Hamza
7808816e9a
import: a value a file gives is one the engine can mean ( #1593 )
...
The importers each had their own spelling of "is this a number we can use", or
only checked that it was finite. Measured before the change:
<cSize:1e300> -> Some(1e300) pt tagged text
<cSize:-5> -> Some(-5.0) pt
<cSize:0> -> Some(0.0) pt
w:spacing before=1e300 -> 5e298 pt indent DOCX
DC-12 asks for a shared clamp helper used by every importer, so there are two
in `doc::attrs`, beside the tab limits:
* `imported_type_size` drops a size the engine cannot mean, so the text keeps
the size its style gives it -- which is what RTF and DOCX already did.
* `imported_measure` clamps an indent, spacing or offset to the widest page.
Clamped rather than dropped on purpose: a file that says "indent everything"
still indents as far as the engine goes, where dropping would quietly restore
the style's own indent and hide what the file asked for.
Both are now used by tagged text, DOCX runs and paragraphs, DOCX numbering
levels and RTF -- the last two had the 0.1-1296 pt range as a literal, which is
the duplication that let tagged text drift.
Control: with the helpers back to a finite-only test, both new tests fail, the
DOCX one printing the 5e298 pt spacing in full.
Stale in the issue: its first location, `docx.rs` `w:sz`, is already clamped --
that came with #1574 . DC-12's remaining holes are the ones fixed here.
2026-10-24 16:17:49 +05:00
Dylan M. Taylor and Claude Sonnet 5.5
8299d26400
chore: replace trivial closures with method paths (Sonar S1612) ( #1714 )
...
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-24 04:30:43 -04:00
Brandon Thomas
8ce126407e
Merge pull request #1310 from Silvertree2010/fix/h572-docx-import-stores-out-of-range
...
Drop out-of-range font sizes on DOCX import
2026-10-23 17:30:10 -04:00
Brandon Thomas
a77660b9fa
Merge pull request #1319 from Silvertree2010/fix/h679-docx-xlsx-xml-parser-silently-dr
...
Keep CDATA text in the DOCX/XLSX XML parser
2026-10-23 17:30:07 -04:00
Brandon Thomas
fc2e42738c
Merge pull request #1397 from Silvertree2010/fix/h951-tagged-text-export-strips-from-s
...
Keep ':', '<', '>' and '\' in Tagged Text style names
2026-10-23 17:28:47 -04:00
Brandon Thomas
a627c88e45
Merge pull request #1396 from Silvertree2010/fix/h957-rtf-import-wraps-out-of-range-un
...
Drop out-of-range RTF \uN values instead of wrapping them
2026-10-23 17:27:14 -04:00
Brandon Thomas
e70fdb4f14
Merge pull request #1607 from Silvertree2010/fix/h1298-rtf-import-doubles-characters-an
...
Skip control words and symbols used as RTF \uN fallbacks
2026-10-23 17:27:11 -04:00
Brandon Thomas and Claude Opus 5.5
96a3290021
Tagged Text: decode only the escapes we write; other backslashes stay literal
...
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com >
2026-10-23 17:16:26 -04:00
Brandon Thomas and Claude Opus 5.5
13d17e7a6d
RTF: a group opening ends the \uN fallback skip
...
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com >
2026-10-23 16:58:25 -04:00
NetterSilver and Claude Haiku 5.5
2ddcf6d2a1
Decode UTF-16 plain text files by BOM on import
...
Found while testing crates/textimport/src/lib.rs:117
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-22 20:08:56 +02:00
NetterSilver and Claude Haiku 5.5
bbb95930a2
Skip control words and symbols used as RTF \uN fallbacks
...
Found while testing crates/textimport/src/rtf.rs:197-243 (the \u handler at :235-242; control words never decrement `pending`)
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-22 19:26:49 +02:00
NetterSilver and Claude Haiku 5.5
e1e9923471
Keep ':', '<', '>' and '\' in Tagged Text style names
...
