get all films bug fix

LoadBalancer.py

@@ -30,12 +30,17 @@ class LoadBalancer:
         self.file_structure = None
         self.previous_file_structure = None  # To keep track of previous content
 
+        # Caches for films and TV shows to avoid returning empty lists in case of errors
+        self.cached_films = []
+        self.cached_tv_shows = {}
+
         # Ensure CACHE_DIR exists
         if not os.path.exists(self.CACHE_DIR):
             os.makedirs(self.CACHE_DIR)
 
-        # Initialize file structure and start prefetching
+        # Initialize file structure, update caches, and start prefetching
         self.file_structure = indexer()
+        self.update_media_cache()
         self.start_prefetching()
 
         # Start polling and file checking in separate threads
@@ -46,11 +51,51 @@ class LoadBalancer:
         # Start periodic tasks
         asyncio.create_task(self.run_periodic_tasks())
 
+    def update_media_cache(self):
+        """Update the cached films and TV shows from the current file_structure.
+           Only update if the new data is non-empty, preserving the last valid cache.
+        """
+        new_films = []
+        new_tv_shows = {}
+
+        for directory in self.file_structure:
+            if directory.get('type') == 'directory':
+                if directory.get('path') == 'films':
+                    for sub_directory in directory.get('contents', []):
+                        if sub_directory.get('type') == 'directory':
+                            new_films.append(sub_directory.get('path'))
+                elif directory.get('path') == 'tv':
+                    for sub_directory in directory.get('contents', []):
+                        if sub_directory.get('type') == 'directory':
+                            show_title = sub_directory.get('path').split('/')[-1]
+                            episodes_list = []
+                            for season_directory in sub_directory.get('contents', []):
+                                if season_directory.get('type') == 'directory':
+                                    season = season_directory.get('path').split('/')[-1]
+                                    for episode in season_directory.get('contents', []):
+                                        if episode.get('type') == 'file':
+                                            episodes_list.append({
+                                                "season": season,
+                                                "episode": episode.get('path').split('/')[-1],
+                                                "path": episode.get('path')
+                                            })
+                            if episodes_list:
+                                new_tv_shows[show_title] = episodes_list
+
+        if new_films:
+            self.cached_films = new_films
+        if new_tv_shows:
+            self.cached_tv_shows = new_tv_shows
+
     async def run_periodic_tasks(self):
         """Run indexer and prefetch functions every 5 minutes."""
         while not self.stop_event.is_set():
-            self.file_structure = indexer()  # Re-run indexer
-            await self.start_prefetching()  # Start prefetching
+            new_file_structure = indexer()  # Re-run indexer
+            # Only update if the new file_structure is non-empty
+            if new_file_structure:
+                self.file_structure = new_file_structure
+                self.update_media_cache()
+            await self.start_prefetching()  # Start prefetching metadata
             await asyncio.sleep(300)  # Sleep for 5 minutes
 
     def start_prefetching(self):
@@ -85,7 +130,8 @@ class LoadBalancer:
                     json_cache_path = os.path.join(self.CACHE_DIR, f"{urllib.parse.quote(original_title)}.json")
                     if not os.path.exists(json_cache_path):
                         tasks.append(fetch_and_cache_json(original_title, title, media_type, year))
-                logging.info(f"Skipping.. {original_title} metadata already cached")
+                    else:
+                        logging.info(f"Skipping.. {original_title} metadata already cached")
         
         # Run all tasks concurrently
         await asyncio.gather(*tasks)
@@ -160,22 +206,13 @@ class LoadBalancer:
             logging.info(f"Instance {instance_url} not found for removal.")
 
     def update_instances_health(self, instance, cache_size):
-        self.instances_health[instance] = {"used":cache_size["cache_size"],
+        self.instances_health[instance] = {"used": cache_size["cache_size"],
                                            "total": "50 GB"}
         logging.info(f"Updated instance {instance} with cache size {cache_size}")
 
     def download_film_to_best_instance(self, title):
         """
-        Downloads a film to the first instance that has more free space on the self.instance_health list variable. 
-        The instance_health looks like this:
-        {
-            "https://unicone-studio-instance1.hf.space": {
-                "total": "50 GB",
-                "used": "3.33 GB"
-            }
-        }
-        Args:
-            title (str): The title of the film.
+        Downloads a film to the first instance that has more free space on the self.instances_health list.
         """
         best_instance = None
         max_free_space = -1
@@ -196,11 +233,10 @@ class LoadBalancer:
             status = result["status"]
             progress_url = f'{best_instance}/api/get/progress/{film_id}'
             response = {
-                "film_id":film_id,
-                "status":status,
-                "progress_url":progress_url
+                "film_id": film_id,
+                "status": status,
+                "progress_url": progress_url
             }
-
             return response
         else:
             logging.error("No suitable instance found for downloading the film.")
@@ -208,18 +244,7 @@ class LoadBalancer:
 
     def download_episode_to_best_instance(self, title, season, episode):
         """
-        Downloads a episode to the first instance that has more free space on the self.instance_health list variable. 
-        The instance_health looks like this:
-        {
-            "https://unicone-studio-instance1.hf.space": {
-                "total": "50 GB",
-                "used": "3.33 GB"
-            }
-        }
-        Args:
-            title (str): The title of the Tv show.
-            season (str): The season of the Tv show.
-            episode (str): The title of the Tv show.
+        Downloads an episode to the first instance that has more free space on the self.instances_health list.
         """
         best_instance = None
         max_free_space = -1
@@ -240,11 +265,10 @@ class LoadBalancer:
             status = result["status"]
             progress_url = f'{best_instance}/api/get/progress/{episode_id}'
             response = {
-                "episode_id":episode_id,
-                "status":status,
-                "progress_url":progress_url
+                "episode_id": episode_id,
+                "status": status,
+                "progress_url": progress_url
             }
-
             return response
         else:
             logging.error("No suitable instance found for downloading the film.")
@@ -253,29 +277,29 @@ class LoadBalancer:
     def find_movie_path(self, title):
         """Find the path of the movie in the JSON data based on the title."""
         for directory in self.file_structure:
-            if directory['type'] == 'directory' and directory['path'] == 'films':
-                for sub_directory in directory['contents']:
-                    if sub_directory['type'] == 'directory':
-                        for item in sub_directory['contents']:
-                            if item['type'] == 'file' and title.lower() in item['path'].lower():
-                                return item['path']
+            if directory.get('type') == 'directory' and directory.get('path') == 'films':
+                for sub_directory in directory.get('contents', []):
+                    if sub_directory.get('type') == 'directory':
+                        for item in sub_directory.get('contents', []):
+                            if item.get('type') == 'file' and title.lower() in item.get('path').lower():
+                                return item.get('path')
         return None
 
     def find_tv_path(self, title):
         """Find the path of the TV show in the JSON data based on the title."""
         for directory in self.file_structure:
-            if directory['type'] == 'directory' and directory['path'] == 'tv':
-                for sub_directory in directory['contents']:
-                    if sub_directory['type'] == 'directory' and title.lower() in sub_directory['path'].lower():
-                        return sub_directory['path']
+            if directory.get('type') == 'directory' and directory.get('path') == 'tv':
+                for sub_directory in directory.get('contents', []):
+                    if sub_directory.get('type') == 'directory' and title.lower() in sub_directory.get('path').lower():
+                        return sub_directory.get('path')
         return None
 
     def get_tv_structure(self, title):
         """Find the path of the TV show in the JSON data based on the title."""
         for directory in self.file_structure:
-            if directory['type'] == 'directory' and directory['path'] == 'tv':
-                for sub_directory in directory['contents']:
-                    if sub_directory['type'] == 'directory' and title.lower() in sub_directory['path'].lower():
+            if directory.get('type') == 'directory' and directory.get('path') == 'tv':
+                for sub_directory in directory.get('contents', []):
+                    if sub_directory.get('type') == 'directory' and title.lower() in sub_directory.get('path').lower():
                         return sub_directory
         return None
 
@@ -284,32 +308,9 @@ class LoadBalancer:
         return title.replace(" ", "_").lower()
 
     def get_all_tv_shows(self):
-        """Get all TV shows from the indexed cache structure JSON file."""
-        tv_shows = {}
-        for directory in self.file_structure:
-            if directory['type'] == 'directory' and directory['path'] == 'tv':
-                for sub_directory in directory['contents']:
-                    if sub_directory['type'] == 'directory':
-                        show_title = sub_directory['path'].split('/')[-1]
-                        tv_shows[show_title] = []
-                        for season_directory in sub_directory['contents']:
-                            if season_directory['type'] == 'directory':
-                                season = season_directory['path'].split('/')[-1]
-                                for episode in season_directory['contents']:
-                                    if episode['type'] == 'file':
-                                        tv_shows[show_title].append({
-                                            "season": season,
-                                            "episode": episode['path'].split('/')[-1],
-                                            "path": episode['path']
-                                        })
-        return tv_shows
+        """Return the cached TV shows."""
+        return self.cached_tv_shows
 
     def get_all_films(self):
-        """Get all films from the indexed cache structure JSON file."""
-        films = []
-        for directory in self.file_structure:
-            if directory['type'] == 'directory' and directory['path'] == 'films':
-                for sub_directory in directory['contents']:
-                    if sub_directory['type'] == 'directory':
-                        films.append(sub_directory['path'])
-        return films
+        """Return the cached films."""
+        return self.cached_films

TODO.md

@@ -1,4 +1,2 @@
-create endpoints for
-* /api/get/film/{title} `Endpoint to get the movie by title.`
-
-* /api/get/tv/{title}/{season}/{episode} `Endpoint to get the episode by title, season and episode.`
+Fix BUGS
+* /api/get/film/all `After some times the list becomes empty`