Adds PowerPoint's Outline/RTF save type: slide titles at outline level 0
(bold, larger), body paragraphs indented at their outline level, bold,
italic and underline kept, non-ASCII as \u escapes. RTF outlines (ours,
PowerPoint's or a word processor's, by outline level or indent) open as
a presentation or drop in as slides.
Co-authored-by: Cursor <cursoragent@cursor.com>
A line that is not a JSON object with a string method, including an HTTP
request line or bytes that are not UTF-8, gets one error reply and the
connection closes before anything after it is read.
The renderer lays every slide background over white (as PowerPoint does),
so a PNG export was always opaque. A new transparentBackground option leaves
the slide background out (both the white underlay and the background fill),
so only what is on the slide is drawn over transparency. Shapes filled with
the background (useBgFill) still show it.
- render: RenderOpts::transparent_background
- engine: file.export {transparentBackground} (PNG only; other formats
refuse it and write nothing), file.render {transparentBackground}
- UI: Transparent background checkbox in the Export dialog (PNG) and in
File > Export > Change File Type > PNG; the web download honours it
- CLI: render --transparent
Paragraph commands (align, case, insert, paste, selectParagraph) iterated
for pi in a.0..=b.0 with both ends taken verbatim from text.select, so a
control-channel selection with caret u64::MAX spun the loop over 2^64
entries on the UI thread, and text.selectParagraph's at.0 + 1 overflow-
panicked in debug builds. Clamp every range at the body's last paragraph
and use saturating_add for the single-index arithmetic.
Regression test: paragraph_commands_clamp_a_hostile_text_selection —
fails on main with 'attempt to add with overflow' at cmd/text.rs:1151,
passes with the clamp.
Found while testing crates/engine/src/cmd/table.rs:291
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
review.reject {index | all} drops revisions from the list without applying
them, so the open deck stays as it is: a rejected inserted slide is never
added, a rejected deleted slide stays, a rejected changed slide keeps its
current content. Accept All skips what was rejected. Rejections are kept on
the compare session (keyed by the open slide id and the revised slide) and
end with End Review; like Compare and End Review, rejecting is not an undo
step, since it leaves the document untouched.
Revisions pane: Reject next to each Accept, and Reject All.
Keep later On Click effects and paragraph builds in separate interactive steps.
Add round-trip and headless show regressions.
Contribution on behalf of @MMexicann.
Found while testing crates/engine/src/cmd/shape.rs:1154 (shape_id, used by connect/sites); same pattern at crates/engine/src/cmd/points.rs:358-359 and :432-433
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Found while testing crates/engine/src/cmd/slide.rs:217-220
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
slides_to_outline only looked at sh.text, which regular text boxes and
placeholders carry. Tables store their text in cells inside
ShapeKind::Table, so convert to .txt or .md silently skipped every
table cell — with exit 0 and no warning.
The outline now walks into tables: each row becomes a tab-separated
line (multi-paragraph cell text is flattened with spaces). Verified
with the issue's python-pptx repro: the four cell texts appear after
the slide title and text box body.
Tests: a deck with a text box and a 2x2 table — the outline contains
all three text sources.
Use the pinned shared PDF writer with explicit font embedding errors, preserving app-owned rendering and current main. All seven local gates and fresh Ubuntu/FreeBSD checks passed on the exact integration tree; paired native/WASM PDF bytes, structure, text and pixels match. Original contributor commits are preserved.
Found while testing crates/engine/src/cmd/richclip.rs:187
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
In release 0.7.0, Cancel in the Preferences dialog (opened via Options
in the File tab) did nothing. This was previously addressed on main by
#757 (#752).
This change adds an egui_kittest regression test verifying that clicking
Options from the File tab opens the Preferences dialog and clicking
Cancel closes it while keeping Backstage open. In addition, the About
dialog's Cancel button now properly closes the dialog instead of being
ignored.
Closes#782.
An empty line inside a paragraph -- between two `a:br`, or after a trailing one
-- was laid out at the size of the paragraph's *first run*. PowerPoint sizes it
by the mark that closes the line: the `a:rPr` of the `a:br` that ends it, or
`a:endParaRPr` for the paragraph's last line. Every line below then sits too
high or too low, by the difference; in the reported poster deck the text drifted
by up to 173pt.
`shape_para` resolved one `lead_size`/`lead_face` per paragraph and
`layout_scaled` used it for every line with no measured cells. A break is a
cell (`\u{b}`) and `break_lines` sets a line's `end` to that break's own index,
so the closing mark is already to hand: take its style, falling back to the
paragraph mark (now carried as `end_size`/`end_face`) for a last line that no
break closes. Any other empty line -- a line of tabs alone -- keeps the old
measure, so only the two cases PowerPoint defines change.
An empty *paragraph* already landed on `end_props` through `lead`, and is
unchanged: it is the control case in the report, and the test for it passes both
before and after.
Measured on the reported fixture, second line top against PowerPoint's own
export: case A 166pt -> 130pt (131 expected), case B 166 -> 130, case C 113 ->
161 (160 expected), and the control case D 130 either way. The 1pt that remains
is the offset the control already had.
Use the pinned shared PDF writer with explicit font embedding errors, preserving app-owned rendering and current main. All seven local gates and fresh Ubuntu/FreeBSD checks passed on the exact integration tree; paired native/WASM PDF bytes, structure, text and pixels match. Original contributor commits are preserved.