Found while testing crates/textimport/src/tagged.rs:35 (also :52)
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-20 18:02:07 +00:00
NetterSilver and Claude Haiku 5.5
6641e7c29f
Drop out-of-range RTF \uN values instead of wrapping them
...
Found while testing crates/textimport/src/rtf.rs:237
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-20 17:44:49 +00:00
NetterSilver and Claude Haiku 5.5
177ac86741
Keep CDATA text in the DOCX/XLSX XML parser
...
Found while testing crates/textimport/src/docx.rs:134-135
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-19 17:41:15 +00:00
NetterSilver and Claude Haiku 5.5
a5ff17356d
Drop out-of-range font sizes on DOCX import
...
Found while testing crates/textimport/src/docx.rs:183-185
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-19 11:56:51 +00:00
Brandon Thomas
cbcf6185b7
Merge pull request #1177 from Silvertree2010/fix/h2234-xlsx-import-appends-phonetic-gui
...
Skip phonetic guide text when importing XLSX strings
2026-10-19 02:50:23 -04:00
Brandon Thomas
71ab5282d9
Merge pull request #1182 from Silvertree2010/fix/h2256-tagged-text-import-accepts-non-f
...
Ignore non-finite <cSize> values in Tagged Text import
2026-10-19 02:50:19 -04:00
Brandon Thomas
242dd4c08d
Merge pull request #1187 from Silvertree2010/fix/h2354-rtf-import-stores-zero-negative
...
Ignore out-of-range font sizes in RTF import
2026-10-19 02:50:11 -04:00
NetterSilver and Claude Haiku 5.5
2cc21e19b2
Ignore non-finite w:sz, w:spacing and w:ind values in DOCX import
...
Found while testing crates/textimport/src/docx.rs:183 (run_attrs, w:sz) and :206-214 (para_attrs, w:spacing/w:ind via the `tw` closure)
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-17 04:55:36 +02:00
NetterSilver and Claude Haiku 5.5
3823d7d942
Ignore out-of-range font sizes in RTF import
...
Found while testing crates/textimport/src/rtf.rs:219
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-17 01:22:17 +00:00
NetterSilver and Claude Haiku 5.5
61387304b6
Ignore non-finite <cSize> values in Tagged Text import
...
Found while testing crates/textimport/src/tagged.rs:173
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-16 23:26:08 +02:00
NetterSilver and Claude Haiku 5.5
81cc1404b8
Skip phonetic guide text when importing XLSX strings
...
Found while testing crates/textimport/src/xlsx.rs:62 and :112 (shared strings are read with El::text, which includes <rPh> children; inline strings use the same call, crates/textimport/src/docx.rs:37)
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-16 20:22:58 +02:00
aiongg
3f9824906f
Merge PR #1029 : docx: an explicit off imports as off
2026-10-16 04:18:38 +00:00
aiongg
5d2c94cb57
Merge PR #985 : End the RTF \uN fallback skip at the end of its group
2026-10-16 04:18:37 +00:00
aiongg
499f5f27c6
Merge PR #1001 : Importers: XML element nesting and DTD group nesting are bounded ( #712 )
...
crates/engine/src/dtd.rs: keep the branch's version; #319 already bounds DTD group nesting (with the same limit of 64) along with the entity budget and matching fixes, so this PR's DTD part is dropped. The XML element limits are taken.
2026-10-16 04:03:53 +00:00
aiongg
3ea3cd97dd
Merge PR #1054 : Keep text of inline content controls in DOCX import
...
crates/textimport/src/docx.rs: keep the branch's page_break_ends_its_paragraph test (already present further down); add the PR's inline content control test.
2026-10-16 04:02:18 +00:00
aiongg
18ae8e4761
Merge upstream main a184db6f8 with our 15 PRs (batch-0) into next
...
