Jan Rous

> t I'm still contemplating what would be the best approach here. I would like to reuse tox, but I would also like to be able to run the three...

I have switched this back to use `tox` to run the commands && perhaps made it such that coverage is exported both for `ferc_to_sqlite` and `pudl_etl` (I still need to...

I've fixed minor typos and tried to make sure that our `coverage` collection works. AFAICT, to make this work, coverage needs to be run with `--concurrency=multiprocessing` option which generates several...

> It looks like the coverage isn't completely working since it dropped from 88% to 59%. This is more coverage than we get from just doing the imports (which IIRC...

> @jdangerx it looks like this PR is working now, in that it runs the ETL before the tests and collects coverage from all of that, but it also doesn't...

I've made some changes, including: - human friendly naming of the action & steps - ditching tox as the intermediary

We have determined that parallelism works as expected (the ETL just got a lot slower), so the only remaining piece is coverage, which still seems to be broken. I will...

I do not know what changed, but it seems that coverage works well now :shrug: so this should be good to go

I think I'm going to loose my mind here. I thought I finally figured the right configuration here only to find out that it still seems to be busted, so...

> @rousik Can you update the main description for this PR to reflect the current situation? It's been through so many changes and the code on `main` has changed so...