="999" w:name="standalone_target"/>
</w:p>"#;
let doc = wordcraft_docx::read(&docx(body, &[], &[])).unwrap();
let p = doc.body.first().unwrap().as_para().unwrap();
let names: Vec<(&'static str, String)> = p
.objects
.iter()
.filter_map(|o| match o {
InlineObject::BookmarkStart { name } => Some(("start", name.clone())),
InlineObject::BookmarkEnd { name } => Some(("end", name.clone())),
_ => None,
})
.collect();
assert_eq!(names, vec![("start", "section_intro".into()), ("end", "section_intro".into()), ("end", "standalone_target".into())]);
}
#[test]
fn zero_effects_stay_zero_rotated() {
const ANGLE_UNITS_PER_DEGREE: i32 = 60_000;
// ECMA-376 §20.4.2.6: zero effect extents add no space around a rotated picture.
for degrees in [0, 90, 180, 270] {
for effects in ["", r#"<wp:effectExtent l="0" t="0" r="0" b="0"/>"#] {
let body = format!(
r#"<w:p><w:r><w:drawing><wp:inline><wp:extent cx="914400" cy="457200"/>{effects}<wp:docPr id="1" name="Synthetic picture"/><a:graphic><a:graphicData uri="http://schemas.openxmlformats.org/drawingml/2006/picture"><pic:pic><pic:blipFill><a:blip r:embed="rId1"/></pic:blipFill><pic:spPr><a:xfrm rot="{}"><a:off x="0" y="0"/><a:ext cx="914400" cy="457200"/></a:xfrm></pic:spPr></pic:pic></a:graphicData></a:graphic></wp:inline></w:drawing></w:r></w:p>"#,
degrees * ANGLE_UNITS_PER_DEGREE
);
let d = wordcraft_docx::read(&docx(&body, &[("rId1", "image", "media/marker.png")], &[("word/media/marker.png", "synthetic")])).unwrap();
let (w, h, float) = paras(&d)[0].objects[0].frame().unwrap();
assert_eq!(float.rot, degrees as f32);
assert_eq!(float.effect, [0.0; 4], "rotation {degrees} must not invent effects");
let (px, py) = float.spin_pad(w, h);
let expected = if degrees % 180 == 0 { (w, h) } else { (h, w) };
assert!((w + 2.0 * px - expected.0).abs() < 0.001);
assert!((h + 2.0 * py - expected.1).abs() < 0.001);
}
}
}
/// The math namespace, declared on the math element itself (the fixture root doesn't).
const M_NS: &str = r#"xmlns:m="http://schemas.openxmlformats.org/officeDocument/2006/math""#;
fn equations(p: &Paragraph) -> Vec<(String, bool)> {
p.objects
.iter()
.filter_map(|o| match o {
InlineObject::Equation { linear, display, .. } => Some((linear.clone(), *display)),
_ => None,
})
.collect()
}
/// #1273: an equation straight in the body (EG_BlockLevelElts allows math content) gets a
/// paragraph of its own instead of vanishing.
#[test]
fn body_level_equation_is_kept() {
let d = read_body(&format!(
r#"<w:p><w:r><w:t>before</w:t></w:r></w:p><m:oMathPara {M_NS}><m:oMath><m:r><m:t>a=b</m:t></m:r></m:oMath></m:oMathPara><m:oMath {M_NS}><m:r><m:t>c</m:t></m:r></m:oMath><w:p><w:r><w:t>after</w:t></w:r></w:p>"#
));
let ps = paras(&d);
assert_eq!(ps.len(), 4, "{:?}", ps.iter().map(|p| p.plain_text()).collect::<Vec<_>>());
assert_eq!(equations(ps[1]), vec![("a=b".to_string(), true)]);
assert_eq!(equations(ps[2]), vec![("c".to_string(), false)]);
}
/// #1273: an equation guarded by markup compatibility (`Requires="m"`) is read as an equation,
/// not replaced by its fallback.
#[test]
fn equation_in_alternate_content_is_read() {
let d = read_body(&format!(
r#"<w:p><mc:AlternateContent {M_NS}><mc:Choice Requires="m"><m:oMath><m:r><m:t>x+1</m:t></m:r></m:oMath></mc:Choice><mc:Fallback><w:r><w:t>[picture]</w:t></w:r></mc:Fallback></mc:AlternateContent></w:p>"#
));
let p = paras(&d)[0];
assert_eq!(equations(p), vec![("x+1".to_string(), false)]);
assert!(!p.plain_text().contains("[picture]"));
}
+386
View File
@@ -0,0 +1,386 @@
//! The glossary document (ECMA-376 §17.12: building blocks, Quick Parts, AutoText): kept byte
//! for byte while unchanged, read into `Document::glossary`, and written from the model once
//! edited. Hand-written fixtures (from the spec), no files produced by Word.
use std::collections::BTreeMap;
use std::io::{Read, Write};
use wordcraft_doc::{Block, BuildingBlock, CharProps, Document, Paragraph, para_block};
const W: &str = "http://schemas.openxmlformats.org/wordprocessingml/2006/main";
const R_NS: &str = "http://schemas.openxmlformats.org/officeDocument/2006/relationships";
const REL: &str = "http://schemas.openxmlformats.org/officeDocument/2006/relationships";
const GLOSSARY_REL: &str = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/glossaryDocument";
const GLOSSARY_CT: &str = "application/vnd.openxmlformats-officedocument.wordprocessingml.document.glossary+xml";
fn zip(entries: &[(&str, Vec<u8>)]) -> Vec<u8> {
let mut zw = zip::ZipWriter::new(std::io::Cursor::new(Vec::new()));
for (name, bytes) in entries {
zw.start_file(*name, zip::write::SimpleFileOptions::default()).unwrap();
zw.write_all(bytes).unwrap();
}
zw.finish().unwrap().into_inner()
}
fn unzip(bytes: &[u8]) -> BTreeMap<String, Vec<u8>> {
let mut z = zip::ZipArchive::new(std::io::Cursor::new(bytes)).unwrap();
let mut out = BTreeMap::new();
for i in 0..z.len() {
let mut f = z.by_index(i).unwrap();
let mut b = Vec::new();
f.read_to_end(&mut b).unwrap();
assert!(out.insert(f.name().to_string(), b).is_none(), "duplicate part {}", f.name());
}
out
}
fn text(files: &BTreeMap<String, Vec<u8>>, name: &str) -> String {
String::from_utf8(files.get(name).unwrap_or_else(|| panic!("no {name} in {:?}", files.keys())).clone()).unwrap()
}
fn rels(list: &[(&str, &str, &str)]) -> String {
let mut s =
String::from(r#"<?xml version="1.0" encoding="UTF-8"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">"#);
for (id, ty, target) in list {
s.push_str(&format!(r#"<Relationship Id="{id}" Type="{ty}" Target="{target}"/>"#));
}
s.push_str("</Relationships>");
s
}
fn content_types(extra: &[(&str, &str)]) -> String {
let mut s = String::from(
r#"<?xml version="1.0" encoding="UTF-8"?><Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types"><Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/><Default Extension="xml" ContentType="application/xml"/><Override PartName="/word/document.xml" ContentType="application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml"/>"#,
);
for (part, ct) in extra {
s.push_str(&format!(r#"<Override PartName="{part}" ContentType="{ct}"/>"#));
}
s.push_str("</Types>");
s
}
/// A package with body text, a main document relating to `doc_rels`, and `extra` parts.
fn package(doc_rels: &[(&str, &str, &str)], extra: &[(&str, String)]) -> Vec<u8> {
let doc = format!(
r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:document xmlns:w="{W}" xmlns:r="{R_NS}"><w:body><w:p><w:r><w:t>Body</w:t></w:r></w:p></w:body></w:document>"#
);
let root = rels(&[("rIdMain", &format!("{REL}/officeDocument"), "word/document.xml")]);
let mut entries: Vec<(&str, Vec<u8>)> = vec![
("_rels/.rels", root.into_bytes()),
("word/document.xml", doc.into_bytes()),
("word/_rels/document.xml.rels", rels(doc_rels).into_bytes()),
];
for (p, c) in extra {
entries.push((p, c.as_bytes().to_vec()));
}
zip(&entries)
}
fn glossary_xml(doc_parts: &str) -> String {
format!(
r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:glossaryDocument xmlns:w="{W}" xmlns:r="{R_NS}"><w:docParts>{doc_parts}</w:docParts></w:glossaryDocument>"#
)
}
const SIGNATURE: &str = r#"<w:docPart><w:docPartPr><w:name w:val="Signature"/><w:style w:val="Normal"/><w:category><w:name w:val="General"/><w:gallery w:val="docParts"/></w:category><w:types><w:type w:val="normal"/></w:types><w:behaviors><w:behavior w:val="content"/></w:behaviors><w:description w:val="A closing line"/><w:guid w:val="{0F1E2D3C-4B5A-6978-8796-A5B4C3D2E1F0}"/></w:docPartPr><w:docPartBody><w:p><w:r><w:t>Kind regards, Ann</w:t></w:r></w:p></w:docPartBody></w:docPart>"#;
const PRICES: &str = r#"<w:docPart><w:docPartPr><w:name w:val="Price table"/><w:category><w:name w:val="Sales"/><w:gallery w:val="autoTxt"/></w:category><w:types><w:type w:val="bbPlcHdr"/><w:type w:val="autoExp"/><w:type w:val="madeUp"/></w:types><w:behaviors><w:behavior w:val="p"/><w:behavior w:val="pg"/></w:behaviors><w:guid w:val="{11111111-2222-3333-4444-555555555555}"/></w:docPartPr><w:docPartBody><w:tbl><w:tblPr><w:tblW w:w="0" w:type="auto"/></w:tblPr><w:tblGrid><w:gridCol w:w="2000"/><w:gridCol w:w="2000"/></w:tblGrid><w:tr><w:tc><w:p><w:r><w:t>Item</w:t></w:r></w:p></w:tc><w:tc><w:p><w:r><w:t>Price</w:t></w:r></w:p></w:tc></w:tr></w:tbl><w:p><w:r><w:t>Prices exclude tax.</w:t></w:r></w:p></w:docPartBody></w:docPart>"#;
const GLOSSARY_STYLES: &str = r#"<w:styles xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:style w:type="paragraph" w:default="1" w:styleId="Normal"><w:name w:val="Normal"/></w:style></w:styles>"#;
const GLOSSARY_WEB: &str =
r#"<w:webSettings xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:optimizeForBrowser/></w:webSettings>"#;
/// A document whose glossary holds a Quick Part and an AutoText entry (with a table), with the
/// glossary's own styles and web settings.
fn with_glossary() -> Vec<u8> {
package(
&[("rId7", GLOSSARY_REL, "glossary/document.xml")],
&[
(
"[Content_Types].xml",
content_types(&[
("/word/glossary/document.xml", GLOSSARY_CT),
("/word/glossary/styles.xml", "application/vnd.openxmlformats-officedocument.wordprocessingml.styles+xml"),
("/word/glossary/webSettings.xml", "application/vnd.openxmlformats-officedocument.wordprocessingml.webSettings+xml"),
]),
),
("word/glossary/document.xml", glossary_xml(&format!("{SIGNATURE}{PRICES}"))),
(
"word/glossary/_rels/document.xml.rels",
rels(&[("rId1", &format!("{REL}/styles"), "styles.xml"), ("rId2", &format!("{REL}/webSettings"), "webSettings.xml")]),
),
("word/glossary/styles.xml", GLOSSARY_STYLES.to_string()),
("word/glossary/webSettings.xml", GLOSSARY_WEB.to_string()),
],
)
}
fn save(doc: &Document) -> BTreeMap<String, Vec<u8>> {
unzip(&wordcraft_docx::write(doc).expect("write"))
}
fn block_text(b: &Block) -> String {
match b {
Block::Para(p) => p.text.clone(),
Block::Table(t) => {
t.rows.iter().flat_map(|r| r.cells.iter()).flat_map(|c| c.blocks.iter()).map(|b| block_text(b)).collect::<Vec<_>>().join("|")
}
}
}
fn entry_text(e: &BuildingBlock) -> String {
e.blocks.iter().map(|b| block_text(b)).collect::<Vec<_>>().join("\n")
}
/// The glossary part a saved package's main document relates to.
fn glossary_target(files: &BTreeMap<String, Vec<u8>>) -> Option<String> {
let doc_rels = text(files, "word/_rels/document.xml.rels");
assert!(doc_rels.matches("/glossaryDocument\"").count() <= 1, "{doc_rels}");
let at = doc_rels.find("/glossaryDocument\"")?;
let rest = doc_rels.get(at..)?;
let t = rest.find("Target=\"")? + "Target=\"".len();
let end = rest.get(t..)?.find('"')?;
Some(format!("word/{}", rest.get(t..t + end)?))
}
#[test]
fn unchanged_glossary_is_carried_byte_for_byte() {
let bytes = with_glossary();
let mut doc = wordcraft_docx::read(&bytes).unwrap();
doc.body.push(para_block(Paragraph::with_text("Added after opening", CharProps::default())));
let files = save(&doc);
let input = unzip(&bytes);
assert_eq!(glossary_target(&files).as_deref(), Some("word/glossary/document.xml"));
for part in ["word/glossary/document.xml", "word/glossary/styles.xml", "word/glossary/webSettings.xml"] {
assert_eq!(files.get(part), input.get(part), "{part} not carried verbatim");
}
let grels = text(&files, "word/glossary/_rels/document.xml.rels");
assert!(grels.contains("styles.xml") && grels.contains("webSettings.xml"), "{grels}");
let ct = text(&files, "[Content_Types].xml");
assert!(ct.contains(&format!(r#"PartName="/word/glossary/document.xml" ContentType="{GLOSSARY_CT}""#)), "{ct}");
assert!(ct.contains(r#"PartName="/word/glossary/styles.xml""#), "{ct}");
let again = wordcraft_docx::read(&wordcraft_docx::write(&doc).unwrap()).unwrap();
assert_eq!(again.glossary.len(), 2);
assert!(again.plain_text(wordcraft_doc::StoryRef::Body).contains("Added after opening"));
}
#[test]
fn glossary_entries_are_read_into_the_model() {
let doc = wordcraft_docx::read(&with_glossary()).unwrap();
let g = &doc.glossary;
assert_eq!(g.iter().map(|e| e.name.as_str()).collect::<Vec<_>>(), ["Signature", "Price table"]);
let sig = &g[0];
assert_eq!((sig.gallery.as_str(), sig.category.as_str()), ("docParts", "General"));
assert_eq!(sig.description, "A closing line");
assert_eq!(sig.guid, "{0F1E2D3C-4B5A-6978-8796-A5B4C3D2E1F0}");
assert_eq!(
(sig.style.as_str(), sig.types.as_slice(), sig.behaviors.as_slice()),
("Normal", &["normal".to_string()][..], &["content".to_string()][..])
);
assert_eq!(entry_text(sig), "Kind regards, Ann");
let prices = &g[1];
assert_eq!((prices.gallery.as_str(), prices.category.as_str()), ("autoTxt", "Sales"));
// Types and behaviours outside the spec's lists aren't kept.
assert_eq!(prices.types, ["bbPlcHdr", "autoExp"]);
assert_eq!(prices.behaviors, ["p", "pg"]);
assert!(matches!(&*prices.blocks[0], Block::Table(_)), "{:?}", prices.blocks);
assert_eq!(entry_text(prices), "Item|Price\nPrices exclude tax.");
// The glossary's text isn't the document's.
assert!(!doc.plain_text(wordcraft_doc::StoryRef::Body).contains("Kind regards"));
}
#[test]
fn edited_glossary_is_written_from_the_model() {
let mut doc = wordcraft_docx::read(&with_glossary()).unwrap();
doc.glossary[0].name = "Sign-off".into();
let num = doc.numbering.add_list(wordcraft_doc::ListKind::Bullet);
let mut listed = Paragraph::with_text("Second line", CharProps::default());
listed.props.numbering = Some(wordcraft_doc::props::NumRef { num, level: 0 });
doc.glossary.push(BuildingBlock {
name: "Address".into(),
gallery: "docParts".into(),
category: "General".into(),
blocks: vec![para_block(Paragraph::with_text("1 Main Street", CharProps::default())), para_block(listed)],
..Default::default()
});
let files = save(&doc);
let target = glossary_target(&files).expect("a glossary relationship");
let g = text(&files, &target);
assert!(g.contains("w:glossaryDocument") && g.contains("Sign-off") && g.contains("1 Main Street"), "{g}");
assert!(!files.values().any(|b| String::from_utf8_lossy(b).contains(r#"w:val="Signature""#)), "the old glossary is gone");
let ct = text(&files, "[Content_Types].xml");
assert!(ct.contains(&format!(r#"PartName="/{target}" ContentType="{GLOSSARY_CT}""#)), "{ct}");
let again = wordcraft_docx::read(&wordcraft_docx::write(&doc).unwrap()).unwrap();
assert_eq!(again.glossary.iter().map(|e| e.name.as_str()).collect::<Vec<_>>(), ["Sign-off", "Price table", "Address"]);
assert_eq!(entry_text(&again.glossary[1]), "Item|Price\nPrices exclude tax.");
assert_eq!(again.glossary[1].gallery, "autoTxt");
assert_eq!(again.glossary[0].guid, "{0F1E2D3C-4B5A-6978-8796-A5B4C3D2E1F0}");
assert_eq!(again.glossary[0].description, "A closing line");
assert_eq!(entry_text(&again.glossary[2]), "1 Main Street\nSecond line");
// Style, types and behaviours survive the rewrite.
for (a, b) in again.glossary.iter().zip(&doc.glossary) {
assert_eq!((&a.style, &a.types, &a.behaviors), (&b.style, &b.types, &b.behaviors), "{}", a.name);
}
assert_eq!(again.glossary[1].types, ["bbPlcHdr", "autoExp"]);
assert_eq!(again.glossary[1].behaviors, ["p", "pg"]);
assert_eq!(again.glossary[0].style, "Normal");
// A list in an entry has its definitions beside the glossary (its own numbering part).
let grels = text(&files, &glossary_rels(&target));
assert!(grels.contains("/numbering\""), "{grels}");
let numbering = grels.split("/numbering\" Target=\"").nth(1).and_then(|r| r.split('"').next()).expect("numbering target");
let numbering = text(&files, &format!("word/{numbering}"));
assert!(numbering.contains("w:abstractNum") && numbering.contains(&format!(r#"w:numId="{num}""#)), "{numbering}");
// Saving the reopened document again keeps its (now unchanged) glossary verbatim.
let twice = save(&again);
assert_eq!(glossary_target(&twice).as_deref(), Some(target.as_str()));
assert_eq!(twice.get(&target), files.get(&target));
}
/// The relationships part of a part in `word/`.
fn glossary_rels(part: &str) -> String {
let (dir, file) = part.rsplit_once('/').unwrap();
format!("{dir}/_rels/{file}.rels")
}
#[test]
fn an_unchanged_glossary_whose_bytes_werent_kept_is_written_from_the_model() {
let mut doc = wordcraft_docx::read(&with_glossary()).unwrap();
// The verbatim copy is gone (over the kept-part caps, say): the model is all there is.
doc.passthrough.remove("word/glossary/document.xml");
let files = save(&doc);
let target = glossary_target(&files).expect("a glossary relationship");
assert!(text(&files, &target).contains("Kind regards, Ann"));
let again = wordcraft_docx::read(&wordcraft_docx::write(&doc).unwrap()).unwrap();
assert_eq!(again.glossary.iter().map(|e| e.name.as_str()).collect::<Vec<_>>(), ["Signature", "Price table"]);
}
#[test]
fn an_external_glossary_relationship_is_passed_over() {
let doc_rels = format!(
r#"<?xml version="1.0" encoding="UTF-8"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId6" Type="{GLOSSARY_REL}" Target="https://example.com/g.xml" TargetMode="External"/><Relationship Id="rId7" Type="{GLOSSARY_REL}" Target="glossary/document.xml"/></Relationships>"#
);
let doc = format!(
r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:document xmlns:w="{W}" xmlns:r="{R_NS}"><w:body><w:p><w:r><w:t>Body</w:t></w:r></w:p></w:body></w:document>"#
);
let bytes = zip(&[
("[Content_Types].xml", content_types(&[("/word/glossary/document.xml", GLOSSARY_CT)]).into_bytes()),
("_rels/.rels", rels(&[("rIdMain", &format!("{REL}/officeDocument"), "word/document.xml")]).into_bytes()),
("word/document.xml", doc.into_bytes()),
("word/_rels/document.xml.rels", doc_rels.into_bytes()),
("word/glossary/document.xml", glossary_xml(SIGNATURE).into_bytes()),
]);
let doc = wordcraft_docx::read(&bytes).unwrap();
assert_eq!(doc.glossary.iter().map(|e| e.name.as_str()).collect::<Vec<_>>(), ["Signature"]);
// The same glossary is the one kept: saved unchanged, it's written back as it was.
let files = save(&doc);
assert_eq!(files.get("word/glossary/document.xml"), unzip(&bytes).get("word/glossary/document.xml"));
}
#[test]
fn a_document_without_glossary_writes_none() {
let files = save(&Document::new());
assert_eq!(glossary_target(&files), None);
assert!(!text(&files, "[Content_Types].xml").contains("glossary"));
// Removing every entry removes the glossary.
let mut doc = wordcraft_docx::read(&with_glossary()).unwrap();
doc.glossary.clear();
let files = save(&doc);
assert_eq!(glossary_target(&files), None);
assert!(!files.keys().any(|k| k.contains("glossary")), "{:?}", files.keys());
assert!(!text(&files, "[Content_Types].xml").contains("glossary"));
assert!(wordcraft_docx::read(&wordcraft_docx::write(&doc).unwrap()).unwrap().glossary.is_empty());
}
#[test]
fn a_new_entry_in_a_new_document_makes_a_glossary() {
let mut doc = Document::new();
doc.glossary.push(BuildingBlock {
name: "Greeting".into(),
gallery: "docParts".into(),
category: "General".into(),
blocks: vec![para_block(Paragraph::with_text("Hello there", CharProps::default()))],
..Default::default()
});
let again = wordcraft_docx::read(&wordcraft_docx::write(&doc).unwrap()).unwrap();
assert_eq!(again.glossary.len(), 1);
assert_eq!(again.glossary[0].name, "Greeting");
assert_eq!(entry_text(&again.glossary[0]), "Hello there");
assert!(again.plain_text(wordcraft_doc::StoryRef::Body).trim().is_empty());
}
/// Hostile glossaries open without a panic; entries that can't be read are skipped.
#[test]
fn hostile_glossaries_never_panic() {
let ct = content_types(&[("/word/glossary/document.xml", GLOSSARY_CT)]);
let one = |g: String| {
package(&[("rId7", GLOSSARY_REL, "glossary/document.xml")], &[("[Content_Types].xml", ct.clone()), ("word/glossary/document.xml", g)])
};
let read = |bytes: Vec<u8>| -> Document {
let doc = wordcraft_docx::read(&bytes).unwrap();
// Saving and reopening must not panic either.
let _ = wordcraft_docx::read(&wordcraft_docx::write(&doc).unwrap()).unwrap();
doc
};
// No body, an empty body, no name, no properties: skipped.
let doc = read(one(glossary_xml(
r#"<w:docPart><w:docPartPr><w:name w:val="NoBody"/></w:docPartPr></w:docPart><w:docPart><w:docPartPr><w:name w:val="Empty"/></w:docPartPr><w:docPartBody/></w:docPart><w:docPart><w:docPartPr/><w:docPartBody><w:p><w:r><w:t>nameless</w:t></w:r></w:p></w:docPartBody></w:docPart><w:docPart><w:docPartBody><w:p/></w:docPartBody></w:docPart>"#,
)));
assert!(doc.glossary.iter().all(|e| !e.name.is_empty() && !e.blocks.is_empty()), "{:?}", doc.glossary);
assert!(doc.glossary.iter().all(|e| e.name != "NoBody"));
// Duplicate names: both read, file order.
let doc = read(one(glossary_xml(&format!("{SIGNATURE}{SIGNATURE}"))));
assert_eq!(doc.glossary.len(), 2);
// Huge counts are capped.
let many: String = (0..12_000)
.map(|i| format!(r#"<w:docPart><w:docPartPr><w:name w:val="e{i}"/></w:docPartPr><w:docPartBody><w:p/></w:docPartBody></w:docPart>"#))
.collect();
let doc = read(one(glossary_xml(&many)));
assert!(doc.glossary.len() <= 10_000, "{}", doc.glossary.len());
// Deep nesting (tables in tables).
let deep = format!(
r#"<w:docPart><w:docPartPr><w:name w:val="Deep"/></w:docPartPr><w:docPartBody>{}<w:p/>{}</w:docPartBody></w:docPart>"#,
"<w:tbl><w:tr><w:tc>".repeat(400),
"</w:tc></w:tr></w:tbl>".repeat(400)
);
let _ = read(one(glossary_xml(&deep)));
// Not a glossary at all, or a wrong content type: nothing read.
let doc = read(one(format!(r#"<w:document xmlns:w="{W}"><w:body><w:p><w:r><w:t>x</w:t></w:r></w:p></w:body></w:document>"#)));
assert!(doc.glossary.is_empty());
let wrong = package(
&[("rId7", GLOSSARY_REL, "glossary/document.xml")],
&[
("[Content_Types].xml", content_types(&[("/word/glossary/document.xml", "image/png")])),
("word/glossary/document.xml", glossary_xml(SIGNATURE)),
],
);
assert!(read(wrong).glossary.is_empty());
// Broken XML.
let _ = read(one("<w:glossaryDocument><w:docParts><w:docPart>".into()));
// A glossary relationship pointing at the main document, or a missing part.
for target in ["document.xml", "missing.xml", "../../../../etc/passwd"] {
let doc = read(package(&[("rId7", GLOSSARY_REL, target)], &[("[Content_Types].xml", ct.clone())]));
assert!(doc.glossary.is_empty(), "{target}");
assert!(!doc.plain_text(wordcraft_doc::StoryRef::Body).contains("Kind regards"));
}
// A glossary relating to itself, or with a glossary of its own.
for own in
[rels(&[("rId1", GLOSSARY_REL, "document.xml")]), rels(&[("rId1", GLOSSARY_REL, "inner.xml"), ("rId2", GLOSSARY_REL, "../document.xml")])]
{
let bytes = package(
&[("rId7", GLOSSARY_REL, "glossary/document.xml")],
&[
("[Content_Types].xml", content_types(&[("/word/glossary/document.xml", GLOSSARY_CT), ("/word/glossary/inner.xml", GLOSSARY_CT)])),
("word/glossary/document.xml", glossary_xml(SIGNATURE)),
("word/glossary/_rels/document.xml.rels", own),
("word/glossary/inner.xml", glossary_xml(PRICES)),
],
);
let doc = read(bytes);
assert_eq!(doc.glossary.iter().map(|e| e.name.as_str()).collect::<Vec<_>>(), ["Signature"]);
}
// Two glossary relationships: the first counts.
let two = package(
&[("rId7", GLOSSARY_REL, "glossary/document.xml"), ("rId8", GLOSSARY_REL, "glossary/document.xml")],
&[("[Content_Types].xml", ct.clone()), ("word/glossary/document.xml", glossary_xml(SIGNATURE))],
);
let doc = read(two);
assert_eq!(doc.glossary.len(), 1);
}
+77
View File
@@ -0,0 +1,77 @@
//! OpenType ligatures (`w14:ligatures`, [MS-DOCX]) in document defaults, styles and runs.
//! Synthetic packages, written here; no external documents are needed.
use std::io::Write;
use wordcraft_doc::props::Ligatures;
use wordcraft_doc::{Document, Paragraph};
fn package(defaults: &str, style: &str, run: &str) -> Vec<u8> {
let ns = r#"xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:w14="http://schemas.microsoft.com/office/word/2010/wordml" xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006" mc:Ignorable="w14""#;
let root = r#"<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="r1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="word/document.xml"/></Relationships>"#;
let rels = r#"<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="s1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/styles" Target="styles.xml"/></Relationships>"#;
let document = format!(
r#"<w:document {ns}><w:body><w:p><w:r><w:rPr><w:rStyle w:val="Fancy"/>{run}</w:rPr><w:t>office affinity</w:t></w:r></w:p></w:body></w:document>"#
);
let styles = format!(
r#"<w:styles {ns}><w:docDefaults><w:rPrDefault><w:rPr>{defaults}</w:rPr></w:rPrDefault></w:docDefaults><w:style w:type="character" w:styleId="Fancy"><w:name w:val="Fancy"/><w:rPr>{style}</w:rPr></w:style></w:styles>"#
);
let mut z = zip::ZipWriter::new(std::io::Cursor::new(Vec::new()));
for (name, data) in [
("_rels/.rels", root),
("word/_rels/document.xml.rels", rels),
("word/document.xml", document.as_str()),
("word/styles.xml", styles.as_str()),
] {
z.start_file(name, zip::write::SimpleFileOptions::default()).unwrap();
z.write_all(data.as_bytes()).unwrap();
}
z.finish().unwrap().into_inner()
}
fn paragraph(d: &Document) -> &Paragraph {
d.body.iter().find_map(|b| b.as_para()).unwrap()
}
fn lig(name: &str) -> Option<Ligatures> {
Some(Ligatures::from_ooxml(name).unwrap())
}
/// Word stores the Font dialog's Ligatures choice as `w14:ligatures`; each level keeps its own
/// value through reading and saving, and an explicit `none` is not the same as no value.
#[test]
fn ligatures_in_defaults_styles_and_runs_survive_saving() {
let el = |v: &str| format!(r#"<w14:ligatures w14:val="{v}"/>"#);
let bytes = package(&el("standardContextual"), &el("none"), &el("historicalDiscretional"));
let mut d = wordcraft_docx::read(&bytes).unwrap();
for generation in 0..3 {
assert_eq!(d.styles.default_chr.ligatures, lig("standardContextual"), "defaults, generation {generation}");
assert_eq!(d.styles.get("Fancy").and_then(|s| s.chr.ligatures), lig("none"), "style, generation {generation}");
let run = &paragraph(&d).runs[0].props;
assert_eq!(run.ligatures, lig("historicalDiscretional"), "run, generation {generation}");
assert_eq!(run.text_effects, None, "not read as a WordArt effect");
d = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
}
}
/// A value [MS-DOCX] doesn't define is ignored, the run keeps inheriting.
#[test]
fn unknown_ligature_values_are_ignored() {
let d = wordcraft_docx::read(&package("", "", r#"<w14:ligatures w14:val="fancy"/>"#)).unwrap();
assert_eq!(paragraph(&d).runs[0].props.ligatures, None);
let d = wordcraft_docx::read(&package("", "", "<w14:ligatures/>")).unwrap();
assert_eq!(paragraph(&d).runs[0].props.ligatures, None);
}
/// Without a value nothing is written, so documents that never asked for ligatures stay as they
/// were.
