Mike-Solar
061aab9180
- full-res source jobs carried only the footage node box: dropping the project mid-flight left the node dangling (crash in the worker's free path) and the node was also freed twice (at renderer creation AND at release). The request now carries an addref'd project copy and the node is freed exactly once; regression test drops the project before the worker runs - the source clock ticked at the SEQUENCE length, so playing footage with an empty sequence froze the source playhead at 0; the source clock now loops at the selected footage's probed duration - oak-cli/oak-worker: DeepSeek's refactor batch (clap migration, engine FFI consumers); the stale exporter-family test flipped to the real contract (mp4 is written) - engine: render_audio smoke test on an empty sequence
oak-cli (Rust)
Headless command-line consumer of the liboakengine C ABI facade — the Rust
rewrite of cli/main.cpp (which stays in the tree until cutover). Same
subcommands, same output format, same exit codes:
| exit | meaning |
|---|---|
| 0 | success |
| 1 | general error (bad project/media file, no sequence, I/O failure) |
| 2 | rendering unavailable or failed (e.g. no GL render backend) |
| 64 | usage error |
Build and test
cargo build --release # binary: target/release/oak-cli
cargo test # unit + integration tests (29 tests)
The crate builds standalone: its only dependency besides clap is the
oakengine rlib (../oakengine), which has no third-party
dependencies.
Subcommands
Every subcommand of the C++ original is implemented:
oak-cli info <project.ove> <start> <end> <out_dir> project name/sequences/footage
oak-cli render <project.ove> <start_seconds> <end_seconds> <out_dir>
oak-cli probe <mediafile>
oak-cli transcode <input_media> <out> [width] [--format ppm|mp4]
Argument validation is faithful to the C++ (invalid start seconds,
invalid width, unknown --format … all exit 64). The output formatters
(src/fmt.rs) reproduce the C++ printf output byte for byte and are
golden-tested against the output captured from the C++ binary on the test
fixtures (tests/project_with_footage.ove, tests/demo.mp4); the PPM and
WAV writers (src/ppm.rs, src/wav.rs) are the exact ports of the C++
write_ppm/write_wav and are unit-tested.
Facade status: everything is currently deferred
All four subcommands depend on facade families that are still deferred
in the oakengine crate (crates/oakengine/src/deferred.rs), so today each
subcommand validates its arguments, then prints a clear "not yet available"
error naming the missing families and the reasons, and exits with the
C++-compatible code — it never crashes and never fakes output:
| subcommand | needs | current behavior |
|---|---|---|
info |
init + node (project/footage) + timeline | "not yet available", exit 1 |
probe |
init + node (footage) | "not yet available", exit 1 |
render |
init + node + timeline + render | "not yet available", exit 2 |
transcode |
init + node + timeline + render + exporter | "not yet available", exit 2 |
The deferral registry is src/deferred.rs (field-for-field in sync with the
facade's own deferred.rs). When a family is wrapped by the facade:
- remove its entry from
src/deferred.rs, - wire the call-through in
src/cmd/using the extern declarations insrc/ffi.rs(verbatim mirrors of the engine headers) and the tested formatters/writers — no manifest or signature change is needed, because the externs resolve against the already-linkedoakfacaderlib.
Layout
src/
main.rs clap surface, --help/-h + unknown-command handling, dispatch
ffi.rs the oakengine_* surface oak-cli consumes (declarations only)
deferred.rs facade-family availability registry (mirror of facade deferred.rs)
fmt.rs golden output formatters (info/probe)
ppm.rs P6 PPM writer (f32/u8 frames)
wav.rs PCM s16 WAV writer (interleaved float samples)
cmd/ per-subcommand validation + deferred gate
tests/cli.rs binary-level tests (exit codes, messages, usage errors)