Marmelator/cv-analysis-service
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 37 Wiki Activity

update serve.py to work with new pyinfra version

Browse Source
This commit is contained in:
Francisco Schulz 2023-02-16 10:47:13 +01:00
parent ad45e2c1da
commit eef371e2a8
1 changed files with 8 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
src/serve.py
View File

@ -22,27 +22,29 @@ def analysis_callback(queue_message: dict):
"dossierId", "fileId", "targetFileExtension", "responseFileExtension", "operation"
)(queue_message)
bucket = PYINFRA_CONFIG.storage_bucket
logging.info(f"Processing {dossier_id=}/{file_id=}, {operation=}.")
logging.info("running operation %s file_id=%s and dossier_id=%s", operation, file_id, dossier_id)
storage = get_storage(PYINFRA_CONFIG)
object_name = f"{dossier_id}/{file_id}.{target_file_ext}"
if storage.exists(bucket, object_name):
should_publish_result = True
object_bytes = gzip.decompress(storage.get_object(bucket, object_name))
analysis_fn = get_analysis_pipeline(operation, CV_CONFIG.table_parsing_skip_pages_without_images)
results = analysis_fn(object_bytes)
logging.info("predictions ready for file_id=%s and dossier_id=%s", file_id, dossier_id)
response = {**queue_message, "data": list(results)}
response = gzip.compress(json.dumps(response).encode())
response_name = f"{dossier_id}/{file_id}.{response_file_ext}"
logging.info("storing predictions for file_id=%s and dossier_id=%s", file_id, dossier_id)
storage.put_object(bucket, response_name, response)
else:
should_publish_result = False
return {"dossierId": dossier_id, "fileId": file_id}
return should_publish_result, {"dossierId": dossier_id, "fileId": file_id}
else:
return None
if __name__ == "__main__":