fix bucket name

Author: kendallwerts

.github/workflows/python-app.yml

@@ -16,6 +16,8 @@ jobs:
     env:
       POETRY_VERSION: "1.3.2"
       POETRY_VENV: "/opt/poetry-venv"
+      S3_BUCKET: ${{ secrets.S3_BUCKET }}
+      AWS_DEFAULT_REGION: us-east-2
 
     steps:
     - uses: actions/checkout@v4

flows/transcribe_meetings.py

@@ -1,14 +1,19 @@
 from prefect import flow
+import os
 
 from db.queries import get_meetings
-from tasks.diarize import diarize_meeting
+from tasks.diarize import diarize_meeting, BUCKET_NAME
 
 
 @flow(log_prints=True)
 def transcribe_meetings():
+    print(f"S3_BUCKET environment variable: {os.getenv('S3_BUCKET')}")
+    print(f"BUCKET_NAME from tasks.diarize: {BUCKET_NAME}")
+
     meetings_to_diarize = get_meetings(video=True, s3_path=True)
     print(f"Found {len(meetings_to_diarize)} meetings to diarize")
     for meeting in meetings_to_diarize:
+        print(f"Processing meeting: {meeting.meeting} with s3_path: {meeting.s3_path}")
         diarize_meeting(meeting)
 
 

tasks/diarize.py

@@ -27,6 +27,13 @@ def download_video_and_put_in_s3(meeting: Meeting):
 
 @task
 def diarize_meeting(meeting: Meeting):
+    if BUCKET_NAME is None:
+        raise ValueError("S3_BUCKET environment variable is not set")
+
+    if meeting.s3_path is None:
+        print(f"Meeting {meeting.meeting} has no s3_path, skipping")
+        return
+
     video_file = get_video_from_s3(BUCKET_NAME, meeting.s3_path)
     if video_file:
         run_diarization(video_file, meeting)