apunkt/SurfSense
1
0
Fork 0
mirror of https://github.com/MODSetter/SurfSense.git synced 2026-05-01 11:56:25 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Merge pull request #336 from sandeeppainuly/fix-294-youtube-transcript-source-not-working-youtubetranscriptapi-has-no-attribute-get-transcript

Browse source 
Fix YouTube transcript API: replace deprecated get_transcript with fetch method
This commit is contained in:
Rohan Verma 2025-09-28 20:51:52 -07:00 • committed by GitHub
commit 4e012845c4
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 5 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

9
surfsense_backend/app/tasks/document_processors/youtube_processor.py
View file

@ -136,13 +136,14 @@ async def add_youtube_video_document(
) )
try: try:
captions = YouTubeTranscriptApi.get_transcript(video_id) ytt_api = YouTubeTranscriptApi()
captions = ytt_api.fetch(video_id)
# Include complete caption information with timestamps # Include complete caption information with timestamps
transcript_segments = [] transcript_segments = []
for line in captions: for line in captions:
start_time = line.get("start", 0) start_time = line.start
duration = line.get("duration", 0) duration = line.duration
text = line.get("text", "") text = line.text
timestamp = f"[{start_time:.2f}s-{start_time + duration:.2f}s]" timestamp = f"[{start_time:.2f}s-{start_time + duration:.2f}s]"
transcript_segments.append(f"{timestamp} {text}") transcript_segments.append(f"{timestamp} {text}")
transcript_text = "\n".join(transcript_segments) transcript_text = "\n".join(transcript_segments)