Remote Python Executor component previously only returned logs of the script execution,  and did not catch errors systematically, making it difficult to reliably decide if execution should continue.   Sometimes, there is also a need to return a structured payload that is produced in the script.   This task addressed these shortcomings.

• component definition updated to have a required status output attribute.  Implementation will now systematically be populated with either SUCCESS of FAILURE as per the True or False response of the script executed, by the remote python executor.
• component implementation will now populate the optional field response-data with a JSON payload that is returned by the script, using a new Python utility function, (automatically installed in all remote python environments) called send_response_data_payload().

The component will automatically populate the output attribute response-data with this payload.

PREREQUISITES
Import the utility script, like so :
from cds_utils.payload_coder import send_response_data_payload
in any script that will call the function send_response_data_payload

 
USAGE
The script should call the utility function before returning.
It can be anywhere in the script.

Example :
if name == 'main':

status = commit(sys.argv[1], sys.argv[2], sys.argv[3], sys.argv[4])
if DEBUG:
    print("commit function returned with status : %s" % status)
if status:
    send_response_data_payload({"étudiant": ["Mélanie", "Joséphine"]})

sys.exit(status)