When a file is re-registered, the old PipelineProgress in Redis
was not cleared, causing the Portal to show 'completed' status
even though the new job was still running.
Now PipelineProgress is deleted when a new job starts processing.
Previously, ingestion_complete only checked ASR and face traces,
causing jobs to complete before TKG was triggered. This resulted in
TKG nodes being 0 even after job completion.
Now ingestion_complete also checks if TKG nodes exist for the file.
The job stays in 'running' status until TKG completes.
The TKG guard was checking if nodes exist before spawning TKG build.
This caused a deadlock when:
1. TKG spawns (async)
2. ingestion_complete checks → false (nodes not yet created)
3. Returns Ok(false), job stays running
4. Next poll: guard checks nodes → no nodes yet (TKG still running)
5. Skips TKG → deadlock!
Since build_tkg uses ON CONFLICT (idempotent), it's safe to call
multiple times. Removed the guard to fix the deadlock.
When a job is deleted from the database (e.g., by unregister),
the worker now checks if the job still exists before processing.
If the job no longer exists, it skips it instead of getting stuck.
This fixes the issue where unregistering a file would leave the
worker stuck trying to process a non-existent job.
Identity agent now checks Qdrant _seeds collection for the file's
seed identity photos before running. If no seeds exist, the agent
is skipped to avoid unnecessary processing.
Flow:
1. Job completes (Face + ASRX done)
2. Check Qdrant _seeds for file_uuid
3. If seeds exist → run identity agent
4. If no seeds → skip identity agent
Previously Rule 1 only created chunks from ASRX segments, merging OCR
text where frame ranges overlapped. OCR text that didn't overlap with
any ASRX segment was ignored.
Now Rule 1 has two phases:
1. Process ASRX segments (merge OCR where overlapping) - existing behavior
2. Create chunks for OCR-only text (frames not covered by ASRX)
OCR-only chunks are grouped by consecutive frames (within 5 frames)
to avoid creating too many single-frame chunks.
Example: ASRX 819 + OCR-only 4 = 823 sentence chunks
Bug: ASRX handler stored pre_chunks as BOTH 'asrx' and 'asr' types.
This caused confusion because Rule 1 queries 'asrx' type, but only
'asr' type existed in the database (asrx type was deleted or never stored).
Fix: Remove the duplicate 'asr' storage (lines 530-542).
ASRX handler now only stores 'asrx' type pre_chunks to workspace SQLite.
PostgreSQL pre_chunks are stored by processor.rs with correct 'asrx' type.
This ensures Rule 1 can find ASRX pre_chunks correctly.
Bug: When regFiles API fails, all files get 'unregistered' status even if
they are actually registered. Also, computed property was using reference
to files.value instead of a copy, which could cause mutation issues.
Fix:
- Fetch scan results FIRST (source of truth for files on disk)
- Use scan API's is_registered field as fallback status
- Only override with regFiles data if file exists in scan results
- Computed property now uses [...files.value] to create a copy
- Skip files from regFiles that don't exist on disk (deleted)
Bug: Scan files were getting status 'registered_scan' which doesn't match
any filter value (unregistered/pending/processing/completed/indexed/unindexed).
When toggling filters on/off, files would disappear because their status
didn't match any valid filter.
Fix:
- Removed 'registered_scan' status entirely
- Fetch regFiles FIRST to get real statuses
- Scan files default to 'unregistered' status
- regFiles overlay with actual status (pending/processing/completed)
- Increased regFiles page_size to 200 for larger libraries
Frontend:
- Add media type filter (全部/影片/照片)
- Add indexed status filter (未入庫/已入庫)
- Show media type column with icons
- Fix status filter to handle indexed/unindexed correctly
- Determine media type from file extension
Backend:
- Add total_chunks field to FileItem API response
- Query chunk counts efficiently in batch with IN clause
- Frontend uses total_chunks to determine is_indexed status
Bug: all_completed only checked existing results, not missing processors.
If a processor (like pose) never created a result row, all_completed would
still return true and mark the job as completed.
Fix: all_completed now checks that every processor in job_processors has
a corresponding completed result. Added logging for missing processors.
