decode_frame_into refuses a frame header larger than the movie picture
limit before sizing any plane, and Movie refuses a ProRes frame whose
header size differs from the track's declared size before decoding it.
Audio source and movie paths stored in a .scraft session go through
io::stored_path (craft-docpath) before they reach the file system: a
relative path is taken from the session's folder, and paths on other
computers, devices and URLs are refused. A refused source is reported
as missing with the reason; a refused movie is marked offline and the
Video track and Video window show why instead of opening it. Import
Session Data and crash recovery resolve the same way.
A track group's name, membership and Edit/Mix flags could only be set
when it was created: nothing in the engine or the UI could change them
afterwards, so a typo in a group name was permanent.
Add track.group_modify {group: id|name, name?, tracks?, edit?, mix?}.
Omitted fields keep their value. `tracks` replaces membership only when
it is an array, so a mistyped param cannot silently swap the members for
whatever happens to be selected; it dedupes, rejects an empty group, and
rejects an empty or duplicate name, since find_group resolves by name.
The group's letter is left alone: it is the stable id in the Groups list.
Dialog::Group gains an optional group id. None creates, as before; Some
pre-fills from the group and dispatches track.group_modify, so one dialog
serves both. Reached from Modify Group... in the Groups list context menu.
OK is now disabled while no track is checked, since both commands reject
an empty group and the old OK could only raise an error.
The command has no menu path, so it adds no menu leaf and the parity
checklist is unchanged.
`cargo xtask ci` failed on Windows before any test ran: the `windows_ignores_case_and_refuses_devices` test moved `upper` into `check` and borrowed it afterwards for the assertion message (E0382), and the recovery test's `paths` is only mutated inside its `#[cfg(unix)]` symlink block, so `-D warnings` rejected the `mut` everywhere else. Neither test is compiled on macOS or Linux, which is why CI never saw them.
With this, `cargo xtask ci` passes on Windows 11 (MSVC, Rust 1.99).
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Picking a session with Open Session or dropping one on the page in the
browser replaced the document as soon as its bytes were read, without the
Save Changes question the native Open goes through, so unsaved edits and
their undo history were lost. The bytes now wait behind that question,
asked when the read finishes: Don't Save opens them, Save saves first and
then opens them, Cancel drops them and leaves the document alone.
Found while testing crates/engine/src/cmd/track_more.rs:370-392 (write_rtp); playback side: crates/mix/src/lib.rs:1526-1527
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/event.rs:338 (edit_marker; params doc at line 97)
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/event_more.rs:416-445
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/event_more.rs:753
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Found while testing crates/engine/src/cmd/midi.rs:129-130
Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The FreeBSD CI VM has no fonts installed, so the "bold" family falls back
to egui's built-in regular face. SENDS A-E and SENDS F-J then measure 47
and 45 pt, inside the 48 pt the measured-fit check allows, so narrow
strips showed them in full next to I A-E and INST, and
mix_narrow_headers failed on FreeBSD.
Narrow strips now always use Pro Tools' short names, whatever the font;
normal strips still shorten a name only when it doesn't fit. A new test
renders the narrow Mix with egui's built-in fonts only, reproducing the
FreeBSD failure on any platform.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Picking a plugin for an insert slot added it but opened nothing, so
every new plugin took an extra click on its slot. As in other DAWs, the
pick now opens it: a hosted plugin's own editor, else its plugin
window (built-in plugins, no audio engine, or an editor that can't open
here, such as CLAP plugins with only an embedded editor on macOS).
The insert menu's picks go through pick_insert, which remembers the
slot when mix.insert succeeds; SoundApp::open_inserted opens it on the
next logic pass, once the audio engine has the instance. Only the menu
opens it, so mix.insert from the CLI, MCP or control channel still
opens nothing. It covers empty and filled slots, in the Mix window and
the Edit window's Inserts column.
Tests: tests/open_on_insert.rs (empty slot, search and Enter,
replacing, Edit window column). menu_search closes the windows the
picks now open before it looks for the menu's Bypass.
Reject oversized batches before any side effects instead of reporting success after truncating calls.
Closes#460
Co-Authored-By: OpenAI Codex <noreply@openai.com>
The Linux font search list in `crates/ui-egui/src/fonts.rs` covered
Debian/Ubuntu (`/usr/share/fonts/truetype/dejavu`, `/usr/share/fonts/TTF`,
`/usr/share/fonts/dejavu`) and Arch (`/usr/local/share/fonts/dejavu`)
for DejaVu, and a single path for Liberation Sans. It did not cover
Fedora's package layout:
- `/usr/share/fonts/dejavu-sans-fonts/DejaVuSans.ttf` (and -Bold)
- `/usr/share/fonts/liberation-sans-fonts/LiberationSans-Regular.ttf`
On Fedora, neither path matched, the `load_from_fontconfig` path returned
None (e.g. the user only has Liberation Sans installed, not DejaVu), and
the UI fell back to egui's built-in fonts. Those don't cover the menu
glyphs (`→ ▸ ▾ ♯ ✕`), so the symbols showed as empty boxes.
Add the two Fedora paths to the REGULAR and BOLD candidates, in the same
order as the existing Debian and Arch paths. `load` still tries
fontconfig first; the direct paths are the fallback.
cargo test -p soundcraft-ui-egui: 105 passed, 0 failed.
Pitch, release velocity, CC number and poly-pressure pitch were masked with
0x7F, so 128 became 0 and 255 became 127. They now clamp to 127, as the
write_smf doc promises and note-on velocity already did.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The submenu under Track › New... opens the same picker as the Mix window's
instrument slot for the first selected instrument track, and is greyed out
with a hint when none is selected.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Brandon Thomas