Fixed
- "Empty Plex trash after scan" worked for TV sections but silently failed for movie sections — log line said
[PLEX] Emptied trash for section Nbut the deleted movie's metadata stayed in Plex. Root cause was timing, not logic: Plex's library scanner is asynchronous, even after/library/sections/{id}/refreshreturns 200 the actual scan runs in the background, and movie sections in particular take ~10-15 s to flag a just-removed file as trashed. We were callingemptyTrash~4 s after triggering the scan — Plex hadn't yet processed the deletion, trash was empty, the API returned 200 (it always does, regardless of whether anything was purged), and we logged success. TV episode removal happened to register fast enough to land inside the same window, hence the asymmetric symptom. Fix:empty_plex_trashnow waits 15 s before issuing the request (default — call site can override). The wait runs as a detached background task so batch-converting 100 files doesn't accumulate 25 minutes of cumulative blocking on the worker — the conversion job completes immediately and the trash cleanup happens async. Errors are still logged via the existing[PLEX]print stream.
Docker images
Pinned to this release:
docker pull ghcr.io/i-ial9000/shrinkerr:0.3.100 # portable (amd64 + arm64, CPU)
docker pull ghcr.io/i-ial9000/shrinkerr:0.3.100-nvenc # x86_64 + NVIDIA GPU (NVENC)
docker pull ghcr.io/i-ial9000/shrinkerr:0.3.100-edge # as above but ffmpeg master
docker pull ghcr.io/i-ial9000/shrinkerr:0.3.100-edge-nvencFloating tags (:latest, :edge, :nvenc, :edge-nvenc) track the most recent release of each lineage.