Handle S3 credential expiration more gracefully

[jrbourbeau]

Closes #353

[github-actions]

👈 Launch a binder notebook on this branch for commit 64c0976

I will automatically update this comment whenever this PR is modified

👈 Launch a binder notebook on this branch for commit a33bf3a

👈 Launch a binder notebook on this branch for commit ab70e65

👈 Launch a binder notebook on this branch for commit ce741a8

👈 Launch a binder notebook on this branch for commit ad85cd4

👈 Launch a binder notebook on this branch for commit 6c342a2

[jrbourbeau]

@MattF-NSIDC @betolink I confirmed the changes here work as expected in practice (i.e. no longer getting the permissions errors reported in the #353)

[MattF-NSIDC]

It may realistically be a while before I can review this! 😢

[mfisher87]

Great work! I have a couple thoughts about readability, but no concerns about the behavior at the moment.

I don't think the suggestions I made are significant enough to warrant a second review, so if you're ready to merge after going through them: 🚀

[betolink]

related to this, just for reference, Alex Mandel from DevSeed mentioned this piece of code that handles more or less the same for the MAAP project https://github.com/MAAP-Project/gedi-subsetter/blob/58c87ab907bc3b121510eeabdeb2b8e63c7df999/src/gedi_subset/maapx.py#L55-L61

[jrbourbeau]

mentioned this piece of code that handles more or less the same for the MAAP project

Nice, thanks for pointing us to that @betolink. Good to know for future reference.

[jrbourbeau]

@mfisher87 I've addressed or responded to all your review comments (thank you for taking the time to review this PR). There are a few minor, subjective points where I think we have differing opinions / styles -- let me know if any of those are blocking comments.

[MattF-NSIDC]

@jrbourbeau thanks! I don't feel any of those differing opinions / styles should be blockers, marked all as resolved! I think this is ready to go 💯

[MattF-NSIDC]

💯