Merge pull request #298 from kettui/fix/respect-metadata-provider-in-album-completeness-job

Honor primary metadata source in album_completeness job and associated repair flow

core/metadata_service.py

@@ -17,6 +17,9 @@ logger = get_logger("metadata_service")
 
 MetadataProvider = Literal["spotify", "itunes", "auto"]
 
+# Ordered by fallback preference. Higher-priority sources appear earlier.
+METADATA_SOURCE_PRIORITY = ('deezer', 'itunes', 'spotify', 'discogs', 'hydrabase')
+
 _client_cache_lock = threading.RLock()
 _client_cache: Dict[str, Any] = {}
 
@@ -62,7 +65,72 @@ def get_primary_client():
 
     This is THE single source of truth for "which client should I call?"
     """
-    return _get_client_for_source(get_primary_source())
+    return get_client_for_source(get_primary_source())
+
+
+def get_source_priority(preferred_source: str):
+    """Return supported sources with the preferred source first."""
+    ordered = []
+
+    if preferred_source in METADATA_SOURCE_PRIORITY:
+        ordered.append(preferred_source)
+
+    for source in METADATA_SOURCE_PRIORITY:
+        if source not in ordered:
+            ordered.append(source)
+
+    return ordered
+
+
+def get_client_for_source(source: str):
+    """Get the client object for an exact metadata source.
+
+    Returns the matching client or None if that source is unavailable.
+    No fallback swaps.
+    """
+    if source == 'spotify':
+        try:
+            import importlib
+            ws = importlib.import_module('web_server')
+            sc = getattr(ws, 'spotify_client', None)
+            if sc and sc.is_spotify_authenticated():
+                return sc
+        except Exception:
+            pass
+        return None
+
+    if source == 'deezer':
+        return get_deezer_client()
+
+    if source == 'discogs':
+        return get_discogs_client()
+
+    if source == 'hydrabase':
+        return get_hydrabase_client(allow_fallback=False)
+
+    if source == 'itunes':
+        return get_itunes_client()
+
+    return None
+
+
+def get_album_tracks_for_source(source: str, album_id: str):
+    """Get album tracks for an exact source.
+
+    Returns Spotify-compatible dict/list data or None.
+    No fallback swaps.
+    """
+    client = get_client_for_source(source)
+    if not client:
+        return None
+
+    try:
+        fetch = getattr(client, 'get_album_tracks_dict', None) if source == 'hydrabase' else getattr(client, 'get_album_tracks', None)
+        if not fetch:
+            return None
+        return fetch(album_id)
+    except Exception:
+        return None
 
 
 def get_deezer_client():
@@ -122,8 +190,12 @@ def get_discogs_client(token: Optional[str] = None):
         return client
 
 
-def get_hydrabase_client():
-    """Return current Hydrabase client if connected, else iTunes fallback."""
+def get_hydrabase_client(allow_fallback: bool = True):
+    """Return current Hydrabase client if connected.
+
+    If allow_fallback is True, return iTunes fallback when Hydrabase is not
+    connected. If False, return None instead.
+    """
     try:
         import importlib
         ws = importlib.import_module('web_server')
@@ -132,7 +204,9 @@ def get_hydrabase_client():
             return client
     except Exception:
         pass
-    return get_itunes_client()
+    if allow_fallback:
+        return get_itunes_client()
+    return None
 
 
 def clear_cached_metadata_clients():
@@ -206,7 +280,7 @@ class MetadataService:
         self.preferred_provider = preferred_provider
         self.spotify = SpotifyClient()
         self._fallback_source = get_primary_source()
-        self.itunes = _get_client_for_source(self._fallback_source)
+        self.itunes = get_client_for_source(self._fallback_source)
 
         self._log_initialization()
 
@@ -378,7 +452,7 @@ class MetadataService:
         self.spotify.reload_config()
         new_source = get_primary_source()
         self._fallback_source = new_source
-        self.itunes = _get_client_for_source(new_source)
+        self.itunes = get_client_for_source(new_source)
         self._log_initialization()
 
 

core/repair_jobs/album_completeness.py

@@ -1,5 +1,10 @@
 """Album Completeness Checker Job — finds albums missing tracks."""
 
+from core.metadata_service import (
+    get_album_tracks_for_source,
+    get_primary_source,
+    get_source_priority,
+)
 from core.repair_jobs import register_job
 from core.repair_jobs.base import JobContext, JobResult, RepairJob
 from utils.logging_config import get_logger
@@ -14,8 +19,9 @@ class AlbumCompletenessJob(RepairJob):
     description = 'Checks if all tracks from albums are present'
     help_text = (
         'Compares the number of tracks you have for each album against the expected total '
-        'from the album tracklist (via Spotify, iTunes, or Deezer). Albums where tracks are '
-        'missing get flagged as findings with details about which tracks are absent.\n\n'
+        'from the active metadata provider first, then other supported sources if needed. '
+        'Albums where tracks are missing get flagged as findings with details about which '
+        'tracks are absent.\n\n'
         'Useful for catching partial downloads or albums where some tracks failed to download. '
         'You can use the Download Missing feature from the album page to fill gaps.\n\n'
         'Settings:\n'
@@ -40,10 +46,13 @@ class AlbumCompletenessJob(RepairJob):
         settings = self._get_settings(context)
         min_tracks = settings.get('min_tracks_for_check', 3)
         min_completion_pct = settings.get('min_completion_pct', 0)
+        primary_source = self._get_primary_source()
 
