ui polish and importer logging tuning

app/tasks/scan.py

@@ -302,6 +302,64 @@ def recover_interrupted_tasks():
     return {'recovered': recovered}
 
 
+@celery.task(name='app.tasks.scan.cleanup_old_tasks')
+def cleanup_old_tasks(days: int = 7):
+    """
+    Clean up old failed and skipped import task records.
+
+    Called periodically by Celery Beat to prevent database bloat.
+    Deletes ImportTask records with status 'failed' or 'skipped'
+    that are older than the specified number of days.
+
+    Also cleans up orphaned ImportBatch records that have no tasks.
+
+    Args:
+        days: Number of days to retain failed/skipped tasks (default: 7)
+
+    Returns:
+        dict with counts of deleted records
+    """
+    from datetime import timedelta
+
+    cutoff = datetime.utcnow() - timedelta(days=days)
+
+    log.info(f"Cleaning up failed/skipped tasks older than {days} days (before {cutoff})")
+
+    # Delete old failed tasks
+    failed_deleted = ImportTask.query.filter(
+        ImportTask.status == 'failed',
+        ImportTask.completed_at < cutoff
+    ).delete(synchronize_session=False)
+
+    # Delete old skipped tasks
+    skipped_deleted = ImportTask.query.filter(
+        ImportTask.status == 'skipped',
+        ImportTask.completed_at < cutoff
+    ).delete(synchronize_session=False)
+
+    db.session.commit()
+
+    # Clean up orphaned batches (batches with no tasks)
+    orphaned_batches = ImportBatch.query.filter(
+        ~ImportBatch.tasks.any()
+    ).all()
+    batches_deleted = len(orphaned_batches)
+    for batch in orphaned_batches:
+        db.session.delete(batch)
+
+    db.session.commit()
+
+    if failed_deleted or skipped_deleted or batches_deleted:
+        log.info(f"Cleanup complete: {failed_deleted} failed tasks, "
+                 f"{skipped_deleted} skipped tasks, {batches_deleted} orphaned batches deleted")
+
+    return {
+        'failed_deleted': failed_deleted,
+        'skipped_deleted': skipped_deleted,
+        'batches_deleted': batches_deleted
+    }
+
+
 @celery.task(name='app.tasks.scan.update_batch_stats')
 def update_batch_stats(batch_id: int):
     """