jarif commited on
Commit
28ad143
·
verified ·
1 Parent(s): 6c43c08

Upload app.py

Browse files
Files changed (1) hide show
  1. app.py +22 -6
app.py CHANGED
@@ -69,8 +69,9 @@ Focus on the main points and key insights. Write in a professional tone.
69
  prompt = PromptTemplate(template=prompt_template, input_variables=["text"])
70
 
71
  def get_youtube_content(url):
72
- """Get content from YouTube video using youtube-transcript-api"""
73
  try:
 
74
  from youtube_transcript_api import YouTubeTranscriptApi
75
  from urllib.parse import urlparse, parse_qs
76
 
@@ -82,11 +83,26 @@ def get_youtube_content(url):
82
  else:
83
  raise ValueError("Not a valid YouTube URL")
84
 
85
- # Get the transcript
86
- transcript_list = YouTubeTranscriptApi.get_transcript(video_id)
87
- transcript_text = ' '.join([entry['text'] for entry in transcript_list])
88
-
89
- # Get video info using requests
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90
  response = requests.get(f"https://www.youtube.com/oembed?url=https://www.youtube.com/watch?v={video_id}&format=json")
91
  if response.status_code == 200:
92
  video_info = response.json()
 
69
  prompt = PromptTemplate(template=prompt_template, input_variables=["text"])
70
 
71
  def get_youtube_content(url):
72
+ """Get content from YouTube video"""
73
  try:
74
+ # First try youtube-transcript-api
75
  from youtube_transcript_api import YouTubeTranscriptApi
76
  from urllib.parse import urlparse, parse_qs
77
 
 
83
  else:
84
  raise ValueError("Not a valid YouTube URL")
85
 
86
+ try:
87
+ # Try getting transcript
88
+ transcript_list = YouTubeTranscriptApi.get_transcript(video_id)
89
+ transcript_text = ' '.join([entry['text'] for entry in transcript_list])
90
+ except:
91
+ # Fallback to yt-dlp for description if transcript fails
92
+ ydl_opts = {
93
+ 'quiet': True,
94
+ 'no_warnings': True,
95
+ 'extract_flat': True,
96
+ }
97
+
98
+ with yt_dlp.YoutubeDL(ydl_opts) as ydl:
99
+ try:
100
+ video_info = ydl.extract_info(url, download=False)
101
+ transcript_text = video_info.get('description', 'No description available')
102
+ except:
103
+ transcript_text = "Could not extract video content."
104
+
105
+ # Get video info
106
  response = requests.get(f"https://www.youtube.com/oembed?url=https://www.youtube.com/watch?v={video_id}&format=json")
107
  if response.status_code == 200:
108
  video_info = response.json()