#[test]
fn no_ligatures_element_without_a_value() {
let d = wordcraft_docx::read(&package("", "", "")).unwrap();
let out = wordcraft_docx::write(&d).unwrap();
let mut z = zip::ZipArchive::new(std::io::Cursor::new(out)).unwrap();
for name in ["word/document.xml", "word/styles.xml"] {
let mut xml = String::new();
std::io::Read::read_to_string(&mut z.by_name(name).unwrap(), &mut xml).unwrap();
assert!(!xml.contains("ligatures"), "{name}");
}
}
+70
View File
@@ -0,0 +1,70 @@
//! Positional tabs (`w:ptab`, ECMA-376 Part 1 §17.3.3.23). Fixtures are hand-written from the spec.
use std::io::{Read, Write};
use wordcraft_doc::InlineObject;
use wordcraft_doc::para::{PTabAlign, PTabBase};
use wordcraft_doc::props::TabLeader;
fn docx(body: &str) -> Vec<u8> {
let root = r#"<?xml version="1.0" encoding="UTF-8"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="word/document.xml"/></Relationships>"#;
let doc = format!(
r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:document xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:body>{body}</w:body></w:document>"#
);
let mut zw = zip::ZipWriter::new(std::io::Cursor::new(Vec::new()));
for (name, bytes) in [("_rels/.rels", root.as_bytes()), ("word/document.xml", doc.as_bytes())] {
zw.start_file(name, zip::write::SimpleFileOptions::default()).unwrap();
zw.write_all(bytes).unwrap();
}
zw.finish().unwrap().into_inner()
}
fn document_xml(bytes: &[u8]) -> String {
let mut z = zip::ZipArchive::new(std::io::Cursor::new(bytes)).unwrap();
let mut s = String::new();
z.by_name("word/document.xml").unwrap().read_to_string(&mut s).unwrap();
s
}
fn ptab(relative_to: &str, alignment: &str, leader: &str) -> String {
format!(r#"<w:r><w:ptab w:relativeTo="{relative_to}" w:alignment="{alignment}" w:leader="{leader}"/></w:r>"#)
}
#[test]
fn positional_tabs_keep_their_alignment_base_and_leader() {
let body = format!(
r#"<w:p><w:r><w:t>A</w:t></w:r>{}<w:r><w:t>B</w:t></w:r>{}<w:r><w:t>C</w:t></w:r>{}<w:r><w:t>D</w:t></w:r></w:p>"#,
ptab("margin", "right", "dot"),
ptab("indent", "center", "none"),
ptab("margin", "left", "underscore"),
);
let d = wordcraft_docx::read(&docx(&body)).unwrap();
let p = d.body.iter().find_map(|b| b.as_para()).unwrap();
assert_eq!(p.plain_text(), "A\tB\tC\tD");
assert_eq!(
p.objects,
vec![
InlineObject::PositionalTab { align: PTabAlign::Right, relative_to: PTabBase::Margin, leader: TabLeader::Dot },
InlineObject::PositionalTab { align: PTabAlign::Center, relative_to: PTabBase::Indent, leader: TabLeader::None },
InlineObject::PositionalTab { align: PTabAlign::Left, relative_to: PTabBase::Margin, leader: TabLeader::Underscore },
]
);
let xml = document_xml(&wordcraft_docx::write(&d).unwrap());
for (rel, align, leader) in [("margin", "right", "dot"), ("indent", "center", "none"), ("margin", "left", "underscore")] {
let want = format!(r#"<w:ptab w:relativeTo="{rel}" w:alignment="{align}" w:leader="{leader}"/>"#);
assert!(xml.contains(&want), "{want} missing from {xml}");
}
assert!(!xml.contains("<w:tab/>"), "{xml}");
}
/// A positional tab whose alignment or base isn't one the spec defines stays an ordinary tab.
#[test]
fn a_positional_tab_with_an_unknown_alignment_is_a_plain_tab() {
let body = format!(r#"<w:p><w:r><w:t>A</w:t></w:r>{}<w:r><w:t>B</w:t></w:r></w:p>"#, ptab("margin", "decimal", "none"));
let d = wordcraft_docx::read(&docx(&body)).unwrap();
let p = d.body.iter().find_map(|b| b.as_para()).unwrap();
assert_eq!(p.plain_text(), "A\tB");
assert!(p.objects.is_empty(), "{:?}", p.objects);
let xml = document_xml(&wordcraft_docx::write(&d).unwrap());
assert!(xml.contains("<w:tab/>") && !xml.contains("w:ptab"), "{xml}");
}
+78
View File
@@ -56,6 +56,8 @@ fn every_char_prop_round_trips() {
let all = CharProps {
style: Some("Strong".into()),
font: Some("Georgia".into()),
// The symbol's own font: the run is written as `w:sym` characters (#2036).
sym_font: Some("Wingdings".into()),
size: Some(10.5),
bold: Some(true),
italic: Some(false),
@@ -75,6 +77,7 @@ fn every_char_prop_round_trips() {
scale: Some(150.0),
position: Some(3.0),
kern: Some(8.0),
ligatures: Some(wordcraft_doc::props::Ligatures { standard: true, historical: true, ..Default::default() }),
outline: Some(true),
shadow: Some(true),
emboss: Some(false),
@@ -110,6 +113,13 @@ fn every_char_prop_round_trips() {
bold_cs: Some(false),
italic_cs: Some(true),
lang_bidi: Some("fa-IR".into()),
east_asian_layout: Some(wordcraft_doc::props::EastAsianLayout {
id: Some(3),
combine: Some(true),
combine_brackets: Some(wordcraft_doc::props::CombineBrackets::Square),
vert: Some(false),
vert_compress: Some(true),
}),
link: None,
ins: None,
del: None,
@@ -165,6 +175,9 @@ fn every_para_prop_round_trips() {
suppress_line_numbers: Some(true),
bidi: Some(false),
drop_cap: None,
drop_cap_margin: None,
drop_cap_distance: None,
drop_cap_font: None,
frame: Some(wordcraft_doc::props::TextFrame {
w: Some(144.0),
h: Some(36.0),
@@ -707,6 +720,32 @@ fn note_never_references_itself() {
assert_eq!(part_text(&r, Some(*id)).trim(), "The note.");
}
/// A note with a custom mark keeps it in the note's own text: a plain run with the mark, not the
/// `w:footnoteRef` / `w:endnoteRef` placeholder that Word replaces by the automatic number.
#[test]
fn custom_note_mark_stays_in_the_note() {
let mut d = Document::new();
let f = d.add_part(PartKind::Footnote, Vec::new());
let e = d.add_part(PartKind::Endnote, Vec::new());
for (id, kind) in [(f, NoteKind::Footnote), (e, NoteKind::Endnote)] {
let mut note = Paragraph::with_text(" The note.", CharProps::default());
note.insert_object(0, InlineObject::NoteRef { kind, id, custom: "*".into() }, &CharProps::default()).unwrap();
d.parts.get_mut(&id).unwrap().blocks = vec![para_block(note)];
}
let mut p = Paragraph::with_text("Text", CharProps::default());
p.insert_object(4, InlineObject::NoteRef { kind: NoteKind::Footnote, id: f, custom: "*".into() }, &CharProps::default()).unwrap();
let end = p.len();
p.insert_object(end, InlineObject::NoteRef { kind: NoteKind::Endnote, id: e, custom: "*".into() }, &CharProps::default()).unwrap();
d.body = vec![para_block(p)];
let bytes = wordcraft_docx::write(&d).expect("write");
for (file, mark) in [("word/footnotes.xml", "<w:footnoteRef/>"), ("word/endnotes.xml", "<w:endnoteRef/>")] {
let xml = part_xml(&bytes, file);
assert!(!xml.contains(mark), "{file}: the custom mark became an automatic number: {xml}");
assert!(xml.contains(r#"<w:t xml:space="preserve">*</w:t>"#), "{file}: {xml}");
}
}
/// A note read from a Word file starts with its own number (a reference to itself, from the
/// `w:footnoteRef` mark). Saving writes that mark back exactly once, with its character style, and
/// never as a `w:footnoteReference`; reading the saved file gives the same note again.
@@ -2606,3 +2645,42 @@ fn page_background_picture_round_trips() {
// Saved again, it is still there.
assert!(rt(&r).settings.page_picture.is_some());
}
/// #2601: Drop Cap Options — a dropped cap with its lines, distance from the text and font, and
/// one in the margin, survive a save: `w:framePr w:dropCap="drop|margin" w:lines w:hSpace` on the
/// letter's own paragraph, the font on the letter's run (ECMA-376 §17.3.1.11). The text after the
/// letter keeps its own font.
#[test]
fn drop_cap_options_round_trip() {
let body = CharProps { font: Some("Liberation Serif".into()), ..Default::default() };
let mut dropped = Paragraph::with_text("Once upon a time", body.clone());
dropped.props.drop_cap = Some(2);
dropped.props.drop_cap_distance = Some(6.0);
dropped.props.drop_cap_font = Some("Liberation Mono".into());
let mut margin = Paragraph::with_text("In the margin", body.clone());
margin.props.drop_cap = Some(4);
margin.props.drop_cap_margin = Some(true);
let d = doc_with(vec![dropped, margin]);
let bytes = wordcraft_docx::write(&d).expect("write");
let mut z = zip::ZipArchive::new(std::io::Cursor::new(&bytes)).unwrap();
let mut xml = String::new();
std::io::Read::read_to_string(&mut z.by_name("word/document.xml").unwrap(), &mut xml).unwrap();
assert!(xml.contains(r#"w:dropCap="drop" w:lines="2" w:hSpace="120""#), "{xml}");
assert!(xml.contains(r#"w:dropCap="margin" w:lines="4""#), "{xml}");
assert!(xml.contains(r#"w:ascii="Liberation Mono""#), "the letter's font on its run: {xml}");
let r = wordcraft_docx::read(&bytes).expect("read");
let got = paras(&r);
assert_eq!(got.len(), 2, "each letter merges back into its paragraph");
let (a, b) = (&got[0], &got[1]);
assert_eq!(a.plain_text(), "Once upon a time");
assert_eq!(
(a.props.drop_cap, a.props.drop_cap_margin, a.props.drop_cap_distance, a.props.drop_cap_font.as_deref()),
(Some(2), None, Some(6.0), Some("Liberation Mono"))
);
assert!(a.runs.iter().all(|r| r.props.font.as_deref() == Some("Liberation Serif")), "the letter keeps the text's font: {:?}", a.runs);
assert_eq!(b.plain_text(), "In the margin");
assert_eq!(
(b.props.drop_cap, b.props.drop_cap_margin, b.props.drop_cap_distance, b.props.drop_cap_font.as_deref()),
(Some(4), Some(true), None, None)
);
}
+30
View File
@@ -133,3 +133,33 @@ fn settings_written_by_wordcraft_keep_no_original() {
assert!(!back.passthrough.contains_key("wordcraft:settings.orig"), "{:?}", back.passthrough.keys());
assert_eq!(text(&save(&back), "word/settings.xml"), text(&save(&doc), "word/settings.xml"));
}
/// Gap #5: `w:doNotTrackMoves` (ECMA-376 §17.15.1.29) turns Track moves off; it is read, written
/// back right after `w:trackRevisions` (in a merged part and a generated one), and left out
/// while moves are tracked.
#[test]
fn do_not_track_moves_round_trips() {
let original = SETTINGS.replace("<w:trackRevisions/>", "<w:trackRevisions/><w:doNotTrackMoves/>");
let mut doc = wordcraft_docx::read(&with_settings(&original)).unwrap();
assert!(!doc.settings.track_moves);
let s = text(&save(&doc), "word/settings.xml");
assert_eq!(s.matches("<w:doNotTrackMoves/>").count(), 1, "{s}");
assert!(pos(&s, "<w:trackRevisions/>") < pos(&s, "<w:doNotTrackMoves/>") && pos(&s, "<w:doNotTrackMoves/>") < pos(&s, "<w:defaultTabStop"));
assert!(s.contains("ClauseSet"), "a merged part: {s}");
doc.settings.track_moves = true;
let s = text(&save(&doc), "word/settings.xml");
assert!(!s.contains("doNotTrackMoves"), "{s}");
// `w:val="false"` keeps moves tracked.
let off = SETTINGS.replace("<w:trackRevisions/>", r#"<w:trackRevisions/><w:doNotTrackMoves w:val="false"/>"#);
assert!(wordcraft_docx::read(&with_settings(&off)).unwrap().settings.track_moves);
// A document of our own: absent by default, written when off, and read back.
let mut doc = wordcraft_doc::Document::new();
assert!(doc.settings.track_moves);
assert!(!text(&save(&doc), "word/settings.xml").contains("doNotTrackMoves"));
doc.settings.track_moves = false;
let s = text(&save(&doc), "word/settings.xml");
assert!(s.contains("<w:doNotTrackMoves/>"), "{s}");
assert!(!wordcraft_docx::read(&wordcraft_docx::write(&doc).unwrap()).unwrap().settings.track_moves);
}
+3 -2
View File
@@ -24,10 +24,11 @@ fn document_xml(bytes: &[u8]) -> String {
s
}
/// The text boxes' stories in body order.
/// The text boxes' stories in body order, a group's members among them
/// (`text_boxes`, not `text_box`, which only sees an object that is itself a box).
fn boxes(d: &Document) -> Vec<u32> {
let mut out = Vec::new();
d.objects_in_reading_order(&mut |o| out.extend(o.text_box()));
d.objects_in_reading_order(&mut |o| out.extend(o.text_boxes()));
out.retain(|id| d.parts.get(id).is_some_and(|p| p.kind == PartKind::TextBox));
out
}
+272
View File
@@ -0,0 +1,272 @@
//! Legacy VML drawings in a `w:pict`: every shape in one, and `v:group` members in the space
//! `coordsize`/`coordorigin` give them. Fixtures are hand-written from ECMA-376 Part 4
//! (VML, §14.1 `v:group`, `coordsize`, `coordorigin`; §14.2 shapes).
use std::io::Write;
use wordcraft_doc::para::ShapeKind;
use wordcraft_doc::{Document, InlineObject, Paragraph};
fn zip(entries: &[(&str, &[u8])]) -> Vec<u8> {
let mut zw = zip::ZipWriter::new(std::io::Cursor::new(Vec::new()));
for (name, bytes) in entries {
zw.start_file(*name, zip::write::SimpleFileOptions::default()).unwrap();
zw.write_all(bytes).unwrap();
}
zw.finish().unwrap().into_inner()
}
/// A document whose body is `body`.
fn read_body(body: &str) -> Document {
let root = r#"<?xml version="1.0" encoding="UTF-8"?><Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="word/document.xml"/></Relationships>"#;
let doc = format!(
r#"<?xml version="1.0" encoding="UTF-8" standalone="yes"?><w:document xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w10="urn:schemas-microsoft-com:office:word"><w:body>{body}</w:body></w:document>"#
);
wordcraft_docx::read(&zip(&[("_rels/.rels", root.as_bytes()), ("word/document.xml", doc.as_bytes())])).unwrap()
}
fn paras(d: &Document) -> Vec<&Paragraph> {
d.body.iter().filter_map(|b| b.as_para()).collect()
}
/// The objects of the body's first paragraph.
fn objects(d: &Document) -> &[InlineObject] {
&paras(d).first().expect("a paragraph").objects
}
fn pict(inner: &str) -> String {
format!(r#"<w:p><w:r><w:pict>{inner}</w:pict></w:r></w:p>"#)
}
fn kinds(objs: &[InlineObject]) -> Vec<ShapeKind> {
objs.iter()
.map(|o| match o {
InlineObject::Shape { kind, .. } => *kind,
o => panic!("{o:?}"),
})
.collect()
}
#[test]
fn every_shape_in_a_pict_is_read() {
// A `w:pict` may hold any number of shapes; each one is its own object. A `v:shapetype` is
// a definition, not a shape, and is not one of them.
let d = read_body(&pict(concat!(
r#"<v:shapetype id="t1" o:spt="202"/>"#,
r#"<v:rect style="width:10pt;height:20pt"/>"#,
r#"<v:oval style="width:30pt;height:40pt"/>"#,
r#"<v:roundrect style="width:50pt;height:60pt"/>"#,
)));
let objs = objects(&d);
assert_eq!(kinds(objs), [ShapeKind::Rectangle, ShapeKind::Ellipse, ShapeKind::RoundedRectangle], "all three, in document order");
let size = |o: &InlineObject| match o {
InlineObject::Shape { w, h, .. } => (*w, *h),
o => panic!("{o:?}"),
};
assert_eq!(objs.iter().map(size).collect::<Vec<_>>(), [(10.0, 20.0), (30.0, 40.0), (50.0, 60.0)], "each keeps its own size");
}
#[test]
fn a_vml_group_keeps_its_members() {
// `coordsize` is the members' space, mapped onto the group's own box; `left`/`top` place a
// member in that space.
let d = read_body(&pict(concat!(
r#"<v:group style="width:100pt;height:50pt" coordsize="1000,500">"#,
r#"<v:rect style="left:0;top:0;width:400;height:500"/>"#,
r#"<v:oval style="left:600;top:100;width:400;height:300"/>"#,
r#"</v:group>"#,
)));
let InlineObject::Group { w, h, ch_w, ch_h, children, canvas, .. } = &objects(&d)[0] else { panic!("{:?}", objects(&d)) };
assert_eq!((*w, *h, *ch_w, *ch_h), (100.0, 50.0, 1000.0, 500.0), "its own box, and its members' space");
assert_eq!(*canvas, None, "a group is not a Drawing Canvas");
assert_eq!(children.len(), 2, "both members");
assert_eq!((children[0].x, children[0].y), (0.0, 0.0));
assert_eq!((children[1].x, children[1].y), (600.0, 100.0), "placed in the members' space, not scaled");
assert_eq!(kinds(&children.iter().map(|c| c.obj.clone()).collect::<Vec<_>>()), [ShapeKind::Rectangle, ShapeKind::Ellipse]);
}
#[test]
fn a_vml_group_shifts_its_members_by_its_coordorigin() {
// `coordorigin` is the top-left of the members' space, so a member's offset is from there.
let d = read_body(&pict(concat!(
r#"<v:group style="width:100pt;height:100pt" coordsize="1000,1000" coordorigin="200,300">"#,
r#"<v:rect style="left:200;top:300;width:100;height:100"/>"#,
r#"<v:rect style="left:700;top:800;width:100;height:100"/>"#,
r#"</v:group>"#,
)));
let InlineObject::Group { ch_w, children, .. } = &objects(&d)[0] else { panic!() };
assert_eq!(*ch_w, 1000.0);
assert_eq!((children[0].x, children[0].y), (0.0, 0.0), "a member at the origin sits at the group's corner");
assert_eq!((children[1].x, children[1].y), (500.0, 500.0));
}
#[test]
fn nested_vml_groups_are_flattened_into_the_outer_one() {
// An inner group's members' space maps onto the inner group's own box, which is itself in
// the outer group's space: 500 wide in a 1000-wide space inside a 1000-wide box halves it.
let d = read_body(&pict(concat!(
r#"<v:group style="width:100pt;height:100pt" coordsize="1000,1000">"#,
r#"<v:rect style="left:100;top:100;width:50;height:50"/>"#,
r#"<v:group style="left:400;top:400;width:500;height:500" coordsize="1000,1000">"#,
r#"<v:oval style="left:200;top:600;width:100;height:100"/>"#,
r#"</v:group>"#,
r#"</v:group>"#,
)));
let InlineObject::Group { children, .. } = &objects(&d)[0] else { panic!() };
assert_eq!(children.len(), 2, "the inner group's member comes into the outer group");
assert_eq!((children[0].x, children[0].y), (100.0, 100.0));
// 400 + 200 × (500/1000) = 500; 400 + 600 × 0.5 = 700.
assert_eq!((children[1].x, children[1].y), (500.0, 700.0), "mapped through the inner group");
let InlineObject::Shape { w, h, .. } = children[1].obj else { panic!() };
assert_eq!((w, h), (50.0, 50.0), "and scaled by it");
}
#[test]
fn a_vml_group_with_no_member_we_can_show_is_left_out() {
// Only a shape definition, and a shape we draw nothing for (no stroke, no fill).
let d = read_body(&pict(concat!(
r#"<v:group style="width:100pt;height:50pt" coordsize="1000,500">"#,
r#"<v:shapetype id="t1" o:spt="202"/>"#,
r#"<v:rect style="left:0;top:0;width:10;height:10" stroked="f" filled="f"/>"#,
r#"</v:group>"#,
)));
assert!(objects(&d).is_empty(), "nothing to show: {:?}", objects(&d));
}
#[test]
fn deeply_nested_vml_groups_stop_at_the_depth_bound() {
// A hostile file nests groups without end; reading one must not run away.
let depth = 64;
let mut inner = r#"<v:rect style="left:0;top:0;width:10;height:10"/>"#.to_string();
for _ in 0..depth {
inner = format!(r#"<v:group style="left:0;top:0;width:100;height:100" coordsize="100,100">{inner}</v:group>"#);
}
let d = read_body(&pict(&format!(r#"<v:group style="width:100pt;height:100pt" coordsize="100,100">{inner}</v:group>"#)));
// Past the bound nothing is read, so the outermost group has no member and is left out.
assert!(objects(&d).is_empty(), "{:?}", objects(&d));
}
#[test]
fn a_vml_group_survives_being_saved() {
let body = pict(concat!(
r#"<v:group style="width:100pt;height:50pt" coordsize="1000,500">"#,
r#"<v:rect style="left:0;top:0;width:400;height:500"/>"#,
r#"<v:oval style="left:600;top:100;width:400;height:300"/>"#,
r#"</v:group>"#,
));
let d = read_body(&body);
let InlineObject::Group { children, .. } = &objects(&d)[0] else { panic!("{:?}", objects(&d)) };
assert_eq!(children.len(), 2, "read with both its members to begin with");
let back = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
assert_eq!(objects(&back), objects(&d), "saved as DrawingML and read again, it is the same group");
}
#[test]
fn a_pict_of_a_rect_a_text_box_and_a_group_all_come_through() {
// The three shapes of issue #1643, which a round trip dropped the last of. The group's
// member gives its size in points, not in the group's own coordinates, as Word writes it.
let d = read_body(&format!(
"{}{}{}",
pict(r##"<v:rect style="position:absolute;width:100pt;height:50pt" fillcolor="#ff0000"/>"##),
pict(
r#"<v:shape style="width:100pt;height:50pt"><v:textbox><w:txbxContent><w:p><w:r><w:t>boxed</w:t></w:r></w:p></w:txbxContent></v:textbox></v:shape>"#
),
pict(concat!(
r#"<v:group style="width:100pt;height:50pt" coordsize="100,50">"#,
r##"<v:rect style="width:50pt;height:20pt" fillcolor="#00ff00"/>"##,
r#"</v:group>"#,
)),
));
let ps = paras(&d);
assert_eq!(ps.len(), 3, "a paragraph each");
assert_eq!(kinds(&ps[0].objects), [ShapeKind::Rectangle]);
assert_eq!(kinds(&ps[1].objects), [ShapeKind::TextBox]);
let InlineObject::Group { children, .. } = &ps[2].objects[0] else { panic!("{:?}", ps[2].objects) };
assert_eq!(kinds(&children.iter().map(|c| c.obj.clone()).collect::<Vec<_>>()), [ShapeKind::Rectangle], "the green rectangle is read");
let InlineObject::Shape { fill, .. } = children[0].obj else { panic!() };
assert_eq!(fill, Some(wordcraft_doc::Rgb(0x00, 0xff, 0x00)), "with its colour");
// And it is still there after a save, which is what the round trip lost.
let back = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
assert_eq!(paras(&back).iter().map(|p| p.objects.len()).collect::<Vec<_>>(), [1, 1, 1]);
let InlineObject::Group { children, .. } = &paras(&back)[2].objects[0] else { panic!() };
assert_eq!(children.len(), 1, "the group keeps its member");
}
/// The text of text box story `id`, its paragraphs joined by newlines.
fn part_text(d: &Document, id: u32) -> String {
d.parts.get(&id).map(|p| p.blocks.iter().filter_map(|b| b.as_para()).map(|p| p.plain_text()).collect::<Vec<_>>().join("\n")).unwrap_or_default()
}
#[test]
fn a_group_of_several_text_boxes_keeps_every_one() {
// Reported on #1643 by @KNUFFIXX, whose fixture is a `_x0000_t202` pair side by side in one
// inline `v:group`: only box A came through a convert. `vml_textbox` recurses to the first
// box with content, so before this every later box in a group was dropped -- and a text box
// is the one member a group used to keep, which is why this case looked half-working.
let shape = |left: u32, height: u32, label: &str| {
let paras: String = (1..=3).map(|n| format!(r#"<w:p><w:r><w:t xml:space="preserve">{label} line {n}</w:t></w:r></w:p>"#)).collect();
format!(
r##"<v:shape style="position:absolute;left:{left};top:100;width:3500;height:{height}" type="#_x0000_t202" filled="false" stroked="true"><v:textbox inset="0,0,0,0"><w:txbxContent>{paras}</w:txbxContent></v:textbox></v:shape>"##
)
};
// The control fixture, and the overflow one where box B is too short for its text; the
// height of a box must not decide whether it is read.
for (height_b, label_b) in [(1600, "Box B (tall)"), (400, "Box B (short)")] {
let d = read_body(&pict(&format!(
concat!(
r##"<v:shapetype id="_x0000_t202" coordsize="21600,21600" o:spt="202" path="m,l,21600r21600,l21600,xe"/>"##,
r##"<v:group style="width:400pt;height:100pt" coordorigin="0,0" coordsize="8000,2000">{}{}</v:group>"##,
),
shape(100, 1600, "Box A (tall)"),
shape(4200, height_b, label_b),
)));
let InlineObject::Group { ch_w, ch_h, children, .. } = &objects(&d)[0] else { panic!("{:?}", objects(&d)) };
assert_eq!((*ch_w, *ch_h), (8000.0, 2000.0), "the group's member space");
assert_eq!(children.len(), 2, "both boxes, with box B {height_b} high");
assert_eq!((children[0].x, children[1].x), (100.0, 4200.0), "side by side");
let stories: Vec<u32> = children.iter().filter_map(|c| c.obj.text_box()).collect();
assert_eq!(stories.len(), 2, "each box has its own story");
assert_ne!(stories[0], stories[1]);
assert_eq!(part_text(&d, stories[0]), "Box A (tall) line 1\nBox A (tall) line 2\nBox A (tall) line 3");
assert_eq!(part_text(&d, stories[1]), format!("{label_b} line 1\n{label_b} line 2\n{label_b} line 3"));
// Saved and read again -- the step the report loses the text at.
let back = wordcraft_docx::read(&wordcraft_docx::write(&d).unwrap()).unwrap();
let InlineObject::Group { children: again, .. } = &objects(&back)[0] else { panic!() };
let back_stories: Vec<u32> = again.iter().filter_map(|c| c.obj.text_box()).collect();
assert_eq!(back_stories.len(), 2, "both boxes survive the save");
assert_eq!(part_text(&back, back_stories[0]), part_text(&d, stories[0]));
assert_eq!(part_text(&back, back_stories[1]), part_text(&d, stories[1]));
}
}
#[test]
fn a_group_member_finds_its_shapetype_beside_the_group() {
// Word writes each `v:shapetype` once, at its first use, as a child of the `w:pict` -- so a
// group's member refers to one that is its *parent's* sibling, not its own. The member here
// carries no geometry and no colours of its own: its diamond shape comes from the
// shapetype's `o:spt` (4) and its fill from the shapetype's `fillcolor`.
let d = read_body(&pict(concat!(
r##"<v:shapetype id="_x0000_t4" o:spt="4" fillcolor="#0000ff"/>"##,
r#"<v:group style="width:100pt;height:50pt" coordsize="1000,500">"#,
r##"<v:shape style="left:0;top:0;width:400;height:500" type="#_x0000_t4"/>"##,
r#"</v:group>"#,
)));
let InlineObject::Group { children, .. } = &objects(&d)[0] else { panic!("{:?}", objects(&d)) };
assert_eq!(children.len(), 1, "the member is read");
let InlineObject::Shape { kind, fill, .. } = children[0].obj else { panic!("{:?}", children[0].obj) };
assert_eq!(kind, ShapeKind::Diamond, "the preset comes from the shapetype's o:spt");
assert_eq!(fill, Some(wordcraft_doc::Rgb(0x00, 0x00, 0xff)), "and the fill from its fillcolor");
}
#[test]
fn a_top_level_shape_still_finds_its_shapetype() {
// The same lookup for a shape that is not in a group, which is what it was written for.
let d = read_body(&pict(concat!(
r##"<v:shapetype id="_x0000_t4" o:spt="4" fillcolor="#00ff00"/>"##,
r##"<v:shape style="width:40pt;height:40pt" type="#_x0000_t4"/>"##,
)));
let InlineObject::Shape { kind, fill, .. } = objects(&d)[0] else { panic!("{:?}", objects(&d)) };
assert_eq!((kind, fill), (ShapeKind::Diamond, Some(wordcraft_doc::Rgb(0x00, 0xff, 0x00))));
}
+3
View File
@@ -82,6 +82,7 @@ Home|Paragraph|Bullets|para.bullets
Home|Paragraph|Numbering|para.numbering
Home|Paragraph|Multilevel List|para.multilevel
Home|Paragraph|Define New Multilevel List|list.define
Home|Paragraph|Multilevel List Library|list.multilevelPreset
Home|Paragraph|Change List Level|para.listLevel
Home|Paragraph|Restart Numbering|para.restartNumbering
Home|Paragraph|Define New Bullet|para.defineBullet
@@ -482,6 +483,8 @@ Editing|Keyboard|Move Paragraph Down|para.moveDown
Editing|Keyboard|Promote|para.promote
Editing|Keyboard|Demote|para.demote
Editing|Keyboard|Toggle Unicode Character|text.toggleUnicode
Editing|Keyboard|Spike|edit.spike
Editing|Keyboard|Insert Spike|edit.insertSpike
Editing|Navigation|Word Left|caret.wordLeft
Editing|Navigation|Word Right|caret.wordRight
Editing|Navigation|Line Start|caret.home
+372 -10
View File
@@ -290,7 +290,7 @@ pub fn style(name: &str) -> &'static CiteStyle {
pub fn specs() -> Vec<CommandSpec> {
vec![
CommandSpec::new("references.sources", "Manage Sources", "References › Citations & Bibliography", sources)
.params(r#"{"add"?: Source, "remove"?: tag} → list"#),
.params(r#"{"add"?: Source, "remove"?: tag, "import"?: path (a Word bibliography file, `b:Sources` / `Sources.xml`)} → list"#),
CommandSpec::new("references.citation", "Insert Citation", "References › Citations & Bibliography", citation).params(r#"{"tag"?: string, "source"?: Source (added if new), "pages"?: string}"#),
CommandSpec::new("references.citationStyle", "Style", "References › Citations & Bibliography", |s, v| {
let st = p::str(v, "style").unwrap_or("APA");
@@ -342,7 +342,7 @@ pub fn specs() -> Vec<CommandSpec> {
sel_result(s)
}),
CommandSpec::new("insert.crossReference", "Cross-reference", "Insert › Links", cross_ref)
.params(r#"{"to": "heading|bookmark|figure|table|equation", "target": string (text / name / number), "show"?: "text|page|number|aboveBelow"} or {} to list targets"#),
.params(r#"{"to": "heading|bookmark|figure|table|equation|label:<Name> (a caption label of your own)", "target": string (text / name / number), "show"?: "text|page|number|aboveBelow"} or {} to list targets"#),
CommandSpec::new("references.noteOptions", "Footnote and Endnote", "References › Footnotes", super::references::note_options).params(
r#"{"kind"?: "footnote|endnote", "pos"?: "pageBottom|beneathText" (footnotes) | "sectEnd|docEnd" (endnotes), "numFmt"?: "decimal|lowerRoman|upperRoman|lowerLetter|upperLetter|…", "numStart"?: 1-32767, "numRestart"?: "continuous|eachSect|eachPage" (each page: footnotes only), "scope"?: "document|section", "footnoteFormat"?: …, "endnoteFormat"?: …} — no settings: the options in effect at the caret"#,
),
@@ -519,18 +519,42 @@ pub fn entry(src: &Source, style_name: &str, n: usize) -> String {
render(if src.journal.is_empty() { st.book } else { st.article }, &field)
}
/// The tag a new source gets, as Word makes it: the first three letters of the first author's
/// name and the last two digits of the year (`Smi20`), with `1`, `2`… added while it is taken.
fn auto_tag(doc: &Document, src: &Source) -> String {
let base = auto_tag_base(src);
let taken = |t: &str| doc.sources.iter().any(|x| x.tag == t);
if !taken(&base) {
return base;
}
(1u32..).map(|n| format!("{base}{n}")).find(|t| !taken(t)).unwrap_or(base)
}
/// The author of a source for matching untagged entries: the editor when there is no author.
fn by_author(x: &Source) -> &str {
if x.author.is_empty() { x.editor.as_str() } else { x.author.as_str() }
}
/// The stem of a new source's tag, before `1`, `2`… are added while it is taken: the first three
/// alphanumerics of the first author's name (nothing of the author's, `Src`) and the last two
/// digits of the year.
fn auto_tag_base(src: &Source) -> String {
let name: String =
authors(&src.author).first().map(|x| x.0.clone()).unwrap_or_default().chars().filter(|c| c.is_alphanumeric()).take(3).collect();
let name = if name.is_empty() { "Src".to_string() } else { name };
let digits: Vec<char> = src.year.chars().filter(char::is_ascii_digit).collect();
let year: String = digits.get(digits.len().saturating_sub(2)..).unwrap_or_default().iter().collect();
format!("{name}{year}")
}
/// Manage Sources: applies `add`, `remove` and `import` (in that order) and returns the
/// document's sources. `import` reads a Word bibliography file (`b:Sources`, #2640).
fn sources(s: &mut Session, v: &Value) -> CmdResult {
if let Some(a) = v.get("add") {
let mut src: Source = serde_json::from_value(a.clone()).map_err(|e| CmdError::Params(e.to_string()))?;
// A new source's tag is made from its author and year; an edit keeps the tag it has.
if src.tag.is_empty() {
let base: String = authors(&src.author)
.first()
.map(|x| x.0.clone())
.unwrap_or_else(|| "Source".into())
.chars()
.filter(|c| c.is_alphanumeric())
.collect();
src.tag = format!("{base}{}", src.year);
src.tag = auto_tag(&s.doc, &src);
}
// Edit Source keeps the source's place in the list and refreshes its citations.
match s.doc.sources.iter().position(|x| x.tag == src.tag) {
@@ -547,9 +571,74 @@ fn sources(s: &mut Session, v: &Value) -> CmdResult {
if let Some(t) = p::str(v, "remove") {
s.doc.sources.retain(|x| x.tag != t);
}
if let Some(path) = p::str(v, "import") {
import_sources(s, path)?;
}
Ok(serde_json::to_value(&s.doc.sources).unwrap_or(Value::Null))
}
/// The largest bibliography sources file imported from disk.
