MatthewFlamm

If all we have to do is copy over the entire `testing_config` folder, then this sounds fairly easy. We need part of that folder now anyway since the top level...

Given that code owners are writing code that impacts disclosure of users sensitive information, I think this deserves documentation. I want to be sure that I'm using it the intended...

One issue is that mypy is run through `pre-commit` that has a different dependency pinning scheme compared to our CI. It would be more stable to use a traditional `requirements-typing.txt`...

However, it is my experience that we have a LOT of errors when running mypy in a typical dev environment. This is because the mypy pre-commit config does not include...

> Actually, no, we'd need nightly builds of the 2.0.0 release branch... `main` is 2.1.0 dev now. 2.0.0 nightly wheels are not available at https://anaconda.org/scientific-python-nightly-wheels/numpy and I cannot easily find...

I agree with this proposal. We will support 2.1 in the future and this allows us to get it into the CI faster, which will prevent more problems.

Yes. I think because the NWS servers are sometimes flaky. You'll need to handle the timeouts in some way.

This is a workaround that works for my use case: ```py from unittest.mock import AsyncMock from tenacity import retry async def test_mock_retry(): amock = AsyncMock() amock.side_effect = [ValueError, None] async...

I think collecting how other packages approach this, or collecting the needs of other projects, will be helpful to define what this should look like. We have a purpose built...

Thank you for the detailed test, I don't have a system for long term testing like this. I will again mark as draft so I can do some detailed logging....