v0.9.0

Added

• Language detection for image-based subtitles (PGS/VobSub) via OCR. On-demand "Detect languages" now handles und PGS (Blu-ray) and VobSub (DVD) subtitles: the track is extracted with mkvextract and OCR'd with pgsrip (tesseract). Multi-script without a separate detection step — OCR runs first with the Latin (eng) model (clean and fast for the common case); if the result can't be identified, it re-OCRs with the CJK/Cyrillic/Arabic packs. The recognized text feeds the existing language detector, and a detected language is written to the file via the v0.8.3 mkvpropedit path, same as text subs. Confidence-gated and fail-open (bad OCR leaves the track und, never mistags). On-demand only — not at scan, not before conversion (OCR is slow). Adds tesseract-ocr + language packs + pgsrip to the images. Verified end-to-end on a real Blu-ray PGS sub (Italian → ita).

Docker images

Pinned to this release:

docker pull ghcr.io/i-ial9000/shrinkerr:0.9.0         # portable (amd64 + arm64, CPU)
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.0-nvenc   # x86_64 + NVIDIA GPU (NVENC)
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.0-edge    # as above but ffmpeg master
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.0-edge-nvenc

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