Optimize TV processor performance with database-first episode lookup

Major performance improvement for manual scans:
- Check NFOGuard database before making Sonarr API calls
- Only query Sonarr history for episodes missing from database
- Filter Sonarr episode processing to only needed episodes
- Add detailed logging for cache hits vs API calls

Expected improvement: 172 API calls -> 0-10 API calls for typical scans

VERSION

@@ -1 +1 @@
-2.0.19
+2.0.20

processors/tv_processor.py

@@ -102,47 +102,70 @@ class TVProcessor:
     
     
     def _gather_episode_dates(self, series_path: Path, imdb_id: str, disk_episodes: Dict[Tuple[int, int], List[Path]]) -> Dict[Tuple[int, int], Tuple[Optional[str], Optional[str], str]]:
-        """Gather episode air dates and date added information"""
+        """Gather episode air dates and date added information with database-first optimization"""
         episode_dates = {}
+        episodes_needing_lookup = []
         
-        # Get data from Sonarr first if available
+        # OPTIMIZATION: Check database first for existing dates
-        sonarr_episodes = self._get_sonarr_episodes(imdb_id)
+        _log("DEBUG", f"Checking database for existing episode dates for {len(disk_episodes)} episodes")
+        db_cache_hits = 0
         
-        # Get data from external sources for missing information
         for (season, episode) in disk_episodes:
-            aired = None
+            # Try database first - this is much faster than API calls
-            dateadded = None
+            db_result = self.db.get_episode_date(imdb_id, season, episode)
-            source = "unknown"
             
-            # Try Sonarr first
+            if db_result and db_result.get('dateadded'):
-            if (season, episode) in sonarr_episodes:
+                # Found in database - use cached data
-                sonarr_data = sonarr_episodes[(season, episode)]
+                aired = db_result.get('aired')
-                aired = sonarr_data.get('airDate')
+                dateadded = db_result.get('dateadded')
-                dateadded = sonarr_data.get('dateAdded')
+                source = db_result.get('source', 'database_cache')
-                if dateadded:
+                episode_dates[(season, episode)] = (aired, dateadded, source)
-                    # Use more specific source to indicate this came from Sonarr import history
+                db_cache_hits += 1
-                    source = "sonarr:history.import"
+                _log("DEBUG", f"Database cache hit for S{season:02d}E{episode:02d}: {dateadded}")
+            else:
+                # Not in database or incomplete - needs lookup
+                episodes_needing_lookup.append((season, episode))
         
-            # Fallback to external sources if needed
+        _log("INFO", f"Database cache hits: {db_cache_hits}/{len(disk_episodes)} episodes. Need API lookup: {len(episodes_needing_lookup)}")
-            if not aired:
-                external_aired = self.external_clients.get_episode_air_date(imdb_id, season, episode)
-                if external_aired:
-                    aired = external_aired
-                    if not dateadded:
-                        source = "external"
         
-            # ONLY use air date as fallback for dateadded if we have no Sonarr import date at all
+        # Only call Sonarr API for episodes not in database
-            # and if we couldn't get any import information from Sonarr history
+        if episodes_needing_lookup:
-            if not dateadded and aired and source != "sonarr:history.import":
+            _log("DEBUG", f"Querying Sonarr for {len(episodes_needing_lookup)} episodes missing from database")
-                dateadded = aired
+            sonarr_episodes = self._get_sonarr_episodes(imdb_id, episodes_needing_lookup)
-                source = f"{source}_fallback" if source != "unknown" else "aired_fallback"
             
-            episode_dates[(season, episode)] = (aired, dateadded, source)
+            # Process episodes that needed lookup
+            for (season, episode) in episodes_needing_lookup:
+                aired = None
+                dateadded = None
+                source = "unknown"
+                
+                # Try Sonarr first
+                if (season, episode) in sonarr_episodes:
+                    sonarr_data = sonarr_episodes[(season, episode)]
+                    aired = sonarr_data.get('airDate')
+                    dateadded = sonarr_data.get('dateAdded')
+                    if dateadded:
+                        source = "sonarr:history.import"
+                
+                # Fallback to external sources if needed
+                if not aired:
+                    external_aired = self.external_clients.get_episode_air_date(imdb_id, season, episode)
+                    if external_aired:
+                        aired = external_aired
+                        if not dateadded:
+                            source = "external"
+                
+                # Use air date as fallback for dateadded if no import date found
+                if not dateadded and aired and source != "sonarr:history.import":
+                    dateadded = aired
+                    source = f"{source}_fallback" if source != "unknown" else "aired_fallback"
+                
+                episode_dates[(season, episode)] = (aired, dateadded, source)
         
         return episode_dates
     
-    def _get_sonarr_episodes(self, imdb_id: str) -> Dict[Tuple[int, int], Dict[str, Any]]:
+    def _get_sonarr_episodes(self, imdb_id: str, episodes_filter: List[Tuple[int, int]] = None) -> Dict[Tuple[int, int], Dict[str, Any]]:
-        """Get episode information from Sonarr including import history"""
+        """Get episode information from Sonarr including import history - optimized to only fetch needed episodes"""
         try:
             series_data = self.sonarr.series_by_imdb(imdb_id)
             if not series_data:
@@ -154,11 +177,20 @@ class TVProcessor:
             
             episodes = self.sonarr.episodes_for_series(series_id)
             
+            # Convert episodes_filter to set for faster lookup
+            filter_set = set(episodes_filter) if episodes_filter else None
+            
             episode_map = {}
+            api_calls_made = 0
+            
             for episode in episodes:
                 season = episode.get('seasonNumber', 0)
                 episode_num = episode.get('episodeNumber', 0)
                 
+                # Skip episodes not in filter if filter is provided
+                if filter_set and (season, episode_num) not in filter_set:
+                    continue
+                
                 if season > 0 and episode_num > 0:
                     # Get basic episode info
                     episode_data = {
@@ -170,6 +202,7 @@ class TVProcessor:
                     episode_id = episode.get('id')
                     if episode_id and episode.get('hasFile'):
                         import_date = self.sonarr.get_episode_import_history(episode_id)
+                        api_calls_made += 1
                         if import_date:
                             episode_data['dateAdded'] = import_date
                             _log("DEBUG", f"Got import date from history for S{season:02d}E{episode_num:02d}: {import_date}")
@@ -183,6 +216,9 @@ class TVProcessor:
                     
                     episode_map[(season, episode_num)] = episode_data
             
+            if filter_set:
+                _log("DEBUG", f"Made {api_calls_made} Sonarr history API calls for filtered episodes (instead of all episodes)")
+            
             return episode_map
             
         except Exception as e: