Commit Graph
3 Commits
Author SHA1 Message Date
dailz c772e4eb0b chore: bump MSRV to 1.87 to match actual API usage
CI / Build + Clippy + Test (push) Failing after 1h10m8s
CI / Security audit (RUSTSEC) (push) Failing after 1m31s
Oracle P2 follow-up. The clippy --fix autofixes earlier in this branch
silently introduced dependencies on APIs newer than the README's
1.70+ claim:
  - u32::is_multiple_of  (stable 1.87)
  - Option::is_none_or   (stable 1.82)

clippy::incompatible_msrv flagged the mismatch once rust-version was
pinned. Bumping the floor to 1.87 is the honest fix — the codebase
genuinely depends on 1.87 features now, and 1.87 has been stable
long enough (current stable is 1.96) that desktop CLI users on stable
Rust already have it.

  - Cargo.toml: rust-version '1.70' -> '1.87'. Comment lists the specific
    APIs that drove the bump and notes that further bumps need to be
    validated against clippy::incompatible_msrv.
  - README.md: Prerequisites line updated to 1.87+ with a brief why.
  - src/state_portal.rs: added the AsRawFd rustc-quirk comment that was
    already in avhw.rs (rustc emits a false 'unused_imports' warning;
    removing it produces E0599). Same known quirk, same documentation
    pattern.
  - src/transform.rs: fixed empty_line_after_doc_comments warning by
    converting the leading // doc-style comment to a //! module-level
    doc comment (which is what it should have been when I rewrote the
    file in commit 145b5d3).

All 79 unit tests + 3 integration tests pass. clippy: 0 errors,
0 incompatible_msrv warnings, 0 empty_line_after_doc_comments warnings.
Remaining warnings are: 1 AsRawFd rustc false-positive (documented),
5 unnecessary_cast FFI false-positives (rustc quirk on pointer casts),
and 8 dead-code items that need product decisions.
2026-06-28 14:39:00 +08:00
dailz 145b5d3e7e chore: design cleanup, dead-code purge, README/doc refresh
Audit-driven follow-up after the SAFETY-debt commit (Oracle steps 6-7).
End state: cargo clippy --release --all-targets still 0 errors; private_interfaces
and type_complexity warnings cleared.

Design cleanups (Oracle step 6):
  - cap_portal.rs: introduce PortalFormatInfo struct to replace the
    Rc<Cell<Option<(u32,u32,u32,u64)>>> cross-callback hand-off. Self-
    documenting struct fields replace positional tuple access at the
    format-change and process callbacks.
  - avhw.rs: import_dma_buf_to_vaapi signature collapses from 8 args
    (fd/width/height/drm_format/modifier/stride/offset) to
    (*mut AVBufferRef, &PwDmaBufFrame). Callers in avhw.rs,
    state_portal.rs, and vaapi_import_bench.rs now pass the frame by
    reference instead of unpacking 7 fields just to repack them. Drops
    the unused width parameter and the too_many_arguments(8/7) warning.
  - state.rs: visibility hygiene. EncConstructionStage and WlrHeadInfo
    downgrade pub -> pub(crate); State.stage field downgrades to
    pub(crate). These are internal state-machine types not exposed
    across the crate boundary; making them pub(crate) clears all
    private_interfaces warnings without leaking more types.

Dead-code purge (Oracle step 7):
  - transform.rs: remove unused Rect struct, transform_basis,
    screen_to_frame, fit_inside_bounds helpers and their 18 dedicated
    tests. Transform enum and transpose_if_transform_transposed remain
    (both are actively used by state.rs and avhw.rs). File shrinks
    from 409 -> 109 lines.

Repository housekeeping (Oracle step 7):
  - .gitignore: add review.json (stray review-tool output that
    regenerates per run).
  - README.md: refresh CLI table to match src/args.rs (now lists
    --backend, --no-persist, --port-as-WebRTC-signaling, --max-bitrate,
    --stats). Add capture-backend explainer + 4 new usage examples.
    Note in README points readers at src/args.rs as the authoritative
    source. Remove stale 'WebTransport, unused in MVP' description.

avhw.rs: AsRawFd import annotated with a rustc-quirk explanation — the
import triggers a false 'unused_imports' warning but E0599 if removed.
Left as-is with explanatory comment rather than chasing the lint.

All 79 remaining unit tests + 3 integration tests still pass. Cargo
build --release clean.
2026-06-28 13:58:52 +08:00
dailz 6d49222de8 feat: Phase 1 MVP with audit fixes — Wayland screen capture + VAAPI encoding
Phase 1 MVP implementation of wl-webrtc: Wayland screen capture tool
with hardware-accelerated VAAPI H.264 encoding and WebTransport output.

Includes all 9 runtime bug fixes from code audit (fix-audit-issues plan):

CRITICAL:
- C2: h264_metadata BSF with repeat_sps/repeat_pps in encode pipeline
- C4: FpsLimit wired as timing gate in on_copy_complete

HIGH:
- C3+A2: DRM device discovery via dmabuf feedback MainDevice event,
  unified resolve_drm_path() helper (CLI > compositor > auto > fallback)
- H2: Separate physical_size (mm) from mode_size (pixels) in wl_output
- H1+A3: Multi-output warning + named-output-not-found error

MEDIUM:
- M5: tv_sec u32->u64 to avoid Y2106 timestamp truncation
- M4: Guard against SHM Buffer event (DMA-BUF only)

Key components:
- src/avhw.rs: FFmpeg VAAPI encoder + filter graph + BSF pipeline
- src/state.rs: Wayland event loop + output negotiation + screencopy
- src/cap_wlr_screencopy.rs: wlr-screencopy capture source
- src/fps_limit.rs: Frame rate limiting with configurable target
- src/transform.rs: Frame format conversion utilities
2026-04-05 23:35:00 +08:00