Conflicts resolved per file (both sides kept unless noted):
crates/engine/src/atomic.rs | 1 | main's: `write_file` removed (main routes command writes through `crate::io::write_file`); its one caller in file.rs now uses `crate::io::write_file`
crates/engine/src/cmd/datamerge/fill.rs | 1 | next's `Graphic` literal (has `clip` and `wrap`, which next's `Graphic` carries)
crates/engine/src/cmd/datamerge/mod.rs | 1 | main's `read_path_capped` (crate::io metadata/read_capped, roots checked); next's capped read is the same behaviour
crates/engine/src/cmd/datamerge/parse.rs | 1 | main's (`crate::io::metadata`, `crate::io::Metadata`)
crates/engine/src/cmd/edit.rs | 1 | next's `spread_param(p, "edit.paste")` (the second argument is the command id on both sides)
crates/engine/src/cmd/export.rs | 3 | main's: EPUB params doc, `toc_title` once (main's position; next's duplicate dropped)
crates/engine/src/cmd/file.rs | 7 | place doc: next's (lists page/rect/autoflow params #803 checks); next's recent-file functions + main's `foreign_document`; next's open (load_bytes, warnings, non-DesignCraft refusal) with main's checked `crate::io::read_file` for the bytes; main's `check_write` be
crates/engine/src/cmd/find.rs | 7 | main's Find scope (FindContext, scope/selection/story, continuation, UTF-8 repair) on next's `designcraft_compose::grep::Grep` (dialect GREP): find uses `find_all` capped at MAX_MATCHES, change uses `replacements`/`find_all` (literal change text not expanded), fin
crates/engine/src/cmd/fonts.rs | 1 | next's (`has_style`, resolved/replacement face); `has_style` already compares styles ignoring spacing and case, which is main's `same_style` fix; main's test kept
crates/engine/src/cmd/interchange.rs | 1 | next's `import_idml_report_with_previews` (previews for unreadable links)
crates/engine/src/cmd/layout.rs | 8 | pages.insert: main's overflow-safe `after` + next's checked `parent_ref`; main's duplicateSpread doc; layout.section: next's `SectionEdit` (already keeps unnamed settings, #717 ) with main's `d.repaginate()` moved into `SectionEdit::apply` (remove and set paths),
crates/engine/src/cmd/links.rs | 5 | next's links.list (no doc clone); main's `text_status` via crate::io; `use std::io::Read` dropped (unused after main's read_capped); test literals main's. `link_missing` (next) uses `crate::io::exists` for the file-guard scan
crates/engine/src/cmd/object.rs | 4 | object.fill/stroke tint-alone: main's (23e1680e7; out-of-range tint is an error, which main's test asserts); object.fit doc: next's (lists `axis`)
crates/engine/src/cmd/paths.rs | 1 | next's (shatai transform on outlines)
crates/engine/src/cmd/place_text.rs | 1 | next's `add_flow_frame` (used by autoflow)
crates/engine/src/cmd/preflight.rs | 2 | next's overset/unplacedStory issue (its `overset_location` already applies the parent transform main's #738 fix adds); next's two test modules kept with main's
crates/engine/src/cmd/scripts.rs | 1 | combined: next's `in_undo_step = false` (#406 ) + main's full rollback of a failed script (#709 )
crates/engine/src/cmd/style.rs | 14 | param docs next's (gradient swatch.create, swatch stops, model colours); create_para next's; rename_style main's structure (rename entry, then `redirect_style_refs`), redirect has next's full reference set (cells, defaults, notes, TOC) + main's never-based-on-it
crates/engine/src/cmd/text.rs | 10 | main's selectSentence/Paragraph/Story + select_around; next's selectionAttrs (mixed) + main's openTypeFeatures/type.feature; main's comments; tracked typing: next's `reserve_command_text` + main's `insert_format_at`; empty-paragraph mark formatting main's (with `
crates/engine/src/cmd/thread.rs | 1 | next's `spread_param(p, "text.thread")`
crates/engine/src/dtd.rs | 1 | next's linear, depth-bounded `expand_parameter_entities` with main's budget (input + 1 MiB, `MAX_EXPANDED_BYTES`) and message ("expand the DTD beyond"); 16 MiB constant removed; both sides' tests pass on message
