Skip to content

Commit

Permalink
Merge pull request #230 from UIUC-Chatbot/refactor-export-fix
Browse files Browse the repository at this point in the history
Conversation Export Bug Fix
  • Loading branch information
star-nox authored Mar 15, 2024
2 parents 34ad712 + 2107aae commit cac7315
Showing 1 changed file with 6 additions and 5 deletions.
11 changes: 6 additions & 5 deletions ai_ta_backend/service/export_service.py
Original file line number Diff line number Diff line change
Expand Up @@ -35,15 +35,16 @@ def export_documents_json(self, course_name: str, from_date='', to_date=''):

response = self.sql.getDocumentsBetweenDates(course_name, from_date, to_date, 'documents')
# add a condition to route to direct download or s3 download
if response.count > 1000:
if response.count > 500:
# call background task to upload to s3

filename = course_name + '_' + str(uuid.uuid4()) + '_documents.zip'
s3_filepath = f"courses/{course_name}/{filename}"
# background task of downloading data - map it with above ID
executor = ProcessPoolExecutor()
executor.submit(self.export_data_in_bg, response, "documents", course_name, s3_filepath)
return {"response": 'Download from S3', "s3_path": s3_filepath}
return {"response": 'Download from S3',
"s3_path": s3_filepath}

else:
# Fetch data
Expand Down Expand Up @@ -206,11 +207,11 @@ def export_convo_history_json(self, course_name: str, from_date='', to_date=''):
from_date (str, optional): The start date for the data export. Defaults to ''.
to_date (str, optional): The end date for the data export. Defaults to ''.
"""
print("Exporting conversation history to csv file...")
print("Exporting conversation history to json file...")

response = self.sql.getDocumentsBetweenDates(course_name, from_date, to_date, 'llm-convo-monitor')

if response.count > 1000:
if response.count > 500:
# call background task to upload to s3
filename = course_name + '_' + str(uuid.uuid4()) + '_convo_history.zip'
s3_filepath = f"courses/{course_name}/{filename}"
Expand All @@ -226,7 +227,7 @@ def export_convo_history_json(self, course_name: str, from_date='', to_date=''):
last_id = response.data[-1]['id']
total_count = response.count

filename = course_name + '_' + str(uuid.uuid4()) + '_convo_history.csv'
filename = course_name + '_' + str(uuid.uuid4()) + '_convo_history.json'
file_path = os.path.join(os.getcwd(), filename)
curr_count = 0
# Fetch data in batches of 25 from first_id to last_id
Expand Down

0 comments on commit cac7315

Please sign in to comment.