         # Fetch all albums with ANY external source ID — not just Spotify
         albums = []
         conn = None
+        has_itunes = False
+        has_deezer = False
         try:
             conn = context.db._get_connection()
             cursor = conn.cursor()
@@ -53,29 +62,44 @@ class AlbumCompletenessJob(RepairJob):
             columns = {row[1] for row in cursor.fetchall()}
             has_itunes = 'itunes_album_id' in columns
             has_deezer = 'deezer_id' in columns
+            has_discogs = 'discogs_id' in columns
+            has_hydrabase = 'soul_id' in columns
 
             # Build SELECT with available source ID columns
             select_cols = [
-                'al.id', 'al.title', 'ar.name', 'al.spotify_album_id', 'al.track_count',
-                'COUNT(t.id) as actual_count', 'al.thumb_url', 'ar.thumb_url',
+                ('al.id', 'album_id'),
+                ('al.title', 'album_title'),
+                ('ar.name', 'artist_name'),
+                ('al.spotify_album_id', 'spotify_album_id'),
+                ('al.track_count', 'track_count'),
+                ('COUNT(t.id)', 'actual_count'),
+                ('al.thumb_url', 'album_thumb_url'),
+                ('ar.thumb_url', 'artist_thumb_url'),
             ]
             if has_itunes:
-                select_cols.append('al.itunes_album_id')
+                select_cols.append(('al.itunes_album_id', 'itunes_album_id'))
             if has_deezer:
-                select_cols.append('al.deezer_id')
+                select_cols.append(('al.deezer_id', 'deezer_album_id'))
+            if has_discogs:
+                select_cols.append(('al.discogs_id', 'discogs_album_id'))
+            if has_hydrabase:
+                select_cols.append(('al.soul_id', 'hydrabase_album_id'))
 
             # WHERE: album has at least one source ID
-            where_parts = []
-            if True:  # spotify always exists
-                where_parts.append("(al.spotify_album_id IS NOT NULL AND al.spotify_album_id != '')")
+            where_parts = ["(al.spotify_album_id IS NOT NULL AND al.spotify_album_id != '')"]
             if has_itunes:
                 where_parts.append("(al.itunes_album_id IS NOT NULL AND al.itunes_album_id != '')")
             if has_deezer:
                 where_parts.append("(al.deezer_id IS NOT NULL AND al.deezer_id != '')")
+            if has_discogs:
+                where_parts.append("(al.discogs_id IS NOT NULL AND al.discogs_id != '')")
+            if has_hydrabase:
+                where_parts.append("(al.soul_id IS NOT NULL AND al.soul_id != '')")
             where_clause = ' OR '.join(where_parts)
 
+            select_sql = ', '.join(f'{expr} AS {alias}' for expr, alias in select_cols)
             cursor.execute(f"""
-                SELECT {', '.join(select_cols)}
+                SELECT {select_sql}
                 FROM albums al
                 LEFT JOIN artists ar ON ar.id = al.artist_id
                 LEFT JOIN tracks t ON t.album_id = al.id
@@ -83,6 +107,7 @@ class AlbumCompletenessJob(RepairJob):
                 GROUP BY al.id
             """)
             albums = cursor.fetchall()
+            column_index = {alias: idx for idx, (_, alias) in enumerate(select_cols)}
         except Exception as e:
             logger.error("Error fetching albums: %s", e, exc_info=True)
             result.errors += 1
@@ -100,27 +125,24 @@ class AlbumCompletenessJob(RepairJob):
         if context.report_progress:
             context.report_progress(phase=f'Checking {total} albums...', total=total)
 
-        # Determine column positions based on what we selected
-        # Fixed: 0=id, 1=title, 2=artist, 3=spotify_id, 4=track_count, 5=actual, 6=album_thumb, 7=artist_thumb
-        itunes_col = 8 if has_itunes else None
-        deezer_col = (9 if has_itunes else 8) if has_deezer else None
-
         for i, row in enumerate(albums):
             if context.check_stop():
                 return result
             if i % 10 == 0 and context.wait_if_paused():
                 return result
 
-            album_id = row[0]
-            title = row[1]
-            artist_name = row[2]
-            spotify_album_id = row[3]
-            db_track_count = row[4]
-            actual_count = row[5]
-            album_thumb = row[6]
-            artist_thumb = row[7]
-            itunes_album_id = row[itunes_col] if itunes_col is not None else None
-            deezer_album_id = row[deezer_col] if deezer_col is not None else None
+            album_id = row[column_index['album_id']]
+            title = row[column_index['album_title']]
+            artist_name = row[column_index['artist_name']]
+            spotify_album_id = row[column_index['spotify_album_id']]
+            db_track_count = row[column_index['track_count']]
+            actual_count = row[column_index['actual_count']]
+            album_thumb = row[column_index['album_thumb_url']]
+            artist_thumb = row[column_index['artist_thumb_url']]
+            itunes_album_id = row[column_index['itunes_album_id']] if 'itunes_album_id' in column_index else None
+            deezer_album_id = row[column_index['deezer_album_id']] if 'deezer_album_id' in column_index else None
+            discogs_album_id = row[column_index['discogs_album_id']] if 'discogs_album_id' in column_index else None
+            hydrabase_album_id = row[column_index['hydrabase_album_id']] if 'hydrabase_album_id' in column_index else None
 
             result.scanned += 1
 
@@ -135,10 +157,16 @@ class AlbumCompletenessJob(RepairJob):
             # If we don't know the expected track count, try to get it from an API
             expected_total = db_track_count
 
+            album_ids = {
+                'spotify': spotify_album_id or '',
+                'itunes': itunes_album_id or '',
+                'deezer': deezer_album_id or '',
+                'discogs': discogs_album_id or '',
+                'hydrabase': hydrabase_album_id or '',
+            }
+
             if not expected_total:
-                expected_total = self._get_expected_total(
-                    context, spotify_album_id, itunes_album_id, deezer_album_id
-                )
+                expected_total = self._get_expected_total(context, primary_source, album_ids)
 
             # Skip singles/EPs based on expected track count (not local count)
             if expected_total and expected_total < min_tracks:
@@ -169,9 +197,7 @@ class AlbumCompletenessJob(RepairJob):
                     continue
 
             # Album is incomplete — try to find which tracks are missing
-            missing_tracks = self._find_missing_tracks(
-                context, album_id, spotify_album_id, itunes_album_id, deezer_album_id
-            )
+            missing_tracks = self._find_missing_tracks(context, primary_source, album_id, album_ids)
 
             if context.report_progress:
                 context.report_progress(
@@ -180,8 +206,6 @@ class AlbumCompletenessJob(RepairJob):
                 )
             if context.create_finding:
                 try:
-                    # Use whichever source ID is available
-                    source_id = spotify_album_id or itunes_album_id or deezer_album_id or ''
                     context.create_finding(
                         job_id=self.job_id,
                         finding_type='incomplete_album',
@@ -198,9 +222,13 @@ class AlbumCompletenessJob(RepairJob):
                             'album_id': album_id,
                             'album_title': title,
                             'artist': artist_name,
+                            'primary_source': primary_source,
+                            'primary_album_id': self._get_album_id_for_source(primary_source, album_ids) or '',
                             'spotify_album_id': spotify_album_id or '',
                             'itunes_album_id': itunes_album_id or '',
                             'deezer_album_id': deezer_album_id or '',
+                            'discogs_album_id': discogs_album_id or '',
+                            'hydrabase_album_id': hydrabase_album_id or '',
                             'expected_tracks': expected_total,
                             'actual_tracks': actual_count,
                             'missing_tracks': missing_tracks,
@@ -220,42 +248,24 @@ class AlbumCompletenessJob(RepairJob):
             context.update_progress(total, total)
 
         logger.info("Completeness check: %d albums checked, %d incomplete found",
-                     result.scanned, result.findings_created)
+                    result.scanned, result.findings_created)
         return result
 
-    def _get_expected_total(self, context, spotify_id, itunes_id, deezer_id):
-        """Try to get the expected track count from any available API source."""
-        # Try Spotify first
-        if spotify_id and context.spotify_client and not context.is_spotify_rate_limited():
-            try:
-                album_data = context.spotify_client.get_album(spotify_id)
-                if album_data:
-                    total = album_data.get('total_tracks', 0)
-                    if total:
-                        return total
-            except Exception:
-                pass
-
-        # Try fallback client (iTunes or Deezer) — both return Spotify-compatible format
-        # Match the ID to the actual client type to avoid passing iTunes ID to Deezer or vice versa
-        if context.itunes_client:
-            is_deezer = type(context.itunes_client).__name__ == 'DeezerClient'
-            primary_id = deezer_id if is_deezer else itunes_id
-            secondary_id = itunes_id if is_deezer else deezer_id
-            for fid in [primary_id, secondary_id]:
-                if not fid:
-                    continue
-                try:
-                    api_tracks = context.itunes_client.get_album_tracks(fid)
-                    if api_tracks and 'items' in api_tracks:
-                        return len(api_tracks['items'])
-                except Exception:
-                    pass
+    def _get_expected_total(self, context, primary_source, album_ids):
+        """Try to get the expected track count from the active metadata provider first."""
+        for source in get_source_priority(primary_source):
+            album_id = self._get_album_id_for_source(source, album_ids)
+            if not album_id:
+                continue
+            api_tracks = self._get_album_tracks(source, album_id)
+            items = self._extract_track_items(api_tracks)
+            if items:
+                return len(items)
 
         return 0
 
-    def _find_missing_tracks(self, context, album_id, spotify_id, itunes_id, deezer_id):
-        """Identify which specific tracks are missing using any available API source."""
+    def _find_missing_tracks(self, context, primary_source, album_id, album_ids):
+        """Identify which specific tracks are missing using the active metadata provider first."""
         # Get track numbers we already have
         owned_numbers = set()
         conn = None
@@ -274,37 +284,23 @@ class AlbumCompletenessJob(RepairJob):
             if conn:
                 conn.close()
 
