Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -488,7 +488,8 @@ def get_youtube_transcript(video_url: str) -> str:
|
|
| 488 |
if not video_id:
|
| 489 |
return f"Error: Could not extract video ID."
|
| 490 |
|
| 491 |
-
|
|
|
|
| 492 |
|
| 493 |
if not transcript_list:
|
| 494 |
return "Error: No transcript found."
|
|
@@ -1320,11 +1321,25 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
|
|
| 1320 |
for item in questions_data:
|
| 1321 |
task_id = item.get("task_id")
|
| 1322 |
question_text = item.get("question")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1323 |
if not task_id or question_text is None:
|
| 1324 |
print(f"Skipping item with missing task_id or question: {item}")
|
| 1325 |
continue
|
| 1326 |
try:
|
| 1327 |
-
|
|
|
|
|
|
|
| 1328 |
answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
|
| 1329 |
results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
|
| 1330 |
except Exception as e:
|
|
@@ -1397,7 +1412,6 @@ with gr.Blocks() as demo:
|
|
| 1397 |
**Disclaimers:**
|
| 1398 |
Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions).
|
| 1399 |
This space provides a basic setup and is intentionally sub-optimal to encourage you to develop your own, more robust solution. For instance for the delay process of the submit button, a solution could be to cache the answers and submit in a seperate action or even to answer the questions in async.
|
| 1400 |
-
Please note that this version requires an OpenAI Key to run.
|
| 1401 |
"""
|
| 1402 |
)
|
| 1403 |
|
|
|
|
| 488 |
if not video_id:
|
| 489 |
return f"Error: Could not extract video ID."
|
| 490 |
|
| 491 |
+
from youtube_transcript_api import YouTubeTranscriptApi
|
| 492 |
+
transcript_list = YouTubeTranscriptApi.list_transcripts(video_id).find_transcript(['en']).fetch()
|
| 493 |
|
| 494 |
if not transcript_list:
|
| 495 |
return "Error: No transcript found."
|
|
|
|
| 1321 |
for item in questions_data:
|
| 1322 |
task_id = item.get("task_id")
|
| 1323 |
question_text = item.get("question")
|
| 1324 |
+
# Download attached files if they exist
|
| 1325 |
+
# Handle file attachments
|
| 1326 |
+
file_info = ""
|
| 1327 |
+
if item.get("file_path"):
|
| 1328 |
+
file_path = item["file_path"]
|
| 1329 |
+
file_info = f"\n\n[Note: Question references file: {file_path}]"
|
| 1330 |
+
|
| 1331 |
+
# Check if file exists
|
| 1332 |
+
if not os.path.exists(file_path):
|
| 1333 |
+
print(f"⚠️ Warning: File {file_path} not found!")
|
| 1334 |
+
file_info += f"\n[WARNING: File not found in current directory]"
|
| 1335 |
+
|
| 1336 |
if not task_id or question_text is None:
|
| 1337 |
print(f"Skipping item with missing task_id or question: {item}")
|
| 1338 |
continue
|
| 1339 |
try:
|
| 1340 |
+
question_with_context = question_text + file_info
|
| 1341 |
+
submitted_answer = agent(question_with_context)
|
| 1342 |
+
#submitted_answer = agent(question_text)
|
| 1343 |
answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
|
| 1344 |
results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
|
| 1345 |
except Exception as e:
|
|
|
|
| 1412 |
**Disclaimers:**
|
| 1413 |
Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions).
|
| 1414 |
This space provides a basic setup and is intentionally sub-optimal to encourage you to develop your own, more robust solution. For instance for the delay process of the submit button, a solution could be to cache the answers and submit in a seperate action or even to answer the questions in async.
|
|
|
|
| 1415 |
"""
|
| 1416 |
)
|
| 1417 |
|