#[cfg(not(target_arch = "wasm32"))]
const MAX_SOURCES_FILE: u64 = 32 << 20;
/// Import the sources of a Word bibliography file (`b:Sources`: Word's own `Sources.xml`) into
/// this document (#2640). A tag already here is left alone — an import never overwrites an
/// edited source; a source with no tag is matched to an existing one by title and author, and
/// otherwise gets a tag the way a new source does.
fn import_sources(s: &mut Session, path: &str) -> Result<(), CmdError> {
#[cfg(target_arch = "wasm32")]
{
let _ = (s, path);
return Err(CmdError::Failed("reading files by path isn't available on the web".into()));
}
#[cfg(not(target_arch = "wasm32"))]
{
// Through `io`, which keeps automation inside its read root (#769).
let bytes = crate::io::read_file_capped(path, MAX_SOURCES_FILE, "sources file").map_err(CmdError::Failed)?;
let file = wordcraft_docx::read_sources(&bytes).map_err(|e| CmdError::Failed(e.to_string()))?;
let mut added = false;
// The tags and (title, author-or-editor) keys already here, once: an import of thousands
// of untagged sources must not rescan the list per source. Tags follow `auto_tag`: the
// stem, then 1, 2… while taken.
let mut tags: std::collections::HashSet<String> = s.doc.sources.iter().map(|x| x.tag.clone()).collect();
let mut keys: std::collections::HashSet<(String, String)> =
s.doc.sources.iter().map(|x| (x.title.clone(), by_author(x).to_string())).collect();
let mut next: std::collections::HashMap<String, u32> = std::collections::HashMap::new();
for mut src in file.sources {
if s.doc.sources.len() >= wordcraft_docx::MAX_SOURCES {
break;
}
if src.tag.is_empty() {
// A source with no tag is matched to one already here by its title and author (or
// editor), as repeated parts of a document are; otherwise it gets a tag the way a
// new source does.
if keys.contains(&(src.title.clone(), by_author(&src).to_string())) {
continue;
}
let base = auto_tag_base(&src);
let n = next.entry(base.clone()).or_insert(0);
let mut tag = if *n == 0 { base.clone() } else { format!("{base}{}", *n) };
while tags.contains(&tag) {
*n += 1;
tag = format!("{base}{}", *n);
}
*n = n.saturating_add(1);
src.tag = tag;
}
if !tags.insert(src.tag.clone()) {
continue;
}
keys.insert((src.title.clone(), by_author(&src).to_string()));
s.doc.sources.push(src);
added = true;
}
if added {
s.touch();
}
Ok(())
}
}
fn citation(s: &mut Session, v: &Value) -> CmdResult {
if v.get("source").is_some() {
sources(s, &json!({"add": v.get("source")}))?;
@@ -906,11 +995,78 @@ fn cross_ref(s: &mut Session, v: &Value) -> CmdResult {
Ok(json!({"inserted": result}))
}
/// The label of a caption: the text before its first field (`Plate ` in `Plate `, `SEQ Plate`,
/// `: A rock`). When the caption starts with its chapter number, the text between that field and
/// the `SEQ` number, unless it is only a separator. Failing both, the `SEQ` identifier names it.
/// A caption without a `SEQ` field (typed by hand) has what comes before its first digit.
fn caption_label(p: &Paragraph, text: &str) -> Option<String> {
let fields: Vec<(usize, &str)> = p
.object_offsets()
.into_iter()
.filter_map(|o| match p.object_at(o) {
Some(InlineObject::Field { instr, .. }) => Some((o, instr.as_str())),
_ => None,
})
.collect();
let seq = fields.iter().position(|(_, instr)| instr.split_whitespace().next().is_some_and(|w| w.eq_ignore_ascii_case("SEQ")));
let label = match seq.and_then(|i| Some((i, fields.get(i)?))) {
Some((i, (seq_at, instr))) => {
// The text in `from..to` with the bookmark marks dropped, if it says anything.
let said = |from: usize, to: usize| {
let t: String = p.text.get(from..to).unwrap_or_default().chars().filter(|c| *c != wordcraft_doc::para::OBJ).collect();
let t = t.trim();
t.chars().any(char::is_alphanumeric).then(|| t.to_string())
};
let first = fields.first().map_or(0, |(o, _)| *o);
let after_previous =
i.checked_sub(1).and_then(|k| fields.get(k)).map_or(0, |(o, _)| o.saturating_add(wordcraft_doc::para::OBJ.len_utf8()));
said(0, first).or_else(|| said(after_previous, *seq_at)).or_else(|| instr.split_whitespace().nth(1).map(|id| id.replace('_', " ")))?
}
None => text.get(..text.find(|c: char| c.is_ascii_digit())?)?.trim().to_string(),
};
(!label.is_empty() && label.chars().count() <= 40).then_some(label)
}
/// The caption paragraphs of the body: (text, position, label).
fn captions(s: &Session) -> Vec<(String, Pos, Option<String>)> {
s.doc
.para_paths(StoryRef::Body)
.into_iter()
.filter_map(|path| {
let p = s.doc.para(StoryRef::Body, &path).filter(|p| p.props.style.as_deref() == Some("Caption"))?;
let text = p.plain_text();
let label = caption_label(p, &text);
Some((text, Pos { story: StoryRef::Body, path, off: 0 }, label))
})
.collect()
}
/// The labels the body's captions use, first seen first: Figure, Table and Equation, and any the
/// person made with Insert Caption › New Label. Each is a cross-reference type, `label:<Name>` (#2940).
pub fn caption_labels(s: &Session) -> Vec<String> {
let mut labels: Vec<String> = Vec::new();
for l in captions(s).into_iter().filter_map(|(_, _, l)| l) {
if !labels.contains(&l) {
labels.push(l);
}
}
labels
}
/// Cross-reference targets of a kind: (shown text or bookmark name, position).
pub fn ref_targets(s: &Session, to: &str) -> Vec<(String, Pos)> {
let mut targets: Vec<(String, Pos)> = Vec::new();
match to {
"bookmark" => targets = s.doc.bookmarks(),
// A caption label of the person's own: the captions that use exactly that label.
t if t.starts_with("label:") => {
let want = t.get("label:".len()..).unwrap_or_default();
targets = captions(s)
.into_iter()
.filter(|(_, _, label)| !want.is_empty() && label.as_deref() == Some(want))
.map(|(text, pos, _)| (text, pos))
.collect();
}
"figure" | "table" | "equation" => {
let label = match to {
"table" => "Table",
@@ -1105,6 +1261,80 @@ mod tests {
assert!(fields(&s).contains(&(format!("REF {name} \\h"), "Figure 1".to_string())), "{:?}", fields(&s));
}
/// #2940: a caption with a label of the person's own (Insert Caption › New Label) is a
/// cross-reference type of its own, `label:<Name>`, next to Figure, Table and Equation.
#[test]
fn cross_reference_to_a_custom_caption_label() {
let mut s = Session::new(wordcraft_doc::Document::from_text("Rock\nSee "));
s.run("select.text", &json!({"text": "Rock"})).unwrap();
s.run("references.caption", &json!({"label": "Plate", "text": "A rock"})).unwrap();
assert_eq!(caption_labels(&s), vec!["Plate".to_string()]);
let list = s.run("insert.crossReference", &json!({"to": "label:Plate"})).unwrap();
assert_eq!(list, json!(["Plate 1: A rock"]));
assert!(ref_targets(&s, "label:Figure").is_empty(), "only that label's captions");
assert!(ref_targets(&s, "label:").is_empty() && ref_targets(&s, "label:Pla").is_empty(), "the whole label");
s.run("caret.docEnd", &json!({})).unwrap();
s.run("insert.crossReference", &json!({"to": "label:Plate", "target": "Plate 1", "show": "number"})).unwrap();
assert!(s.doc.plain_text(StoryRef::Body).ends_with("See Plate 1"), "{}", s.doc.plain_text(StoryRef::Body));
assert!(fields(&s).iter().any(|(i, r)| i.starts_with("REF _Ref") && r == "Plate 1"), "{:?}", fields(&s));
}
/// #2940: the label comes from the caption's structure (the text before its `SEQ` number), so
/// letter and roman numbering, which have no digit, still name it; a multi-word label too.
#[test]
fn caption_labels_do_not_need_a_digit() {
let mut s = Session::new(wordcraft_doc::Document::from_text("Rock\nGem\nSand\nSee "));
for (text, label, format) in [("Rock", "Plate", "ALPHABETIC"), ("Gem", "Photo Plate", "ROMAN"), ("Sand", "Chart", "ARABIC")] {
s.run("select.text", &json!({"text": text})).unwrap();
s.run("references.caption", &json!({"label": label, "text": text, "format": format})).unwrap();
}
assert_eq!(caption_labels(&s), ["Plate", "Photo Plate", "Chart"]);
assert_eq!(ref_targets(&s, "label:Plate").len(), 1);
assert_eq!(ref_targets(&s, "label:Photo Plate").len(), 1);
assert_eq!(ref_targets(&s, "label:Chart").len(), 1);
s.run("caret.docEnd", &json!({})).unwrap();
s.run("insert.crossReference", &json!({"to": "label:Photo Plate", "target": "Photo Plate I", "show": "number"})).unwrap();
assert!(s.doc.plain_text(StoryRef::Body).ends_with("See Photo Plate I"), "{}", s.doc.plain_text(StoryRef::Body));
}
/// #2940: a caption may start with its chapter number (`STYLEREF`) and carry its label text
/// between that and the `SEQ` number; the text is the label, not the `SEQ` identifier.
#[test]
fn caption_label_after_a_chapter_number_field() {
let mut s = Session::new(wordcraft_doc::Document::from_text("Rock\nSee "));
let field = |instr: &str| InlineObject::Field {
instr: instr.into(),
result: String::new(),
locked: false,
dirty: false,
code: false,
form: String::new(),
};
let plain = CharProps::default();
// STYLEREF, " Specimen ", SEQ Figure, ": A rock"
let mut imported = Paragraph::new().styled("Caption");
imported.insert_object(0, field("STYLEREF 1 \\s"), &plain).unwrap();
imported.insert_text(imported.len(), " Specimen ", &plain).unwrap();
imported.insert_object(imported.len(), field("SEQ Figure \\* ARABIC"), &plain).unwrap();
imported.insert_text(imported.len(), ": A rock", &plain).unwrap();
s.doc.insert_block(StoryRef::Body, &Path::top(1), Block::Para(imported)).unwrap();
// A generated caption with a chapter number: label, STYLEREF, separator, SEQ.
s.run("select.text", &json!({"text": "Rock"})).unwrap();
s.run("references.caption", &json!({"label": "Plate", "chapter": 1, "separator": "-"})).unwrap();
// No label text at all: the SEQ identifier names it, after the chapter number's separator.
let mut bare = Paragraph::new().styled("Caption");
bare.insert_object(0, field("STYLEREF 1 \\s"), &plain).unwrap();
bare.insert_text(bare.len(), "-", &plain).unwrap();
bare.insert_object(bare.len(), field("SEQ Photo_Plate"), &plain).unwrap();
let end = s.doc.body.len();
s.doc.insert_block(StoryRef::Body, &Path::top(end), Block::Para(bare)).unwrap();
let labels = caption_labels(&s);
assert!(labels.contains(&"Specimen".to_string()), "{labels:?}");
assert!(labels.contains(&"Plate".to_string()), "{labels:?}");
assert!(labels.contains(&"Photo Plate".to_string()), "{labels:?}");
assert!(!labels.iter().any(|l| l == "-" || l.is_empty() || l == "Figure"), "{labels:?}");
}
/// #661: References › Table of Authorities › Update Table picks up citations marked after the table.
#[test]
fn table_of_authorities_updates() {
@@ -1185,6 +1415,52 @@ mod tests {
assert_eq!(crate::cmd::references::bookmark_text(&s.doc, &name).as_deref(), Some("Kilns"));
}
fn src_of(author: &str, year: &str) -> Value {
json!({"author": author, "year": year, "title": "T"})
}
#[test]
fn editing_author_and_year_keeps_the_tag_and_updates_the_citation_text() {
let mut s = Session::new(wordcraft_doc::Document::new());
s.run("references.sources", &json!({"add": src_of("Smith, Ann", "2020")})).unwrap();
assert_eq!(s.doc.sources[0].tag, "Smi20");
s.run("references.citation", &json!({"tag": "Smi20"})).unwrap();
// What the Edit Source dialog sends: the unchanged tag with the new author and year.
let mut edited = src_of("Jones, Bo", "2021");
edited["tag"] = json!("Smi20");
s.run("references.sources", &json!({"add": edited})).unwrap();
assert_eq!(s.doc.sources.len(), 1);
assert_eq!(s.doc.sources[0].tag, "Smi20");
let f = fields(&s);
assert_eq!(f[0].0, "CITATION Smi20");
assert!(f[0].1.contains("Jones") && f[0].1.contains("2021"), "{:?}", f[0].1);
}
#[test]
fn a_new_sources_tag_is_three_letters_and_two_year_digits() {
let mut s = Session::new(wordcraft_doc::Document::new());
for (a, y) in [
("Kim", "2000"),
("Kim", "2000"),
("Kim", "2000"),
("Lee", "2019"),
("Li", "1999"),
("", ""),
("", ""),
("Ünal, Ece", "7"),
("O'Neil", "20x5"),
] {
s.run("references.sources", &json!({"add": src_of(a, y)})).unwrap();
}
let tags: Vec<_> = s.doc.sources.iter().map(|x| x.tag.as_str()).collect();
assert_eq!(tags, ["Kim00", "Kim001", "Kim002", "Lee19", "Li99", "Src", "Src1", "Üna7", "ONe05"]);
let mut s = Session::new(wordcraft_doc::Document::new());
let mut a = src_of("Lee", "2019");
a["tag"] = json!("MyTag");
s.run("references.sources", &json!({"add": a})).unwrap();
assert_eq!(s.doc.sources[0].tag, "MyTag");
}
fn two_sources(s: &mut Session, a_title: &str) {
let a = json!({"tag": "A", "author": "Rivera, Alex; Chen, Mei", "title": a_title, "year": "2021", "publisher": "Harbor Books", "city": "Portland"});
let b = json!({"tag": "B", "author": "Okafor, Ngozi", "title": "Kilns and Clay", "year": "2019", "journal": "Studio Quarterly", "volume": "4", "pages": "10-20"});
@@ -1226,6 +1502,92 @@ mod tests {
assert_eq!(s.bib_style, "Turabian", "an unknown style keeps the current one");
}
/// #2640: Manage Sources › Import reads a Word bibliography file (`b:Sources`: Word's own
/// `Sources.xml`) into the document. An untagged source gets a tag like a new one; a tag
/// already here is left alone; the whole import is one undo step; and a file that isn't
/// `b:Sources` is an error, not a silent success.
#[test]
fn import_sources_reads_a_word_bibliography_file() {
let dir = std::env::temp_dir().join(format!("wc-import-sources-{}", std::process::id()));
let _ = std::fs::remove_dir_all(&dir);
std::fs::create_dir_all(&dir).unwrap();
let path = dir.join("Sources.xml");
let xml = br#"<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<b:Sources xmlns:b="http://schemas.openxmlformats.org/officeDocument/2006/bibliography"
xmlns="http://schemas.openxmlformats.org/officeDocument/2006/bibliography"
SelectedStyle="\APA.XSL" StyleName="APA">
<b:Source><b:Tag>Riv21</b:Tag><b:SourceType>Book</b:SourceType>
<b:Author><b:Author><b:NameList><b:Person><b:Last>Rivera</b:Last><b:First>Alex</b:First></b:Person></b:NameList></b:Author></b:Author>
<b:Title>Shared Spaces</b:Title><b:Year>2021</b:Year><b:City>Portland</b:City><b:Publisher>Harbor Books</b:Publisher>
</b:Source>
<b:Source><b:SourceType>Book</b:SourceType></b:Source>
<b:Source><b:SourceType>JournalArticle</b:SourceType>
<b:Author><b:Author><b:NameList><b:Person><b:Last>Okafor</b:Last><b:First>Ngozi</b:First></b:Person></b:NameList></b:Author></b:Author>
<b:Title>Kilns and Clay</b:Title><b:JournalName>Studio Quarterly</b:JournalName><b:Year>2019</b:Year>
</b:Source>
</b:Sources>"#;
std::fs::write(&path, xml).unwrap();
let path = path.to_string_lossy().to_string();
let mut s = Session::new(wordcraft_doc::Document::new());
s.run("references.sources", &json!({"import": path})).unwrap();
assert_eq!(s.doc.sources.len(), 2, "the source with no content is dropped");
assert_eq!(s.doc.sources[0].tag, "Riv21");
assert_eq!((s.doc.sources[0].author.as_str(), s.doc.sources[0].title.as_str()), ("Rivera, Alex", "Shared Spaces"));
assert_eq!(
(s.doc.sources[1].tag.as_str(), s.doc.sources[1].kind.as_str()),
("Oka19", "article"),
"an untagged source gets a tag like a new one"
);
assert!(s.dirty, "the document is marked edited");
// The import is one undo step, and importing again adds nothing.
s.run("edit.undo", &json!({})).unwrap();
assert!(s.doc.sources.is_empty(), "one undo takes the whole import back");
s.run("references.sources", &json!({"import": path})).unwrap();
s.doc.sources[0].title = "Shared Spaces, revised".into();
s.run("references.sources", &json!({"import": path})).unwrap();
assert_eq!(s.doc.sources.len(), 2, "a tag already here is not added again");
assert_eq!(s.doc.sources[0].title, "Shared Spaces, revised", "an import never overwrites an edited source");
assert_eq!(s.doc.sources[1].tag, "Oka19", "the untagged source keeps the tag it was given");
// Anything that isn't b:Sources is refused.
let bad = dir.join("not-sources.xml");
std::fs::write(&bad, "<html><body>no</body></html>").unwrap();
let e = s.run("references.sources", &json!({"import": bad.to_string_lossy()})).unwrap_err();
assert!(e.to_string().contains("not a Word document"), "{e}");
let _ = std::fs::remove_dir_all(&dir);
}
/// #2640: untagged sources that share a tag stem get the tag, then `1`, `2`… in import order,
/// and a second import adds none of them (the untagged one matches by title).
#[test]
fn an_import_tags_each_untagged_source_once() {
let dir = std::env::temp_dir().join(format!("wc-import-tags-{}", std::process::id()));
let _ = std::fs::remove_dir_all(&dir);
std::fs::create_dir_all(&dir).unwrap();
let path = dir.join("Sources.xml");
std::fs::write(
&path,
r#"<b:Sources xmlns:b="http://schemas.openxmlformats.org/officeDocument/2006/bibliography">
<b:Source><b:SourceType>Book</b:SourceType><b:Title>First Work</b:Title><b:Year>2021</b:Year>
<b:Author><b:Author><b:NameList><b:Person><b:Last>Rivera</b:Last><b:First>Alex</b:First></b:Person></b:NameList></b:Author></b:Author></b:Source>
<b:Source><b:SourceType>Book</b:SourceType><b:Title>Second Work</b:Title><b:Year>2021</b:Year>
<b:Author><b:Author><b:NameList><b:Person><b:Last>Rivera</b:Last><b:First>Alex</b:First></b:Person></b:NameList></b:Author></b:Author></b:Source>
<b:Source><b:SourceType>Book</b:SourceType><b:Title>Third Work</b:Title></b:Source>
</b:Sources>"#,
)
.unwrap();
let path = path.to_string_lossy().to_string();
let mut s = Session::new(wordcraft_doc::Document::new());
s.run("references.sources", &json!({"import": path})).unwrap();
let tags: Vec<_> = s.doc.sources.iter().map(|x| x.tag.as_str()).collect();
assert_eq!(tags, ["Riv21", "Riv211", "Src"], "{tags:?}");
s.run("references.sources", &json!({"import": path})).unwrap();
assert_eq!(s.doc.sources.len(), 3, "the tags are here and the untagged one matches by title");
let _ = std::fs::remove_dir_all(&dir);
}
#[test]
fn numbered_styles_count_in_order_of_first_citation() {
let mut s = Session::new(wordcraft_doc::Document::new());
+7
View File
@@ -159,6 +159,13 @@ pub(crate) fn dispatch(s: &mut Session, id: &str, mutates: bool, run: fn(&mut Se
}
return super::sel_result(s);
}
// Backspace over whole rows, whole columns or the whole table deletes them, as in Word
// (Delete only clears them, as for any other cells).
if id == "text.backspace"
&& let Some(r) = super::table::backspace_cells(s)
{
return r;
}
let Some(segs) = s.column_segments().map(<[(Pos, Pos)]>::to_vec) else { return run(s, v) };
let cells = s.cell_range().is_some();
match id {
+81 -10
View File
@@ -142,12 +142,27 @@ pub fn days_from_civil(y: i64, m: u32, d: u32) -> i64 {
era * 146_097 + doe - 719_468
}
/// The built-in name of the placeholder style. Its id is localized: German Word writes
/// `Platzhaltertext`, English `PlaceholderText`.
const PLACEHOLDER_STYLE_NAME: &str = "Placeholder Text";
/// The id of the document's placeholder style: the one named "Placeholder Text", else the
/// English id.
fn placeholder_style_id(doc: &wordcraft_doc::Document) -> String {
doc.styles
.styles
.iter()
.find(|st| st.kind == StyleKind::Character && st.name.eq_ignore_ascii_case(PLACEHOLDER_STYLE_NAME))
.map(|st| st.id.clone())
.unwrap_or_else(|| PLACEHOLDER_STYLE.into())
}
/// The placeholder style, added to the document when it hasn't one.
fn ensure_placeholder_style(s: &mut Session) {
if s.doc.styles.get(PLACEHOLDER_STYLE).is_none() {
if s.doc.styles.get(&placeholder_style_id(&s.doc)).is_none() {
s.doc.styles.styles.push(Style {
id: PLACEHOLDER_STYLE.into(),
name: "Placeholder Text".into(),
name: PLACEHOLDER_STYLE_NAME.into(),
kind: StyleKind::Character,
based_on: Some("DefaultParagraphFont".into()),
chr: CharProps { color: Some(TextColor::Rgb(Rgb(0x66, 0x66, 0x66))), ..Default::default() },
@@ -161,13 +176,13 @@ fn ensure_placeholder_style(s: &mut Session) {
/// Formatting for placeholder text based on `base`.
fn placeholder_props(s: &mut Session, base: &CharProps) -> CharProps {
ensure_placeholder_style(s);
CharProps { style: Some(PLACEHOLDER_STYLE.into()), link: None, ..base.clone() }
CharProps { style: Some(placeholder_style_id(&s.doc)), link: None, ..base.clone() }
}
/// `base` without the placeholder style (for real content).
fn content_props(base: &CharProps) -> CharProps {
/// `base` without the placeholder style (for real content), whatever its localized id.
fn content_props(doc: &wordcraft_doc::Document, base: &CharProps) -> CharProps {
let mut p = base.clone();
if p.style.as_deref() == Some(PLACEHOLDER_STYLE) {
if p.style.as_deref().is_some_and(|id| id == PLACEHOLDER_STYLE || id == placeholder_style_id(doc)) {
p.style = None;
}
p
@@ -248,7 +263,7 @@ fn content_base(s: &Session, r: &ControlRange) -> CharProps {
.para_at(&a)
.map(|p| if r.is_empty() { p.props_of_char(r.start.off).clone() } else { p.props_of_char(a.off).clone() })
.unwrap_or_default();
content_props(&props)
content_props(&s.doc, &props)
}
// ---------------------------------------------------------------------------------------------
@@ -299,7 +314,14 @@ pub fn prepare_edit(s: &mut Session) -> Result<Option<CharProps>, CmdError> {
if !r.control.showing_placeholder || !r.contains(&b) {
return Ok(None);
}
let props = content_base(s, &r);
// Text typed in place of the placeholder takes the control's own run properties (Word, and
// ECMA-376 §17.5.2.27), else the content's formatting without the placeholder style. The
// showing placeholder's own character style is that style, whatever a producer called it
// (Word localizes only the id; another program may localize the name too).
let props = match &r.control.run_props {
Some(rp) => content_props(&s.doc, rp),
None => CharProps { style: None, ..content_base(s, &r) },
};
s.sel = Selection { anchor: r.content_start(), focus: r.end.clone() };
if let Some(c) = s.doc.control_mut(&r.start) {
c.showing_placeholder = false;
@@ -423,7 +445,7 @@ fn insert(s: &mut Session, v: &Value, kind: ControlKind) -> CmdResult {
// An inline control: around the selection (text and pictures), else at the caret with its
// placeholder or value.
let wraps = a != b && a.path == b.path && matches!(ctl.kind, ControlKind::RichText | ControlKind::Text { .. } | ControlKind::Picture);
let base = s.doc.para_at(&a).map(|p| content_props(&p.props_of_char(a.off).clone())).unwrap_or_default();
let base = s.doc.para_at(&a).map(|p| content_props(&s.doc, &p.props_of_char(a.off).clone())).unwrap_or_default();
if wraps {
s.doc.insert_object(&b, InlineObject::ControlEnd, &base)?;
let start = a.clone();
@@ -464,7 +486,7 @@ fn insert_block(s: &mut Session, mut ctl: ContentControl, a: &Pos, b: &Pos) -> C
let last = if b.path.parent() == a.path.parent() { b.path.clone() } else { a.path.clone() };
ctl.block = true;
let empty = a.path == last && s.doc.para(story, &a.path).is_some_and(|p| p.is_empty());
let base = s.doc.para(story, &a.path).map(|p| content_props(&p.props_of_char(0).clone())).unwrap_or_default();
let base = s.doc.para(story, &a.path).map(|p| content_props(&s.doc, &p.props_of_char(0).clone())).unwrap_or_default();
let item = matches!(ctl.kind, ControlKind::RepeatingSection { .. }).then(|| {
let mut i = ContentControl::new(ControlKind::RepeatingSectionItem);
i.block = true;
@@ -748,6 +770,55 @@ mod editing_tests {
run(s, "developer.control", json!({}))
}
/// A form from German Word names its placeholder style `Platzhaltertext` ("Placeholder Text"),
/// and each control's `w:rPr` gives the style entered text takes (`Eingabe`, bold black).
/// Text typed over the placeholder takes the control's run properties, as in Word (17 of 17
/// values filled in Word did); without them it drops the placeholder style whatever its id.
