@Marvin run_db_sheet_job() and process_results() are both tasks, used as below.

# Run scheduled jobs
        results = run_db_sheets_job.with_options(persist_result=True, result_storage=result_persistence_location).map(scheduled)

        # Process the results, allowing Futures from mapped jobs to resolve
        to_notify = process_results(results=quote(results))

I want process_results() to wait for all results from run_db_sheets_job() but I see this in the log:

12:14:21.520 | INFO    | Flow run 'cocky-degu' - Created task run 'run_db_sheets_job-0' for task 'run_db_sheets_job'
12:14:21.522 | INFO    | Flow run 'cocky-degu' - Submitted task run 'run_db_sheets_job-0' for execution.
12:14:21.825 | INFO    | Flow run 'cocky-degu' - Created task run 'process_results-0' for task 'process_results'
12:14:21.827 | INFO    | Flow run 'cocky-degu' - Executing 'process_results-0' immediately...

It looks like process_results() does not complete until all the mapped jobs in run_db_sheets_job() have an outcomes but I would like to understand this better. How do the internal states of results and to_notify evolve over the course of the flow both tasks are embedded in?

Your message was too long by 2 tokens - please shorten it and try again. For reference, here's your message at the allowed limit:

run_db_sheet_job() and process_results() are both tasks, used as below.

# Run scheduled jobs
        results = run_db_sheets_job.with_options(persist_result=True, result_storage=result_persistence_location).map(scheduled)

        # Process the results, allowing Futures from mapped jobs to resolve
        to_notify = process_results(results=quote(results))

I want process_results() to wait for all results from run_db_sheets_job() but I see this in the log:

12:14:21.520 | INFO    | Flow run 'cocky-degu' - Created task run 'run_db_sheets_job-0' for task 'run_db_sheets_job'
12:14:21.522 | INFO    | Flow run 'cocky-degu' - Submitted task run 'run_db_sheets_job-0' for execution.
12:14:21.825 | INFO    | Flow run 'cocky-degu' - Created task run 'process_results-0' for task 'process_results'
12:14:21.827 | INFO    | Flow run 'cocky-degu' - Executing 'process_results-0' immediately...

It looks like process_results() does not complete until all the mapped jobs in run_db_sheets_job() have an outcomes but I would like to understand this better. How do the internal states of results and to_notify evolve over the course of the flow both tasks are embedded