Upload folder using huggingface_hub

modules/dropbox/audio.py

@@ -82,6 +82,8 @@ def list_dropbox_folder_hierarchy(dbx: dropbox.Dropbox, base_path: str = ""):
 audio_cache: dict[tuple[str, int, str], dict] = {}
 CACHE_TTL = timedelta(hours=3, minutes=30)  # refresh before 4h expiry
 
+AUDIO_LIST_CACHE_TTL = timedelta(hours=24)
+audio_list_cache = {}  # {(scripture_name): {"entries": [...], "expiry": datetime}}
 
 async def get_audio_urls(req: AudioRequest):
     base_path = f"/{req.scripture_name}/audio"
@@ -90,20 +92,29 @@ async def get_audio_urls(req: AudioRequest):
     urls = {}
     now = datetime.now(timezone.utc)
 
-    try:
-        # Initial listing (first batch of up to 500)
-        result = dbx.files_list_folder(base_path)
-        entries = result.entries
+    # --- 1️⃣ Check if folder listing is cached ---
+    cache_entry = audio_list_cache.get(req.scripture_name)
+    if cache_entry and cache_entry["expiry"] > now:
+        entries = cache_entry["entries"]
+    else:
+        # Fetch fresh listing from Dropbox
+        try:
+            result = dbx.files_list_folder(base_path)
+            entries = result.entries
 
-        # Continue listing if there are more entries
-        while result.has_more:
-            result = dbx.files_list_folder_continue(result.cursor)
-            entries.extend(result.entries)
+            while result.has_more:
+                result = dbx.files_list_folder_continue(result.cursor)
+                entries.extend(result.entries)
 
-    except dropbox.exceptions.ApiError:
-        raise HTTPException(status_code=404, detail="Audio directory not found")
+            audio_list_cache[req.scripture_name] = {
+                "entries": entries,
+                "expiry": now + AUDIO_LIST_CACHE_TTL,
+            }
 
-    # Filter files matching the prefix
+        except dropbox.exceptions.ApiError:
+            raise HTTPException(status_code=404, detail="Audio directory not found")
+
+    # --- 2️⃣ Filter matching files ---
     matching_files = [
         entry
         for entry in entries
@@ -113,10 +124,10 @@ async def get_audio_urls(req: AudioRequest):
     if not matching_files:
         raise HTTPException(status_code=404, detail="No audio files found")
 
+    # --- 3️⃣ Generate or reuse cached URLs ---
     for entry in matching_files:
         filename = entry.name
-        file_type = filename[len(prefix) :].rsplit(".", 1)[0]
-
+        file_type = filename[len(prefix):].rsplit(".", 1)[0]
         cache_key = (req.scripture_name, req.global_index, file_type)
 
         cached = audio_cache.get(cache_key)
@@ -135,6 +146,7 @@ async def get_audio_urls(req: AudioRequest):
     return urls
 
 
+
 async def cleanup_audio_url_cache(interval_seconds: int = 600):
     """Periodically remove expired entries from audio_cache."""
     while True: