Added
- Live progress for image-sub OCR detection. The on-demand "Detect languages" action now streams coarse stages over websocket ("Extracting subtitle track N…", "OCR (Latin)…", "OCR (CJK/Cyrillic/Arabic)…") so the minutes-long PGS/VobSub OCR shows what it's doing instead of a silent wait.
- Dedicated "Unknown language" filter in the scanner, listing titles with any
undaudio or subtitle track (separate from the audio-cleanup filter), with a count. - Plex notified after a track-language change. When detection writes corrected language tags to a file, Shrinkerr triggers a Plex library refresh for that item so Plex re-reads them. Gated on the default-on
plex_notify_on_lang_changesetting; reuses the existing rename/rescan path; fail-open. - Tooltips on the title-detail icon buttons (codec badge, converted indicator, ignore, remove-from-list, move-to-trash) — each now explains what it does on hover.
Docker images
Pinned to this release:
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.1 # portable (amd64 + arm64, CPU)
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.1-nvenc # x86_64 + NVIDIA GPU (NVENC)
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.1-edge # as above but ffmpeg master
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.1-edge-nvencFloating tags (:latest, :edge, :nvenc, :edge-nvenc) track the most recent release of each lineage.