-        # Try Spotify first
         api_tracks = None
-        if spotify_id and context.spotify_client and not context.is_spotify_rate_limited():
-            try:
-                api_tracks = context.spotify_client.get_album_tracks(spotify_id)
-            except Exception as e:
-                logger.debug("Error getting Spotify album tracks for %s: %s", spotify_id, e)
+        for source in get_source_priority(primary_source):
+            source_album_id = self._get_album_id_for_source(source, album_ids)
+            if not source_album_id:
+                continue
+            api_tracks = self._get_album_tracks(source, source_album_id)
+            if self._extract_track_items(api_tracks):
+                break
 
-        # Try fallback client (iTunes or Deezer)
-        if not api_tracks or 'items' not in (api_tracks or {}):
-            if context.itunes_client:
-                is_deezer = type(context.itunes_client).__name__ == 'DeezerClient'
-                primary_id = deezer_id if is_deezer else itunes_id
-                secondary_id = itunes_id if is_deezer else deezer_id
-                for fid in [primary_id, secondary_id]:
-                    if not fid:
-                        continue
-                    try:
-                        api_tracks = context.itunes_client.get_album_tracks(fid)
-                        if api_tracks and 'items' in api_tracks:
-                            break
-                    except Exception as e:
-                        logger.debug("Error getting fallback album tracks for %s: %s", fid, e)
-
-        if not api_tracks or 'items' not in api_tracks:
+        items = self._extract_track_items(api_tracks)
+        if not items:
             return []
 
-        # Both Spotify, iTunes, and Deezer return the same format:
+        # All supported provider responses expose the same core fields once normalized.
         # items[].track_number, items[].name, items[].disc_number, items[].id, items[].artists
         missing_tracks = []
-        for item in api_tracks['items']:
+        for item in items:
             tn = item.get('track_number')
             if tn and tn not in owned_numbers:
                 track_artists = []
@@ -317,6 +313,9 @@ class AlbumCompletenessJob(RepairJob):
                     'track_number': tn,
                     'name': item.get('name', ''),
                     'disc_number': item.get('disc_number', 1),
+                    'source': item.get('_source', primary_source),
+                    'source_track_id': item.get('id', ''),
+                    'track_id': item.get('id', ''),
                     'spotify_track_id': item.get('id', ''),
                     'duration_ms': item.get('duration_ms', 0),
                     'artists': track_artists,
@@ -331,6 +330,35 @@ class AlbumCompletenessJob(RepairJob):
         merged.update(cfg)
         return merged
 
+    def _get_primary_source(self) -> str:
+        """Return the active metadata source used for source prioritization."""
+        try:
+            return get_primary_source()
+        except Exception:
+            return 'deezer'
+
+    def _get_album_id_for_source(self, source: str, album_ids: dict) -> str:
+        return album_ids.get(source, '')
+
+    def _get_album_tracks(self, source: str, album_id: str):
+        """Fetch album tracks from a specific source."""
+        try:
+            return get_album_tracks_for_source(source, album_id)
+        except Exception as e:
+            logger.debug("Error getting %s album tracks for %s: %s", source.capitalize(), album_id, e)
+            return None
+
+    def _extract_track_items(self, api_tracks):
+        """Normalize album track responses to a list of item dicts."""
+        if not api_tracks:
+            return []
+        if isinstance(api_tracks, dict):
+            items = api_tracks.get('items') or []
+            return items if items else []
+        if isinstance(api_tracks, list):
+            return api_tracks
+        return []
+
     def estimate_scope(self, context: JobContext) -> int:
         conn = None
         try:
@@ -346,6 +374,10 @@ class AlbumCompletenessJob(RepairJob):
                 where_parts.append("(itunes_album_id IS NOT NULL AND itunes_album_id != '')")
             if 'deezer_id' in columns:
                 where_parts.append("(deezer_id IS NOT NULL AND deezer_id != '')")
+            if 'discogs_id' in columns:
+                where_parts.append("(discogs_id IS NOT NULL AND discogs_id != '')")
+            if 'soul_id' in columns:
+                where_parts.append("(soul_id IS NOT NULL AND soul_id != '')")
 
             cursor.execute(f"""
                 SELECT COUNT(*) FROM albums

core/repair_worker.py

@@ -18,6 +18,11 @@ from difflib import SequenceMatcher
 from datetime import datetime, timedelta
 from typing import Any, Dict, List, Optional, Tuple
 
+from core.metadata_service import (
+    get_album_tracks_for_source,
+    get_source_priority,
+    get_primary_source,
+)
 from core.repair_jobs import get_all_jobs
 from core.repair_jobs.base import JobContext, JobResult, RepairJob
 from utils.logging_config import get_logger
@@ -104,7 +109,6 @@ class RepairWorker:
         self._on_job_finish = None   # (job_id, status, result) -> None
 
