A demo app to explore patterns for MP in flask and celery apps.
This will publish a message to a python multiprocessing Queue when a page is loaded or a task is ran.
When the flask/celery app is started it will also start a Process which will read from the queue. Once it has read the batch size (which is set to 3 here) it will print the messages it has received from the queue.
It's the most strpped down version of this pattern. See the sentry_sdk for a more full featured example.
You should be able to run gunicorn flask_app:app to start the flask app.
Then, view the demo page at http://127.0.0.1:8000 and after 3 page loads it will print the queue contents:
['Page loaded', 'Page loaded', 'Page loaded']
You should be able to run celery -A celery_app worker --loglevel info to
start a celery worker process.
Then, running python celery_app.py will dispatch the add task.
After the 3rd execution of a task it will print the queue contents:
['Task run', 'Task run', 'Task run']