crates/engine/src/guard.rs | 2 | main's whole-transaction snapshot (docs, active, journal; #742 ) + next's `in_undo_step` restore; tests both
crates/engine/src/lib.rs | 5 | LoadedGraphic doc next's; next's `new_document_preview`; execute: main's `discard_stale_find_scope()` before next's nested-undo run; modules both (tests_autosize, tests_file_guard, tests_params, tests_roots)
crates/engine/src/recovery.rs | 1 | next's test (superset of main's assertion)
crates/engine/src/tests.rs | 4 | moved tests deduplicated (identical on both sides); next's preflight and direct-selection/kerning tests + main's style-delete, locked-frame and baseline-fit tests
crates/engine/src/tests_idml.rs | 3 | main's order (drop cap and margins tests identical on both sides, once each)
crates/ui-egui/src/canvas.rs | 3 | next's screen_options/DisplayKey block and tests kept; two tests that auto-merge had added twice (parent_guides_on_a_page_are_not_picked_there, baseline_grid_lines_are_the_ones_on_the_page) kept once; main's `typing`/`screen_at`/`click` helpers and its tests clicks
crates/ui-egui/src/control.rs | 3 | main's `count` check and pending budget (control queue + synthetic queue); exact-budget test runs both sides' cases (click 255 on 9,492 pending; drag 9492×255)
crates/ui-egui/src/cursors.rs | 1 | main's (same Pen cursor set)
crates/ui-egui/src/dialogs.rs | 14 | dialog body (newDocument … alert match) is next's text, with main's Find/Change scope reset (`set_field`, `reset_find_scope`, `find_change_params`) and main's Create Guides / Margins and Columns arms ported in; `Dialog::new` ruby section (next) + main's `set_fi
crates/ui-egui/src/i18n.rs | 7 | ka and pl modules, languages, statics and lookups both; `category_first` covers uk, ru, pl, ka and `count_label` keeps next's Czech plurals; shared TABLE union; next's duplicate rows for Auto, Number:, Gutter:, Top:, Bottom:, Inside:, Outside:, Left:, Right: dropped
crates/ui-egui/src/i18n/ar.rs | 2 | union of rows, no duplicate keys
crates/ui-egui/src/i18n/it.rs | 1 | union of rows, no duplicate keys
crates/ui-egui/src/i18n/ja.rs | 1 | union of rows, no duplicate keys
crates/ui-egui/src/i18n/pt_br.rs | 1 | union of rows, no duplicate keys
crates/ui-egui/src/i18n/uk.rs | 1 | union of rows, no duplicate keys
crates/ui-egui/src/icons.rs | 2 | both tests (main's typography icon, next's distribute icons)
crates/ui-egui/src/lib.rs | 11 | UiState: next's tabs_panel + main's last_color/last_gradient; `cursors` field once; next's SVG paste and paste-on-spread-in-view; next's extensionless document open; opens_as_document next's comment; tests both (next's control bar / key-name tests, main's rich black
crates/ui-egui/src/menus.rs | 15 | app.language doc next's text with pl and ka; UI_COMMANDS both (next's Tabs, Toggle Presentation, Apply None text; main's panel commands, Apply Color, Apply Gradient); Edit › Interface Language stays out (next moved it to Preferences); Type menu main's order with
crates/ui-egui/src/panels/glyphs.rs | 1 | next's (already has main's scroll-bar fix); the scroll-bar test module auto-merge had added twice kept once
crates/ui-egui/src/panels/layers.rs | 3 | next's drag-and-drop `object_rows`, with main's features ported into it: selection reveal (layer opens, row scrolls into view once), selected rows highlighted (also a group whose member is selected), double-click rename in place (`object.rename` queued like
crates/ui-egui/src/panels/mod.rs | 6 | tabs + typography modules and color_panel_tests; SelInfo has wrap_side/wrap_contour and wrap_offsets (offsets from the graphic's own wrap when it is selected); next's swatch_picker returning the chip response; next's content_selection_tests
crates/ui-egui/src/panels/pages.rs | 1 | next's (parents dragged by id)