Also fixed:
- any_pending now checks all expected processors, not just existing results
- Added missing_processors detection and logging
- Skip chunks where both ASRX text and OCR text are empty
- Use count-based chunk_id instead of index to avoid gaps
- This ensures PostgreSQL and Qdrant chunk counts match
- Added text_content field to SearchResult and SemanticSearchResult
- Added get_chunk_by_id_no_embedding for keyword results without embedding requirement
- Fixed search_bm25 to use position-based ranking for CJK/Korean content
- Fixed sqlx column mapping with explicit alias
- Skip text_match filter for keyword-only results
- Use text_content as fallback when summary is empty
- Frontend calls /api/v1/face-thumbnail?uuid=...&frame=...
- Backend only had /api/v1/file/:file_uuid/thumbnail
- Added compat route and uuid field to ThumbQuery
- ingestion_complete query used file_uuid column which is always NULL
- Changed to JOIN processor_results with monitor_jobs on job_id
- All stuck jobs now complete successfully
- Removed trace_chunks field from PostgresStats struct
- Removed trace_chunks query from get_file_stats and get_ingestion_status
- Fixed OCR fetch_ocr_texts to compute frames from start_time*FPS
- Updated scan.rs to use separate count_nodes/count_edges functions
- get_pipeline_progress_handler now queries actual DB counts
- Fixed processor_results query (requires JOIN with monitor_jobs)
- Card progress bar and right-click content now consistent
- tkg_nodes has no edge_type column, query was failing silently
- Split into count_nodes(node_type) and count_edges(edge_type)
- Fixed text_region → text_trace node type name
- Also: OCR frame fix in rule1 (end_frame computed from end_time+FPS)
- get_file_identities: UNION face_detections + file_identities
- list_identities: add file_bindings from file_identities table
- Add back /api/v1/traces/unassigned route
- Total count query now includes file_identities
Frontend can now:
- Filter pending identities by file_uuid
- Filter pending faces (unassigned traces) by file_uuid
- skin_tone is a person attribute (like height), not trace attribute
- Remove build_skin_tone_trace_nodes function
- Remove skin_tone_trace_nodes from TkgResult and API response
- Remove skin_tone_trace from documentation tables
- Fix trace_id type mismatch (INT4 vs i64) with explicit ::bigint cast
- Change build_face_track_nodes to use from_pg version
- Add skin_tone_trace_nodes to API response
- Add #[derive(Serialize)] to TkgResult
- Fix Unicode panic in text label truncation
- Add push_existing_embeddings.py script
- Add public health routes at /api/v1/health, /api/v1/health/detailed, /api/v1/health/consistency
- Make health functions and response types public
- Public routes bypass auth middleware (unlike protected /api/v1/* routes)
Fix qdrant_request() to properly handle empty dict {} as body.
Python's 'if body' evaluates to False for empty dict, causing EOF error.
Changed:
- data = json.dumps(body).encode() if body is not None else None
Also cleaned up count_seeds() to use consistent body passing.
TKG Helper (scripts/utils/tkg_helper.py):
- mark_face_track_suggested(): Mark node as 'suggested' with pending identity info
- mark_face_track_confirmed(): Mark node as 'confirmed' with identity_ref
- mark_face_track_stranger(): Mark node as 'stranger' with stranger_ref
- batch_mark_suggestions(): Batch mark multiple traces
- batch_mark_strangers(): Batch mark stranger clusters
- get_face_track_nodes(): Get all face_track nodes for a file
- get_pending_face_tracks(): Get nodes with status='pending'
- get_suggested_face_tracks(): Get nodes with status='suggested'
Identity Matcher updates:
- Add --mark-tkg flag to update TKG nodes after matching
- Integrates with tkg_helper for batch operations
Node properties schema:
- status: pending | suggested | confirmed | stranger
- pending_identity_name/uuid/id: suggested identity info
- suggested_by: tmdb | propagation | manual
- confidence: matching score
- identity_ref: confirmed identity reference
- Add ensure_seeds_collection(): create _seeds collection (512D, Cosine)
- Add push_seed_embedding(): push identity seed with payload {identity_id, uuid, name, source, file_uuid, trace_id, tmdb_id}
- Add get_seeds(): get all seeds (optional source filter)
- Add search_seeds(): cosine search against seeds
- Add delete_seed(): delete seed by identity_id
- Add count_seeds(): count seeds (optional source filter)
- Add get_trace_representatives(): get 3 representatives per trace for multi-angle matching
- Add get_trace_centroid(): get centroid embedding for a trace
- Add update_identity_in_faces(): update identity_id/uuid for all face points with trace_id
Point ID strategy: identity_id directly as point_id for _seeds collection
All functions tested successfully
