CI update

[aburrell]

Description

Addresses #120 by updating the CI. Also addresses: https://github.com/orgs/pysat/projects/2

Type of change

• Bug fix (non-breaking change which fixes an issue)

How Has This Been Tested?

Must be tested on GitHub Actions.

Test Configuration:

• Operating system: GitHub Actions
• Version number: Python 3.8, 3.9, 3.10
• Any details about your local setup that are relevant: main pysat

Checklist:

• Make sure you are merging into the develop (not main) branch
• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in downstream modules
• Add a note to CHANGELOG.md, summarizing the changes

If this is a release PR, replace the first item of the above checklist with the
release checklist on the pysat wiki:
https://github.com/pysat/pysat/wiki/Checklist-for-Release

[jklenzing]

From the GA log for the nep29 job:

Installing collected packages: pytz, six, pyparsing, numpy, python-dateutil, packaging, pandas, xarray
  Attempting uninstall: numpy
    Found existing installation: numpy 1.20.0
    Uninstalling numpy-1.20.0:
      Successfully uninstalled numpy-1.20.0
Successfully installed numpy-1.23.4 packaging-21.3 pandas-1.5.1 pyparsing-3.0.9 python-dateutil-2.8.2 pytz-2022.5 six-1.16.0 xarray-2022.3.0

I'm not sure why it's updating numpy after installing the requested version. I have a new structure up at sami2py/sami2py#175 that fixes the issue.

Sorry about the confusion on this update. I think I was looking at the wrong part of the log before.

[aburrell]

Ok @jklenzing what do you think? The alternative is to limit the pandas version to one that works with numpy 1.20.0 (probably 1.4.x).

[jklenzing]

Ok @jklenzing what do you think? The alternative is to limit the pandas version to one that works with numpy 1.20.0 (probably 1.4.x).

I'm happy enough with this. In January, numpy 1.21 becomes the new minimum, so it becomes moot at that point. I think we need to discuss how we proceed across pysat packages at the next meeting. At the moment, I'm leaning toward the separate requirements file like in sami2py. Probably the right way is to implement docker at some point.