         # Lazy client accessors
-        self._spotify_client = None
         self._itunes_client = None
         self._mb_client = None
         self._acoustid_client = None
@@ -150,13 +154,12 @@ class RepairWorker:
     # ------------------------------------------------------------------
     @property
     def spotify_client(self):
-        if self._spotify_client is None:
-            try:
-                from core.spotify_client import SpotifyClient
-                self._spotify_client = SpotifyClient()
-            except Exception as e:
-                logger.error("Failed to initialize SpotifyClient: %s", e)
-        return self._spotify_client
+        try:
+            from core.metadata_service import get_client_for_source
+            return get_client_for_source('spotify')
+        except Exception as e:
+            logger.error("Failed to resolve shared Spotify client: %s", e)
+            return None
 
     @property
     def itunes_client(self):
@@ -1743,7 +1746,9 @@ class RepairWorker:
             track_number = mt.get('track_number', 0)
             disc_number = mt.get('disc_number', 1)
             track_artists = mt.get('artists', [])
-            spotify_track_id = mt.get('spotify_track_id', '')
+            source = mt.get('source', '') or 'spotify'
+            source_track_id = mt.get('source_track_id', '') or mt.get('track_id', '') or mt.get('spotify_track_id', '')
+            spotify_track_id = mt.get('spotify_track_id', '') or (source_track_id if source == 'spotify' else '')
             artist_search = track_artists[0] if track_artists else artist_name
 
             if not track_name:
@@ -1809,7 +1814,7 @@ class RepairWorker:
                     logger.warning("File operation failed for '%s': %s", track_name, result.get('error'))
 
             # Phase 4: Wishlist fallback
-            if spotify_track_id:
+            if source_track_id:
                 try:
                     # Build album images from finding thumb URL
                     album_images = []
@@ -1818,7 +1823,7 @@ class RepairWorker:
                         album_images = [{'url': album_thumb, 'height': 300, 'width': 300}]
 
                     wishlist_data = {
-                        'id': spotify_track_id,
+                        'id': source_track_id,
                         'name': track_name,
                         'artists': [{'name': a} for a in track_artists] if track_artists else [{'name': artist_name}],
                         'album': {
@@ -1832,6 +1837,7 @@ class RepairWorker:
                         'duration_ms': mt.get('duration_ms', 0),
                         'track_number': track_number,
                         'disc_number': disc_number,
+                        'uri': f"{source}:track:{source_track_id}" if source and source_track_id else '',
                     }
                     source_info = {
                         'album_title': album_title,
@@ -1839,6 +1845,8 @@ class RepairWorker:
                         'track_number': track_number,
                         'disc_number': disc_number,
                         'spotify_album_id': spotify_album_id,
+                        'source': source,
+                        'source_track_id': source_track_id,
                         'is_album': True,
                         'reason': 'album_completeness_auto_fill',
                     }
@@ -1860,7 +1868,7 @@ class RepairWorker:
                     track_details.append({'track': track_name, 'status': 'skipped', 'reason': f'wishlist error: {e}'})
             else:
                 skipped_count += 1
-                track_details.append({'track': track_name, 'status': 'skipped', 'reason': 'no spotify_track_id for wishlist'})
+                track_details.append({'track': track_name, 'status': 'skipped', 'reason': 'no source_track_id for wishlist'})
 
         # Build result message
         parts = []
@@ -1885,11 +1893,18 @@ class RepairWorker:
 
     def _refetch_missing_tracks(self, album_id, details):
         """Re-fetch missing track list from APIs when the stored list is empty."""
+        configured_primary_source = get_primary_source()
         spotify_album_id = details.get('spotify_album_id', '')
         itunes_album_id = details.get('itunes_album_id', '')
         deezer_album_id = details.get('deezer_album_id', '')
-        logger.debug("Refetch missing tracks for album %s: spotify=%s, itunes=%s, deezer=%s",
-                      album_id, spotify_album_id, itunes_album_id, deezer_album_id)
+        discogs_album_id = details.get('discogs_album_id', '')
+        hydrabase_album_id = details.get('hydrabase_album_id', '')
+        primary_source = details.get('primary_source') or configured_primary_source
+        logger.debug(
+            "Refetch missing tracks for album %s: primary=%s spotify=%s itunes=%s deezer=%s discogs=%s hydrabase=%s",
+            album_id, primary_source, spotify_album_id, itunes_album_id, deezer_album_id, discogs_album_id,
+            hydrabase_album_id
+        )
 
         # Get track numbers we already own
         owned_numbers = set()
@@ -1909,32 +1924,27 @@ class RepairWorker:
             if conn:
                 conn.close()
 
-        # Try Spotify first
+        current_source = primary_source
         api_tracks = None
-        if spotify_album_id:
-            try:
-                sp = self.spotify_client
-                if sp and hasattr(sp, 'get_album_tracks'):
-                    api_tracks = sp.get_album_tracks(spotify_album_id)
-            except Exception as e:
-                logger.debug("Refetch: Spotify album tracks failed for %s: %s", spotify_album_id, e)
+        album_sources = {
+            'spotify': spotify_album_id,
+            'itunes': itunes_album_id,
+            'deezer': deezer_album_id,
+            'discogs': discogs_album_id,
+            'hydrabase': hydrabase_album_id,
+        }
 
