The parent-page regression covers a placement a page inherits. The other placement the export
walks reach is an object anchored in a story: the IDML copy gets a variant that shows every
layer, an external-only copy is refused, and PDF/A-2b rasterises it and warns. The nested-story
regression that used to cover this case also asserted placements anchored inside a table cell and
a footnote, which this pass does not reach; this one keeps the anchored case alone.
Both exports walk Document::placed_items: document and parent spreads, groups, and the objects
anchored in a story. That is the walk the parent-page and anchored fix introduced. The
nested-story traversal (table cells, notes, endnotes), Document::update_placed_items and the
fail-closed refusals that went with it are not part of it: that is the traversal change on its
own, not the placements this pass carries. update_items keeps its reach, so the two exports clear
the choice where the placement actually is again.
A choice anchored inside a table cell's or a note's text therefore stays on the document and is
not carried into the embedded IDML copy, as on main; the refusal that turned that silent
difference into an export error goes with the traversal it guarded, since asking the walk's own
predicate it could never fire. The stale-field half keeps: has_applicable_pdf_layer_overrides
still asks for a graphic whose asset is a PDF, so relinking a placed PDF to an image no longer
makes the package report name a choice that does not exist.
Tests: the nested-placement regression and the relink integration test of the removed refusal
go; the parent-page regression, the stale-field predicate test and the format 4 cell-only reader
test stay. The comments no longer claim every placement counts.
Graphic gained an optional layer_options field, so the test-only literals in the PDF
crate and in the engine's optional-content support need it. No behaviour changes.
The workspace file guard scans test modules too when they are declared `pub(crate) mod` (it only
skips a plain `mod`), so a temp file written inside pdflayers.rs' tests failed
no_crate_touches_the_file_system_outside_the_checked_readers. The unit test now builds the state a
relink leaves behind - layer fields on an item whose asset is an image - and asserts the refusal
predicate ignores it while all three exports run. The user path (a real links.relink to a PNG) moves
to tests/pdf_optional_content.rs, where the guard exempts fixtures written by test files.
Pin the final egui, winit and shared UI revisions and reconcile current main.
Preserve document-epoch input and screenshot boundaries, host action ordering,
and Windows path regression coverage. Add click/End batch characterization.
Validation: all nine Linux CI stages pass (3230 tests, zero failures); Windows
CI and bounded native editing workflows pass on the same production source.
a_missing_link_shown_by_its_preview_is_still_reported builds its link
under std::env::temp_dir(). On macOS and Linux that returns TMPDIR as
set, so a relative TMPDIR gave a relative link, which document_path
refuses in a document without a path, and the test failed with
"(the path is not valid)". The test now passes the temporary folder
through std::path::absolute first. The preflight code is unchanged.
Comment-only: the refusal in both exports cannot fire while the walk and the predicate cover the
same placements, so the comments now say it is a tripwire for a future nesting construct and ask
nobody to narrow the walk to match it.
The refusal guard asked whether any item still had the layer fields, while the export walks
also require a graphic whose asset is a PDF. Relinking a placed PDF to an image keeps those
fields on the item (links.relink replaces the asset in place), so a stale field blocked IDML
and PDF/A-2b exports that used to work. The guard, and the package report, now ask the same
question the walks ask: has_applicable_pdf_layer_overrides requires the PDF asset too, so a
stale choice is simply ignored and the tripwire still fires if a future construct ever puts a
real choice where the walk cannot reach it.
Regression: a_stale_choice_on_a_relinked_image_does_not_block_exports places a layered PDF,
hides a layer, relinks that asset to a PNG and exports IDML, PDF/A-2b and plain PDF - all
succeed, and the document reports the stale field while the applicable helper does not.
* EPUB: Split Document in a paragraph style's Export Tagging
* EPUB: Split Document as a document setting (off, one style, or export tags)
The document stores EPUB Export Options › Split Document (DocSettings.epub_split):
off (one content file, the default), a single paragraph style, or the
paragraph styles' Export Tagging › Split Document flags. document.epubSplit sets
it as one undo step; file.exportEpub takes a split that overrides it for one
export. A Style setting follows a renamed style and a deleted style's
replacement. IDML reads and writes EPubExportPreference BreakDocument and
ParagraphStyleName. A split before a list item or preformatted line closes the
open list or block first, so each content file is balanced.
* EPUB Export Options dialog with Split Document before Export EPUB
File › Export EPUB… opens EPUB Export Options, seeded from the document's
Split Document setting: a Split Document checkbox with Single Paragraph Style
(and its style) or Based on Paragraph Style Export Tags. OK stores a changed
setting through document.epubSplit, then continues to the save dialog; Cancel
does nothing. app.exportEpub with a path skips the dialog. Docs and the
progress log describe the setting.