/// WordCraft kept the German placeholder style, so entries stayed grey.
#[test]
fn typing_over_a_localized_placeholder_takes_the_controls_formatting() {
use wordcraft_doc::props::{CharProps, Rgb, TextColor};
use wordcraft_doc::styles::{Style, StyleKind};
let style =
|id: &str, name: &str, chr: CharProps| Style { id: id.into(), name: name.into(), kind: StyleKind::Character, chr, ..Default::default() };
let typed_style = |run_props: Option<CharProps>| {
let mut s = Session::new(wordcraft_doc::Document::new());
let grey = CharProps { color: Some(TextColor::Rgb(Rgb(0x80, 0x80, 0x80))), ..Default::default() };
s.doc.styles.styles.push(style("Platzhaltertext", "Placeholder Text", grey));
s.doc.styles.styles.push(style("Eingabe", "Eingabe", CharProps { bold: Some(true), ..Default::default() }));
run(&mut s, "text.insert", json!({"text": "Name: "}));
run(&mut s, "developer.plainText", json!({"tag": "name"}));
let start = s.doc.control_at(&s.sel.focus).unwrap().start;
s.doc.control_mut(&start).unwrap().run_props = run_props;
// The placeholder uses the document's own placeholder style; none is added.
let r = s.doc.control_at(&s.sel.focus).unwrap();
let shown = s.doc.para_at(&r.content_start()).unwrap().props_of_char(r.content_start().off).style.clone();
assert_eq!(shown.as_deref(), Some("Platzhaltertext"));
assert_eq!(s.doc.styles.styles.iter().filter(|st| st.name == "Placeholder Text").count(), 1);
run(&mut s, "caret.right", json!({}));
run(&mut s, "text.insert", json!({"text": "Ada"}));
let r = s.doc.control_at(&s.sel.focus).unwrap();
s.doc.para_at(&r.content_start()).unwrap().props_of_char(r.content_start().off).style.clone()
};
let eingabe = CharProps { style: Some("Eingabe".into()), ..Default::default() };
assert_eq!(typed_style(Some(eingabe)).as_deref(), Some("Eingabe"), "the control's run properties");
assert_eq!(typed_style(None), None, "no placeholder style left");
// A producer that localizes the style's name as well: the showing placeholder's own
// style goes all the same.
let mut s = Session::new(wordcraft_doc::Document::new());
let grey = CharProps { color: Some(TextColor::Rgb(Rgb(0x80, 0x80, 0x80))), ..Default::default() };
s.doc.styles.styles.push(style("TexteEspaceReserve", "Texte de l'espace réservé", grey));
run(&mut s, "text.insert", json!({"text": "Nom : "}));
run(&mut s, "developer.plainText", json!({"tag": "nom"}));
let r = s.doc.control_at(&s.sel.focus).unwrap();
s.doc.format_range(&r.content_start(), &r.end, &|c| c.style = Some("TexteEspaceReserve".into())).unwrap();
run(&mut s, "caret.right", json!({}));
run(&mut s, "text.insert", json!({"text": "Ada"}));
let r = s.doc.control_at(&s.sel.focus).unwrap();
let typed = s.doc.para_at(&r.content_start()).unwrap().props_of_char(r.content_start().off).style.clone();
assert_eq!(typed, None, "a placeholder style localized in id and name");
}
#[test]
fn typing_replaces_the_placeholder_and_emptying_brings_it_back() {
let mut s = Session::new(wordcraft_doc::Document::new());
+300 -13
View File
@@ -22,7 +22,7 @@ use wordcraft_doc::{CharProps, Pos, StoryRef};
use wordcraft_geom::Rect;
use wordcraft_layout::{DocLayout, Placed};
use super::ink_shape::{InkShape, recognise};
use super::ink_shape::{ArrowParts, InkShape, arrow_parts, pair_parts, recognise};
use super::pos_json;
use crate::{CmdError, CmdResult, CommandSpec, Session, p};
@@ -461,9 +461,26 @@ fn stroke(s: &mut Session, v: &Value) -> CmdResult {
extra: Default::default(),
};
let at = place(s, &layout, obj, page, (fx, fy), (first[0], first[1]))?;
// Ink to Shape: a stroke that looks like a shape becomes it (in the same undo step).
let shape = if s.view.draw.ink_to_shape && tool != InkTool::Highlighter { convert_at(s, &at)? } else { None };
Ok(json!({"pos": pos_json(&at), "page": page, "rect": [fx, fy, fw, fh], "shape": shape}))
// Ink to Shape: a stroke that looks like a shape becomes it, and a line and a head drawn as two
// strokes become one arrow (in the same undo step).
let (at, shape, rect) = if s.view.draw.ink_to_shape && tool != InkTool::Highlighter {
match pair_with_neighbour(s, page, &at)? {
Some(arrow) => {
// The arrow is the line drawn earlier (or just now): its frame, not the stroke's.
s.touch();
let r = s.layout().object(&arrow, page).map(|o| o.rect);
let rect = r.map_or([fx, fy, fw, fh], |r| [r.x, r.y, r.w, r.h]);
(arrow, Some("arrow"), rect)
}
None => {
let shape = convert_at(s, &at)?;
(at, shape, [fx, fy, fw, fh])
}
}
} else {
(at, None, [fx, fy, fw, fh])
};
Ok(json!({"pos": pos_json(&at), "page": page, "rect": rect, "shape": shape}))
}
/// Insert floating `obj` with its top-left at `(x, y)` on `page`, anchored to the body paragraph
@@ -676,28 +693,141 @@ fn lasso_color(s: &mut Session, v: &Value) -> CmdResult {
// --- Ink to Shape --------------------------------------------------------------------------
/// Turn the lasso-selected strokes that look like shapes into those shapes (one undo step).
/// Turn the lasso-selected strokes that look like shapes into those shapes, and a selected line
/// and arrowhead drawn as two strokes into one arrow (one undo step).
fn convert_selection(s: &mut Session, _: &Value) -> CmdResult {
let sel = lasso_required(s)?;
let mut converted = Vec::new();
for pos in &sel.items {
let layout = s.layout();
let pieces = arrow_pieces(s, &layout, sel.page, |pos, _| sel.items.contains(pos));
let (mut used, mut heads, mut converted): (Vec<Pos>, Vec<Pos>, Vec<(Pos, &'static str)>) = (Vec::new(), Vec::new(), Vec::new());
// Each piece was recognised once; pairing them is cheap geometry, even for a crowded page.
for line in pieces.iter().filter(|k| k.parts.line.is_some()) {
for head in pieces.iter().filter(|k| !k.parts.heads.is_empty()) {
if line.pos == head.pos || used.contains(&line.pos) || used.contains(&head.pos) || !near(line.rect, head.rect) {
continue;
}
if let Some(arrow) = pair_parts(&line.parts, &head.parts) {
set_shape(s, &line.pos, &local_to(&arrow, line.rect))?;
used.extend([line.pos.clone(), head.pos.clone()]);
heads.push(head.pos.clone());
converted.push((line.pos.clone(), "arrow"));
}
}
}
for pos in sel.items.iter().filter(|p| !used.contains(p)) {
if let Some(name) = convert_at(s, pos)? {
converted.push(json!({"pos": pos_json(pos), "shape": name}));
converted.push((pos.clone(), name));
}
}
if converted.is_empty() {
return Err(CmdError::Failed("none of the selected strokes looks like a shape".into()));
}
keep_lasso(s, sel);
// The heads went into their arrows; last first, so the earlier ones keep their place.
heads.sort();
for pos in heads.iter().rev() {
delete_object(s, pos)?;
}
let items = sel.items.iter().filter(|p| !heads.contains(p)).map(|p| after_deleting(p, &heads)).collect();
let converted: Vec<Value> = converted.iter().map(|(p, name)| json!({"pos": pos_json(&after_deleting(p, &heads)), "shape": name})).collect();
keep_lasso(s, LassoSelection { items, ..sel });
Ok(json!({"count": converted.len(), "converted": converted}))
}
/// Where `pos` is once the objects at `deleted` (none of them `pos`) are gone.
fn after_deleting(pos: &Pos, deleted: &[Pos]) -> Pos {
let before = deleted.iter().filter(|d| d.story == pos.story && d.path == pos.path && d.off < pos.off).count();
Pos { off: pos.off.saturating_sub(before.saturating_mul(OBJ.len_utf8())), ..pos.clone() }
}
/// A stroke (or an Ink to Shape line or triangle) on a page that may be one half of a two-stroke
/// arrow: where it is, when it was drawn (ink's drawing order) and what it can be in an arrow.
struct ArrowPiece {
pos: Pos,
rect: Rect,
order: Option<u32>,
parts: ArrowParts,
}
/// Whether two objects laid out at `a` and `b` are close enough to make an arrow (a cheap test
/// before recognising them).
fn near(a: Rect, b: Rect) -> bool {
let reach = 0.2 * a.w.max(a.h).max(b.w).max(b.h);
reach.is_finite() && a.expand(reach).intersects(&b.expand(reach))
}
/// The pieces of two-stroke arrows on `page` that `keep` (given the position and laid-out rect)
/// accepts: unturned, unflipped pen and pencil strokes, and the open two-point lines and closed
/// triangles Ink to Shape makes. `keep` runs first, so only kept pieces are recognised.
fn arrow_pieces(s: &Session, layout: &DocLayout, page: usize, keep: impl Fn(&Pos, &Rect) -> bool) -> Vec<ArrowPiece> {
let Some(pg) = layout.pages.get(page) else { return Vec::new() };
let mut out = Vec::new();
for it in &pg.items {
let Placed::Object { rect, spin, story, path, off, .. } = it else { continue };
let pos = Pos { story: *story, path: path.clone(), off: *off };
if !spin.is_identity() || !keep(&pos, rect) {
continue;
}
let Some(InlineObject::Shape { kind, freeform: Some(f), float, .. }) = s.doc.para_at(&pos).and_then(|p| p.object_at(*off)) else { continue };
let Some(fp) = f.paths.first().filter(|_| f.paths.len() == 1) else { continue };
let pen = matches!(f.ink, Some(InkTool::Pen | InkTool::Pencil)) && !s.view.hide_ink;
let shape = f.ink.is_none() && *kind == ShapeKind::Freeform && matches!((fp.pts.len(), fp.closed), (2, false) | (3, true));
if !(pen || shape) || float.rot != 0.0 || float.flip_h || float.flip_v {
continue;
}
let Some((mut pts, closed)) = f.placed(rect.x, rect.y, rect.w, rect.h).into_iter().next() else { continue };
if closed && let Some(&first) = pts.first() {
pts.push(first);
}
out.push(ArrowPiece { pos, rect: *rect, order: f.order, parts: arrow_parts(&pts) });
}
out
}
/// Ink to Shape while drawing: when the stroke just drawn at `new` and another on the page are a
/// line and an arrowhead (either drawn first), make the line an arrow and drop the head. Returns
/// where the arrow is.
fn pair_with_neighbour(s: &mut Session, page: usize, new: &Pos) -> Result<Option<Pos>, CmdError> {
s.touch(); // lay out the new stroke where it was just placed
let layout = s.layout();
let Some(mine) = layout.object(new, page).map(|o| o.rect) else { return Ok(None) };
// Only the new stroke and its neighbours are recognised, not the whole page.
let mut pieces = arrow_pieces(s, &layout, page, |pos, rect| pos == new || near(mine, *rect));
let Some(at) = pieces.iter().position(|k| k.pos == *new) else { return Ok(None) };
let me = pieces.swap_remove(at);
// The most recently drawn first (ink records its drawing order; shapes and ink from files
// don't, and come after, latest on the page first), so the winner doesn't depend on layout.
pieces.sort_by(|a, b| b.order.cmp(&a.order).then_with(|| b.pos.cmp(&a.pos)));
for other in &pieces {
for (line, head) in [(other, &me), (&me, other)] {
if let Some(arrow) = pair_parts(&line.parts, &head.parts) {
set_shape(s, &line.pos, &local_to(&arrow, line.rect))?;
delete_object(s, &head.pos)?;
return Ok(Some(after_deleting(&line.pos, std::slice::from_ref(&head.pos))));
}
}
}
Ok(None)
}
/// `shape` (in page points) in the frame of the object laid out at `rect`.
fn local_to(shape: &InkShape, rect: Rect) -> InkShape {
let m = |[x, y]: [f32; 2]| [x - rect.x, y - rect.y];
match shape {
InkShape::Arrow { a, b, head_at_end, head_at_start } => {
InkShape::Arrow { a: m(*a), b: m(*b), head_at_end: *head_at_end, head_at_start: *head_at_start }
}
InkShape::Line { a, b } => InkShape::Line { a: m(*a), b: m(*b) },
InkShape::Rectangle { x, y, w, h } => InkShape::Rectangle { x: x - rect.x, y: y - rect.y, w: *w, h: *h },
InkShape::Ellipse { x, y, w, h } => InkShape::Ellipse { x: x - rect.x, y: y - rect.y, w: *w, h: *h },
InkShape::Polygon(c) => InkShape::Polygon(c.iter().copied().map(m).collect()),
}
}
/// Replace the ink stroke at `pos` by the shape it looks like, in its colour and thickness and
/// where it was drawn. `None` (nothing changed) when it isn't a pen or pencil stroke, it is
/// turned or flipped, or it doesn't look like a shape.
fn convert_at(s: &mut Session, pos: &Pos) -> Result<Option<&'static str>, CmdError> {
let para = s.doc.para_mut(pos.story, &pos.path)?;
let Some(InlineObject::Shape { kind, w, h, fill, float, freeform, extra, .. }) = para.object_at_mut(pos.off) else { return Ok(None) };
let Some(InlineObject::Shape { w, h, float, freeform, .. }) = s.doc.para_at(pos).and_then(|p| p.object_at(pos.off)) else { return Ok(None) };
let Some(f) = freeform.as_deref().filter(|f| matches!(f.ink, Some(InkTool::Pen | InkTool::Pencil)) && f.paths.len() == 1) else {
return Ok(None);
};
@@ -707,13 +837,24 @@ fn convert_at(s: &mut Session, pos: &Pos) -> Result<Option<&'static str>, CmdErr
// The stroke in its frame's own points (the frame may have been resized since).
let Some((pts, _)) = f.placed(0.0, 0.0, *w, *h).into_iter().next() else { return Ok(None) };
let Some(found) = recognise(&pts) else { return Ok(None) };
set_shape(s, pos, &found)?;
Ok(Some(found.name()))
}
/// Make the drawing at `pos` the shape `found` (given in the drawing's own frame), keeping its
/// colour and thickness, where it is.
fn set_shape(s: &mut Session, pos: &Pos, found: &InkShape) -> Result<(), CmdError> {
let para = s.doc.para_mut(pos.story, &pos.path)?;
let Some(InlineObject::Shape { kind, w, h, fill, float, freeform, extra, .. }) = para.object_at_mut(pos.off) else {
return Err(CmdError::Failed("no drawing there".into()));
};
let [bx, by, bw, bh] = found.bounds();
let (bw, bh) = (bw.max(1.0), bh.max(1.0));
let outline = |pts: Vec<[f32; 2]>, closed: bool| {
Some(Arc::new(Freeform { w: bw, h: bh, paths: vec![FreePath { pts, closed }], ..Default::default() }.sanitized()))
};
let local = |c: &[[f32; 2]]| c.iter().map(|[x, y]| [x - bx, y - by]).collect::<Vec<_>>();
let (new_kind, geometry) = match &found {
let (new_kind, geometry) = match found {
InkShape::Rectangle { .. } => (ShapeKind::Rectangle, None),
InkShape::Ellipse { .. } => (ShapeKind::Ellipse, None),
InkShape::Line { a, b } => (ShapeKind::Freeform, outline(local(&[*a, *b]), false)),
@@ -735,7 +876,7 @@ fn convert_at(s: &mut Session, pos: &Pos) -> Result<Option<&'static str>, CmdErr
float.x += bx;
float.y += by;
para.touch();
Ok(Some(found.name()))
Ok(())
}
// --- Ink Replay ----------------------------------------------------------------------------
@@ -1017,6 +1158,152 @@ mod tests {
assert!(extra.arrow_end && !extra.arrow_start && float.flip_h, "{extra:?} {float:?}");
}
/// Every drawing in the body, in document order.
fn drawings(s: &Session) -> Vec<InlineObject> {
s.doc
.body
.iter()
.filter_map(|b| b.as_para())
.flat_map(|p| p.objects.iter())
.filter(|o| matches!(o, InlineObject::Shape { .. }))
.cloned()
.collect()
}
/// A straight stroke from (100, 300) to (300, 300).
fn shaft_stroke() -> Vec<[f32; 2]> {
(0..=40).map(|i| [100.0 + i as f32 * 5.0, 300.0]).collect()
}
#[test]
fn ink_to_shape_turns_a_line_and_a_head_drawn_after_it_into_one_arrow_in_one_undo_step() {
let mut s = session();
s.run("draw.inkToShape", &json!({"value": true})).unwrap();
let r = s.run("draw.stroke", &json!({"page": 0, "points": shaft_stroke(), "color": "C00000", "width": 2})).unwrap();
assert_eq!(r["shape"], "line");
let steps = s.undo_depth();
// An open V at the right-hand end, drawn as a second stroke.
let head = [[274.0, 285.0], [287.0, 292.5], [300.0, 300.0], [287.0, 307.5], [274.0, 315.0]];
let r = s.run("draw.stroke", &json!({"page": 0, "points": head})).unwrap();
assert_eq!(r["shape"], "arrow");
assert_eq!(s.undo_depth(), steps + 1, "paired in one undo step");
assert!(ink(&s).is_empty(), "the head stroke is gone");
let all = drawings(&s);
let [InlineObject::Shape { kind, w, stroke, stroke_width, extra, float, .. }] = all.as_slice() else { panic!("one arrow: {all:?}") };
assert_eq!((*kind, *stroke, *stroke_width), (ShapeKind::StraightConnector, Some(wordcraft_doc::Rgb(0xC0, 0, 0)), 2.0));
assert!(extra.arrow_end && !extra.arrow_start && !float.flip_h && !float.flip_v, "{extra:?} {float:?}");
assert!((w - 200.0).abs() < 1.0, "{w}");
let at: wordcraft_doc::Pos = serde_json::from_value(r["pos"].clone()).unwrap();
assert!(matches!(s.doc.para_at(&at).and_then(|p| p.object_at(at.off)), Some(InlineObject::Shape { kind: ShapeKind::StraightConnector, .. })));
let rect = s.layout().object(&at, 0).unwrap().rect;
assert!((rect.x - 100.0).abs() < 1.0 && (rect.y - 300.0).abs() < 1.0, "where the line was: {rect:?}");
let got: Vec<f32> = r["rect"].as_array().unwrap().iter().map(|v| v.as_f64().unwrap() as f32).collect();
assert!(
got.len() == 4 && [rect.x, rect.y, rect.w, rect.h].iter().zip(&got).all(|(a, b)| (a - b).abs() < 0.5),
"the arrow's rect, not the deleted head's: {got:?} vs {rect:?}"
);
// It survives a DOCX save and open.
let back = wordcraft_docx::read(&wordcraft_docx::write(&s.doc).unwrap()).unwrap();
let again: Vec<_> = back.body.iter().filter_map(|b| b.as_para()).flat_map(|p| p.objects.iter()).collect();
assert!(
again
.iter()
.any(|o| matches!(o, InlineObject::Shape { kind: ShapeKind::StraightConnector, extra, .. } if extra.arrow_end && !extra.arrow_start)),
"{again:?}"
);
// Undo brings back the line, without the head.
s.run("edit.undo", &json!({})).unwrap();
let all = drawings(&s);
let [InlineObject::Shape { kind: ShapeKind::Freeform, extra, .. }] = all.as_slice() else { panic!("the line: {all:?}") };
assert!(!extra.arrow_end);
assert!(ink(&s).is_empty());
}
#[test]
fn ink_to_shape_pairs_a_closed_triangle_head_at_the_start_and_a_head_drawn_first() {
let mut s = session();
s.run("draw.inkToShape", &json!({"value": true})).unwrap();
s.run("draw.stroke", &json!({"page": 0, "points": shaft_stroke()})).unwrap();
// A closed triangle at the left end, where the line began: the arrow points left.
let tri = [[100.0, 300.0], [113.0, 292.5], [126.0, 285.0], [126.0, 300.0], [126.0, 315.0], [113.0, 307.5], [100.0, 300.0]];
let r = s.run("draw.stroke", &json!({"page": 0, "points": tri})).unwrap();
assert_eq!(r["shape"], "arrow");
let all = drawings(&s);
let [InlineObject::Shape { kind: ShapeKind::StraightConnector, extra, float, .. }] = all.as_slice() else { panic!("one arrow: {all:?}") };
assert!(extra.arrow_end && float.flip_h && !float.flip_v, "tip at the left: {extra:?} {float:?}");
// Head first (it stays ink on its own), then the shaft that ends at its apex.
let mut s = session();
s.run("draw.inkToShape", &json!({"value": true})).unwrap();
let r = s.run("draw.stroke", &json!({"page": 0, "points": [[274.0, 285.0], [300.0, 300.0], [274.0, 315.0]]})).unwrap();
assert!(r["shape"].is_null());
let r = s.run("draw.stroke", &json!({"page": 0, "points": shaft_stroke()})).unwrap();
assert_eq!(r["shape"], "arrow");
assert!(ink(&s).is_empty());
let all = drawings(&s);
let [InlineObject::Shape { kind: ShapeKind::StraightConnector, extra, float, .. }] = all.as_slice() else { panic!("one arrow: {all:?}") };
assert!(extra.arrow_end && !float.flip_h, "{extra:?} {float:?}");
// A V far from the line's ends stays ink, and the line stays a line.
let mut s = session();
s.run("draw.inkToShape", &json!({"value": true})).unwrap();
s.run("draw.stroke", &json!({"page": 0, "points": shaft_stroke()})).unwrap();
let r = s.run("draw.stroke", &json!({"page": 0, "points": [[174.0, 385.0], [200.0, 400.0], [174.0, 415.0]]})).unwrap();
assert!(r["shape"].is_null());
assert_eq!(ink(&s).len(), 1);
assert!(drawings(&s).iter().all(|o| !matches!(o, InlineObject::Shape { kind: ShapeKind::StraightConnector, .. })));
}
#[test]
fn converting_a_lasso_selection_pairs_a_line_and_a_head_stroke_into_an_arrow() {
let mut s = session();
s.run("draw.stroke", &json!({"page": 0, "points": shaft_stroke(), "color": "0070C0"})).unwrap();
s.run("draw.stroke", &json!({"page": 0, "points": [[274.0, 285.0], [300.0, 300.0], [274.0, 315.0]]})).unwrap();
s.run("draw.stroke", &json!({"page": 0, "points": rect_stroke(100.0, 400.0, 120.0, 60.0)})).unwrap();
assert_eq!(ink(&s).len(), 3);
let steps = s.undo_depth();
s.run("draw.lasso", &json!({"page": 0, "points": [[80, 260], [340, 260], [340, 480], [80, 480]]})).unwrap();
let r = s.run("draw.inkToShape", &json!({})).unwrap();
let names: Vec<&str> = r["converted"].as_array().unwrap().iter().filter_map(|c| c["shape"].as_str()).collect();
assert!(names.contains(&"arrow") && names.contains(&"rectangle"), "{names:?}");
assert!(ink(&s).is_empty(), "the head stroke went into the arrow");
let all = drawings(&s);
assert_eq!(all.len(), 2, "{all:?}");
let arrow = all.iter().find(|o| matches!(o, InlineObject::Shape { kind: ShapeKind::StraightConnector, .. }));
let Some(InlineObject::Shape { extra, stroke, .. }) = arrow else { panic!("{all:?}") };
assert!(extra.arrow_end && *stroke == Some(wordcraft_doc::Rgb(0, 0x70, 0xC0)));
// The selection still holds what's left: deleting it removes both shapes.
s.run("draw.lassoDelete", &json!({})).unwrap();
assert!(drawings(&s).is_empty());
s.run("edit.undo", &json!({})).unwrap();
assert_eq!(drawings(&s).len(), 2);
s.run("edit.undo", &json!({})).unwrap();
assert_eq!(ink(&s).len(), 3, "one undo step brings all three strokes back");
assert_eq!(s.undo_depth(), steps);
}
#[test]
fn converting_a_crowded_lasso_selection_recognises_each_stroke_a_bounded_number_of_times() {
let mut s = session();
s.run("draw.stroke", &json!({"page": 0, "points": shaft_stroke()})).unwrap();
s.run("draw.stroke", &json!({"page": 0, "points": [[274.0, 285.0], [300.0, 300.0], [274.0, 315.0]]})).unwrap();
// A stack of 40 lines close together: every pair is near, none makes an arrow.
let n = 40;
for i in 0..n {
let y = 420.0 + i as f32 * 2.0;
let pts: Vec<[f32; 2]> = (0..=40).map(|k| [100.0 + k as f32 * 5.0, y]).collect();
s.run("draw.stroke", &json!({"page": 0, "points": pts})).unwrap();
}
s.run("draw.lasso", &json!({"page": 0, "points": [[80, 260], [340, 260], [340, 560], [80, 560]]})).unwrap();
super::super::ink_shape::RECOGNISED.with(|c| c.set(0));
let r = s.run("draw.inkToShape", &json!({})).unwrap();
let calls = super::super::ink_shape::RECOGNISED.with(|c| c.get());
let strokes = n + 2;
assert!(calls <= 2 * strokes, "{calls} recognitions for {strokes} strokes");
let names: Vec<&str> = r["converted"].as_array().unwrap().iter().filter_map(|c| c["shape"].as_str()).collect();
assert_eq!(names.iter().filter(|n| **n == "arrow").count(), 1, "{names:?}");
assert_eq!(names.iter().filter(|n| **n == "line").count(), n, "{names:?}");
assert!(ink(&s).is_empty());
}
#[test]
fn ink_replay_plays_strokes_in_the_order_they_were_drawn() {
let mut s = session();
+65 -8
View File
@@ -61,6 +61,18 @@ pub fn specs() -> Vec<CommandSpec> {
.pure(),
CommandSpec::new("edit.cut", "Cut", "Home › Clipboard", cut).key("Mod+X").when(has_sel),
CommandSpec::new("edit.copy", "Copy", "Home › Clipboard", copy).key("Mod+C / Mod+Insert").pure(),
// Word's Spike (Ctrl+F3, Ctrl+Shift+F3; ⌘ on the Mac): the AutoText entry "Spike", so Quick
// Parts and AutoText insert it without emptying it, as in Word.
CommandSpec::new("edit.spike", "Spike", "Editing", spike).key("Mod+F3").when(has_sel),
CommandSpec::new("edit.insertSpike", "Insert Spike", "Editing", insert_spike).key("Mod+Shift+F3").when(|s| {
if s.math.is_some() {
Some("leave the equation first")
} else if s.building_blocks.contains_key(SPIKE) {
None
} else {
Some("the Spike is empty")
}
}),
CommandSpec::new("edit.paste", "Paste", "Home › Clipboard", paste).key("Mod+V / Shift+Insert").params(r#"{"text"?: string, "html"?: string, "rtf"?: string}"#),
CommandSpec::new("edit.pasteText", "Paste: Keep Text Only", "Home › Clipboard › Paste", paste_text)
.params(r#"{"text"?: string}"#)
@@ -167,13 +179,25 @@ fn copy(s: &mut Session, _: &Value) -> CmdResult {
}
/// What Copy and Cut return: the clipboard's text, and the copy as HTML (equations as MathML,
/// #816) for the host to put on the system clipboard too, so other programs keep the formatting.
/// #816) for the host to put on the system clipboard too, so other programs keep the formatting;
/// and for a copy of one equation alone, its MathML (#1297).
pub(crate) fn copy_result(s: &mut Session) -> Value {
s.clipboard_html = match &s.clipboard {
Some(f) => wordcraft_formats::html::export_fragment(&f.to_document()),
None => String::new(),
};
if s.clipboard_html.is_empty() { json!({"text": s.clipboard_text}) } else { json!({"text": s.clipboard_text, "html": s.clipboard_html}) }
let mut r = json!({"text": s.clipboard_text});
if !s.clipboard_html.is_empty() {
r["html"] = json!(s.clipboard_html);
}
// A copy of one equation alone, as MathML for the clipboard formats equation editors read
// ("MathML" on Windows, `public.mathml.presentation` on macOS; #1297). Word looks for them
// before anything else when it pastes; a copy with text around the equation doesn't offer
// them, so pasting it keeps the text.
if let Some((linear, display)) = s.clipboard.as_ref().and_then(|f| f.sole_equation()) {
r["mathml"] = json!(wordcraft_formats::mathml::linear_to_mathml(linear, display));
}
r
}
fn cut(s: &mut Session, v: &Value) -> CmdResult {
@@ -186,11 +210,42 @@ fn cut(s: &mut Session, v: &Value) -> CmdResult {
Ok(r)
}
/// Paste keeps the source formatting: our own copy while the system clipboard still holds it,
/// else another program's HTML or RTF (#687), else the system clipboard's text.
fn paste(s: &mut Session, v: &Value) -> CmdResult {
let frag = super::paste::paste_fragment(s, v)?;
/// The AutoText entry the Spike collects into.
const SPIKE: &str = "Spike";
/// Ctrl+F3: cut the selection to the end of the Spike. The clipboard is left alone.
fn spike(s: &mut Session, _: &Value) -> CmdResult {
let (a, b) = s.sel.ordered();
// Column selection collapses the selection before the command runs.
if a == b {
return Err(CmdError::Disabled("edit.spike: nothing is selected".into()));
}
let piece = s.doc.copy_range(&a, &b);
if piece.blocks.is_empty() {
return Err(CmdError::Disabled("edit.spike: this selection can't be collected".into()));
}
delete_selection(s)?;
super::controls::restore_placeholder(s)?;
s.building_blocks.entry(SPIKE.to_string()).or_default().append_piece(piece);
sel_result(s)
}
/// Ctrl+Shift+F3: insert everything the Spike collected, in order, and empty it.
fn insert_spike(s: &mut Session, _: &Value) -> CmdResult {
let frag = s.building_blocks.get(SPIKE).cloned().ok_or_else(|| CmdError::Disabled("edit.insertSpike: the Spike is empty".into()))?;
super::paste::insert(s, frag)?;
s.building_blocks.remove(SPIKE);
sel_result(s)
}
/// Paste keeps the source formatting: our own copy while the system clipboard still holds it,
/// else another program's HTML or RTF (#687). The system clipboard's text has none, so it pastes
/// as Keep Text Only does.
fn paste(s: &mut Session, v: &Value) -> CmdResult {
match super::paste::paste_content(s, v)? {
super::paste::Pasted::Fragment(frag) => super::paste::insert(s, frag)?,
super::paste::Pasted::Text(text) => super::paste::insert_text(s, &text)?,
}
sel_result(s)
}
@@ -644,7 +699,8 @@ fn replace_all(s: &mut Session, v: &Value) -> CmdResult {
super::paste_text(s, &with)?;
continue;
}
let props = s.doc.para_at(&a).map(|p| p.props_of_char(a.off).clone()).unwrap_or_default();
// Replacing a symbol gives the new text the symbol's formatting, not its own font.
let props = s.doc.para_at(&a).map(|p| p.props_of_char(a.off).clone().without_symbol()).unwrap_or_default();
let mut at = s.doc.delete_range(&a, &b)?;
for (k, part) in with.split(['\r', '\u{2029}']).enumerate() {
if k > 0 {
@@ -708,7 +764,8 @@ fn goto(s: &mut Session, v: &Value) -> CmdResult {
fn painter_pick(s: &mut Session) {
let f = s.sel.ordered().0;
let chr = s.doc.para_at(&f).map(|p| p.props_of_char(f.off).clone()).unwrap_or_default();
// A symbol's own font isn't formatting to paint onto text.
let chr = s.doc.para_at(&f).map(|p| p.props_of_char(f.off).clone().without_symbol()).unwrap_or_default();
let para = s.doc.para_at(&f).map(|p| p.props.clone()).unwrap_or_default();
s.painter = Some((chr, para, false));
}
+50
View File
@@ -891,4 +891,54 @@ mod tests {
assert_eq!(s.run("equation.get", &json!({})).unwrap()["latex"], want, "{tex}");
}
}
#[test]
fn typed_multi_letter_scripts_stay_whole() {
// #1609, Word for Mac 16.113.4: `V_dd` + Space leaves "dd" entirely in the subscript.
for typed in ["V_dd", "V_(dd)"] {
let mut s = session();
s.run("insert.equation", &json!({})).unwrap();
s.run("equation.type", &json!({"text": typed})).unwrap();
s.run("equation.type", &json!({"text": " "})).unwrap();
let nodes = nodes_of(&s, &s.math.as_ref().unwrap().at).unwrap();
assert_eq!(nodes.len(), 1, "{typed}: {nodes:#?}");
assert!(
matches!(&nodes[0], MNode::Script { sub, .. } if matches!(sub.as_slice(), [MNode::Run(r)] if r.text == "dd")),
"{typed}: {nodes:#?}"
);
}
let mut s = session();
s.run("insert.equation", &json!({})).unwrap();
s.run("equation.type", &json!({"text": "x^ab "})).unwrap();
let nodes = nodes_of(&s, &s.math.as_ref().unwrap().at).unwrap();
assert!(matches!(&nodes[0], MNode::Script { sup, .. } if matches!(sup.as_slice(), [MNode::Run(r)] if r.text == "ab")), "{nodes:#?}");
}
#[test]
fn a_built_script_keeps_the_letter_typed_after_it() {
// #1609 review: `x_i y ` built x_{iy}: the letter after a script is its own item.
let mut s = session();
s.run("insert.equation", &json!({})).unwrap();
s.run("equation.type", &json!({"text": "x_i y "})).unwrap();
let nodes = nodes_of(&s, &s.math.as_ref().unwrap().at).unwrap();
assert_eq!(nodes.len(), 2, "{nodes:#?}");
assert!(matches!(&nodes[0], MNode::Script { sub, .. } if matches!(sub.as_slice(), [MNode::Run(r)] if r.text == "i")), "{nodes:#?}");
assert!(matches!(&nodes[1], MNode::Run(r) if r.text.trim() == "y"), "{nodes:#?}");
let mut s = session();
s.run("insert.equation", &json!({})).unwrap();
s.run("equation.type", &json!({"text": "a_i b_j "})).unwrap();
let nodes = nodes_of(&s, &s.math.as_ref().unwrap().at).unwrap();
let subs: Vec<&str> = nodes
.iter()
.filter_map(|n| match n {
MNode::Script { sub, .. } => match sub.as_slice() {
[MNode::Run(r)] => Some(r.text.as_str()),
_ => None,
},
_ => None,
})
.collect();
assert_eq!(subs, ["i", "j"], "{nodes:#?}");
assert!(nodes.iter().all(|n| matches!(n, MNode::Script { .. }) || matches!(n, MNode::Run(r) if r.text.trim().is_empty())), "{nodes:#?}");
}
}
+48 -2
View File
@@ -188,11 +188,17 @@ fn shown_text(s: &mut Session, f: &FieldAt) -> String {
};
let l = s.layout();
let page = l.caret(&f.at).and_then(|c| l.pages.get(c.page));
let section = page.map(|p| (p.section + 1) as u32).unwrap_or(1);
let section_pages = page
.map(|p| l.pages.iter().filter(|pg| pg.section == p.section).count() as u32)
.unwrap_or_else(|| u32::try_from(l.pages.len()).unwrap_or(u32::MAX));
let page_format = page.and_then(|p| s.doc.sections().get(p.section).map(|(_, sec)| sec.page_num_format)).unwrap_or_default();
let ctx = wordcraft_layout::fields::FieldCtx {
page: page.map(|p| p.number).unwrap_or(1),
page_format,
pages: u32::try_from(l.pages.len()).unwrap_or(u32::MAX),
section_pages: u32::try_from(l.pages.len()).unwrap_or(u32::MAX),
section: 1,
section_pages,
section,
title: s.doc.core.title.as_str().into(),
author: s.doc.core.creator.as_str().into(),
filename: s.path.as_ref().and_then(|p| p.file_name()).map(|n| n.to_string_lossy().to_string()).unwrap_or_default().into(),
@@ -293,6 +299,30 @@ mod tests {
out
}
/// #1765: ASK and SET fields draw nothing on the page; their stored results are kept (and
/// saved), and code view still shows them.
#[test]
fn ask_and_set_fields_draw_nothing() {
let mut s = Session::new(wordcraft_doc::Document::new());
s.run("insert.field", &json!({"instr": "ASK Nominativo \"Inserire nome\" \\d \"N\"", "result": "«NOME» «COGNOME»"})).unwrap();
s.run("insert.field", &json!({"instr": "SET Codice \"COD\"", "result": "COD"})).unwrap();
s.run("text.insert", &json!({"text": "Title"})).unwrap();
let shown = page_text(&mut s);
assert!(!shown.contains("NOME") && !shown.contains("COD"), "{shown}");
let para = s.doc.para(wordcraft_doc::StoryRef::Body, &wordcraft_doc::Path::top(0)).unwrap();
let results: Vec<&str> = para
.objects
.iter()
.filter_map(|o| match o {
InlineObject::Field { result, .. } => Some(result.as_str()),
_ => None,
})
.collect();
assert_eq!(results, ["«NOME» «COGNOME»", "COD"]);
s.run("view.fieldCodes", &json!({})).unwrap();
assert!(page_text(&mut s).contains("{ SET Codice \"COD\" }"));
}
/// Alt+F9 shows every field's code in braces; Shift+F9 flips just the selected one; F9 shows
/// its result again.
#[test]
@@ -384,4 +414,20 @@ mod tests {
s.run("caret.docStart", &json!({})).unwrap();
assert!(s.run("fields.lock", &json!({})).is_err());
}
#[test]
fn unlink_uses_section_context() {
let mut s = Session::new(wordcraft_doc::Document::new());
// Insert section break so there are 2 sections
s.run("layout.break", &json!({"kind": "nextPage"})).unwrap();
// In section 2: insert SECTION and SECTIONPAGES fields
s.run("insert.field", &json!({"instr": "SECTION"})).unwrap();
s.run("text.insert", &json!({"text": " of "})).unwrap();
s.run("insert.field", &json!({"instr": "SECTIONPAGES"})).unwrap();
// Select both and unlink
s.run("select.all", &json!({})).unwrap();
s.run("fields.unlink", &json!({})).unwrap();
let text = s.doc.plain_text(wordcraft_doc::StoryRef::Body);
assert!(text.contains("2 of 1"), "unlinked text should reflect section 2: {text}");
}
}
+12 -3
View File
@@ -80,7 +80,12 @@ pub fn specs() -> Vec<CommandSpec> {
.params(r#"{"password": string (Word documents are saved encrypted with it) | null or "" (saved unencrypted)}"#)
.pure(),
CommandSpec::new("file.exportPdf", "Export PDF", "File › Export", |s, v| {
let path = p::req_str(v, "path")?;
// No path (the command search, say): the app asks where to save, as File › Export
// does, instead of failing on the required parameter (#2313).
let Some(path) = p::str(v, "path") else {
s.ui_requests.push(json!({"open": "exportPdf"}));
return Ok(json!({"exported": false}));
};
let path = if path.to_ascii_lowercase().ends_with(".pdf") { path.to_string() } else { format!("{path}.pdf") };
let opts = export_options(s, v);
crate::io::save_path_with(std::path::Path::new(&path), &s.doc, None, &opts).map_err(CmdError::Failed)?;
@@ -100,7 +105,7 @@ pub fn specs() -> Vec<CommandSpec> {
.pure(),
CommandSpec::new("file.printPdf", "Print to PDF", "File › Print", crate::print::print_pdf_to_file)
.params(
r#"{"path": string, "range"?: "all" | "current" | "selection" (the pages the selection touches) | "custom", "pages"?: "1-3,5" (implies custom), "parity"?: "all" | "odd" | "even", "copies"?: 1-999, "collate"?: bool (default true)}"#,
r#"{"path": string, "range"?: "all" | "current" | "selection" (the pages the selection touches) | "custom", "pages"?: "1-3,5" (implies custom), "parity"?: "all" | "odd" | "even", "copies"?: 1-999, "collate"?: bool (default true), "paper"?: "none" | "Letter" | "Legal" | "A4" | … (a Layout › Size name: scale every page to this paper, centred; the document keeps its size)}"#,
)
.pure(),
CommandSpec::new("file.print", "Print", "File", |s, _| {
@@ -334,7 +339,11 @@ pub fn export_options(s: &Session, v: &Value) -> crate::io::ExportOptions {
}
fn export_png(s: &mut Session, v: &Value) -> CmdResult {
let path = p::req_str(v, "path")?;
// No path: the app asks where to save, as file.exportPdf does (#2313).
let Some(path) = p::str(v, "path") else {
s.ui_requests.push(json!({"open": "exportPng"}));
return Ok(json!({"exported": false}));
};
let page = p::u64(v, "page").unwrap_or(1).max(1) as usize - 1;
let scale = p::f32(v, "scale").unwrap_or(2.0).clamp(0.1, 8.0);
let l = s.export_layout_with(export_options(s, v).markup);
+2
View File
@@ -685,6 +685,8 @@ pub fn state(s: &Session) -> Value {
"caps": all.iter().all(|r| r.caps),
"hidden": all.iter().all(|r| r.hidden),
"spacing": same(&|r| json!(r.spacing)),
// Kerning for fonts: the smallest kerned size, points; 0 = off.
"kern": same(&|r| json!(r.kern)),
"color": same(&|r| json!(match r.color { TextColor::Auto => "auto".to_string(), TextColor::Rgb(c) => c.hex() })),
"highlight": same(&|r| json!(r.highlight.map(|c| c.hex()))),
"style": style,
+317 -36
View File
@@ -4,8 +4,10 @@
//! between its ends; a closed one is an ellipse when its points sit on the ellipse inscribed in
//! its bounds (low RMS of the normalised radius), else a polygon from its corners (the closed
//! path simplified with Ramer–Douglas–Peucker, near-straight and crowded corners merged) when
//! that polygon stays close to every point. An arrow is a line-like shaft plus a head of two barbs
//! drawn in the same stroke (shaft first or head first). Anything else (scribbles, curves, letters) stays ink.
//! that polygon stays close to every point. An arrow is a line-like shaft plus a head drawn in the
//! same stroke (shaft first or head first): two barbs, or a small closed triangle round the tip.
//! [`pair_arrow`] makes one arrow of two strokes: a line, and a head (an open V or a small closed
//! triangle) whose apex sits at one of its ends. Anything else (scribbles, curves, letters) stays ink.
use wordcraft_doc::freeform::MAX_POINTS;
@@ -55,15 +57,17 @@ const MIN_SIZE: f32 = 6.0;
/// Most corners a polygon may have.
pub const MAX_CORNERS: usize = 8;
#[cfg(test)]
thread_local! {
/// How many times [`recognise`] ran on this thread (tests check the work stays bounded).
pub static RECOGNISED: std::cell::Cell<usize> = const { std::cell::Cell::new(0) };
}
/// The shape `pts` draws, if it is one.
pub fn recognise(pts: &[[f32; 2]]) -> Option<InkShape> {
// Finite points only, without repeats.
let mut p: Vec<[f32; 2]> = Vec::with_capacity(pts.len().min(MAX_POINTS));
for q in pts.iter().take(MAX_POINTS).filter(|[x, y]| x.is_finite() && y.is_finite() && x.abs() < 1e6 && y.abs() < 1e6) {
if p.last() != Some(q) {
p.push(*q);
}
}
#[cfg(test)]
RECOGNISED.with(|c| c.set(c.get().saturating_add(1)));
let p = clean(pts);
let (first, last) = (*p.first()?, *p.last()?);
if p.len() < 2 {
return None;
@@ -83,8 +87,10 @@ pub fn recognise(pts: &[[f32; 2]]) -> Option<InkShape> {
return Some(InkShape::Line { a: first, b: last });
}
}
// An arrow: a shaft and a head with both barbs, drawn in one stroke (either end first).
if let Some(arrow) = arrow(&p).or_else(|| arrow(&p.iter().rev().copied().collect::<Vec<_>>())) {
// An arrow: a shaft and a head (both barbs, or a closed triangle), drawn in one stroke
// (either end first).
let rev: Vec<[f32; 2]> = p.iter().rev().copied().collect();
if let Some(arrow) = arrow(&p).or_else(|| arrow(&rev)).or_else(|| closed_arrow(&p)).or_else(|| closed_arrow(&rev)) {
return Some(arrow);
}
// Other shapes are closed: the stroke ends near where it began and goes around.
@@ -126,27 +132,7 @@ const BARB_ANGLE: (f32, f32) = (15.0, 75.0);
/// similar angles and lengths. (A stroke drawn head first is tried reversed by the caller.)
fn arrow(p: &[[f32; 2]]) -> Option<InkShape> {
let tail = *p.first()?;
// The tip: start at the first point about as far from the start as any (a return that
// overshoots it by a point or two must not move it), then follow the stroke on while it still
// moves away from the start (densely sampled strokes have shaft points inside that window).
// The first barb heads back, so the walk stops at the real tip and never reaches an overshoot;
// jitter of under half a point doesn't stop it early.
let dmax = p.iter().map(|q| dist(tail, *q)).fold(0.0, f32::max);
let mut ti = p.iter().position(|q| dist(tail, *q) >= dmax - (0.03 * dmax).max(2.0))?;
let mut best = dist(tail, *p.get(ti)?);
for (i, q) in p.iter().enumerate().skip(ti + 1) {
let d = dist(tail, *q);
if d >= best {
(ti, best) = (i, d);
} else if d < best - 0.5 {
break;
}
}
let tip = *p.get(ti)?;
let shaft = dist(tail, tip);
if shaft < MIN_SIZE || p.get(..=ti)?.iter().any(|q| seg_distance(tail, tip, *q) > (0.06 * shaft).max(1.5)) {
return None;
}
let (ti, tip, shaft) = shaft_tip(p)?;
// The head: the stroke after the tip leaves it twice, once for each barb.
let head = p.get(ti..)?;
let reach = head.iter().map(|q| dist(tip, *q)).fold(0.0, f32::max);
@@ -172,20 +158,166 @@ fn arrow(p: &[[f32; 2]]) -> Option<InkShape> {
}
let &[(l1, b1), (l2, b2)] = barbs.as_slice() else { return None };
// The pen returns to the tip between the barbs.
if gap > 0.35 * l1.min(l2) {
if gap > 0.35 * l1.min(l2) || shaft <= 0.0 {
return None;
}
// Each barb as a vector from the tip, against the direction back along the shaft.
head_fits(tail, tip, tip, b1, b2).then_some(InkShape::Arrow { a: tail, b: tip, head_at_end: true, head_at_start: false })
}
/// An arrow with a closed triangle head, drawn shaft first: the shaft runs from the start to the
/// tip like [`arrow`]'s, then the pen goes round a small triangle (tip, one barb end, the other)
/// and ends back at the tip. (Drawn head first, the caller tries it reversed.)
fn closed_arrow(p: &[[f32; 2]]) -> Option<InkShape> {
let tail = *p.first()?;
let (ti, tip, _) = shaft_tip(p)?;
let head = p.get(ti..)?;
// The corners: the point farthest from the tip, then the one farthest from that side.
let b1 = head.iter().copied().fold(tip, |m, q| if dist(tip, q) > dist(tip, m) { q } else { m });
let b2 = head.iter().copied().fold(tip, |m, q| if seg_distance(tip, b1, q) > seg_distance(tip, b1, m) { q } else { m });
let (l1, l2) = (dist(tip, b1), dist(tip, b2));
if l1.min(l2) <= 0.0 || dist(*head.last()?, tip) > 0.35 * l1.min(l2) {
return None;
}
// Every point of the head is on the triangle.
let tri = [tip, b1, b2];
if head.iter().any(|q| polygon_distance(&tri, *q) > (0.15 * l1).max(1.5)) {
return None;
}
head_fits(tail, tip, tip, b1, b2).then_some(InkShape::Arrow { a: tail, b: tip, head_at_end: true, head_at_start: false })
}
/// Whether a head whose barbs run from `apex` to `b1` and `b2` fits a shaft from `tail` to `tip`:
/// barbs 5–50 % of the shaft, each 15–75° off the way back along it, on opposite sides of it,
/// and of similar length and angle.
fn head_fits(tail: [f32; 2], tip: [f32; 2], apex: [f32; 2], b1: [f32; 2], b2: [f32; 2]) -> bool {
let shaft = dist(tail, tip);
let (l1, l2) = (dist(apex, b1), dist(apex, b2));
if !(shaft > 0.0 && l1 > 0.0 && l2 > 0.0) {
return false;
}
// Each barb as a vector from the apex, against the direction back along the shaft.
let back = [tail[0] - tip[0], tail[1] - tip[1]];
let angle_side = |e: [f32; 2], len: f32| {
let u = [e[0] - tip[0], e[1] - tip[1]];
let u = [e[0] - apex[0], e[1] - apex[1]];
let angle = ((back[0] * u[0] + back[1] * u[1]) / (shaft * len)).clamp(-1.0, 1.0).acos().to_degrees();
(angle, back[0] * u[1] - back[1] * u[0])
};
let ((a1, s1), (a2, s2)) = (angle_side(b1, l1), angle_side(b2, l2));
let ok = |l: f32, a: f32| (BARB_LEN.0 * shaft..=BARB_LEN.1 * shaft).contains(&l) && (BARB_ANGLE.0..=BARB_ANGLE.1).contains(&a);
let similar = l1.max(l2) <= 2.5 * l1.min(l2) && (a1 - a2).abs() <= 35.0;
(ok(l1, a1) && ok(l2, a2) && s1 * s2 < 0.0 && similar).then_some(InkShape::Arrow { a: tail, b: tip, head_at_end: true, head_at_start: false })
ok(l1, a1) && ok(l2, a2) && s1 * s2 < 0.0 && similar
}
/// How far a two-stroke arrow's head apex may sit from the line's end, as a fraction of the line.
const PAIR_REACH: f32 = 0.12;
/// What a stroke can be in a two-stroke arrow, worked out once (one recognition): the ends of the
/// straight line it is, if it is one, and the ways it can be an arrowhead on its own.
#[derive(Debug, Clone, Default, PartialEq)]
pub struct ArrowParts {
pub line: Option<([f32; 2], [f32; 2])>,
pub heads: Vec<[[f32; 2]; 3]>,
}
/// `pts` as a possible half of a two-stroke arrow (see [`pair_parts`]).
pub fn arrow_parts(pts: &[[f32; 2]]) -> ArrowParts {
let p = clean(pts);
let found = recognise(&p);
let line = match found {
Some(InkShape::Line { a, b }) => Some((a, b)),
_ => None,
};
ArrowParts { line, heads: heads(&p, found.as_ref()) }
}
/// One arrow from two strokes: `line`, a straight stroke, and `head`, an open V (one barb end,
/// the apex, the other end) or a small closed triangle, whose apex is at one end of the line and
/// whose barbs point back along it (the single-stroke head limits). The arrow runs from the
/// line's other end to that end; `None` when they don't make one.
pub fn pair_arrow(line: &[[f32; 2]], head: &[[f32; 2]]) -> Option<InkShape> {
pair_parts(&arrow_parts(line), &arrow_parts(head))
}
/// [`pair_arrow`] on strokes already recognised by [`arrow_parts`]: cheap geometry only.
pub fn pair_parts(line: &ArrowParts, head: &ArrowParts) -> Option<InkShape> {
let (a, b) = line.line?;
let mut best: Option<(f32, InkShape)> = None;
for &[apex, b1, b2] in &head.heads {
for (tail, tip) in [(a, b), (b, a)] {
let off = dist(apex, tip);
let near = off <= (PAIR_REACH * dist(tail, tip)).max(3.0);
if near && best.as_ref().is_none_or(|(d, _)| off < *d) && head_fits(tail, tip, apex, b1, b2) {
best = Some((off, InkShape::Arrow { a: tail, b: tip, head_at_end: true, head_at_start: false }));
}
}
}
best.map(|(_, arrow)| arrow)
}
/// The ways the cleaned points `p` (which recognise as `found`) can be an arrowhead on their own,
/// as `[apex, barb end, barb end]`: each corner of a closed triangle, or the corner of an open V
/// whose two arms are straight.
fn heads(p: &[[f32; 2]], found: Option<&InkShape>) -> Vec<[[f32; 2]; 3]> {
if let Some(InkShape::Polygon(c)) = found
&& let &[x, y, z] = c.as_slice()
{
return vec![[x, y, z], [y, z, x], [z, x, y]];
}
let (Some(&f), Some(&l)) = (p.first(), p.last()) else { return Vec::new() };
// The apex: the point farthest from the chord between the ends.
let Some((ai, apex)) = p.iter().copied().enumerate().fold(None, |m: Option<(usize, [f32; 2])>, (i, q)| match m {
Some((_, b)) if seg_distance(f, l, b) >= seg_distance(f, l, q) => m,
_ => Some((i, q)),
}) else {
return Vec::new();
};
let straight = |a: [f32; 2], b: [f32; 2], arm: Option<&[[f32; 2]]>| {
let len = dist(a, b);
len >= 1.0 && arm.is_some_and(|arm| arm.iter().all(|q| seg_distance(a, b, *q) <= (0.12 * len).max(1.5)))
};
if ai == 0 || ai + 1 >= p.len() || !straight(f, apex, p.get(..=ai)) || !straight(apex, l, p.get(ai..)) {
return Vec::new();
}
vec![[apex, f, l]]
}
/// Finite, not absurdly far points only (at most [`MAX_POINTS`]), without repeats.
fn clean(pts: &[[f32; 2]]) -> Vec<[f32; 2]> {
let mut p: Vec<[f32; 2]> = Vec::with_capacity(pts.len().min(MAX_POINTS));
for q in pts.iter().take(MAX_POINTS).filter(|[x, y]| x.is_finite() && y.is_finite() && x.abs() < 1e6 && y.abs() < 1e6) {
if p.last() != Some(q) {
p.push(*q);
}
}
p
}
/// A shaft drawn from the start of `p`: the index of its tip, the tip and the shaft's length;
/// `None` when it is too short or not straight.
fn shaft_tip(p: &[[f32; 2]]) -> Option<(usize, [f32; 2], f32)> {
let tail = *p.first()?;
// The tip: start at the first point about as far from the start as any (a return that
// overshoots it by a point or two must not move it), then follow the stroke on while it still
// moves away from the start (densely sampled strokes have shaft points inside that window).
// The first barb heads back, so the walk stops at the real tip and never reaches an overshoot;
// jitter of under half a point doesn't stop it early.
let dmax = p.iter().map(|q| dist(tail, *q)).fold(0.0, f32::max);
let mut ti = p.iter().position(|q| dist(tail, *q) >= dmax - (0.03 * dmax).max(2.0))?;
let mut best = dist(tail, *p.get(ti)?);
for (i, q) in p.iter().enumerate().skip(ti + 1) {
let d = dist(tail, *q);
if d >= best {
(ti, best) = (i, d);
} else if d < best - 0.5 {
break;
}
}
let tip = *p.get(ti)?;
let shaft = dist(tail, tip);
if shaft < MIN_SIZE || p.get(..=ti)?.iter().any(|q| seg_distance(tail, tip, *q) > (0.06 * shaft).max(1.5)) {
return None;
}
Some((ti, tip, shaft))
}
/// Left, top, width and height of `pts` (zero for none).
@@ -686,4 +818,153 @@ mod tests {
let b = tip_of(&stroke);
assert!(b.is_some_and(|b| dist(b, tip) < 1.0), "overshoot head first: {b:?}");
}
/// A shaft from `tail` to `tip`, then a closed triangle head drawn round the tip in the same
/// stroke (tip, one barb end, the other, back to the tip); `n` shaft points, `m` per side.
fn closed_arrow_n(tail: [f32; 2], tip: [f32; 2], barb: f32, spread_deg: f32, n: usize, m: usize) -> Vec<[f32; 2]> {
let [b1, b2] = barb_ends(tail, tip, barb, spread_deg);
let mut pts = line(tail, tip, n);
pts.extend(line(tip, b1, m));
pts.extend(line(b1, b2, m));
pts.extend(line(b2, tip, m));
pts.push(tip);
pts
}
/// The ends of a head's two barbs at `tip`, `barb` long, `spread_deg` either side of the shaft.
fn barb_ends(tail: [f32; 2], tip: [f32; 2], barb: f32, spread_deg: f32) -> [[f32; 2]; 2] {
let back = (tail[1] - tip[1]).atan2(tail[0] - tip[0]);
let end = |sign: f32| {
let a = back + sign * spread_deg.to_radians();
[tip[0] + barb * a.cos(), tip[1] + barb * a.sin()]
};
[end(1.0), end(-1.0)]
}
#[test]
fn arrows_with_a_closed_triangle_head_in_one_stroke_are_recognised_both_ways() {
for (tail, tip) in
[([20.0, 100.0], [180.0, 100.0]), ([180.0, 100.0], [20.0, 100.0]), ([100.0, 180.0], [100.0, 20.0]), ([150.0, 120.0], [30.0, 30.0])]
{
for stroke in [closed_arrow_n(tail, tip, 25.0, 30.0, 24, 5), closed_arrow_n(tail, tip, 20.0, 25.0, dist(tail, tip).round() as usize, 15)]
{
let Some(InkShape::Arrow { a, b, head_at_end, head_at_start }) = recognise(&stroke) else {
panic!("{tail:?}->{tip:?}: {:?}", recognise(&stroke))
};
assert!(head_at_end && !head_at_start);
assert!(dist(a, tail) < 2.0 && dist(b, tip) < 2.0, "{a:?} {b:?} for {tail:?} {tip:?}");
// The loop round the head first, then the shaft: the same arrow.
let mut rev = stroke;
rev.reverse();
let Some(InkShape::Arrow { a, b, .. }) = recognise(&rev) else { panic!("head first {tail:?}->{tip:?}: {:?}", recognise(&rev)) };
assert!(dist(a, tail) < 2.0 && dist(b, tip) < 2.0, "head first {a:?} {b:?}");
}
}
// The head size limits hold: 5 to 50 % of a 160 pt shaft, 15 to 75 degrees.
let (tail, tip) = ([20.0, 100.0], [180.0, 100.0]);
let is_arrow = |l: f32, deg: f32| matches!(recognise(&closed_arrow_n(tail, tip, l, deg, 24, 5)), Some(InkShape::Arrow { .. }));
assert!(is_arrow(12.0, 30.0) && is_arrow(70.0, 30.0) && is_arrow(25.0, 20.0) && is_arrow(25.0, 70.0));
assert!(!is_arrow(6.0, 30.0) && !is_arrow(90.0, 30.0) && !is_arrow(25.0, 8.0) && !is_arrow(25.0, 85.0));
}
#[test]
fn triangles_without_a_shaft_stay_triangles() {
// Closed, and closed with the pen running on past the start along the first side.
let tri = around(&[[0.0, 80.0], [45.0, 0.0], [90.0, 80.0]], 15);
assert_eq!(recognise(&tri).map(|s| s.name()), Some("triangle"));
let mut over = tri.clone();
over.extend(line([0.0, 80.0], [45.0, 0.0], 15).into_iter().skip(1).take(8));
assert!(!matches!(recognise(&over), Some(InkShape::Arrow { .. })), "{:?}", recognise(&over));
// A long thin triangle is not a shaft with a head.
let thin = around(&[[0.0, 0.0], [160.0, 10.0], [0.0, 20.0]], 20);
assert!(!matches!(recognise(&thin), Some(InkShape::Arrow { .. })), "{:?}", recognise(&thin));
// A shaft with a triangle as big as the shaft is not an arrow.
let (tail, tip) = ([20.0, 100.0], [180.0, 100.0]);
assert!(!matches!(recognise(&closed_arrow_n(tail, tip, 150.0, 30.0, 24, 5)), Some(InkShape::Arrow { .. })));
// Hostile points around a closed head never panic.
let mut junk = closed_arrow_n(tail, tip, 25.0, 30.0, 24, 5);
junk.insert(30, [f32::NAN, f32::INFINITY]);
assert!(matches!(recognise(&junk), Some(InkShape::Arrow { .. })));
}
/// A head drawn on its own: an open V (one barb end, the apex, the other end) or a closed
/// triangle (apex, barb end, barb end, apex), `m` points per side.
fn v_head(apex: [f32; 2], [b1, b2]: [[f32; 2]; 2], m: usize) -> Vec<[f32; 2]> {
let mut pts = line(b1, apex, m);
pts.extend(line(apex, b2, m));
pts.push(b2);
pts
}
fn triangle_head(apex: [f32; 2], [b1, b2]: [[f32; 2]; 2], m: usize) -> Vec<[f32; 2]> {
let mut pts = line(apex, b1, m);
pts.extend(line(b1, b2, m));
pts.extend(line(b2, apex, m));
pts.push(apex);
pts
}
#[test]
fn a_head_stroke_at_a_line_end_pairs_into_an_arrow() {
for (tail, tip) in [([20.0, 100.0], [180.0, 100.0]), ([180.0, 160.0], [40.0, 30.0]), ([100.0, 20.0], [100.0, 180.0])] {
let shaft = line(tail, tip, 40).into_iter().chain([tip]).collect::<Vec<_>>();
let ends = barb_ends(tail, tip, 25.0, 30.0);
for (what, head) in [("V", v_head(tip, ends, 8)), ("triangle", triangle_head(tip, ends, 8)), ("bare V", vec![ends[0], tip, ends[1]])] {
let Some(InkShape::Arrow { a, b, head_at_end, head_at_start }) = pair_arrow(&shaft, &head) else {
panic!("{what} {tail:?}->{tip:?}: {:?}", pair_arrow(&shaft, &head))
};
assert!(head_at_end && !head_at_start);
assert!(dist(a, tail) < 1.0 && dist(b, tip) < 1.0, "{what}: {a:?} {b:?}");
// Drawn the other way round, the head is still found at its end.