-        # Try fallback client (iTunes/Deezer)
-        if not api_tracks or 'items' not in (api_tracks or {}):
-            itunes = self.itunes_client
-            if itunes:
-                is_deezer = type(itunes).__name__ == 'DeezerClient'
-                primary_id = deezer_album_id if is_deezer else itunes_album_id
-                secondary_id = itunes_album_id if is_deezer else deezer_album_id
-                for fid in [primary_id, secondary_id]:
-                    if not fid:
-                        continue
-                    try:
-                        api_tracks = itunes.get_album_tracks(fid)
-                        if api_tracks and 'items' in api_tracks:
-                            break
-                    except Exception as e:
-                        logger.debug("Refetch: fallback album tracks failed for %s: %s", fid, e)
+        for source in get_source_priority(primary_source):
+            fid = album_sources.get(source, '')
+            if not fid:
+                continue
+            try:
+                api_tracks = get_album_tracks_for_source(source, fid)
+                if api_tracks and 'items' in (api_tracks or {}):
+                    current_source = source
+                    break
+            except Exception as e:
+                logger.debug("Refetch: %s album tracks failed for %s: %s", source, fid, e)
 
         if not api_tracks or 'items' not in api_tracks:
             return []
@@ -1953,6 +1963,9 @@ class RepairWorker:
                     'track_number': tn,
                     'name': item.get('name', ''),
                     'disc_number': item.get('disc_number', 1),
+                    'source': current_source or 'spotify',
+                    'source_track_id': item.get('id', ''),
+                    'track_id': item.get('id', ''),
                     'spotify_track_id': item.get('id', ''),
                     'duration_ms': item.get('duration_ms', 0),
                     'artists': track_artists,

database/music_database.py

@@ -6426,10 +6426,10 @@ class MusicDatabase:
             with self._get_connection() as conn:
                 cursor = conn.cursor()
 
-                # Use Spotify track ID as unique identifier
+                # Use track ID as unique identifier. Field name stays legacy-compatible.
                 track_id = spotify_track_data.get('id')
                 if not track_id:
-                    logger.error("Cannot add track to wishlist: missing Spotify track ID")
+                    logger.error("Cannot add track to wishlist: missing track ID")
                     return False
 
                 track_name = spotify_track_data.get('name', 'Unknown Track')