crates/ui-egui/src/panels/properties.rs | 4 | next's swatch popup (Apply To, Tint); main's four wrap offsets then next's Wrap To / Contour Type; test fixtures main's (`..Default::default()`, no unused builder)
crates/ui-egui/src/panels/swatches.rs | 3 | main's Delete Swatch button + next's New Swatch comment; next's draggable rows; click: main's PICKED + fill/stroke in front, with next's no-apply on double-click
crates/ui-egui/src/toolbar.rs | 2 | main's Apply Color / Gradient / None wells and functions; next's apply_none doc line
crates/compose/src/lib.rs | 5 | next's line loop kept (mark_leading + last_line_mark, wrap segments/carry, grid gyoudori); main's mark-chars base_leading block (auto-merged) stays beside it. `fits`: next's #106 rule (baseline fit for horizontal frames, full descent for paths, vertical text, footnote
crates/compose/src/tests.rs | 5 | next's tests (tombstone set, sliver/one-sided/contour/clipping wrap tests, borders, lists, kenten, grid, anchored wrap, balance, shaped frames) plus main's an_empty_last_paragraph_isnt_overset; main's wrap_object_beside_one_narrow_column_leaves_the_other_alone re-ad
crates/doc/src/assets.rs | 1 | main's doc comment (same meaning)
crates/doc/src/build.rs | 3 | parent_page_for: next's parent_page_for_side (RTL-aware); single_sided_page_shows_first_parent_page test kept once, at main's position
crates/doc/src/item.rs | 1 | next's rotated_group_bounds_follow_the_children_outlines test kept
crates/doc/src/lib.rs | 1 | next's kerning_increment field kept; main's "Bulleted & Numbered" wording
crates/doc/src/page.rs | 1 | next's outer_page_indices + parent_page_for_side (integration of parent-margins with rtl-master-sides); main's parent_page_index dropped, its only caller (build.rs) uses parent_page_for_side
crates/doc/src/slice.rs | 5 | extract: next's `marked()` (get-based, fix_notes/fix_endnotes/fix_marks; main's version indexed slices and could panic); insert_story: main's take() limit on endnote/editorial payload copies with next's id comment, redundant `let ne`/`let nn` dropped; main's excess_note
crates/geom/src/corners.rs | 1 | next's box_outline_rounds_only_closed_corners test kept
crates/render/src/damage.rs | 1 | next's wrapping-anchored-object damage test kept
crates/render/src/fx.rs | 2 | feather: next's choke/corners sigma+join inside main's structure (body and clip at device resolution, with_filters(.., None)); inner glow centre: next's blend `mix` with main's clip argument (bp, xf)
crates/render/src/lib.rs | 3 | next's layer comment; next's front baseline grid draw after items; gradient match on g.kind (main's tuple scrutinee did not match the single-pattern arms)
crates/render/src/text.rs | 5 | paint_glyphs: main's paint_stroke closure structure with next's stroke overprint (overprinting helper, stroke `(TextPaint, bool)`); callers pass next's overprint_fill/overprint_stroke
crates/render/tests/placed_pdf.rs | 1 | main's `near` helper version
crates/tools/src/lib.rs | 1 | main's (PenConvert already declared below, avoids a duplicate variant)
crates/tools/src/pen.rs | 1 | main's module doc
crates/tools/src/select.rs | 1 | next's Drag::Corner variant (live corners) kept
crates/tools/src/tests.rs | 4 | next's tests (content drag, live corners, table border drag, eyedropper matchFrom, fitting handles, pen cursors once); eyedropper_ignores_inactive keeps next's samples_image check (main's exact-JSON compare lacks matchFrom) with main's `..Default::default()` Asset lit
crates/tools/src/text.rs | 2 | next's table border drag (BorderDrag, border_at) kept, then main's comment on locked text frames
crates/tools/src/xform.rs | 2 | main's explicit `Vec<&Item>` for shown children; next's eyedropper matchFrom (integration 796e9d713#4694 backport) and adds main's krilla patch; main's skylartaylor winit left out (see UNSURE 1)
Cargo.lock | 1 | union: next's craft-atomic and craft-cms entries plus main's craft-pdf entry; needs `cargo` normalisation
README.md | 1 | Open bullet combines INDD import via indd-utils (next) and editable PowerPoint export (main); main's paragraph break kept
contributors/people.toml | 1 | both people kept (switch87 from next, Nmarello from main)
apps/designcraft-cli/src/main.rs | 4 | next's `Export` enum dispatch kept; main's .pptx export added as `Export::Pptx` (run prints slides and warnings, script runs file.exportPptx); main's `--pdf-options` null-as-{} parse; usage lines list .pptx
apps/designcraft/src/main.rs | 3 | `startup.leave()` right after `app.ui` (main #380 ) and next's macOS `sync_movable` kept after the IME discard; next's Apple Events install kept; main's macOS Dock Quit `with_close_requested_on_terminate` block left out (needs the skylartaylor winit; UNSURE 1)
crates/fonts/src/fontdb.rs | 9 | FONT6 aliases (next) combined with #911 WWS names (main): ScannedFace and both CatalogEntry constructors carry `wws` and `aliases`; `canonical_family` falls back to WWS family names; `exact_style_traits` canonicalises then tries typographic, then WWS names (`named_st
crates/format/src/lib.rs | 1 | tests: both sides kept (next's r7a nullable-default tests; main's oversized ZIP entry/asset/entry-count tests)
crates/idml/src/import.rs | 11 | main's bounded `Package` reader under next's `import_idml_report_with_previews` (metadata read through `pkg.part`); next's `tab_list_repaired` kept; ruler origin: next's `settings.ruler_origin`, and main's guide import reads it (main's private `RulerOrigin` enum and
crates/idml/src/tests.rs | 6 | item-level union: all next tests plus main's new ones (ruler guides, package limits, CJK leading model, liquid page rules module, mojikumi); no test dropped
crates/mcp/src/backend.rs | 1 | test comment merged from both
crates/pdf/src/tests.rs | 1 | tests: both sides kept
crates/pdf/src/text.rs | 4 | next's (skew `tan`, cluster `start`, ruby overprint warning); main had no change here beyond base
crates/textimport/src/docx.rs | 5 | next's (ListCounter integration for list levels, page-break test position); main's side was the older form of the same code
crates/epub/src/lib.rs | 30 | non-test code is next's (StoryWriter holds img_src/images/no_alt/heading_ids; covers main's #787/#791/#782 fixes, plus #314 group transforms, anchored images in images_without_alt, endnote story skipped in the TOC); tests: next's set plus main's extra css_weight_style a
docs/control-protocol.md | 2 | next's rows (ui.paste, count 1–255, documentGrid) and wording
docs/file-format-parity.md | 1 | rows combined (next's ICC notes, main's TIFF indexed colour)
docs/gaps.md | 2 | status change notes unioned; Flex Layout removed from missing (main implemented it), next's Parent Options sentence kept
docs/localization-parity.md | 4 | Polish and Georgian rows; 5 languages beyond the twelve; file list includes cs and ka; both history rows
docs/roadmap.md | 1 | M1 next's (#186 fixed on next), M2 main's IME wording
docs/target-app-parity.md | 6 | next's status line; File formats row next's, UI/UX row main's (IME no longer failing); UI line next's plus main's Swatches detail; one Page tool line (main's fuller one); Load styles and main's fuller Pathfinder line both kept, main's duplicate short Pathfinder line r
docs/ui-parity.md | 1 | IME row main's, Font menu row next's
i18n/fr.rs: five rows dropped that main's shared table now has (Row, Open Path, Flex Layout, Current Page Number, Rulers).
2026-10-16 03:09:22 +00:00
NetterSilver and Claude Haiku 5.5
108427e216
Keep text of inline content controls in DOCX import
...
Found while testing crates/textimport/src/docx.rs:354
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-15 15:24:11 +00:00
NetterSilver and Claude Haiku 5.5
61d625bd90
Keep '-' after an RTF control word when no digit follows
...