let mut rev = head.clone();
rev.reverse();
assert!(matches!(pair_arrow(&shaft, &rev), Some(InkShape::Arrow { b, .. }) if dist(b, tip) < 1.0), "{what} reversed");
let mut back = shaft.clone();
back.reverse();
assert!(matches!(pair_arrow(&back, &head), Some(InkShape::Arrow { b, .. }) if dist(b, tip) < 1.0), "{what}, shaft drawn tip first");
}
// A triangle given as its corners (an Ink to Shape triangle), apex anywhere in the ring.
let [b1, b2] = barb_ends(tail, tip, 25.0, 30.0);
assert!(matches!(pair_arrow(&shaft, &[b1, b2, tip, b1]), Some(InkShape::Arrow { b, .. }) if dist(b, tip) < 1.0), "corners");
// An apex a few points short of the end still pairs.
let d = [(tip[0] - tail[0]) / dist(tail, tip), (tip[1] - tail[1]) / dist(tail, tip)];
let near = [tip[0] - 6.0 * d[0] + 3.0 * d[1], tip[1] - 6.0 * d[1] - 3.0 * d[0]];
assert!(pair_arrow(&shaft, &v_head(near, barb_ends(tail, near, 25.0, 30.0), 8)).is_some(), "near the end");
}
}
#[test]
fn heads_away_from_the_end_the_wrong_way_or_the_wrong_size_dont_pair() {
let (tail, tip) = ([20.0, 100.0], [180.0, 100.0]);
let shaft = line(tail, tip, 40).into_iter().chain([tip]).collect::<Vec<_>>();
let head_at = |apex: [f32; 2], barb: f32, deg: f32| v_head(apex, barb_ends(tail, apex, barb, deg), 8);
// Far from either end: mid-shaft, and 40 pt past the tip.
assert_eq!(pair_arrow(&shaft, &head_at([100.0, 100.0], 25.0, 30.0)), None);
assert_eq!(pair_arrow(&shaft, &head_at([220.0, 100.0], 25.0, 30.0)), None);
assert_eq!(pair_arrow(&shaft, &head_at([180.0, 140.0], 25.0, 30.0)), None);
// Pointing back at the shaft (barbs forward), too big, too small, too narrow, too wide.
let forward = v_head(tip, barb_ends([340.0, 100.0], tip, 25.0, 30.0), 8);
assert_eq!(pair_arrow(&shaft, &forward), None);
for (barb, deg) in [(100.0, 30.0), (5.0, 30.0), (25.0, 5.0), (25.0, 85.0)] {
assert_eq!(pair_arrow(&shaft, &head_at(tip, barb, deg)), None, "barb {barb} at {deg} degrees");
}
// Lopsided: both barbs on one side.
assert_eq!(pair_arrow(&shaft, &[[155.0, 87.0], tip, [158.0, 80.0]]), None);
// The "line" must be a line, the head a head.
let ends = barb_ends(tail, tip, 25.0, 30.0);
let zig: Vec<[f32; 2]> = (0..9).map(|i| [20.0 + i as f32 * 20.0, if i % 2 == 0 { 100.0 } else { 120.0 }]).collect();
assert_eq!(pair_arrow(&zig, &v_head(tip, ends, 8)), None);
assert_eq!(pair_arrow(&shaft, &[[150.0, 80.0], [170.0, 120.0], [175.0, 80.0], [180.0, 120.0]]), None, "a scribble");
assert_eq!(pair_arrow(&shaft, &around(&[[150.0, 60.0], [190.0, 60.0], [190.0, 140.0], [150.0, 140.0]], 8)), None, "a box");
// Hostile input never panics.
let mut junk = v_head(tip, ends, 8);
junk.insert(2, [f32::NAN, 1.0]);
junk.push([f32::INFINITY, 0.0]);
assert!(pair_arrow(&shaft, &junk).is_some());
assert_eq!(pair_arrow(&[], &[]), None);
assert_eq!(pair_arrow(&shaft, &[[180.0, 100.0]]), None);
let huge: Vec<[f32; 2]> = (0..MAX_POINTS * 2).map(|i| [(i % 977) as f32, ((i * 31) % 541) as f32]).collect();
let _ = pair_arrow(&huge, &huge);
let _ = pair_arrow(&shaft, &huge);
}
}
+166 -18
View File
@@ -106,11 +106,9 @@ pub fn specs() -> Vec<CommandSpec> {
.key(super::mac_or("Ctrl+Shift+T", "Alt+Shift+T")),
CommandSpec::new("insert.pageField", "Insert Page Field", "Insert › Text › Quick Parts", |s, _| field(s, &json!({"instr": "PAGE"})))
.key(super::mac_or("Ctrl+Shift+P", "Alt+Shift+P")),
CommandSpec::new("insert.dropCap", "Drop Cap", "Insert › Text", |s, v| {
let lines = p::u64(v, "lines").unwrap_or(3).min(10) as u8;
super::para::fmt(s, &|p| p.drop_cap = Some(lines))
})
.params(r#"{"lines"?: n (0 = none)}"#),
CommandSpec::new("insert.dropCap", "Drop Cap", "Insert › Text", drop_cap)
.params(r#"{"position"?: "none"|"dropped"|"margin" (default dropped), "lines"?: 1..10 (default 3; 0 = none), "distance"?: pt from the text (0..1584, default 0), "font"?: family (omit for the text's font)}"#)
.when(drop_cap_off),
CommandSpec::new("insert.horizontalLine", "Horizontal Line", "Home › Paragraph › Borders", |s, _| {
super::para::fmt(s, &|p| {
p.borders = Some(wordcraft_doc::props::Borders { bottom: Some(wordcraft_doc::props::Border::single(1.5)), ..Default::default() })
@@ -159,6 +157,11 @@ fn table(s: &mut Session, v: &Value) -> CmdResult {
let id = s.doc.styles.find(st).map(|x| x.id.clone()).ok_or_else(|| CmdError::Params(format!("no table style `{st}`")))?;
t.props.style = Some(id);
}
// Like Word, a table inserted in a right-to-left paragraph is a right-to-left table (#2634):
// its first cell stands at the right, so Tab, which follows reading order, moves leftwards.
if s.doc.para_at(&s.sel.focus).is_some_and(|p| s.doc.styles.resolve_para(&p.props).bidi) {
t.props.bidi_visual = true;
}
let story = s.sel.focus.story;
let path = insert_block_at_caret(s, Block::Table(t))?;
let mut first = path.0.clone();
@@ -343,8 +346,11 @@ fn shape(s: &mut Session, v: &Value) -> CmdResult {
if super::canvas::add_member(s, obj.clone())?.is_some() {
return sel_result(s);
}
// Word draws a new shape beside what is selected and never deletes it: the shape just
// inserted (which is selected) stays, and so does selected text (#2019). It goes after the
// selection and becomes the selection. (A picture does replace the selection, as in Word.)
let props = s.typing_props();
let at = delete_selection(s)?;
let (_, at) = s.sel.ordered();
let end = s.doc.insert_object(&at, obj, &props)?;
s.sel = Selection { anchor: at, focus: end };
sel_result(s)
@@ -537,21 +543,30 @@ fn bookmark(s: &mut Session, v: &Value) -> CmdResult {
sel_result(s)
}
/// The header (or footer) of the caret's section in the slot the caret's page uses (first page,
/// even page or the default), made when there is none: with Different First Page on, a header
/// inserted on page 1 is the one page 1 shows (#2826).
fn hf_part(s: &mut Session, header: bool) -> u32 {
let block = s.sel.focus.path.0.first().copied().unwrap_or(0) as usize;
let sect = s.doc.section_mut(block).clone();
let existing = if header { sect.headers.default } else { sect.footers.default };
let layout = s.layout();
let page = if s.sel.focus.story == StoryRef::Body { layout.page_of(&s.sel.focus) } else { Some(s.page_hint) };
let slot = page.and_then(|pg| layout.pages.get(pg)).map(|pg| pg.hf_slot).unwrap_or_default();
fn slot_of(slot: wordcraft_layout::HfSlot, set: &mut wordcraft_doc::section::HeaderSet) -> &mut Option<u32> {
match slot {
wordcraft_layout::HfSlot::First => &mut set.first,
wordcraft_layout::HfSlot::Even => &mut set.even,
wordcraft_layout::HfSlot::Default => &mut set.default,
}
}
let mut sect = s.doc.section_mut(block).clone();
let existing = *slot_of(slot, if header { &mut sect.headers } else { &mut sect.footers });
if let Some(id) = existing.filter(|id| s.doc.parts.contains_key(id)) {
return id;
}
let style = if header { "Header" } else { "Footer" };
let id = s.doc.add_part(if header { PartKind::Header } else { PartKind::Footer }, vec![para_block(Paragraph::new().styled(style))]);
let sect = s.doc.section_mut(block);
if header {
sect.headers.default = Some(id);
} else {
sect.footers.default = Some(id);
}
*slot_of(slot, if header { &mut sect.headers } else { &mut sect.footers }) = Some(id);
id
}
@@ -590,7 +605,12 @@ fn header_footer(s: &mut Session, v: &Value, header: bool) -> CmdResult {
}
fn edit_hf(s: &mut Session, v: &Value, header: bool) -> CmdResult {
let page = p::u64(v, "page").map(|page| usize::try_from(page).unwrap_or(usize::MAX));
let mut page = p::u64(v, "page").map(|page| usize::try_from(page).unwrap_or(usize::MAX));
if page.is_none() && s.sel.focus.story == StoryRef::Body {
// The caret's page: with Different First Page on, page 1 edits its own header (#2826).
let focus = s.sel.focus.clone();
page = s.layout().page_of(&focus);
}
let shown = match page {
Some(page) => shown_hf(s, page, header)?,
None => {
@@ -867,10 +887,27 @@ fn symbol(s: &mut Session, v: &Value) -> CmdResult {
return Err(CmdError::Params("font: a family name".into()));
}
let base = s.typing_props();
let mut props = base.clone();
props.font = Some(font.to_string());
s.pending = Some(props);
let r = type_text(s, c);
// A symbol font's character remembers it as its own font (`w:sym`), so changing the text's
// font later keeps the symbol; other fonts are ordinary run formatting. `w:sym` holds a 16-bit
// code, so a character beyond the Basic Multilingual Plane stays ordinary text in that font.
// Each character decides for itself, so a mixed string keeps its symbols and its text apart.
let font_is_symbol = wordcraft_doc::symbol_fonts::is_symbol_font(font);
let mut r = Ok(());
let mut buf = [0u8; 4];
for ch in c.chars() {
let mut props = base.clone();
let private_use = ('\u{F020}'..='\u{F0FF}').contains(&ch);
if u32::from(ch) <= 0xFFFF && (font_is_symbol || private_use) {
props.sym_font = Some(font.to_string());
} else {
props.font = Some(font.to_string());
}
s.pending = Some(props);
r = type_text(s, ch.encode_utf8(&mut buf));
if r.is_err() {
break;
}
}
// Typing on after the symbol goes on in the text's own font.
s.pending = Some(base);
r?;
@@ -974,6 +1011,29 @@ mod tests {
assert!(s.run("insert.editHeader", &json!({"page": u64::MAX})).is_err());
}
/// #2826: with Different First Page on, Insert › Header and Edit Header with the caret on
/// page 1 give page 1 its own header; page 2 keeps the default one.
#[test]
fn different_first_page_header_is_reachable_from_page_one() {
let mut s = two_pages();
s.run("insert.header", &json!({"text": "Head"})).unwrap();
s.run("insert.closeHeader", &json!({})).unwrap();
let head = s.doc.last_section.headers.default.unwrap();
s.run("layout.differentFirstPage", &json!({"value": true})).unwrap();
assert_eq!(s.layout().pages[0].header_story, None, "page 1 shows the empty first-page header");
s.sel = Selection::caret(s.doc.start_of(StoryRef::Body));
let first = s.run("insert.header", &json!({"preset": "title"})).unwrap()["story"].as_u64().unwrap() as u32;
assert_ne!(first, head);
assert_eq!(s.doc.last_section.headers.first, Some(first));
let pages = s.layout().pages.clone();
assert_eq!((pages[0].header_story, pages[1].header_story), (Some(first), Some(head)));
// Edit Footer from page 1 (no page given): page 1's own footer, not the default.
s.run("insert.closeHeader", &json!({})).unwrap();
s.sel = Selection::caret(s.doc.start_of(StoryRef::Body));
let foot = s.run("insert.editFooter", &json!({})).unwrap()["story"].as_u64().unwrap() as u32;
assert_eq!((s.doc.last_section.footers.first, s.doc.last_section.footers.default), (Some(foot), None));
}
fn two_pages() -> Session {
let mut s = Session::new(wordcraft_doc::Document::new());
s.run("text.insert", &json!({"text": "One"})).unwrap();
@@ -1114,6 +1174,27 @@ mod tests {
}
/// Insert › Symbol with a font: only the symbol takes it, the caret's font carries on after.
#[test]
fn a_mixed_symbol_string_keeps_symbols_and_text_apart() {
let mut s = Session::new(wordcraft_doc::Document::from_text("a"));
s.run("caret.docEnd", &json!({})).unwrap();
s.run("insert.symbol", &json!({"char": "\u{F041}A", "font": "My Custom Font"})).unwrap();
let p = s.doc.para(StoryRef::Body, &Path::top(0)).unwrap();
assert_eq!(p.props_of_char(1).sym_font.as_deref(), Some("My Custom Font"), "the private-use glyph");
assert_eq!(p.props_of_char(4).sym_font, None, "the plain letter");
assert_eq!(p.props_of_char(4).font.as_deref(), Some("My Custom Font"));
}
#[test]
fn an_astral_symbol_stays_text_in_its_font() {
let mut s = Session::new(wordcraft_doc::Document::from_text("a"));
s.run("caret.docEnd", &json!({})).unwrap();
s.run("insert.symbol", &json!({"char": "\u{1F600}", "font": "Wingdings"})).unwrap();
let p = s.doc.para(StoryRef::Body, &Path::top(0)).unwrap();
assert_eq!(p.props_of_char(1).sym_font, None, "w:sym can't hold it");
assert_eq!(p.props_of_char(1).font.as_deref(), Some("Wingdings"));
}
#[test]
fn symbol_takes_its_own_font() {
let mut s = Session::new(wordcraft_doc::Document::from_text("ab"));
@@ -1141,3 +1222,70 @@ mod tests {
assert_eq!(date_picture("DATE"), None);
}
}
/// Why Insert › Drop Cap is off at the caret, as in Word: in a table cell, in an empty paragraph
/// and in one that starts with a space.
fn drop_cap_off(s: &Session) -> Option<&'static str> {
if s.sel.focus.path.cell().is_some() {
return Some("drop caps can't go in a table");
}
let first = s.doc.para_at(&s.sel.focus).and_then(|p| p.text.chars().next());
match first {
None => Some("the paragraph is empty"),
Some(c) if c.is_whitespace() => Some("the paragraph starts with a space"),
Some(_) => None,
}
}
/// Insert › Drop Cap and Drop Cap Options: none, or the letter dropped into the text or hung in
/// the margin, `lines` tall, `distance` points from the text, in `font` (unset: the text's).
/// None takes all of it away, the letter's font with it.
fn drop_cap(s: &mut Session, v: &Value) -> CmdResult {
let lines = v.get("lines").and_then(Value::as_f64).filter(|l| l.is_finite());
let none = p::str(v, "position") == Some("none") || lines == Some(0.0);
let margin = p::str(v, "position") == Some("margin");
let lines = lines.map_or(3.0, |l| l.round().clamp(1.0, 10.0)) as u8;
let distance = p::f32(v, "distance").map_or(0.0, |d| d.clamp(0.0, wordcraft_doc::resolve::MAX_DROP_CAP_DISTANCE));
// A family name, not a paragraph of text.
let font = p::str(v, "font").map(str::trim).filter(|f| !f.is_empty()).map(|f| f.chars().take(64).collect::<String>());
super::para::fmt(s, &|p| {
p.drop_cap = (!none).then_some(lines);
p.drop_cap_margin = (!none && margin).then_some(true);
p.drop_cap_distance = (!none && distance > 0.0).then_some(distance);
p.drop_cap_font = font.clone().filter(|_| !none);
})
}
/// Insert › Drop Cap at the caret: the menu's highlight and the Drop Cap dialog's values.
#[derive(Clone, Debug, PartialEq)]
pub struct DropCapState {
/// `none`, `dropped` or `margin`.
pub position: &'static str,
/// Lines to drop (3 when there is no drop cap).
pub lines: u8,
/// Distance from the text, points.
pub distance: f32,
/// The letter's own font, if it has one.
pub font: Option<String>,
/// The font of the paragraph's first character.
pub text_font: String,
}
pub fn drop_cap_state(s: &Session) -> DropCapState {
let para = s.doc.para_at(&s.sel.focus);
let props = para.map(|p| p.props.clone()).unwrap_or_default();
let rp = s.doc.styles.resolve_para(&props);
let first = para.map(|p| p.props_at(0).clone()).unwrap_or_default();
let text_font = s.doc.styles.resolve_char_in(props.style.as_deref(), None, &first).font;
DropCapState {
position: match (rp.drop_cap, rp.drop_cap_margin) {
(0, _) => "none",
(_, true) => "margin",
_ => "dropped",
},
lines: if rp.drop_cap == 0 { 3 } else { rp.drop_cap },
distance: if rp.drop_cap == 0 { 0.0 } else { rp.drop_cap_distance },
font: rp.drop_cap_font.filter(|_| rp.drop_cap > 0),
text_font,
}
}
+2 -2
View File
@@ -46,11 +46,11 @@ fn para_props(s: &Session) -> ParaProps {
fn char_props(s: &Session) -> CharProps {
let (a, b) = s.sel.ordered();
if a == b {
return s.typing_props();
return s.caret_props();
}
match s.doc.para_at(&a) {
Some(p) if a.off < p.len() => p.props_of_char(a.off).clone(),
_ => s.typing_props(),
_ => s.caret_props(),
}
}
+227
View File
@@ -23,6 +23,9 @@ pub fn specs() -> Vec<CommandSpec> {
CommandSpec::new("list.get", "List Definition", "Home › Paragraph › Multilevel List", get)
.params(r#"{"num"?: list id (default: the caret's list)} → {"num", "levels": [as list.define]}"#)
.pure(),
CommandSpec::new("list.multilevelPreset", "Multilevel List Library", "Home › Paragraph › Multilevel List", preset).params(
r#"{"preset": "headings" (1 / 1.1 / 1.1.1) | "article" (Article I. / Section 1.01 / (a)) | "outlineHeadings" (I. / A. / 1.) | "chapter" (Chapter 1, lower levels unnumbered)} — a new list whose nine levels are linked to Heading 1–9: every paragraph in those styles is numbered, selected paragraphs already in a Heading style keep it, and any other selected paragraph becomes Heading 1 (the first level)"#,
),
]
}
@@ -226,6 +229,96 @@ fn define(s: &mut Session, v: &Value) -> CmdResult {
Ok(r)
}
/// The heading-linked lists of the Multilevel List gallery's library, by preset id: level `i`
/// is linked to Heading `i + 1`.
fn preset_levels(preset: &str) -> Option<Vec<Level>> {
use NumFormat as F;
let levels = (0..LEVELS)
.map(|i| {
let n = i + 1;
// Numbers start at the margin; each level's text a little further in.
let at = |indent: f32| Level { indent, hanging: indent, style: Some(format!("Heading{n}")), ..Level::default() };
let (format, text, indent) = match preset {
"headings" => (F::Decimal, (1..=n).map(|k| format!("%{k}")).collect::<Vec<_>>().join("."), 21.6 + 7.2 * i as f32),
"article" => match i {
0 => (F::UpperRoman, "Article %1.".to_string(), 0.0),
1 => (F::DecimalZero, "Section %1.%2".to_string(), 0.0),
_ => {
let f = [F::LowerLetter, F::LowerRoman, F::Decimal, F::LowerLetter, F::LowerRoman, F::LowerLetter, F::LowerRoman];
let t = if i < 4 {
format!("(%{n})")
} else if i < 7 {
format!("%{n})")
} else {
format!("%{n}.")
};
(f.get(i - 2).copied().unwrap_or(F::Decimal), t, 36.0 + 7.2 * (i - 2) as f32)
}
},
"outlineHeadings" => {
let f = [
F::UpperRoman,
F::UpperLetter,
F::Decimal,
F::LowerLetter,
F::Decimal,
F::LowerLetter,
F::LowerRoman,
F::LowerLetter,
F::LowerRoman,
];
let t = match i {
0..=2 => format!("%{n}."),
3 => format!("%{n})"),
_ => format!("(%{n})"),
};
(f.get(i).copied().unwrap_or(F::Decimal), t, 36.0 * n as f32)
}
"chapter" if i == 0 => (F::Decimal, "Chapter %1".to_string(), 0.0),
"chapter" => (F::None, String::new(), 0.0),
_ => return None,
};
// "Section 1.01" shows the article number in digits (legal numbering).
Some(Level { format, text, legal: preset == "article" && i == 1, ..at(indent) })
})
.collect::<Option<Vec<_>>>()?;
Some(levels)
}
/// A Multilevel List gallery preset: a new list linked to Heading 1–9. The heading styles carry
/// the number, so every heading (and every heading added later) is numbered. Paragraphs in a
/// Heading style keep their level; any other paragraph under the caret or selection (Normal,
/// Quote, Title…) becomes Heading 1, as in Word.
fn preset(s: &mut Session, v: &Value) -> CmdResult {
let id = p::req_str(v, "preset")?;
let levels = preset_levels(id).ok_or_else(|| CmdError::Params("`preset` must be headings, article, outlineHeadings or chapter".into()))?;
let num = s.doc.numbering.try_add_levels(None, levels)?;
let mut linked = Vec::new();
for n in 1..=LEVELS as u8 {
let id = super::para::ensure_heading_style(s, n);
if let Some(st) = s.doc.styles.get_mut(&id) {
st.para.numbering = Some(NumRef { num, level: n - 1 });
linked.push(id);
}
}
touch_styles(s, &linked)?;
super::para::fmt(s, &|pp| {
// A heading keeps its style and level; any other paragraph (Normal, Quote, Title…)
// becomes the first level's heading, Heading 1. The number comes from the style.
if !pp.style.as_ref().is_some_and(|st| linked.contains(st)) {
pp.style = linked.first().cloned();
}
pp.numbering = None;
pp.indent_left = None;
pp.indent_first = None;
pp.indent_left_chars = None;
pp.indent_first_chars = None;
})?;
let mut r = sel_result(s)?;
r["num"] = json!(num);
Ok(r)
}
/// Lay out again every paragraph in one of `nums`.
fn touch_lists(s: &mut Session, nums: &[u32]) -> Result<(), CmdError> {
touch_where(s, &|p| p.props.numbering.is_some_and(|n| nums.contains(&n.num)))
@@ -309,4 +402,138 @@ mod tests {
// Programmatic calls without levels are an error, not a dialog.
assert!(s.run("list.define", &json!({})).is_err());
}
/// Each body paragraph's number as laid out (`None`: no number), and its style.
fn numbered(s: &mut Session) -> Vec<(Option<String>, String)> {
let l = s.layout();
let labels: Vec<Option<String>> = l
.pages
.iter()
.flat_map(|p| &p.items)
.filter_map(|i| if let Placed::Lines { para, l0: 0, .. } = i { Some(para.label.as_ref().map(|lb| lb.text.clone())) } else { None })
.collect();
let styles = s.doc.body.iter().filter_map(|b| b.as_para()).map(|p| p.props.style.clone().unwrap_or_else(|| "Normal".into()));
labels.into_iter().zip(styles).collect()
}
fn n(label: Option<&str>, style: &str) -> (Option<String>, String) {
(label.map(str::to_string), style.to_string())
}
/// Paragraphs "One" (Heading 1), "Two" (Heading 2), "Three" (`third`), the caret in "One".
fn headings(third: &str) -> Session {
let mut s = Session::new(Document::new());
for (i, (t, st)) in [("One", "Heading1"), ("Two", "Heading2"), ("Three", third)].iter().enumerate() {
if i > 0 {
s.run("text.newParagraph", &json!({})).unwrap();
}
s.run("text.insert", &json!({"text": t})).unwrap();
s.run("para.style", &json!({"style": st})).unwrap();
}
s.sel = crate::Selection::caret(wordcraft_doc::Pos::body(0, 0));
s
}
/// #1793: the Multilevel List gallery's heading presets number headings through their styles
/// (Word: One=1, Two=1.1, Three=2), keep the styles, and survive a save to .docx and back.
#[test]
fn a_heading_preset_numbers_headings_and_keeps_their_styles() {
let mut s = headings("Heading1");
s.run("list.multilevelPreset", &json!({"preset": "headings"})).unwrap();
let want = [n(Some("1"), "Heading1"), n(Some("1.1"), "Heading2"), n(Some("2"), "Heading1")];
assert_eq!(numbered(&mut s), want);
// The number comes from the style, so a heading added later is numbered too.
let h2 = s.doc.styles.get("Heading2").and_then(|st| st.para.numbering).unwrap();
assert_eq!(h2.level, 1);
let back = wordcraft_docx::read(&wordcraft_docx::write(&s.doc).unwrap()).unwrap();
assert_eq!(back.styles.get("Heading2").and_then(|st| st.para.numbering).map(|x| x.level), Some(1), "numPr on the style");
let mut b = Session::new(back);
assert_eq!(numbered(&mut b), want);
}
/// #1793: the other heading presets of Word's List Library, level by level.
#[test]
fn every_heading_preset_numbers_three_heading_levels() {
for (preset, want) in [
("headings", [Some("1"), Some("1.1"), Some("1.1.1")]),
("article", [Some("Article I."), Some("Section 1.01"), Some("(a)")]),
("outlineHeadings", [Some("I."), Some("A."), Some("1.")]),
("chapter", [Some("Chapter 1"), None, None]),
] {
let mut s = headings("Heading3");
s.run("list.multilevelPreset", &json!({"preset": preset})).unwrap();
let got: Vec<Option<String>> = numbered(&mut s).into_iter().map(|x| x.0).collect();
assert_eq!(got, want.map(|x| x.map(str::to_string)), "{preset}");
}
let mut s = headings("Heading3");
assert!(s.run("list.multilevelPreset", &json!({"preset": "nope"})).is_err());
assert!(s.run("list.multilevelPreset", &json!({})).is_err());
}
/// Paragraphs (text, style id), the selection from paragraph `from` to the end of `to`.
fn doc(paras: &[(&str, &str)], from: usize, to: usize) -> Session {
let mut s = Session::new(Document::new());
for (i, (t, st)) in paras.iter().enumerate() {
if i > 0 {
s.run("text.newParagraph", &json!({})).unwrap();
}
s.run("text.insert", &json!({"text": t})).unwrap();
s.run("para.style", &json!({"style": st})).unwrap();
}
let end = paras.get(to).map_or(0, |p| p.0.len());
s.sel = crate::Selection { anchor: wordcraft_doc::Pos::body(from, 0), focus: wordcraft_doc::Pos::body(to, end) };
s
}
/// #1793 (Word for Mac 16.113.4, 2026-10-16): with a caret or a selection, a heading preset
/// leaves paragraphs already in a Heading style at their own style and level; every other
/// selected paragraph (Normal, Quote, Title…) becomes Heading 1.
#[test]
fn a_heading_preset_keeps_headings_and_makes_other_paragraphs_heading_1() {
let (h1, h2, h3) = ("Heading1", "Heading2", "Heading3");
let cases: Vec<(Vec<(&str, &str)>, usize, usize, Vec<(Option<&str>, &str)>)> = vec![
// All selected.
(vec![("One", h1), ("Two", h2), ("Three", h3)], 0, 2, vec![(Some("1"), h1), (Some("1.1"), h2), (Some("1.1.1"), h3)]),
// No Heading 1 above.
(vec![("Two", h2), ("Three", h3)], 0, 1, vec![(Some("1.1"), h2), (Some("1.1.1"), h3)]),
(vec![("Body", "Normal"), ("Three", h3)], 0, 1, vec![(Some("1"), h1), (Some("1.1.1"), h3)]),
(vec![("One", h1), ("Body", "Normal"), ("Two", h2)], 0, 2, vec![(Some("1"), h1), (Some("2"), h1), (Some("2.1"), h2)]),
(vec![("One", h1), ("Two", h2), ("Body", "Normal")], 0, 2, vec![(Some("1"), h1), (Some("1.1"), h2), (Some("2"), h1)]),
// The caret in Two.
(vec![("One", h1), ("Two", h2), ("Three", h3)], 1, 1, vec![(Some("1"), h1), (Some("1.1"), h2), (Some("1.1.1"), h3)]),
(vec![("Quote", "Quote"), ("Two", h2)], 0, 1, vec![(Some("1"), h1), (Some("1.1"), h2)]),
(vec![("Title", "Title")], 0, 0, vec![(Some("1"), h1)]),
// Body text after the selection stays unnumbered.
(vec![("One", h1), ("After", "Normal")], 0, 0, vec![(Some("1"), h1), (None, "Normal")]),
];
for (paras, from, to, want) in cases {
let mut s = doc(&paras, from, to);
s.run("list.multilevelPreset", &json!({"preset": "headings"})).unwrap();
let want: Vec<_> = want.into_iter().map(|(l, st)| n(l, st)).collect();
assert_eq!(numbered(&mut s), want, "{paras:?} {from}..={to}");
}
}
/// #1793 (Word for Mac): the caret alone in a Normal paragraph makes it Heading 1 numbered 1.
#[test]
fn a_heading_preset_on_a_lone_normal_paragraph_makes_it_heading_1() {
let mut s = doc(&[("Body", "Normal")], 0, 0);
s.sel = crate::Selection::caret(wordcraft_doc::Pos::body(0, 2));
s.run("list.multilevelPreset", &json!({"preset": "headings"})).unwrap();
assert_eq!(numbered(&mut s), [n(Some("1"), "Heading1")]);
}
/// #1793: one Undo takes the preset back: paragraph styles and the heading styles' numbers.