* EPUB: the navigation test exports one content file with Split Document off
* EPUB export from the menu passes its options (cover, split) to the engine
* EPUB: Split Document on a style's EPUB tag map counts when the style says false; a caller's cover: false stays
Kept separate from the parent-page fix and from the all-on export fix: this is the traversal
remediation on its own.
The detection helpers already walked parent spreads, groups and the stories in table cells,
notes and endnotes, while the two exports walked document spreads only, and the write side
(Document::update_items) did not reach placements nested in those stories either. So a choice
on a frame anchored in a cell's text or in a footnote asked for the newer file format and was
named by the package report, yet the IDML copy carried the file's defaults. The walk was
narrow on main as well (see the probe evidence), so this is pre-existing for the older
hidden-list state; it becomes a wrong claim only because the paired PR added the variants and
said every placement counts.
Document::placed_items recurses those stories (depth budget 64, like the detection side) and
a new Document::update_placed_items applies the change where the placement actually is,
copying only the branches holding a match. update_items is untouched, so unrelated commands
keep their reach. Both exports also refuse rather than fall back when a document has a choice
but the walk found no placement to carry it.
Regression: nested_placements_keep_their_layer_choices_in_saves_and_exports runs a placement
anchored in a table cell's text, then in a footnote: the save keeps the choice, the embedded
IDML copy shows every layer, external-only IDML is rejected, PDF/A-2b export succeeds.
A package's IDML copy links the placed files, so a per-placement Object Layer Options
choice cannot travel in it, while the packaged DesignCraft document keeps the choice. The
report now says so instead of leaving the two copies to differ silently.
The graphic-layer test covered a mixed document. A save whose only layer settings sit in a
table cell has to choose version 4 too, asserted against the saved metadata.
The IDML layer-variant pass and the PDF/A-2b rasterisation pass both walked only
document-spread items, so a choice on a parent-page placement (or an anchored one) was
replaced by the file's defaults on export, and an external-only IDML export succeeded
when it could not carry the choice. Both now cover every placement through
Document::placed_items, and the choice is cleared where the placement actually is.
Regression: a parent-page placement with an explicit all-on choice keeps all layers in
embedded IDML, external-only IDML rejects it, and PDF/A-2b rasterises it.
Graphic gained an optional layer_options field, so the test-only literals in the PDF
crate and in the engine's optional-content support need it. No behaviour changes.
Preserve contributor commits and adapt optional-content registration and table-cell graphic calls to the current PDF exporter refactors. Keep current-main roadmap corrections.
All nine cargo xtask ci gates passed on tree d0c21f347e: 3209 tests passed, 6 skipped, 16 WASM-safe crates. The exact PR head also passed 118 PDF unit/integration tests. Independent production-export controls: 8 flat/null cases succeeded; 7 unreadable Prev values refused, 4 compatibility controls succeeded; all 12 emitted PDFs pass qpdf.
This PR is independent of the format-v4 pair (#1647/#1671), which remains held for a reproduced parent-placement IDML issue.
note.delete removes the note's anchor (3 bytes) from the story, and
note.convertToText replaces it with the note's text. Both left a caret or
selection in the story text at its old byte offsets. With "copy" selected
after a note in "Draft copy.", Delete Note left the selection at 9..13 in
an 11-byte story, and Convert Note to Text with the note's text "ab" left
"opy." selected.
Both commands now move a caret or selection in the story's own text by
the change in length when it comes after the anchor. A caret before the
anchor stays where it is, and a caret in a table cell or in another story
is unchanged.
Editing one keyword in File Info re-split the whole field, so a stored
keyword holding a comma or semicolon became two. A stored keyword that is
still in the field whole now stays one keyword.
An Export PDF title equal to the document's name was dropped, so the PDF
took File Info's title. The dialog's title now starts empty, and a title
given there is always the PDF's title; without one the file is named
after the document.
file.info refuses parameters that are not an object; null reads the
fields like {}.
File Info's OK stores the fields in the document the dialog was opened
for, even when another document became active meanwhile, and stores
nothing once that document has closed.
The new command note.split {story, id, at} splits an editorial note at
byte `at` of its text. The note keeps the text before that point, and
the rest moves into a new note by the same author whose anchor directly
follows the first one. A split point at the start or end of the text,
or inside a character, is refused. The Notes panel shows each note's
text in a read-only field, and its Split Note button splits the note at
the insertion point in that field.
footnote.delete and endnote.delete removed the reference character from
the story but left a caret or selection in the story text at its old
byte offsets. Everything after the reference moved back by the
reference's length, so the caret ended up three bytes further along the
text. A caret or selection after the reference now moves back with the
text, and one before it stays where it is.
footnote.setText left a caret in the replaced footnote at its old
offset, which could fall past the end of the new text or inside a
character. Backspace there then failed with an internal error. The
caret now goes to the end of the new text.