Found while testing crates/textimport/src/rtf.rs:160
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com >
2026-10-15 16:23:55 +02:00
midasmatterhorn
937e948a3d
docx: an explicit off imports as off
...
ST_OnOff (ECMA-376 17.17.4) is a boolean plus the words on and off, but
the toggle reader only counted 0/false/none as off — so <w:b w:val="off"/>
(a legal spelling written by non-Word producers) imported as bold, and
the same for <w:i>, <w:strike>, <w:caps>, <w:isLgl> and <w:u>.
Add "off" to both off-lists. Regression test:
off_valued_toggles_stay_off — on main the off run imports bold; with the
patch it stays regular and an on run stays bold.
2026-10-15 15:54:09 +03:00
wstock and Claude Fable 5.1
02bc1e38aa
Importers: XML element nesting and DTD group nesting are bounded ( #712 )
...
The DOCX and IDML element-tree builders accepted any depth, and the
tree is then walked and dropped recursively; a DTD content model's
groups recursed without a budget. Each parser now refuses nesting past
a fixed budget with an error, before any recursive consumer runs.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com >
Claude-Session: https://claude.ai/code/session_011K1peyrh5MdZ6ttHkVq8Gh
2026-10-15 07:13:51 +00:00
NetterSilver and Claude Haiku 5.5
ec510b9069
End the RTF \uN fallback skip at the end of its group
...
Found while testing crates/textimport/src/rtf.rs:107
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com >
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com >
2026-10-15 04:39:17 +00:00
Brandon Thomas
0923c8247a
Merge PR #797
...
# Conflicts:
# crates/textimport/src/docx.rs
2026-10-14 22:57:54 -04:00
Brandon Thomas
bfd2c59867
Merge PR #687
...
# Conflicts:
# crates/textimport/src/docx.rs
2026-10-14 22:57:24 -04:00
Brandon Thomas
8d254120e7
Merge PR #658
2026-10-14 22:05:11 -04:00
aiongg
4e3a5364e5
Merge docx-tables: fix(textimport): Word table cells skip deleted text and field codes; Normal comes in as its own style ( #721 , #722 )
2026-10-13 14:31:03 +00:00
aiongg
debec75166
fix(textimport): Word table cells skip deleted text and field codes; Normal comes in as its own style ( #721 , #722 )
...
Table cell paragraphs go through the same run and inline path as body paragraphs, so deleted and moved-from text and field instructions stay out of cells, and field results, tabs and breaks come in.
Word's Normal imports as a paragraph style named Normal whose definition is w:docDefaults with Normal's own properties on top; paragraphs without a style use Word's default paragraph style. A styleMap entry can still map Normal to [Basic Paragraph].
2026-10-13 14:23:08 +00:00
aiongg
afd2f45b69
Merge pr/673: fix(import): preserve XLSX numeric and cached text values ( #673 )
2026-10-13 12:31:43 +00:00
aiongg
5f2110e7e5
Merge pr/687: Bump quick-xml to 0.41 ( #687 )
2026-10-13 12:13:14 +00:00
natocool2 and Claude Opus 5.5
5a5e97b0a6
Bump quick-xml to 0.41 (RUSTSEC-2026-0194, RUSTSEC-2026-0195)
...
The .docx/.xlsx text import iterates attributes with quick-xml's default duplicate check, so one crafted
start tag with N attributes cost O(N^2) (RUSTSEC-2026-0194, fixed in quick-xml 0.41). Bumping also clears
RUSTSEC-2026-0195 (NsReader namespace allocation), which this workspace does not reach (no NsReader).
unescape_value() is deprecated in 0.41 (clippy -D warnings); normalized_value(Implicit1_0) is what the
deprecated method now does. Regression test: an element with 100,000 attributes parses in < 5 s.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com >
2026-10-13 09:07:47 +01:00
storybored379 and Claude Opus 5.5
5e22a3dbee
fix(import): show XLSX numbers as Excel's General format does
...
Round cached doubles to 15 significant digits, so =0.1+0.2 reads 0.3,
and write very large or small values with an exponent (1E+300).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com >
2026-10-12 19:50:14 -07:00
storybored379 and OpenAI Codex
d27eff31cd
fix(import): preserve XLSX numeric and cached text values
...
Keep small and high-precision numbers intact in table placement and data merge. Honor cached formula strings as text, including leading zeros and whitespace.
Co-Authored-By: OpenAI Codex <noreply@openai.com >
2026-10-12 18:29:37 -07:00
aiongg
e7ad004f4d
Merge main: mirror gradient midpoints when reversing; preserve leading empty Tagged Text paragraphs
2026-10-12 23:56:21 +00:00
Jason Brashear
91263c912f
fix(textimport): preserve leading empty Tagged Text paragraphs
...
Preserve paragraph breaks after a ParaStyle tag starts the story so leading empty paragraphs and their styles survive Tagged Text import.
Validated individually and after #638 with all six cargo xtask ci gates (1,278 tests passed, four existing tests ignored, 15 WASM compilation checks), six independent paragraph/header/Unicode probes, and an actual export-to-placement command smoke test.
Refs #641
2026-10-12 17:21:52 -06:00
aiongg
6483a43ba8
Integration (docx-lists x #360 list levels): Word levels set list_level (ilvl + 1) and lvlRestart; parent numbers in a label are ^k level references, written as text only where the model's level numbers differ from Word's (legal labels, unnumbered levels); Start At from a simulated per-level ListCounter; tests nested_levels_and_restarts adjusted, parent_numbers_the_model_lacks_are_text
2026-10-12 22:07:29 +00:00
aiongg
8ce45eaef6
Merge docx-lists: Word import: bullets and numbering from numbering.xml ( #626 )
...
Conflict: crates/textimport/src/docx.rs: tests both
Conflict: ROADMAP.md: logs both
2026-10-12 22:04:39 +00:00
aiongg
7515dfeaf5
Word import: bullets and numbering from numbering.xml ( #626 )
...
Paragraphs with w:numPr, directly or through their paragraph style, become
bullet or numbered paragraphs: Word's number formats, label text, start
numbers and overrides, level restarts, the text after the label and the
level's indents. Parent levels in a label are written as text. Unreadable or
hostile list definitions leave plain paragraphs.
2026-10-12 20:59:17 +00:00
aiongg
25aa2735b8
Merge copy-notes: Text: footnotes insert into a story whose references lack footnotes
...
Conflict: crates/doc/src/slice.rs: tests module keeps next's append_story overflow test and the branch's copy/paste and hostile-marker tests
2026-10-11 13:04:05 +00:00
aiongg
fe8a724da4
Text: footnotes insert into a story whose references lack footnotes
2026-10-11 12:54:42 +00:00
8146737e97
Text: a column, frame or page break ends its paragraph ( #285 )
...
A break character is stored as the last character of the paragraph it
ends, followed by the usual '\n' separator. The story model, paragraph
ranges, styles, editing and every exporter keep splitting paragraphs on
'\n' only, so the change stays in the places that create, lay out or
write break characters:
- IDML import writes the break and ends the paragraph, so the range that
follows keeps its own paragraph style. NextOddPage and NextEvenPage map
to new odd and even page break characters; IDML export writes a
paragraph that ends in a break as that one Br with its
ParagraphBreakType, so export and re-import give the same story.
- Word and RTF import, Type > Insert Break Character (with odd and even
page breaks added) and typing end the paragraph after a break. Typing
between a break and its return goes to the next paragraph; deleting
either deletes both.
- Layout: a break that ends a paragraph leaves no empty line; the next
paragraph starts in the next column, frame or page, as a paragraph start
option would. Page breaks and Next Page start options move to the next
frame on a later page; odd and even ones to the next page with that
page number parity (no such frame: overset). A break inside a paragraph
from an older document still moves the rest of that paragraph on.
Storing the break as a paragraph separator of its own would have changed
every place that counts or splits paragraphs on '\n'.
Co-authored-by: Brandon Thomas <echelon@gmail.com >
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com >
2026-10-10 03:36:47 -04:00
Hamza