Update module and add Supadata fallback option

stellarnode · stellarnode · commit 4f02febc2f7d · 2025-06-19T11:36:00.000+03:00
diff --git a/youtubevideotranscriptbot/Dockerfile_template b/youtubevideotranscriptbot/Dockerfile_template
@@ -1,6 +1,6 @@
 # This is Dockerfile
 # Use an official Python runtime as a parent image
-FROM python:3.9-slim
+FROM python:3.10-slim
 
 # Set the working directory in the container
 WORKDIR /app
diff --git a/youtubevideotranscriptbot/requirements.txt b/youtubevideotranscriptbot/requirements.txt
@@ -1,6 +1,6 @@
 python-telegram-bot==20.3
 google-api-python-client==2.104.0
-youtube-transcript-api==1.0.3 # 1.0.3 # 0.6.2 #1.0.3 # 0.6.1
+youtube-transcript-api==1.1.0 # 1.0.3 # 0.6.2 #1.0.3 # 0.6.1
 deep-translator==1.11.4
 mysql-connector-python==8.1.0
 isodate==0.6.1
@@ -9,3 +9,5 @@ tiktoken==0.9.0
 deepseek==1.0.0
 aiofiles>=0.4.0
 amplitude-analytics==1.1.5
+supadata==1.2.1
+
diff --git a/youtubevideotranscriptbot/run_bot.sh b/youtubevideotranscriptbot/run_bot.sh
@@ -14,6 +14,8 @@ fi
 # Build the Docker image
 echo "Building the Docker image..."
 docker build -t ${IMAGE_TAG} .
+# Use the line below if you want to build without cache
+# docker build --no-cache -t ${IMAGE_TAG} .
 
 # Run the container with the volume and auto-restart
 echo "Starting the container..."
diff --git a/youtubevideotranscriptbot/transcript.py b/youtubevideotranscriptbot/transcript.py
@@ -7,7 +7,8 @@
 import os
 import logging
 from translate import translate_text  
-from config import MODEL_TO_USE 
+from config import MODEL_TO_USE, SUPADATA_API_KEY
+from supadata import Supadata, SupadataError
 from config import PROXY_USERNAME
 from config import PROXY_PASSWORD
 from model_params import get_model_params
@@ -153,11 +154,11 @@ async def save_transcripts(transcript_list, base_filename, transcript_properties
         #     continue
 
         logger.info(f"Fetching transcript for language: {language} ({normalized_language_code})")
-        logger.info(f"[Get Single Transcript] Attempting to fetch with(out) proxy")
+        logger.info(f"[Get Single Transcript] Attempting to fetch without proxy")
         
         try:
             # transcript_data = transcript.fetch()
-            transcript_data = ytt_api_proxied.fetch(transcript_properties.get('video_id'), languages=[language_code])
+            transcript_data = ytt_api.fetch(transcript_properties.get('video_id'), languages=[language_code])
             logger.info(f"Transcript data from fetch attempt: {str(transcript_data)[:500]}")
         except Exception as e:
             logger.warning(f"Fetch without proxy failed for {language}: {e}")
@@ -185,7 +186,23 @@ async def save_transcripts(transcript_list, base_filename, transcript_properties
                         transcript_data = None
 
         if not transcript_data:
-            continue
+            logger.error(f"Failed to fetch transcript for {language} with proxy after all attempts.")
+            logger.info(f"[Supadata] Trying to fetch the transcript with Supadata for {language}.")
+
+            try:
+                # Initialize the client
+                supadata = Supadata(api_key=SUPADATA_API_KEY)
+                # Get YouTube transcript with Spanish language preference
+                transcript_data = supadata.youtube.transcript(video_id=transcript_properties.get('video_id'), lang=language_code)
+                if transcript_data:
+                    logger.info(f"[Supadata] Got transcript for {language_code} from Supadata: {str(transcript_data)[:500]}")
+                    transcript_data = transcript_data.content
+                else:
+                    logger.error(f"[Supadata] No transcript data found for {language_code}.")
+                    continue
+            except SupadataError as e:
+                logger.error(f"[Supadata] Failed to fetch transcript from Supadata for {language_code}: {e}")
+                continue
 
         try:
             formatted_transcript = formatter.format_transcript(transcript_data)
diff --git a/youtubevideotranscriptbot/youtube_api.py b/youtubevideotranscriptbot/youtube_api.py
@@ -43,4 +43,5 @@ def get_channel_subscribers(channel_id):
     response = request.execute()
     if response['items']:
         return response['items'][0]['statistics']['subscriberCount']
-    return 'N/A'
+    return 'N/A'
+