Four tests from #1678 fail on Windows. Each one builds its paths, or the
text it expects, from a folder resolved with `std::fs::canonicalize`,
which adds the `\\?\` prefix on Windows. `AutomationRoots::describe` and
the rejection messages show the roots without that prefix, so the
expected text does not match. Windows does not resolve `..` in a path
with the prefix, so the checks refuse every such path as one they cannot
resolve, including the ones the tests expect to pass from one root into
another. The code the tests check is unchanged.
The engine tests take the resolved folder through `file_access::plain`,
which the roots use for the text they show. The CLI test strips the
prefix the same way.
Two designcraft-engine tests fail on Windows on main. The code they
test is unchanged.
eps_relink_update_and_folder_relink_retain_source built the expected
link as dir.join("new/replacement.eps"), which on Windows ends in
new/replacement.eps, while links.relinkFolder stores the folder
joined to the file name, ending in new\replacement.eps. The test now
joins the folder and the file name separately.
only_local_graphic_files_are_read_from_links allowed a read of the
link only at the link's own text. For the link /dev/zero the import
checks C:\dev\zero on Windows, because craft-docpath joins a
root-relative path to the document's drive. The test now allows the
path that crate::io::document_path resolves the link to.
Open #1444 by AndreRese makes the same change to the relink test and
a similar one to the IDML test.
* Type: story Optical Margin Alignment with a size, and Ignore Optical Margin (engine, IDML)
A story's Optical Margin Alignment applies to every paragraph except those
with Ignore Optical Margin; hang amounts scale with the story's size.
story.options sets it, type.para and paragraph styles take
ignoreOpticalMargin. IDML imports and exports StoryPreference
OpticalMarginAlignment/OpticalMarginSize and IgnoreEdgeAlignment. The
per-paragraph opticalMargin switch keeps hanging at the text's own size.
Part of #947.
* Type: the Story panel, and Ignore Optical Margin in the Paragraph panel menu and Paragraph Style Options
Window > Type & Tables > Story sets Optical Margin Alignment and its size
for the stories of the text selection or the selected text frames through
story.options, and is disabled without either. type.selectionAttrs
reports the stories' values as storyOptions, null where they differ.
The Paragraph panel menu toggles ignoreOpticalMargin, and Paragraph Style
Options > Indents and Spacing has an Ignore Optical Margin checkbox.
Part of #947.
* ROADMAP: the story Optical Margin Alignment entry names its default and scope
Go to Footnote Reference, Insert Footnote and Delete Footnote (when the
caret is in the deleted footnote) moved the caret without clearing the
pending typing format. Formatting chosen at the old insertion point and
not yet typed then came back when the caret returned there, for example
in a new footnote that takes a deleted footnote's id. The three
commands now clear it, as text.select and text.move do.
The new commands note.previous and note.next move the insertion point
to the previous or next editorial note in its story and put the caret
after the note's anchor. Previous Note skips the note whose anchor the
caret directly follows, so repeated use reaches every note. Each item
is unavailable when no note lies in its direction in the story, or when
the insertion point is in a table cell or a footnote. The items follow
the Notes panel item in Type › Notes, with translations in the catalogs
that already translate Next Document and Previous Document.
skylartaylor's review: a PDF can list readable layers yet fail the visibility
rewrite. On a synthetic 33-section xref history, object.layerOptions with
useDefaults:false and an empty hidden list was accepted (only hidden names
triggered #1670's rewrite check), the canvas could not draw the choice and
PDF export then failed.
Every explicit choice now runs one actual pdf_hide_layers rewrite per
distinct source before any item is edited or undo is recorded; this subsumes
the separate layerless check. Resetting to file defaults (useDefaults:true
or the legacy empty list) needs no variant and bypasses the check. #1670's
fail-closed canvas handling is kept alongside the explicit-override state: a
stored explicit choice (including all-on) whose source can no longer be
rewritten draws nothing rather than the unmodified source, and export errors.
The new regression test fails before this change (the all-on request is
accepted) and covers both selection orders, atomicity, undo depth, reset
bypass and the stale all-on fail-closed path.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
{"offset": [null, null, null, null]} parsed as an offset edit, so the custom
wrap boundary (from IDML) was dropped although no side changed. An array with
no number in it is no offset edit. Found by CodeRabbit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
skylar taylor-barrick
dotCooCoo
aiongg
Brandon Thomas
Andee CollardandClaude Opus 5.5
moppeeerandClaude Opus 5.5