#[test]
fn undo_takes_a_heading_preset_back_in_one_step() {
let mut s = doc(&[("Body", "Normal"), ("Two", "Heading2")], 0, 1);
let before = numbered(&mut s);
s.run("list.multilevelPreset", &json!({"preset": "headings"})).unwrap();
assert_ne!(numbered(&mut s), before);
s.run("edit.undo", &json!({})).unwrap();
assert_eq!(numbered(&mut s), before);
for n in 1..=9 {
assert_eq!(s.doc.styles.get(&format!("Heading{n}")).and_then(|st| st.para.numbering), None, "Heading{n}");
}
}
}
+47 -8
View File
@@ -226,7 +226,16 @@ fn apply_filter(s: &mut Session) {
if m.rows.is_empty() {
m.preview = false;
}
refresh(s);
refresh_shown(s);
}
/// [`refresh`] for the non-mutating recipient-list commands: in a protected document the merge
/// fields' shown results stay as they are, since these commands skip the protection checks and
/// undo (Preview Results and the record buttons, which mutate, show the new list's values).
fn refresh_shown(s: &mut Session) {
if !super::protect::enforced(&s.doc) {
refresh(s);
}
}
/// The canonical field name for `f` (any case), or a params error.
@@ -338,7 +347,8 @@ pub fn specs() -> Vec<CommandSpec> {
CommandSpec::new("mailings.editRecipients", "Edit Recipient List", "Mailings › Start Mail Merge", |s, v| {
if v.get("rows").and_then(Value::as_array).is_some() {
// Edited records keep the list's filter and sort (rules on removed fields drop).
let (filter, sort) = (std::mem::take(&mut s.merge.filter), std::mem::take(&mut s.merge.sort));
// Copies, so a list that fails to load leaves the filter and sort as they were.
let (filter, sort) = (s.merge.filter.clone(), s.merge.sort.clone());
let r = recipients(s, &json!({"rows": v.get("rows"), "fields": v.get("fields")}))?;
s.merge.filter = filter;
s.merge.sort = sort;
@@ -604,11 +614,21 @@ fn recipients(s: &mut Session, v: &Value) -> CmdResult {
let named = ext.as_deref().is_some_and(wordcraft_formats::sheet::is_extension);
let text = wordcraft_formats::txt::is_text(&bytes);
let spreadsheet = wordcraft_formats::sheet::detect(&bytes).is_some() || (named && !text);
if named && !spreadsheet && decode(&bytes, v)?.trim_start().starts_with('<') {
return Err(CmdError::Failed(
"this spreadsheet file is a web page (HTML), not a workbook; open it in a spreadsheet program, save it as .xlsx or .csv and select it again"
.into(),
));
if named && !spreadsheet {
let text = decode(&bytes, v)?;
let text = text.trim_start().trim_start_matches('\u{feff}');
if text.starts_with("<?xml") {
return Err(CmdError::Failed(
"this spreadsheet file is an XML spreadsheet, not a workbook; open it in a spreadsheet program, save it as .xlsx or .csv and select it again"
.into(),
));
}
if text.starts_with('<') {
return Err(CmdError::Failed(
"this spreadsheet file is a web page (HTML), not a workbook; open it in a spreadsheet program, save it as .xlsx or .csv and select it again"
.into(),
));
}
}
if spreadsheet {
use wordcraft_formats::sheet::{Pick, read_rows, sheet_names};
@@ -647,7 +667,7 @@ fn recipients(s: &mut Session, v: &Value) -> CmdResult {
s.merge.filter.clear();
s.merge.sort.clear();
s.merge.record = 0;
refresh(s);
refresh_shown(s);
let mut r = json!({"fields": s.merge.headers, "records": s.merge.rows.len()});
if !sheet.is_null() {
r["sheet"] = sheet;
@@ -1703,6 +1723,10 @@ mod tests {
assert_eq!(r["fields"], json!(["Name", "City"]));
let e = s.run("mailings.recipients", &json!({"path": "export.xls", "data": b64(b"<html><table><tr><td>x</td></tr></table></html>")}));
assert!(matches!(e, Err(CmdError::Failed(m)) if m.contains("web page")));
// #1287: an Excel 2003 XML spreadsheet named .xls says what it is.
let xml = b"<?xml version=\"1.0\"?>\n<Workbook xmlns=\"urn:schemas-microsoft-com:office:spreadsheet\"></Workbook>";
let e = s.run("mailings.recipients", &json!({"path": "export.xls", "data": b64(xml)}));
assert!(matches!(e, Err(CmdError::Failed(m)) if m.contains("XML spreadsheet")));
assert!(
s.run("mailings.recipients", &json!({"path": "junk.xls", "data": b64(&[0xD0, 0xCF, 0x11, 0xE0, 0xA1, 0xB1, 0x1A, 0xE1, 0, 1])})).is_err()
);
@@ -1811,6 +1835,21 @@ mod tests {
assert_eq!(names(&s), ["X", "Y"]);
}
/// #1287: an Edit Recipient List that fails to load keeps the list's filter and sort.
#[test]
fn failed_edit_recipients_keeps_filter_and_sort() {
let mut s = people();
s.run("mailings.filterRecipients", &json!({"rules": [{"field": "Age", "op": "greater", "value": 20}], "sort": [{"field": "City"}]})).unwrap();
let before = names(&s);
// Array rows without `fields` fail to load.
let e = s.run("mailings.editRecipients", &json!({"rows": [["X", "Y"]]}));
assert!(matches!(e, Err(CmdError::Params(_))));
assert_eq!(s.merge.filter.len(), 1);
assert_eq!(s.merge.sort.len(), 1);
assert_eq!(names(&s), before);
assert_eq!(s.run("mailings.editRecipients", &json!({})).unwrap()["rules"].as_array().map(Vec::len), Some(1));
}
/// Review fix for #885: sorting a column that mixes numbers and text is a total order
/// (numbers first, numerically; then text; ties keep list order) and never panics.
#[test]
+15 -2
View File
@@ -287,9 +287,17 @@ fn insert_text(s: &mut Session, text: &str, enter: bool) -> Result<(), CmdError>
if let Some(p) = controls::prepare_edit(s)? {
props = p;
}
// Pasted text is text: pending properties left by AutoCorrect don't give it a symbol's font (#2036).
if !enter {
props = props.without_symbol();
}
delete_selection(s)?;
// Typed text is new: never deleted, and not a formatting change of the text around it.
props.fmt_change = None;
// Its direction comes from its own letters, as on screen, not from the run it's typed into:
// Latin typed into reopened Hebrew must not be saved as right-to-left (#1543). The DOCX
// writer marks the right-to-left pieces itself.
props.rtl = None;
if s.doc.settings.track_changes {
props.ins = Some(new_revision(s, RevisionKind::Insert));
props.del = None;
@@ -361,9 +369,14 @@ pub fn split_para(s: &mut Session, at: &Pos) -> Result<Pos, CmdError> {
/// [`split_para`]; without `enter` an empty list item is split like any other paragraph.
fn split_para_as(s: &mut Session, at: &Pos, enter: bool) -> Result<Pos, CmdError> {
// `num: 0` means "explicitly not in a list", so it isn't a list item.
// `num: 0` means "explicitly not in a list", so it isn't a list item. The list may come from
// the paragraph's style (List Number 2 in a Word file) rather than the paragraph itself.
let (at_end, style, empty_list) = match s.doc.para_at(at) {
Some(p) => (at.off >= p.len(), p.props.style.clone(), p.props.numbering.filter(|n| enter && n.num != 0 && p.is_empty())),
Some(p) => (
at.off >= p.len(),
p.props.style.clone(),
p.props.numbering.or_else(|| s.doc.styles.resolve_para(&p.props).numbering).filter(|n| enter && n.num != 0 && p.is_empty()),
),
None => return Err(CmdError::Failed("no paragraph at caret".into())),
};
if let Some(n) = empty_list {
+21 -2
View File
@@ -53,7 +53,7 @@ fn in_use(s: &Session, rev: u32) -> bool {
/// done) when it isn't such a cut.
pub fn cut_tracked(s: &mut Session) -> Result<bool, CmdError> {
let (a, b) = s.sel.ordered();
if !s.doc.settings.track_changes || a == b || a.story != b.story || !move_sized(&s.clipboard_text) {
if !s.doc.settings.track_changes || !s.doc.settings.track_moves || a == b || a.story != b.story || !move_sized(&s.clipboard_text) {
return Ok(false);
}
s.doc.revisions.push(Revision::new(RevisionKind::Delete, s.author.clone(), now_iso()));
@@ -71,7 +71,7 @@ pub fn cut_tracked(s: &mut Session) -> Result<bool, CmdError> {
/// returns the revision for the pasted text (the destination). `None` = an ordinary insertion.
pub fn take_paste(s: &mut Session, frag: &Fragment) -> Option<u32> {
let m = s.pending_move.take()?;
if m.author != s.author || frag.plain_text() != m.text || !in_use(s, m.rev) {
if !s.doc.settings.track_moves || m.author != s.author || frag.plain_text() != m.text || !in_use(s, m.rev) {
return None;
}
let src = s.doc.revisions.get(m.rev as usize).filter(|r| r.kind == RevisionKind::Delete && r.move_name.is_none())?;
@@ -240,6 +240,25 @@ mod tests {
assert_eq!(k, ["delete", "insert"]);
}
/// Gap #5: with Track moves off (`w:doNotTrackMoves`), a tracked cut and paste of a
/// paragraph is an ordinary deletion and insertion, with no move names.
#[test]
fn with_track_moves_off_a_cut_and_paste_is_a_deletion_and_an_insertion() {
let mut s = Session::new(Document::new());
s.run("document.setText", &json!({"text": "First one.\nSecond one.\nThird one."})).unwrap();
s.run("review.trackChanges", &json!({"value": true})).unwrap();
s.run("review.trackingOptions", &json!({"trackMoves": false})).unwrap();
s.run("select.range", &json!({"anchor": {"block": 1, "off": 0}, "focus": {"block": 2, "off": 0}})).unwrap();
s.run("edit.cut", &json!({})).unwrap();
s.run("caret.set", &json!({"pos": {"block": 0, "off": 0}})).unwrap();
s.run("edit.paste", &json!({})).unwrap();
assert_eq!(texts(&s), ["Second one.", "First one.", "Second one.", "Third one."]);
assert!(s.doc.revisions.iter().all(|r| r.move_name.is_none()), "{:?}", s.doc.revisions);
let k = kinds(&mut s);
assert!(k.iter().all(|(k, m)| (k == "insert" || k == "delete") && m.is_null()), "{k:?}");
assert!(k.iter().any(|(k, _)| k == "insert") && k.iter().any(|(k, _)| k == "delete"), "{k:?}");
}
#[test]
fn accepting_one_end_of_a_move_accepts_both() {
let mut s = moved();
+472 -43
View File
@@ -4,6 +4,7 @@
//! caret). Picture adjustments re-encode the bitmap (originals are kept for Reset Picture).
use serde_json::{Value, json};
use wordcraft_doc::graphic::GraphicKind;
use wordcraft_doc::para::{Float, InlineObject, Wrap, WrapText};
use wordcraft_doc::props::{Rgb, TextVert, VAlign};
use wordcraft_doc::{Pos, StoryRef, TextBody};
@@ -56,9 +57,17 @@ fn has_movable(s: &Session) -> Option<&'static str> {
_ => has_object(s),
}
}
/// Word's own ink in a group is kept in the form a group member takes; it can't stand on its own
/// yet, so such a group stays whole.
const INK_GROUP: &str = "a group holding Word ink can't be ungrouped yet";
/// Does group `o` hold a member that can't stand on its own (Word's ink)?
fn holds_ink(o: &InlineObject) -> bool {
matches!(o, InlineObject::Group { children, .. } if children.iter().any(|c| matches!(&c.obj, InlineObject::Graphic { graphic, .. } if graphic.kind == GraphicKind::Ink)))
}
fn has_group(s: &Session) -> Option<&'static str> {
match selected(s) {
Some((_, o)) if o.is_canvas() => Some("a drawing canvas can't be ungrouped"),
Some((_, o)) if holds_ink(&o) => Some(INK_GROUP),
Some((_, InlineObject::Group { .. })) => None,
_ => Some("select a group first"),
}
@@ -82,7 +91,11 @@ fn has_shape(s: &Session) -> Option<&'static str> {
pub fn specs() -> Vec<CommandSpec> {
vec![
CommandSpec::new("picture.size", "Size", "Picture Format › Size", size).params(r#"{"width"?: pt, "height"?: pt, "lockAspect"?: bool, "scale"?: percent}"#).when(has_sizable),
CommandSpec::new("picture.size", "Size", "Picture Format › Size", size)
.params(
r#"{"width"?: pt, "height"?: pt, "lockAspect"?: bool, "scale"?: percent, "original"?: bool} (scale: of the current size, or with original of the picture's original size, its pixels at 96 ppi less what is cropped; original alone restores that size; one side alone keeps the proportions unless lockAspect is false; 1–4000 pt: when the proportions are kept, a side past 4000 pt shrinks the other with it)"#,
)
.when(has_sizable),
CommandSpec::new("picture.crop", "Crop", "Picture Format › Size", crop)
.params(
r#"{"left"?, "top"?, "right"?, "bottom"? (fractions of the source 0–0.45; a side left out is uncropped), "width"?: pt, "height"?: pt (the new display size, in the same undo step; a floating picture also moves by what a left/top crop trims, so the visible part stays put)}"#,
@@ -151,32 +164,11 @@ pub fn specs() -> Vec<CommandSpec> {
.params(r#"{"color": "RRGGBB"}"#)
.when(has_picture),
CommandSpec::new("picture.effects", "Artistic Effects", "Picture Format › Adjust", |s, v| {
// One effect at a time, as in Word: a second one replaces the first (#2052).
let e = p::str(v, "effect").unwrap_or("blur").to_string();
adjust(s, move |img| match e.as_str() {
"blur" => img.blur(3.0),
"sharpen" => img.unsharpen(2.0, 2),
"invert" => {
let mut i = img;
i.invert();
i
}
"posterize" => {
let mut rgba = img.to_rgba8();
for px in rgba.pixels_mut() {
for c in px.0.iter_mut().take(3) {
*c = (*c / 64) * 85;
}
}
image::DynamicImage::ImageRgba8(rgba)
}
"pixelate" => {
let (w, h) = (img.width().max(1), img.height().max(1));
img.resize_exact((w / 12).max(1), (h / 12).max(1), image::imageops::FilterType::Nearest).resize_exact(w, h, image::imageops::FilterType::Nearest)
}
_ => img,
})
relook(s, move |l| l.effect = (e != "none").then_some(e))
})
.params(r#"{"effect": "blur|sharpen|invert|posterize|pixelate"}"#)
.params(r#"{"effect": "blur|sharpen|invert|posterize|pixelate|none"} (replaces the picture's effect)"#)
.when(has_picture),
CommandSpec::new("picture.transparency", "Transparency", "Picture Format › Adjust", |s, v| {
let t = p::f32(v, "percent").unwrap_or(50.0).clamp(0.0, 100.0);
@@ -230,17 +222,22 @@ pub fn specs() -> Vec<CommandSpec> {
.params(r#"{"path"?: string, "data"?: base64}"#)
.when(has_picture),
CommandSpec::new("picture.style", "Picture Styles", "Picture Format › Picture Styles", |s, v| {
// A style replaces the picture's style and border, as Word's gallery does (#2052).
let style = p::str(v, "style").unwrap_or("simpleFrame").to_string();
adjust(s, move |img| picture_style(img, &style))
relook(s, move |l| {
l.style = Some(style);
l.border = None;
})
})
.params(r#"{"style": "simpleFrame|thickFrame|rounded|softEdge|shadow"}"#)
.params(r#"{"style": "simpleFrame|thickFrame|rounded|softEdge|shadow"} (replaces the picture's style and border)"#)
.when(has_picture),
CommandSpec::new("picture.border", "Picture Border", "Picture Format › Picture Styles", |s, v| {
let c = p::str(v, "color").and_then(Rgb::parse).unwrap_or(Rgb::BLACK);
let w = p::u64(v, "width").unwrap_or(4).clamp(1, 200) as u32;
adjust(s, move |img| frame(img, w, [c.0, c.1, c.2, 255]))
// Replaces the picture's border, keeping its style (#2052).
relook(s, move |l| l.border = Some((c, w)))
})
.params(r#"{"color"?: "RRGGBB", "width"?: px}"#)
.params(r#"{"color"?: "RRGGBB", "width"?: px} (replaces the picture's border)"#)
.when(has_picture),
CommandSpec::new("arrange.rotate", "Rotate", "Layout › Arrange", |s, v| {
// Turns and flips are the object's angle and mirroring (DrawingML `a:xfrm`): the
@@ -721,15 +718,39 @@ fn copy_to(s: &mut Session, v: &Value) -> CmdResult {
/// text it lands on flows around it, and it moves with that text. On a page where no paragraph
/// starts, it anchors to the nearest line and is positioned on the page. Returns the object's
/// position afterwards.
///
/// A drop across pages can be thrown back: while the object sits on its old page it wraps text
/// away from the paragraph it is anchored to, so once it has moved, that paragraph can move back
/// onto the old page itself — and the object goes with it. Dropping near the top of the next
/// page does it every time (the anchor is that page's first paragraph, which fits on the old page
/// again once the object's wrap is gone). So a cross-page drop is anchored again until the object
/// really stays on the page asked for: the later passes see the pagination as it now is. A
/// same-page move (a resize, a nudge) places once, and a drop that still won't stay keeps its
/// last place rather than failing.
fn move_object(s: &mut Session, pos: Pos, page: Option<u64>, x: f32, y: f32, (w, h): (f32, f32)) -> Result<Pos, CmdError> {
let cur = s.layout().object(&pos, s.page_hint).ok_or_else(|| CmdError::Failed("the object isn't laid out".into()))?.page;
let target = page.map_or(cur, |n| usize::try_from(n).unwrap_or(usize::MAX));
let tries = if target == cur { 1 } else { 3 };
let mut at = pos;
for _ in 0..tries {
at = move_object_once(s, &at, target, x, y, (w, h))?;
s.touch();
if s.layout().object(&at, target).is_some_and(|hit| hit.page == target) {
break;
}
}
Ok(at)
}
/// [`move_object`]'s one placement: move the object's character to the anchor under (x, y) on
/// `page` and offset it from there.
fn move_object_once(s: &mut Session, pos: &Pos, page: usize, x: f32, y: f32, (w, h): (f32, f32)) -> Result<Pos, CmdError> {
use wordcraft_doc::para::Anchor;
if pos.path.depth() > 0 || !movable_story(s, pos.story) {
return Err(CmdError::Disabled("only objects in the body text, headers and footers can be moved".into()));
}
let story = pos.story;
let layout = s.layout();
let cur = layout.object(&pos, s.page_hint).ok_or_else(|| CmdError::Failed("the object isn't laid out".into()))?;
let page = page.map_or(cur.page, |n| usize::try_from(n).unwrap_or(usize::MAX));
let pg = layout.pages.get(page).ok_or_else(|| CmdError::Params(format!("no page {page}")))?;
// Keep a corner of it on the page.
let x = x.max(12.0 - w).min(pg.w - 12.0);
@@ -741,11 +762,11 @@ fn move_object(s: &mut Session, pos: Pos, page: Option<u64>, x: f32, y: f32, (w,
};
// Move its character to the anchor (unless it's already in that paragraph).
let at = if anchor.story == pos.story && anchor.path == pos.path && !on_page {
pos
pos.clone()
} else {
let obj = s.doc.para_at(&pos).and_then(|p| p.object_at(pos.off)).cloned().ok_or_else(|| CmdError::Failed("object vanished".into()))?;
let obj = s.doc.para_at(pos).and_then(|p| p.object_at(pos.off)).cloned().ok_or_else(|| CmdError::Failed("object vanished".into()))?;
let len = wordcraft_doc::para::OBJ.len_utf8();
s.doc.delete_range(&pos, &Pos { off: pos.off + len, ..pos.clone() })?;
s.doc.delete_range(pos, &Pos { off: pos.off + len, ..pos.clone() })?;
let mut at = anchor;
if at.story == pos.story && at.path == pos.path && at.off > pos.off {
at.off = at.off.saturating_sub(len);
@@ -985,35 +1006,83 @@ fn crop(s: &mut Session, v: &Value) -> CmdResult {
})
}
/// A picture's original size: its image's pixels at 96 ppi (the size Insert › Pictures gives it)
/// or a metafile's own frame, before any crop.
#[derive(Clone, Copy, Debug, PartialEq)]
pub struct OriginalSize {
/// Points.
pub w: f32,
pub h: f32,
/// A bitmap's pixels (None for a metafile).
pub pixels: Option<(u32, u32)>,
}
/// The original size of the picture showing `media`, read from the image its Picture Format
/// adjustments started from (Compress Pictures may have resampled the current one); None when the
/// image can't be read.
pub fn original_size(s: &Session, media: &str) -> Option<OriginalSize> {
let key = s.originals.get(media).map_or(media, String::as_str);
let bytes = s.doc.media.get(key).or_else(|| s.doc.media.get(media))?;
if let Some((pw, ph)) = wordcraft_render::image_size(bytes) {
return Some(OriginalSize { w: pw as f32 * 0.75, h: ph as f32 * 0.75, pixels: Some((pw, ph)) });
}
let (w, h, _) = wordcraft_render::metafile_size(bytes)?;
Some(OriginalSize { w, h, pixels: None })
}
/// What a picture's crop leaves of its original size: the size Reset gives back and what a scale
/// of the original counts from. None for anything but a picture whose image can be read.
pub fn cropped_original_size(s: &Session, o: &InlineObject) -> Option<(f32, f32)> {
let InlineObject::Image { media, crop, .. } = o else { return None };
let orig = original_size(s, media)?;
let part = |lo: f32, hi: f32| (1.0 - lo - hi).max(0.1);
Some((orig.w * part(crop[0], crop[2]), orig.h * part(crop[1], crop[3])))
}
/// Size: `scale` is a percentage of the current size, or with `original` of what the crop leaves
/// of the original size (`original` alone gives that back: the Size dialog's Reset); then `width`
/// and `height`, where one alone keeps the proportions unless `lockAspect` is false.
fn size(s: &mut Session, v: &Value) -> CmdResult {
let (_, o) = selected(s).ok_or_else(|| CmdError::Disabled("no object".into()))?;
let (w0, h0) = obj_size(&o);
let lock = p::bool(v, "lockAspect").unwrap_or(true);
let (mut w, mut h) = (w0, h0);
if let Some(sc) = p::f32(v, "scale") {
let (mut w, mut h) = obj_size(&o);
let scale = p::f32(v, "scale");
if p::bool(v, "original").unwrap_or(false) {
let (ow, oh) =
cropped_original_size(s, &o).ok_or_else(|| CmdError::Disabled("only a picture whose image can be read has an original size".into()))?;
let k = scale.unwrap_or(100.0) / 100.0;
(w, h) = (ow * k, oh * k);
} else if let Some(sc) = scale {
w *= sc / 100.0;
h *= sc / 100.0;
}
match (p::f32(v, "width"), p::f32(v, "height")) {
let (nw, nh) = (p::f32(v, "width"), p::f32(v, "height"));
match (nw, nh) {
(Some(nw), Some(nh)) => {
w = nw;
h = nh;
}
(Some(nw), None) => {
if lock && w0 > 0.0 {
h = h0 * nw / w0;
if lock && w > 0.0 {
h *= nw / w;
}
w = nw;
}
(None, Some(nh)) => {
if lock && h0 > 0.0 {
w = w0 * nh / h0;
if lock && h > 0.0 {
w *= nh / h;
}
h = nh;
}
_ => {}
}
let (w, h) = (w.clamp(1.0, 4000.0), h.clamp(1.0, 4000.0));
// A size that keeps the proportions shrinks on both sides to fit, rather than flattening.
let keep = match (nw, nh) {
(Some(_), Some(_)) => false,
(None, None) => true,
_ => lock,
};
let (w, h) = fit_size(w, h, keep);
// Any drawing: a group's members scale with its frame (its child space is kept); a chart or
// SmartArt graphic lays itself out again at the new size.
let theme = s.doc.settings.theme_colors.clone();
@@ -1024,6 +1093,27 @@ fn size(s: &mut Session, v: &Value) -> CmdResult {
})
}
/// An object size within 1–4000 pt. With `keep`, both sides scale by one factor, the one nearest
/// 1 that brings them within the limits, so the proportions hold; a shape too long and thin for any
/// such factor is clamped side by side.
pub fn fit_size(w: f32, h: f32, keep: bool) -> (f32, f32) {
let side = |x: f32| if x.is_nan() { 1.0 } else { x.clamp(1.0, 4000.0) };
if keep && w.is_finite() && h.is_finite() && w > 0.0 && h > 0.0 {
let (lo, hi) = ((1.0 / w).max(1.0 / h), (4000.0 / w).min(4000.0 / h));
if lo <= hi {
let k = if lo > 1.0 {
lo
} else if hi < 1.0 {
hi
} else {
1.0
};
return (side(w * k), side(h * k));
}
}
(side(w), side(h))
}
/// Apply an image operation to the selected picture's bitmap (stored as a new PNG).
fn adjust(s: &mut Session, f: impl Fn(image::DynamicImage) -> image::DynamicImage) -> CmdResult {
let (_, o) = selected(s).ok_or_else(|| CmdError::Disabled("no picture".into()))?;
@@ -1064,6 +1154,97 @@ fn adjust_media(s: &mut Session, base: &str, f: impl Fn(image::DynamicImage) ->
})
}
/// What Artistic Effects, Picture Styles and Picture Border drew on a picture's bitmap: the
/// bitmap they started from and the effect, style and border on it, so the next one draws again
/// from that bitmap instead of over the last result. Picking a second style replaces the first,
/// as in Word, instead of framing the frame (#2052). Kept in [`Session::picture_looks`] under the
/// media key of the drawn bitmap.
#[derive(Clone, Debug, Default)]
pub struct PictureLook {
/// The media key of the bitmap before any of them.
pub base: String,
pub effect: Option<String>,
pub style: Option<String>,
pub border: Option<(Rgb, u32)>,
/// The drawn bitmap: a media key that names other bytes later (undo then a new picture
/// reusing the key) is not this look's.
pub drawn: std::sync::Arc<Vec<u8>>,
}
/// Change the selected picture's effect, style or border and draw it again from the bitmap
/// under them (effect, then style, then border).
fn relook(s: &mut Session, change: impl FnOnce(&mut PictureLook)) -> CmdResult {
let (_, o) = selected(s).ok_or_else(|| CmdError::Disabled("no picture".into()))?;
let InlineObject::Image { media, .. } = o else { return Err(CmdError::Disabled("not a picture".into())) };
let cur = s.doc.media.get(&media).cloned();
let mut look = s
.picture_looks
.get(&media)
.filter(|l| cur.as_ref().is_some_and(|c| std::sync::Arc::ptr_eq(c, &l.drawn)) && s.doc.media.contains_key(&l.base))
.cloned()
.unwrap_or_else(|| PictureLook { base: media.clone(), ..Default::default() });
change(&mut look);
if look.effect.is_none() && look.style.is_none() && look.border.is_none() {
// Nothing left on it: the bitmap as it was.
let base = look.base.clone();
return with_obj(s, |o| {
if let InlineObject::Image { media, .. } = o {
*media = base.clone();
}
});
}
let l = look.clone();
let out = adjust_media(s, &look.base, move |img| {
let mut img = match &l.effect {
Some(e) => artistic_effect(img, e),
None => img,
};
if let Some(st) = &l.style {
img = picture_style(img, st);
}
if let Some((c, w)) = l.border {
img = frame(img, w, [c.0, c.1, c.2, 255]);
}
img
})?;
if let Some((_, InlineObject::Image { media, .. })) = selected(s)
&& let Some(drawn) = s.doc.media.get(&media).cloned()
{
s.picture_looks.insert(media, PictureLook { drawn, ..look });
}
Ok(out)
}
fn artistic_effect(img: image::DynamicImage, effect: &str) -> image::DynamicImage {
match effect {
"blur" => img.blur(3.0),
"sharpen" => img.unsharpen(2.0, 2),
"invert" => {
let mut i = img;
i.invert();
i
}
"posterize" => {
let mut rgba = img.to_rgba8();
for px in rgba.pixels_mut() {
for c in px.0.iter_mut().take(3) {
*c = (*c / 64) * 85;
}
}
image::DynamicImage::ImageRgba8(rgba)
}
"pixelate" => {
let (w, h) = (img.width().max(1), img.height().max(1));
img.resize_exact((w / 12).max(1), (h / 12).max(1), image::imageops::FilterType::Nearest).resize_exact(
w,
h,
image::imageops::FilterType::Nearest,
)
}
_ => img,
}
}
fn reset(s: &mut Session, _: &Value) -> CmdResult {
let (_, o) = selected(s).ok_or_else(|| CmdError::Disabled("no picture".into()))?;
let InlineObject::Image { media, .. } = o else { return Err(CmdError::Disabled("not a picture".into())) };
@@ -1359,6 +1540,9 @@ fn group(s: &mut Session, _: &Value) -> CmdResult {
fn ungroup(s: &mut Session, _: &Value) -> CmdResult {
let (pos, obj) = selected(s).ok_or_else(|| CmdError::Disabled("select a group first".into()))?;
let InlineObject::Group { w, h, float, .. } = &obj else { return Err(CmdError::Disabled("select a group first".into())) };
if holds_ink(&obj) {
return Err(CmdError::Disabled(INK_GROUP.into()));
}
// Its offsets, explicit: the members are placed from the same anchor.
let pos = if float.wrap == Wrap::Inline {
if pos.path.depth() > 0 || !movable_story(s, pos.story) {
@@ -1594,6 +1778,41 @@ mod tests {
assert_eq!(s.run("arrange.selectionPane", &json!({})).unwrap().as_array().map(Vec::len), Some(2));
}
/// #2019: Insert › Shapes adds each new shape beside the selected one, as Word draws it, so
/// shapes inserted in a row are all kept (the second used to replace the first, which stays
/// selected after its insert). Selected text stays too, and the new shape is selected.
#[test]
fn shapes_inserted_in_a_row_are_all_kept_and_keep_selected_text() {
let objects = |s: &mut Session| s.run("arrange.selectionPane", &json!({})).unwrap().as_array().map(Vec::len);
let mut s = Session::new(wordcraft_doc::Document::new());
for kind in ["rectangle", "ellipse", "star"] {
s.run("insert.shape", &json!({"kind": kind})).unwrap();
}
assert_eq!(objects(&mut s), Some(3));
let mut s = Session::new(wordcraft_doc::Document::from_text("Kiln room"));
s.run("select.text", &json!({"text": "Kiln"})).unwrap();
s.run("insert.shape", &json!({"kind": "rectangle"})).unwrap();
let text = s.doc.plain_text(StoryRef::Body);
assert!(text.starts_with("Kiln") && text.ends_with(" room"), "{text:?}");
assert_eq!(objects(&mut s), Some(1));
let (a, b) = s.sel.ordered();
let obj = wordcraft_doc::para::OBJ.len_utf8();
assert_eq!((a.off, b.off), (4, 4 + obj), "the new shape, after the text, is selected");
}
/// Insert › Pictures does replace a selected picture, as Word does (checked in Word for
/// Windows while fixing #2019): only shapes are added beside the selection.
