This is a workhorse for Transcribers Of Reddit to chug through tasks queued up for processing. This shall act as an experiment to rewrite the Transcribers of Reddit bots using Celery task runner framework as a basis.
It is currently in pre-alpha and is only made publicly available for feedback from select individuals (you already know who you are).
$ git clone git@github.com:GrafeasGroup/tor_worker.git tor_worker
$ cd tor_worker
$ pip install --process-dependency-links -e .
Set these environment variables ahead of time:
praw_username
- The username of the Reddit account to loginpraw_password
- The password of the Reddit account to loginpraw_client_id
- The client key of the registered Reddit apppraw_client_secret
- The client secret of the registered Reddit app
Invoke the worker locally, like so:
$ celery worker --app=tor_worker.app -l info -Q 'default' --autoscale '10,1' -n 'main@%h' -B -E
Once built, where it says 'default'
above, we will add in the username-specific queues that match up with the PRAW username. We will also start one worker per PRAW username.