v0.7.9

Changed

• Scanner runs ffprobe in parallel (default 4 concurrent). Pre-v0.7.9 scans were serial — a 120K-file library took ~12 h on a NUC because every ffprobe runs at ~300 ms. v0.7.9 pre-probes files in asyncio.Semaphore-bounded chunks, then runs the existing serial classify+emit loop with cached probe results. ~4× speedup on cold scans, no new DB-write contention with concurrent queue workers (probes are subprocess I/O only; writes still flow through the single-writer batched path). Tune via SHRINKERR_SCAN_CONCURRENCY (env var, default 4); set to 1 to restore serial behavior.

Docker images

Pinned to this release:

docker pull ghcr.io/i-ial9000/shrinkerr:0.7.9         # portable (amd64 + arm64, CPU)
docker pull ghcr.io/i-ial9000/shrinkerr:0.7.9-nvenc   # x86_64 + NVIDIA GPU (NVENC)
docker pull ghcr.io/i-ial9000/shrinkerr:0.7.9-edge    # as above but ffmpeg master
docker pull ghcr.io/i-ial9000/shrinkerr:0.7.9-edge-nvenc

Floating tags (:latest, :edge, :nvenc, :edge-nvenc) track the most recent release of each lineage.