junte/Model
1
0
Fork 0
mirror of https://github.com/Hestia-Homes/Model.git synced 2026-06-08 11:17:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

don't bombard yet

Browse source
This commit is contained in:
Jun-te Kim 2026-02-12 18:05:29 +00:00
parent d07fc351a5
commit dac676f538
1 changed files with 13 additions and 10 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

23
backend/postcode_splitter/main.py
View file

@ -63,7 +63,9 @@ def parse_s3_uri(s3_uri: str) -> tuple[str, str]:
raise ValueError(f"Could not parse S3 URI") from e
def upload_batch_to_s3(batch_df: pd.DataFrame, task_id: str, sub_task_id: str, bucket_name: str = None) -> str:
def upload_batch_to_s3(
batch_df: pd.DataFrame, task_id: str, sub_task_id: str, bucket_name: str = None
) -> str:
"""
Upload batch DataFrame to S3 as CSV.
@ -87,7 +89,9 @@ def upload_batch_to_s3(batch_df: pd.DataFrame, task_id: str, sub_task_id: str, b
try:
file_name = f"{datetime.now().isoformat()}_{str(uuid4())[:8]}"
file_key = f"ara_postcode_splitter_batches/{task_id}/{sub_task_id}/{file_name}.csv"
file_key = (
f"ara_postcode_splitter_batches/{task_id}/{sub_task_id}/{file_name}.csv"
)
success = save_csv_to_s3(batch_df, bucket_name, file_key)
@ -128,10 +132,11 @@ def send_to_address2uprn_queue(task_id: str, sub_task_id: str, s3_uri: str) -> s
"s3_uri": s3_uri,
}
response = sqs_client.send_message(
QueueUrl=queue_url,
MessageBody=json.dumps(message_body),
)
# Don't run on sqs yet
# response = sqs_client.send_message(
# QueueUrl=queue_url,
# MessageBody=json.dumps(message_body),
# )
logger.info(
f"Sent message to address2UPRN queue. "
@ -174,7 +179,7 @@ def create_batch_and_send_to_address2uprn(
"sub_task_id": batch_sub_task_id,
"batch_size": len(batch_rows),
"s3_uri": s3_uri,
}
},
)
logger.info(f"Created batch subtask {created_batch_sub_task_id}")
@ -241,9 +246,7 @@ def handler(event, context):
logger.info(f"Created subtask {subtask_id} for task {task_id}")
# Mark subtask as in progress
subtask_interface.update_subtask_status(
subtask_id, "in progress"
)
subtask_interface.update_subtask_status(subtask_id, "in progress")
logger.info(f"Marked subtask {subtask_id} as in progress")
# Read CSV from S3