Update app.py

app.py

@@ -4,6 +4,8 @@ import google.generativeai as genai
 from dotenv import load_dotenv
 from pytube import YouTube
 from PyPDF2 import PdfReader
+from youtube_transcript_api import YouTubeTranscriptApi, TranscriptsDisabled, NoTranscriptFound
+from urllib.parse import urlparse, parse_qs
 
 load_dotenv()
 genai.configure(api_key=os.getenv("GEMINI_API_KEY"))
@@ -75,27 +77,47 @@ Content:\n\n{text[:5000]}
         return f"Error reading PDF: {e}"
 
 
+def get_video_id(url):
+    try:
+        parsed_url = urlparse(url)
+        if parsed_url.hostname in ['www.youtube.com', 'youtube.com']:
+            return parse_qs(parsed_url.query)['v'][0]
+        elif parsed_url.hostname == 'youtu.be':
+            return parsed_url.path[1:]
+    except Exception:
+        return None
+
 def summarize_youtube(url):
     try:
-        yt = YouTube(url)
-        caption = yt.captions.get_by_language_code('en')
+        video_id = get_video_id(url)
+        if not video_id:
+            return "❌ Invalid YouTube URL."
 
-        if not caption:
-            return "❌ No English captions available for this video."
+        # Fetch English transcript
+        transcript_list = YouTubeTranscriptApi.list_transcripts(video_id)
+        transcript = None
+        try:
+            transcript = transcript_list.find_transcript(['en'])  # Try manual transcript first
+        except NoTranscriptFound:
+            transcript = transcript_list.find_transcript(['en']).fetch()  # Try auto-generated
 
-        # Captions are in XML/TTML format – convert to plain text
-        caption_text = caption.generate_srt_captions()
+        transcript_text = " ".join([t['text'] for t in transcript.fetch()])
 
-        if not caption_text.strip():
-            return "❌ Could not extract meaningful caption text."
+        if len(transcript_text.strip()) == 0:
+            return "❌ Transcript is empty."
 
-        prompt = f"Summarize the key educational points in the following YouTube video transcript:\n\n{caption_text[:5000]} in markdown points. Use Bullet points and sections."
+        prompt = f"Summarize the key educational points in the following YouTube video transcript:\n\n{transcript_text[:5000]} in markdown format and Organize into sections."
         response = model.generate_content(prompt)
         return response.text
 
+    except TranscriptsDisabled:
+        return "❌ Transcripts are disabled for this video."
+    except NoTranscriptFound:
+        return "❌ No English transcript available for this video."
     except Exception as e:
         return f"❌ Error summarizing video: {e}"
 
+
 with gr.Blocks() as demo:
     gr.Markdown("# 📘 Study Buddy\nYour smart AI learning assistant!")