tests/test_album_completeness_job.py

@@ -0,0 +1,305 @@
+import sys
+import types
+
+
+if "spotipy" not in sys.modules:
+    spotipy = types.ModuleType("spotipy")
+
+    class _DummySpotify:
+        def __init__(self, *args, **kwargs):
+            pass
+
+    oauth2 = types.ModuleType("spotipy.oauth2")
+
+    class _DummyOAuth:
+        def __init__(self, *args, **kwargs):
+            pass
+
+    spotipy.Spotify = _DummySpotify
+    oauth2.SpotifyOAuth = _DummyOAuth
+    oauth2.SpotifyClientCredentials = _DummyOAuth
+    spotipy.oauth2 = oauth2
+    sys.modules["spotipy"] = spotipy
+    sys.modules["spotipy.oauth2"] = oauth2
+
+if "config.settings" not in sys.modules:
+    config_pkg = types.ModuleType("config")
+    settings_mod = types.ModuleType("config.settings")
+
+    class _DummyConfigManager:
+        def get(self, key, default=None):
+            return default
+
+    settings_mod.config_manager = _DummyConfigManager()
+    config_pkg.settings = settings_mod
+    sys.modules["config"] = config_pkg
+    sys.modules["config.settings"] = settings_mod
+
+from core.repair_jobs.album_completeness import AlbumCompletenessJob
+import core.repair_jobs.album_completeness as album_completeness_module
+
+
+class _FakeCursor:
+    def __init__(self, owned_track_numbers):
+        self._owned_track_numbers = owned_track_numbers
+        self._last_query = ""
+
+    def execute(self, query, params=None):
+        self._last_query = query
+        return self
+
+    def fetchall(self):
+        if "SELECT track_number" in self._last_query:
+            return [(track_number,) for track_number in self._owned_track_numbers]
+        return []
+
+
+class _FakeConnection:
+    def __init__(self, owned_track_numbers):
+        self._cursor = _FakeCursor(owned_track_numbers)
+
+    def cursor(self):
+        return self._cursor
+
+    def close(self):
+        return None
+
+
+class _FakeDB:
+    def __init__(self, owned_track_numbers):
+        self._owned_track_numbers = owned_track_numbers
+
+    def _get_connection(self):
+        return _FakeConnection(self._owned_track_numbers)
+
+
+class _FakeSpotifyClient:
+    def __init__(self, track_count=5):
+        self.track_count = track_count
+        self.calls = []
+
+    def is_spotify_authenticated(self):
+        return True
+
+    def get_album_tracks(self, album_id):
+        self.calls.append(album_id)
+        return {
+            "items": [
+                {"id": f"sp-{i}", "name": f"Spotify Track {i}", "track_number": i, "disc_number": 1, "artists": []}
+                for i in range(1, self.track_count + 1)
+            ]
+        }
+
+
+class _FakeDeezerClient:
+    def __init__(self, track_count=2):
+        self.track_count = track_count
+        self.calls = []
+
+    def is_authenticated(self):
+        return True
+
+    def get_album_tracks(self, album_id):
+        self.calls.append(album_id)
+        return {
+            "items": [
+                {"id": f"dz-{i}", "name": f"Deezer Track {i}", "track_number": i, "disc_number": 1, "artists": []}
+                for i in range(1, self.track_count + 1)
+            ]
+        }
+
+
+class _FakeITunesClient:
+    def __init__(self):
+        self.calls = []
+
+    def is_authenticated(self):
+        return True
+
+    def get_album_tracks(self, album_id):
+        self.calls.append(album_id)
+        return {"items": []}
+
+
+class _FakeDiscogsClient:
+    def __init__(self, track_count=3):
+        self.track_count = track_count
+        self.calls = []
+
+    def get_album_tracks(self, album_id):
+        self.calls.append(album_id)
+        return {
+            "items": [
+                {"id": f"dg-{i}", "name": f"Discogs Track {i}", "track_number": i, "disc_number": 1, "artists": []}
+                for i in range(1, self.track_count + 1)
+            ]
+        }
+
+
+class _FakeHydrabaseClient:
+    def __init__(self, track_count=4):
+        self.track_count = track_count
+        self.calls = []
+
+    def is_connected(self):
+        return True
+
+    def get_album_tracks_dict(self, album_id):
+        self.calls.append(album_id)
+        return {
+            "items": [
+                {"id": f"hy-{i}", "name": f"Hydrabase Track {i}", "track_number": i, "disc_number": 1, "artists": []}
+                for i in range(1, self.track_count + 1)
+            ]
+        }
+
+
+def test_album_completeness_uses_primary_provider_first(monkeypatch):
+    job = AlbumCompletenessJob()
+    context = types.SimpleNamespace(
+        db=_FakeDB(owned_track_numbers={1}),
+        spotify_client=_FakeSpotifyClient(track_count=5),
+        is_spotify_rate_limited=lambda: False,
+    )
+
+    deezer_client = _FakeDeezerClient(track_count=2)
+    itunes_client = _FakeITunesClient()
+    calls = []
+    album_ids = {
+        "spotify": "spotify-album",
+        "itunes": "itunes-album",
+        "deezer": "deezer-album",
+        "discogs": "",
+        "hydrabase": "",
+    }
+
+    monkeypatch.setattr(album_completeness_module, "get_primary_source", lambda: "deezer")
+    monkeypatch.setattr(
+        album_completeness_module,
+        "get_album_tracks_for_source",
+        lambda source, album_id: (
+            calls.append((source, album_id)) or
+            (
+                deezer_client.get_album_tracks(album_id) if source == "deezer" else
+                itunes_client.get_album_tracks(album_id) if source == "itunes" else
+                context.spotify_client.get_album_tracks(album_id) if source == "spotify" else
+                {"items": []}
+            )
+        )
+    )
+
+    expected_total = job._get_expected_total(context, "deezer", album_ids)
+    missing_tracks = job._find_missing_tracks(context, "deezer", 42, album_ids)
+
+    assert expected_total == 2
+    assert calls == [("deezer", "deezer-album"), ("deezer", "deezer-album")]
+    assert deezer_client.calls == ["deezer-album", "deezer-album"]
+    assert context.spotify_client.calls == []
+    assert itunes_client.calls == []
+
+    assert [track["track_number"] for track in missing_tracks] == [2]
+    assert missing_tracks[0]["source"] == "deezer"
+    assert missing_tracks[0]["source_track_id"] == "dz-2"
+    assert missing_tracks[0]["spotify_track_id"] == "dz-2"
+
+
+def test_album_completeness_supports_discogs_primary(monkeypatch):
+    job = AlbumCompletenessJob()
+    context = types.SimpleNamespace(
+        db=_FakeDB(owned_track_numbers={1}),
+        spotify_client=_FakeSpotifyClient(track_count=5),
+        is_spotify_rate_limited=lambda: False,
+    )
+
+    discogs_client = _FakeDiscogsClient(track_count=3)
+    itunes_client = _FakeITunesClient()
+    deezer_client = _FakeDeezerClient(track_count=2)
+    calls = []
+    album_ids = {
+        "spotify": "spotify-album",
+        "itunes": "itunes-album",
+        "deezer": "deezer-album",
+        "discogs": "discogs-release",
+        "hydrabase": "",
+    }
+
+    monkeypatch.setattr(album_completeness_module, "get_primary_source", lambda: "discogs")
+    monkeypatch.setattr(
+        album_completeness_module,
+        "get_album_tracks_for_source",
+        lambda source, album_id: (
+            calls.append((source, album_id)) or
+            (
+                discogs_client.get_album_tracks(album_id) if source == "discogs" else
+                itunes_client.get_album_tracks(album_id) if source == "itunes" else
+                deezer_client.get_album_tracks(album_id) if source == "deezer" else
+                context.spotify_client.get_album_tracks(album_id) if source == "spotify" else
+                {"items": []}
+            )
+        )
+    )
+
+    expected_total = job._get_expected_total(context, "discogs", album_ids)
+    missing_tracks = job._find_missing_tracks(context, "discogs", 42, album_ids)
+
+    assert expected_total == 3
+    assert calls == [("discogs", "discogs-release"), ("discogs", "discogs-release")]
+    assert discogs_client.calls == ["discogs-release", "discogs-release"]
+    assert context.spotify_client.calls == []
+    assert itunes_client.calls == []
+    assert deezer_client.calls == []
+
+    assert [track["track_number"] for track in missing_tracks] == [2, 3]
+    assert missing_tracks[0]["source"] == "discogs"
+    assert missing_tracks[0]["source_track_id"] == "dg-2"
+
+
+def test_album_completeness_supports_hydrabase_primary(monkeypatch):
+    job = AlbumCompletenessJob()
+    context = types.SimpleNamespace(
+        db=_FakeDB(owned_track_numbers={1}),
+        spotify_client=_FakeSpotifyClient(track_count=5),
+        is_spotify_rate_limited=lambda: False,
+    )
+
+    hydrabase_client = _FakeHydrabaseClient(track_count=4)
+    itunes_client = _FakeITunesClient()
+    deezer_client = _FakeDeezerClient(track_count=2)
+    calls = []
+    album_ids = {
+        "spotify": "spotify-album",
+        "itunes": "itunes-album",
+        "deezer": "deezer-album",
+        "discogs": "",
+        "hydrabase": "soul-album",
+    }
+
+    monkeypatch.setattr(album_completeness_module, "get_primary_source", lambda: "hydrabase")
+    monkeypatch.setattr(
+        album_completeness_module,
+        "get_album_tracks_for_source",
+        lambda source, album_id: (
+            calls.append((source, album_id)) or
+            (
+                hydrabase_client.get_album_tracks_dict(album_id) if source == "hydrabase" else
+                itunes_client.get_album_tracks(album_id) if source == "itunes" else
+                deezer_client.get_album_tracks(album_id) if source == "deezer" else
+                context.spotify_client.get_album_tracks(album_id) if source == "spotify" else
+                {"items": []}
+            )
+        )
+    )
+
+    expected_total = job._get_expected_total(context, "hydrabase", album_ids)
+    missing_tracks = job._find_missing_tracks(context, "hydrabase", 42, album_ids)
+
+    assert expected_total == 4
+    assert calls == [("hydrabase", "soul-album"), ("hydrabase", "soul-album")]
+    assert hydrabase_client.calls == ["soul-album", "soul-album"]
+    assert context.spotify_client.calls == []
+    assert itunes_client.calls == []
+    assert deezer_client.calls == []
+
+    assert [track["track_number"] for track in missing_tracks] == [2, 3, 4]
+    assert missing_tracks[0]["source"] == "hydrabase"
+    assert missing_tracks[0]["source_track_id"] == "hy-2"

webui/static/script.js

@@ -64051,7 +64051,15 @@ function _renderFindingDetail(f) {
         case 'incomplete_album':
             if (d.artist) rows.push(['Artist', d.artist]);
             if (d.album_title) rows.push(['Album', d.album_title]);
-            if (d.spotify_album_id) rows.push(['Spotify ID', d.spotify_album_id]);
+            if (d.primary_source && d.primary_album_id) {
+                const primaryLabel = d.primary_source.charAt(0).toUpperCase() + d.primary_source.slice(1);
+                rows.push([`${primaryLabel} ID`, d.primary_album_id]);
+                if (d.spotify_album_id && d.primary_source !== 'spotify') {
+                    rows.push(['Spotify ID', d.spotify_album_id]);
+                }
+            } else if (d.spotify_album_id) {
+                rows.push(['Spotify ID', d.spotify_album_id]);
+            }
             let incHtml = media + _gridRows(rows);
             const actual = d.actual_tracks || 0, expected = d.expected_tracks || 0;
             if (expected > 0) {
@@ -64065,6 +64073,7 @@ function _renderFindingDetail(f) {
                 incHtml += `<div class="repair-detail-sublist">${d.missing_tracks.map(t => `
                     <div class="repair-detail-subitem">
                         <strong>#${t.track_number || '?'} ${_escFinding(t.name || t.title || 'Unknown')}</strong>
+                        ${t.source && t.source !== 'spotify' ? `<span>Source: ${_escFinding(t.source)}${t.source_track_id ? ` · ID: ${_escFinding(t.source_track_id)}` : ''}</span>` : ''}
                         ${t.duration_ms ? `<span>Duration: ${Math.round(t.duration_ms / 1000)}s</span>` : ''}
                     </div>`).join('')}</div>`;
             }