#[test]
fn an_inserted_picture_replaces_a_selected_picture() {
let data = super::super::insert::base64_encode(&png(20, 10));
let mut s = Session::new(wordcraft_doc::Document::new());
s.run("insert.picture", &json!({"data": data})).unwrap();
s.sel = crate::Selection { anchor: Pos::body(0, 0), focus: Pos::body(0, wordcraft_doc::para::OBJ.len_utf8()) };
s.run("insert.picture", &json!({"data": data})).unwrap();
assert_eq!(s.run("arrange.selectionPane", &json!({})).unwrap().as_array().map(Vec::len), Some(1));
}
#[test]
fn picture_pipeline() {
let mut s = Session::new(wordcraft_doc::Document::new());
@@ -1669,6 +1888,59 @@ mod tests {
assert_eq!(rgba(&s), orig, "reset still finds the original");
}
/// #2052: a picture style picked after another replaces it (Thick Frame then Simple Frame is
/// only the simple frame), a border replaces the border and keeps the style, an artistic
/// effect replaces the effect, and transparency 50% then 10% is 10%, as in Word.
#[test]
fn picture_styles_borders_and_effects_replace_the_last_one() {
let picture = || {
let mut s = Session::new(wordcraft_doc::Document::new());
s.run("insert.picture", &json!({"data": super::super::insert::base64_encode(&png(80, 40))})).unwrap();
s
};
let bitmap = |s: &Session| {
let (_, o) = selected(s).unwrap();
let InlineObject::Image { media, .. } = o else { panic!("a picture") };
image::load_from_memory(&s.doc.media.get(&media).cloned().unwrap()).unwrap().to_rgba8()
};
let after = |cmds: &[(&str, Value)]| {
let mut s = picture();
for (id, v) in cmds {
s.run(id, v).unwrap();
}
bitmap(&s)
};
let (thick, simple) = (json!({"style": "thickFrame"}), json!({"style": "simpleFrame"}));
assert_eq!(after(&[("picture.style", thick.clone()), ("picture.style", simple.clone())]), after(&[("picture.style", simple.clone())]));
let (wide, thin) = (json!({"width": 8}), json!({"width": 2}));
assert_eq!(
after(&[("picture.style", thick.clone()), ("picture.border", wide.clone()), ("picture.border", thin.clone())]),
after(&[("picture.style", thick.clone()), ("picture.border", thin.clone())]),
"a border replaces the border and keeps the style"
);
assert_eq!(
after(&[("picture.border", wide), ("picture.style", simple.clone())]),
after(&[("picture.style", simple)]),
"a style replaces the border too"
);
assert_eq!(
after(&[("picture.effects", json!({"effect": "invert"})), ("picture.effects", json!({"effect": "posterize"}))]),
after(&[("picture.effects", json!({"effect": "posterize"}))])
);
assert_eq!(after(&[("picture.effects", json!({"effect": "invert"})), ("picture.effects", json!({"effect": "none"}))]), bitmap(&picture()));
let fifty_then_ten = after(&[("picture.transparency", json!({"percent": 50})), ("picture.transparency", json!({"percent": 10}))]);
assert!(fifty_then_ten.pixels().all(|p| p.0[3] == 229), "255 * 0.9");
// Undo then a new edit reusing the drawn bitmap's media key starts from what is there.
let mut s = picture();
s.run("picture.style", &thick).unwrap();
s.run("edit.undo", &json!({})).unwrap();
s.run("picture.effects", &json!({"effect": "invert"})).unwrap();
s.run("picture.border", &json!({"width": 2})).unwrap();
let (w, h) = bitmap(&s).dimensions();
assert_eq!((w, h), (84, 44), "the effect and a 2 px border, no thick frame from before the undo");
}
/// #513: Wrap Text › Fix Position on Page / Move with Text re-anchor a floating picture
/// vertically without moving it; one undo step each; an inline picture can't.
#[test]
@@ -1925,6 +2197,67 @@ mod tests {
assert_eq!(crop, [0.1, 0.2, 0.05, 0.0]);
}
/// #2573: Size from the original. `original` alone gives the picture back its original size
/// (its pixels at 96 ppi, less what is cropped) and `scale` is a percentage of it; Compress
/// Pictures doesn't change what the original is. Only a picture has one.
#[test]
fn picture_size_counts_from_the_original() {
let mut s = Session::new(wordcraft_doc::Document::new());
let data = super::super::insert::base64_encode(&png(40, 20));
s.run("insert.picture", &json!({"data": data})).unwrap();
let size = |s: &Session| obj_size(&selected(s).unwrap().1);
assert_eq!(size(&s), (30.0, 15.0), "40×20 px at 96 ppi");
s.run("picture.size", &json!({"width": 300.0, "height": 100.0, "lockAspect": false})).unwrap();
s.run("picture.compress", &json!({"maxPixels": 32})).unwrap();
let Some((_, InlineObject::Image { media, .. })) = selected(&s) else { panic!("a picture") };
assert_eq!(wordcraft_render::image_size(&s.doc.media[&media]), Some((32, 16)), "compressed");
assert_eq!(original_size(&s, &media), Some(OriginalSize { w: 30.0, h: 15.0, pixels: Some((40, 20)) }));
s.run("picture.size", &json!({"original": true})).unwrap();
assert_eq!(size(&s), (30.0, 15.0));
s.run("picture.size", &json!({"original": true, "scale": 250})).unwrap();
assert_eq!(size(&s), (75.0, 37.5));
// Half the width cropped away: half the original width is left.
s.run("picture.crop", &json!({"left": 0.25, "right": 0.25})).unwrap();
s.run("picture.size", &json!({"original": true})).unwrap();
assert_eq!(size(&s), (15.0, 15.0));
s.run("edit.undo", &json!({})).unwrap();
assert_eq!(size(&s), (75.0, 37.5), "one undo step");
s.run("insert.shape", &json!({"kind": "rectangle"})).unwrap();
assert!(s.run("picture.size", &json!({"original": true})).is_err(), "a shape has no original size");
}
/// #2573: a size past 4000 pt (or under 1 pt) that keeps the proportions scales as a whole
/// instead of flattening against the limit; both sides given are clamped each on its own.
#[test]
fn picture_size_past_the_limit_keeps_the_proportions() {
let mut s = Session::new(wordcraft_doc::Document::new());
let data = super::super::insert::base64_encode(&png(40, 20));
s.run("insert.picture", &json!({"data": data})).unwrap();
let size = |s: &Session| obj_size(&selected(s).unwrap().1);
s.run("picture.size", &json!({"width": 9000.0})).unwrap();
assert_eq!(size(&s), (4000.0, 2000.0));
s.run("picture.size", &json!({"scale": 1000})).unwrap();
assert_eq!(size(&s), (4000.0, 2000.0));
s.run("picture.size", &json!({"width": 9000.0, "height": 10.0})).unwrap();
assert_eq!(size(&s), (4000.0, 10.0));
s.run("picture.size", &json!({"height": 30.0, "lockAspect": false})).unwrap();
assert_eq!(size(&s), (4000.0, 30.0));
// Below 1 pt as well: a 1% scale of 50×200 pt is 1×4 pt, not 1×2.
s.run("picture.size", &json!({"width": 50.0, "height": 200.0})).unwrap();
s.run("picture.size", &json!({"scale": 1})).unwrap();
assert_eq!(size(&s), (1.0, 4.0));
// Too long and thin for one factor: side by side. NaN and infinity stay in range.
assert_eq!(fit_size(0.5, 5000.0, true), (1.0, 4000.0));
assert_eq!(fit_size(f32::NAN, 10.0, true), (1.0, 10.0));
assert_eq!(fit_size(f32::INFINITY, 10.0, true), (4000.0, 10.0));
// Hostile values never leave the 1–4000 pt range.
for v in [json!({"scale": -50}), json!({"width": -1.0}), json!({"original": "yes", "scale": 1e30}), json!({"height": 1e30})] {
s.run("picture.size", &v).unwrap();
let (w, h) = size(&s);
assert!((1.0..=4000.0).contains(&w) && (1.0..=4000.0).contains(&h), "{v}: {w}×{h}");
}
}
/// A crop handle drop: `picture.crop` with the new size is one edit, one undo step.
#[test]
fn picture_crop_with_size_is_one_undo_step() {
@@ -2209,4 +2542,100 @@ mod tests {
let got = anchor_paragraph(&s.doc, &layout, 1, (mid, top + 1.0), StoryRef::Body);
assert_eq!(got, Some(wordcraft_doc::Path::top(plain)));
}
/// A Word file whose only paragraph holds a group of a rectangle and Word's own ink (with its
/// picture fallback).
fn group_with_word_ink() -> Vec<u8> {
use std::io::Write;
let ns = r#"xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships" xmlns:wp="http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing" xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main" xmlns:pic="http://schemas.openxmlformats.org/drawingml/2006/picture" xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006" xmlns:wps="http://schemas.microsoft.com/office/word/2010/wordprocessingShape" xmlns:wpg="http://schemas.microsoft.com/office/word/2010/wordprocessingGroup" xmlns:w14="http://schemas.microsoft.com/office/word/2010/wordml" xmlns:wpi="http://schemas.microsoft.com/office/word/2010/wordprocessingInk""#;
let rect = r#"<wps:wsp><wps:cNvPr id="2" name="Box"/><wps:cNvSpPr/><wps:spPr><a:xfrm><a:off x="0" y="0"/><a:ext cx="914400" cy="914400"/></a:xfrm><a:prstGeom prst="rect"><a:avLst/></a:prstGeom></wps:spPr><wps:bodyPr/></wps:wsp>"#;
let ink = r#"<mc:AlternateContent><mc:Choice Requires="wpi"><w14:contentPart bwMode="auto" r:id="rInk"><w14:nvContentPartPr><w14:cNvPr id="3" name="Ink 3"/><w14:cNvContentPartPr/></w14:nvContentPartPr><w14:xfrm><a:off x="914400" y="0"/><a:ext cx="914400" cy="457200"/></w14:xfrm></w14:contentPart></mc:Choice><mc:Fallback><pic:pic><pic:nvPicPr><pic:cNvPr id="3" name="Ink 3"/><pic:cNvPicPr/></pic:nvPicPr><pic:blipFill><a:blip r:embed="rPreview"/></pic:blipFill><pic:spPr><a:xfrm><a:off x="914400" y="0"/><a:ext cx="914400" cy="457200"/></a:xfrm><a:prstGeom prst="rect"><a:avLst/></a:prstGeom></pic:spPr></pic:pic></mc:Fallback></mc:AlternateContent>"#;
let group = format!(
r#"<w:drawing><wp:anchor behindDoc="0" locked="0" layoutInCell="1" allowOverlap="1" relativeHeight="1" simplePos="0" distT="0" distB="0" distL="0" distR="0"><wp:simplePos x="0" y="0"/><wp:positionH relativeFrom="column"><wp:posOffset>0</wp:posOffset></wp:positionH><wp:positionV relativeFrom="paragraph"><wp:posOffset>0</wp:posOffset></wp:positionV><wp:extent cx="1828800" cy="914400"/><wp:wrapNone/><wp:docPr id="1" name="Group 1"/><a:graphic><a:graphicData uri="http://schemas.microsoft.com/office/word/2010/wordprocessingGroup"><wpg:wgp><wpg:cNvGrpSpPr/><wpg:grpSpPr><a:xfrm><a:off x="0" y="0"/><a:ext cx="1828800" cy="914400"/><a:chOff x="0" y="0"/><a:chExt cx="1828800" cy="914400"/></a:xfrm></wpg:grpSpPr>{rect}{ink}</wpg:wgp></a:graphicData></a:graphic></wp:anchor></w:drawing>"#
);
let rel = |id: &str, ty: &str, target: &str| {
format!(r#"<Relationship Id="{id}" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/{ty}" Target="{target}"/>"#)
};
let parts = [
("_rels/.rels", format!(r#"<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">{}</Relationships>"#, rel("rId1", "officeDocument", "word/document.xml"))),
("[Content_Types].xml", r#"<Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types"><Override PartName="/word/ink/ink1.xml" ContentType="application/inkml+xml"/><Default Extension="png" ContentType="image/png"/></Types>"#.to_string()),
("word/document.xml", format!(r#"<w:document {ns}><w:body><w:p><w:r>{group}</w:r></w:p></w:body></w:document>"#)),
(
"word/_rels/document.xml.rels",
format!(r#"<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">{}{}</Relationships>"#, rel("rInk", "customXml", "ink/ink1.xml"), rel("rPreview", "image", "media/preview.png")),
),
("word/ink/ink1.xml", r#"<ink xmlns="http://www.w3.org/2003/InkML"><trace>0 0, 1 1</trace></ink>"#.to_string()),
];
let mut zw = zip::ZipWriter::new(std::io::Cursor::new(Vec::new()));
for (name, xml) in parts {
zw.start_file(name, zip::write::SimpleFileOptions::default()).unwrap();
zw.write_all(xml.as_bytes()).unwrap();
}
zw.start_file("word/media/preview.png", zip::write::SimpleFileOptions::default()).unwrap();
zw.write_all(&png(8, 4)).unwrap();
zw.finish().unwrap().into_inner()
}
fn ink_members(doc: &wordcraft_doc::Document) -> usize {
let groups: Vec<_> = doc.body.iter().filter_map(|b| b.as_para()).flat_map(|p| p.objects.iter()).collect();
let nested = groups
.iter()
.filter_map(|o| match o {
InlineObject::Group { children, .. } => Some(children.iter().filter(|c| matches!(c.obj, InlineObject::Graphic { .. })).count()),
_ => None,
})
.sum::<usize>();
let alone = groups.iter().filter(|o| matches!(o, InlineObject::Graphic { .. })).count();
nested + alone
}
/// Gap 2 review: a group holding Word's own ink can't be ungrouped (its ink can't stand alone
/// yet), so open → Ungroup → save → reopen keeps the ink in the group with its part.
#[test]
fn ungroup_refuses_a_group_holding_word_ink_and_the_ink_survives_saving() {
use std::io::Read;
let mut s = Session::new(wordcraft_docx::read(&group_with_word_ink()).unwrap());
assert_eq!(ink_members(&s.doc), 1, "the fixture's ink is read into the group");
let pos = Pos::body(0, 0);
s.sel = Selection { anchor: pos.clone(), focus: Pos { off: pos.off + wordcraft_doc::para::OBJ.len_utf8(), ..pos } };
assert!(matches!(selected(&s), Some((_, InlineObject::Group { .. }))));
let before = wordcraft_docx::write(&s.doc).unwrap();
assert!(has_group(&s).is_some(), "Ungroup is greyed out");
let err = s.run("arrange.ungroup", &json!({})).unwrap_err();
assert!(matches!(err, CmdError::Disabled(_)), "{err:?}");
let err = ungroup(&mut s, &json!({})).unwrap_err();
assert!(matches!(err, CmdError::Disabled(_)), "{err:?}");
let bytes = wordcraft_docx::write(&s.doc).unwrap();
let mut z = zip::ZipArchive::new(std::io::Cursor::new(bytes.clone())).unwrap();
let mut xml = String::new();
z.by_name("word/document.xml").unwrap().read_to_string(&mut xml).unwrap();
assert_eq!(xml.matches("<w14:contentPart").count(), 1, "{xml}");
assert!(xml.contains("<mc:Fallback>"), "{xml}");
assert_eq!(bytes, before, "the refused ungroup leaves the document unchanged");
let names: Vec<String> = z.file_names().map(String::from).collect();
assert_eq!(names.iter().filter(|n| n.starts_with("word/ink/")).count(), 1, "{names:?}");
let mut rels = String::new();
z.by_name("word/_rels/document.xml.rels").unwrap().read_to_string(&mut rels).unwrap();
assert_eq!(rels.matches("/customXml\"").count(), 1, "one ink relationship: {rels}");
let back = wordcraft_docx::read(&bytes).unwrap();
assert_eq!(ink_members(&back), 1);
}
/// Review: only Word's ink pins a group; a group of charts or diagrams ungroups as before.
#[test]
fn only_ink_members_pin_a_group() {
let s = Session::new(wordcraft_docx::read(&group_with_word_ink()).unwrap());
let pos = Pos::body(0, 0);
let mut s = s;
s.sel = Selection { anchor: pos.clone(), focus: Pos { off: pos.off + wordcraft_doc::para::OBJ.len_utf8(), ..pos } };
let Some((_, mut group)) = selected(&s) else { panic!("no group") };
assert!(holds_ink(&group));
let InlineObject::Group { children, .. } = &mut group else { panic!("not a group") };
for c in children.iter_mut() {
if let InlineObject::Graphic { graphic, .. } = &mut c.obj {
std::sync::Arc::make_mut(graphic).kind = GraphicKind::Chart;
}
}
assert!(!holds_ink(&group), "a group of charts is not pinned");
}
}
+172 -13
View File
@@ -27,7 +27,7 @@ pub fn specs() -> Vec<CommandSpec> {
CommandSpec::new("layout.size", "Size", "Layout › Page Setup", size).params(r#"{"name"?: "Letter|Legal|A4|…", "width"?: pt, "height"?: pt} (each side clamped to 7.2–1584 pt, 0.1–22 in; the section keeps its orientation)"#),
CommandSpec::new("layout.columns", "Columns", "Layout › Page Setup", columns)
.params(
r#"{"count": 1-12, "space"?: pt, "separator"?: bool, "preset"?: "left|right", "widths"?: [[width pt, space after pt], …] (unequal columns; sets the count), "apply"?: "section|document|forward"}"#,
r#"{"count": 1-12, "space"?: pt, "separator"?: bool, "preset"?: "left|right", "widths"?: [[width pt, space after pt], …] (unequal columns; sets the count), "apply"?: "section|document|forward|selection"} (default: selection when text is selected in the main text — section breaks around it — else section)"#,
),
CommandSpec::new("layout.break", "Breaks", "Layout › Page Setup", breaks)
.params(r#"{"kind": "page|column|textWrapping|nextPage|continuous|evenPage|oddPage"}"#),
@@ -215,20 +215,46 @@ fn manual_hyphenation(s: &mut Session, v: &Value) -> CmdResult {
Ok(best.map(|(_, v)| v).unwrap_or_else(|| json!({"done": true})))
}
fn block_of(s: &Session) -> usize {
s.sel.focus.path.0.first().copied().unwrap_or(0) as usize
fn section_of_part(s: &Session, id: u32) -> usize {
let sections = s.doc.sections();
for (i, (_, sp)) in sections.iter().enumerate() {
if [sp.headers.default, sp.headers.first, sp.headers.even, sp.footers.default, sp.footers.first, sp.footers.even].contains(&Some(id)) {
return i;
}
}
for (bi, block) in s.doc.body.iter().enumerate() {
if let wordcraft_doc::Block::Para(p) = &**block {
for o in &p.objects {
if matches!(o, wordcraft_doc::para::InlineObject::NoteRef { id: nid, .. } if *nid == id) {
return s.doc.section_index_of(bi);
}
}
}
}
0
}
fn section_index_of_focus(s: &Session) -> usize {
match s.sel.focus.story {
wordcraft_doc::StoryRef::Body => s.doc.section_index_of(s.sel.focus.path.0.first().copied().unwrap_or(0) as usize),
wordcraft_doc::StoryRef::Part(id) => section_of_part(s, id),
}
}
pub fn sect(s: &Session) -> SectionProps {
let i = s.doc.section_index_of(block_of(s));
let i = section_index_of_focus(s);
s.doc.sections().get(i).map(|(_, x)| (*x).clone()).unwrap_or_default()
}
/// The last block of each section of the selection.
fn selected_section_ends(s: &Session) -> Vec<usize> {
let (a, b) = s.sel.ordered();
let (i0, i1) = (a.path.0.first().copied().unwrap_or(0) as usize, b.path.0.first().copied().unwrap_or(0) as usize);
let ends: Vec<usize> = s.doc.sections().iter().map(|(e, _)| *e).collect();
if a.story != wordcraft_doc::StoryRef::Body || b.story != wordcraft_doc::StoryRef::Body {
let i = section_index_of_focus(s);
return vec![ends.get(i).copied().unwrap_or(0)];
}
let (i0, i1) = (a.path.0.first().copied().unwrap_or(0) as usize, b.path.0.first().copied().unwrap_or(0) as usize);
let first = ends.iter().position(|e| i0 <= *e).unwrap_or(0);
let last = ends.iter().position(|e| i1 <= *e).unwrap_or(first);
(first..=last).map(|k| ends.get(k).copied().unwrap_or(0)).collect()
@@ -474,7 +500,9 @@ fn margins(s: &mut Session, v: &Value) -> CmdResult {
)
}
};
let gutter = p::f32(v, "gutter");
// Negative margins count as zero before the fit check, so one can't offset another.
let (t, b, l, r) = (t.max(0.0), b.max(0.0), l.max(0.0), r.max(0.0));
let gutter = p::f32(v, "gutter").map(|g| g.max(0.0));
let cur = sect(s);
if !margins_fit((cur.page_w, cur.page_h), (t, b, l, r), gutter.unwrap_or(cur.gutter)) {
return Err(CmdError::Params("margins leave no room for text".into()));
@@ -485,12 +513,12 @@ fn margins(s: &mut Session, v: &Value) -> CmdResult {
s.doc.settings.mirror_margins = on;
}
with_sect(s, |x| {
x.margin_top = t.max(0.0);
x.margin_bottom = b.max(0.0);
x.margin_left = l.max(0.0);
x.margin_right = r.max(0.0);
x.margin_top = t;
x.margin_bottom = b;
x.margin_left = l;
x.margin_right = r;
if let Some(g) = gutter {
x.gutter = g.max(0.0);
x.gutter = g;
}
})
}
@@ -569,7 +597,9 @@ fn columns(s: &mut Session, v: &Value) -> CmdResult {
// Equal columns keep one spacing; unequal ones remember the first gap for a later switch back.
let space = widths.first().map(|w| w.1).unwrap_or(space);
let cols = Columns { count: count as u32, space, separator: sep, widths };
match p::str(v, "apply").unwrap_or("section") {
// With text selected the gallery works on that text, as in Word (#2024).
let default = if selection_sectionable(s) { "selection" } else { "section" };
match p::str(v, "apply").unwrap_or(default) {
"section" => with_sect(s, |x| x.columns = cols.clone()),
"document" => {
let ends: Vec<usize> = s.doc.sections().iter().map(|(e, _)| *e).collect();
@@ -585,10 +615,61 @@ fn columns(s: &mut Session, v: &Value) -> CmdResult {
breaks(s, &json!({"kind": "continuous"}))?;
with_sect(s, |x| x.columns = cols.clone())
}
x => Err(CmdError::Params(format!("unknown apply `{x}` (section, document or forward)"))),
"selection" => {
selection_to_sections(s)?;
with_sect(s, |x| x.columns = cols.clone())
}
x => Err(CmdError::Params(format!("unknown apply `{x}` (section, document, forward or selection)"))),
}
}
/// Whether the selection can become sections of its own: text selected in the main text, outside
/// tables.
pub fn selection_sectionable(s: &Session) -> bool {
let (a, b) = s.sel.ordered();
!s.sel.is_collapsed() && [a, b].iter().all(|p| p.story == StoryRef::Body && p.path.depth() == 0)
}
/// Columns for the selected text (#2024), as Word does it: a continuous section break where the
/// selection starts and one where it ends, splitting paragraphs there, unless that end already
/// meets a section boundary (the document's start or end, or a section break). The selection keeps
/// its text, which is then the section(s) between.
fn selection_to_sections(s: &mut Session) -> Result<(), CmdError> {
if !selection_sectionable(s) {
return Err(CmdError::Params("selected text in the main text, outside tables, is needed".into()));
}
let old = s.sel.clone();
let (a, mut b) = old.ordered();
let block = |p: &Pos| p.path.0.first().copied().unwrap_or(0) as usize;
let len = |s: &Session, p: &Pos| s.doc.para_at(p).map(|q| q.len()).unwrap_or(0);
let ends: Vec<usize> = s.doc.sections().iter().map(|(e, _)| *e).collect();
let after_break = |i: usize| i.checked_sub(1).is_some_and(|j| ends.contains(&j));
// A selection that took a section break with it ends before that break.
if b.off == 0 && block(&b) > block(&a) && after_break(block(&b)) {
let prev = Pos { path: wordcraft_doc::Path::top(block(&b) - 1), off: 0, ..b.clone() };
b = Pos { off: len(s, &prev), ..prev };
}
let last_block = s.doc.body.len().saturating_sub(1);
let end_break = !(b.off == len(s, &b) && (ends.contains(&block(&b)) || block(&b) == last_block));
let start_break = !(a.off == 0 && (block(&a) == 0 || after_break(block(&a))));
// The end first: the start's position doesn't move.
if end_break {
s.sel = Selection::caret(b.clone());
breaks(s, &json!({"kind": "continuous"}))?;
}
let (first, last) = if start_break {
s.sel = Selection::caret(a.clone());
breaks(s, &json!({"kind": "continuous"}))?;
let first = s.sel.focus.clone();
let last = shifted_by_split(&b, &a, &first);
(first, last)
} else {
(a, b)
};
s.sel = if old.anchor <= old.focus { Selection { anchor: first, focus: last } } else { Selection { anchor: last, focus: first } };
Ok(())
}
/// `[[width, space after], …]` in points: 1–12 columns, each at least 0.25", that fit the text
/// width (`tw`).
fn column_widths(v: &Value, tw: f32) -> Result<Vec<(f32, f32)>, CmdError> {
@@ -956,6 +1037,84 @@ mod tests {
assert!(s.run("layout.columns", &json!({"count": 2, "apply": "elsewhere"})).is_err());
}
/// Each section's paragraphs (a section break's own paragraph shows as "") and column count.
fn sections_of(s: &Session) -> Vec<(Vec<String>, u32)> {
let mut from = 0;
let mut out = Vec::new();
for (end, x) in s.doc.sections() {
let paras = (from..=end).map(|i| s.doc.para_at(&Pos::body(i, 0)).map(|p| p.text.clone()).unwrap_or_default()).collect();
out.push((paras, x.columns.count));
from = end + 1;
}
out
}
fn select(s: &mut Session, from: (usize, usize), to: (usize, usize)) {
s.sel = Selection { anchor: Pos::body(from.0, from.1), focus: Pos::body(to.0, to.1) };
}
fn strs(v: &[&str]) -> Vec<String> {
v.iter().map(|x| x.to_string()).collect()
}
/// Columns on selected text (#2024), as observed in Word: a continuous section break before
/// and after the selection, the columns on the section between, the rest left alone.
#[test]
fn columns_apply_to_the_selected_text() {
let five = || Session::new(wordcraft_doc::Document::from_text("p1\np2\nalpha beta gamma\np4\np5"));
// A whole paragraph (a triple-click takes its mark): each break gets a paragraph of its own.
let mut s = five();
select(&mut s, (2, 0), (3, 0));
let whole = s.selected_text();
s.run("layout.columns", &json!({"count": 2})).unwrap();
assert_eq!(sections_of(&s), vec![(strs(&["p1", "p2", ""]), 1), (strs(&["alpha beta gamma", ""]), 2), (strs(&["p4", "p5"]), 1)]);
assert!(s.doc.sections().iter().skip(1).all(|(_, x)| x.start == SectionStart::Continuous));
assert_eq!(s.selected_text(), whole, "the paragraph stays selected");
// The same paragraph again: its section changes, no new breaks.
select(&mut s, (3, 0), (4, 0));
s.run("layout.columns", &json!({"count": 3})).unwrap();
assert_eq!(sections_of(&s).iter().map(|(_, n)| *n).collect::<Vec<_>>(), vec![1, 3, 1]);
// Taking the section break along changes nothing more.
select(&mut s, (3, 0), (5, 0));
s.run("layout.columns", &json!({"count": 2})).unwrap();
assert_eq!(sections_of(&s).iter().map(|(_, n)| *n).collect::<Vec<_>>(), vec![1, 2, 1]);
// One undo step takes the breaks and the columns away.
let mut s = five();
select(&mut s, (2, 0), (3, 0));
s.run("layout.columns", &json!({"count": 2})).unwrap();
s.run("edit.undo", &json!({})).unwrap();
assert_eq!(sections_of(&s), vec![(strs(&["p1", "p2", "alpha beta gamma", "p4", "p5"]), 1)]);
// Words inside a paragraph: the breaks split it there.
let mut s = five();
select(&mut s, (2, 10), (2, 6));
s.run("layout.columns", &json!({"count": 2})).unwrap();
assert_eq!(sections_of(&s), vec![(strs(&["p1", "p2", "alpha "]), 1), (strs(&["beta"]), 2), (strs(&[" gamma", "p4", "p5"]), 1)]);
assert_eq!(s.selected_text(), "beta");
assert!(s.sel.anchor > s.sel.focus, "a backward selection stays backward");
// At the document's start no break goes before it, at its end none after.
let mut s = five();
select(&mut s, (0, 0), (1, 0));
s.run("layout.columns", &json!({"count": 2})).unwrap();
assert_eq!(sections_of(&s), vec![(strs(&["p1", ""]), 2), (strs(&["p2", "alpha beta gamma", "p4", "p5"]), 1)]);
let mut s = five();
select(&mut s, (4, 0), (4, 2));
s.run("layout.columns", &json!({"count": 2})).unwrap();
assert_eq!(sections_of(&s), vec![(strs(&["p1", "p2", "alpha beta gamma", "p4", ""]), 1), (strs(&["p5"]), 2)]);
// Select All: the whole document, no breaks.
let mut s = five();
s.run("select.all", &json!({})).unwrap();
s.run("layout.columns", &json!({"count": 2})).unwrap();
assert_eq!(sections_of(&s), vec![(strs(&["p1", "p2", "alpha beta gamma", "p4", "p5"]), 2)]);
// Without a selection the caret's section changes; "selection" then needs one.
let mut s = five();
s.run("layout.columns", &json!({"count": 2})).unwrap();
assert_eq!(sections_of(&s).len(), 1);
assert!(s.run("layout.columns", &json!({"count": 2, "apply": "selection"})).is_err());
}
/// #407 Line Numbers options: start, count by, distance and restart; options alone keep the
/// restart mode, `none` turns numbering off.
#[test]

Some files were not shown because too many files have changed in this diff Show More