Martin BallandClaude Opus 5.5 4963e30eed #3692: a row's gridAfter is bounded like the DOCX reader's ...

Audit finding 35: Table::cols() added a deserialized row's grid_after
(u32) unchecked, and layout_table sized its column vectors from it, so a
small native JSON document could ask for a 16 GB grid (and overflow the
sum on 32-bit wasm). Row::grid_after() now clamps to MAX_COLS as
first_grid() does and the importer does; cols() sums with saturating
adds; layout reads the bounded value. Test:
tests_budget::a_huge_grid_after_is_bounded_like_the_docx_readers.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

2026-10-25 21:01:25 -04:00

..

balloons.rs

#3706: author colours ranked once per drawing, not per run or comment

2026-10-26 00:19:16 +00:00

contour.rs

Pictures: one decoder checks the header against the budget, progressive JPEG coefficients included (#3644)

2026-10-24 18:01:45 -01:00

display.rs

#3692: review — linear-time tests count lookup steps instead of timing them

2026-10-26 00:19:17 +00:00

fields.rs

Merge PR #4224: Fields: \* Upper, \* Lower, \* Caps and \* FirstCap change the result, and CardText is lowercase

2026-10-25 04:54:09 +00:00

hit.rs

#3692: review — linear-time tests count lookup steps instead of timing them

2026-10-26 00:19:17 +00:00

kinsoku.rs

Layout: automatic space between Asian text and Latin text or digits (autoSpaceDE, autoSpaceDN) (#2358) (#2397)

2026-10-18 04:54:39 -04:00

lib.rs

#3692: a row's gridAfter is bounded like the DOCX reader's

2026-10-25 21:01:25 -04:00

math.rs

#3692: review — linear-time tests count lookup steps instead of timing them

2026-10-26 00:19:17 +00:00

outline.rs

Integration: track t07

2026-10-22 01:47:46 +00:00

para.rs

#3692: review — linear-time tests count lookup steps instead of timing them

2026-10-26 00:19:17 +00:00

styleref.rs

STYLEREF review follow-ups: headers stay cached, switch labels in our own words (#1286) (#1294)

2026-10-14 17:22:36 -04:00

table.rs

#3692: a row's gridAfter is bounded like the DOCX reader's

2026-10-25 21:01:25 -04:00

tests_autospace.rs

Layout: automatic space between Asian text and Latin text or digits (autoSpaceDE, autoSpaceDN) (#2358) (#2397)

2026-10-18 04:54:39 -04:00

tests_bidi.rs

Integration: track t07

2026-10-22 01:47:46 +00:00

tests_budget.rs

#3692: a row's gridAfter is bounded like the DOCX reader's

2026-10-25 21:01:25 -04:00

tests_complex.rs

Khmer: ignore w14:ligatures when shaping Khmer text (#3921)

2026-10-23 11:18:30 -04:00

tests_grid.rs

Document grid: lines per page and characters per line (#391) (#401)

2026-10-11 15:29:13 -04:00

tests_ligature.rs

egui-037: reconcile main and protect IME undo and wrapped layout

2026-10-25 16:15:05 +01:00

tests_linear.rs

#3692: review — linear-time tests count lookup steps instead of timing them

2026-10-26 00:19:17 +00:00

tests_nbsp.rs

Layout: a no-break space is as wide as a space, not as the font's own U+00A0 glyph (#4210)

2026-10-25 10:23:25 -04:00

tests_numeral.rs

Keep Arabic-Indic digits as typed in Arabic numeral mode (#3329)

2026-10-25 10:44:31 -04:00

tests_pages.rs

Layout: a multiple-spaced line stays on the page while its text fits, as in Word

2026-10-22 15:10:32 -05:00

tests_styleref.rs

Hebrew: field results like STYLEREF draw right to left; smart quotes match Word (#1670) (#2007)

2026-10-18 04:42:54 -04:00

tests_table_contact.rs

#3977: use exact line height for table contact

2026-10-24 10:32:07 +02:00

tests_textbox.rs

M-egui037: fix fontless tests and document migration behavior

2026-10-25 15:24:23 +01:00

tests_vertical.rs

#2038: the same for turned text in a vertical section

2026-10-20 09:28:50 +02:00

tests.rs

#3692: proof_text walks the runs once

2026-10-26 00:19:16 +00:00

textbox.rs

Fit inline text boxes with Resize shape to fit text to their text (#2912)

2026-10-18 11:52:32 -04:00

vertical.rs

Vertical text for sections (w:textDirection) (#491) (#496)

2026-10-11 19:45:49 -04:00

wordart.rs

Shape fills: gradients and transparency (#1500) (#1605)

2026-10-21 19:12:26 -04:00

work.rs

#3692: review — linear-time tests count lookup steps instead of timing them

2026-10-26 00:19:17 +00:00