Add helper methods for RFC 3339 datetime handling

[philvarner]

Related Issue(s): n/a

Description:

There are a number of places in the STAC ecosystem where datetimes are converted two and from RFC 3339 strings. Many of these places are doing this incorrectly. These changes aim to add a correct implementations to:

• rfc3339_str_to_datetime: parse ONLY a valid RFC 3339 string to a datetime
• str_to_interval: parse an interval string into a datetime tuple
• now_in_utc: create a datetime with UTC set as the datetime (since utcnow() inexplicably does not do this)
• now_to_rfc3339_str: create an RFC3339 string representing now

My primary motivation is to replace the numerous uses of the stac_pydantic constant DATETIME_RFC339(sic) with uses of these methods. Other libraries that depend on pystac will also use them, and there are likely several places even within pystac that these could be used instead of the current code.

One downside to this is that the tests need the pytest parameterized decorator to run, and this doesn't work in the class-based test modules that unittest requires. I ran the existing tests with pytest and it seems to run them all fine, so we could change to using pytest to run the test without needing to convert all the tests from unittest to pytest.

PR Checklist:

• Code is formatted (run pre-commit run --all-files)
• Tests pass (run scripts/test)
• Documentation has been updated to reflect changes, if applicable
• This PR maintains or improves overall codebase code coverage.
• Changes are added to the CHANGELOG. See the docs for information about adding to the changelog.

[codecov-commenter]

Codecov Report

Base: 90.21% // Head: 90.21% // Increases project coverage by +0.00% 🎉

Coverage data is based on head (e7a4d0d) compared to base (2d01225).
Patch coverage: 100.00% of modified lines in pull request are covered.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #760   +/-   ##
=======================================
  Coverage   90.21%   90.21%           
=======================================
  Files          47       47           
  Lines        6090     6094    +4     
  Branches      915      915           
=======================================
+ Hits         5494     5498    +4     
  Misses        414      414           
  Partials      182      182           

Impacted Files	Coverage Δ
pystac/utils.py	95.55% <100.00%> (+0.13%)	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[philvarner]

It may be more appropriate to put str_to_interval in stac-fastapi, since the datetime interval format is only used by query parameters.

[philvarner]

looks like i need to write some stubs for ciso8601

[duckontheweb]

looks like i need to write some stubs for ciso8601

@philvarner I think it would also be fine to just ignore these in the mypy.ini file as we've done for other libraries without stubs.

[philvarner]

looks like i need to write some stubs for ciso8601

@philvarner I think it would also be fine to just ignore these in the mypy.ini file as we've done for other libraries without stubs.

Okay, I'll do that. Weird thing is that it looks like ciso8601 has stubs defined. I haven't used mypy that much, so I don't really know how the stubs are supposed to be resolved, etc.

[duckontheweb]

looks like i need to write some stubs for ciso8601

@philvarner I think it would also be fine to just ignore these in the mypy.ini file as we've done for other libraries without stubs.

Okay, I'll do that. Weird thing is that it looks like ciso8601 has stubs defined. I haven't used mypy that much, so I don't really know how the stubs are supposed to be resolved, etc.

Yeah, that is peculiar. I'll look into this a bit more and see if there is a better way to solve this besides ignoring the import.

[philvarner]

looks like i need to write some stubs for ciso8601

@philvarner I think it would also be fine to just ignore these in the mypy.ini file as we've done for other libraries without stubs.

Okay, I'll do that. Weird thing is that it looks like ciso8601 has stubs defined. I haven't used mypy that much, so I don't really know how the stubs are supposed to be resolved, etc.

Yeah, that is peculiar. I'll look into this a bit more and see if there is a better way to solve this besides ignoring the import.

I needed to add it to the requirements-test.txt and setup.py validation requirements.

[lossyrob]

One downside to this is that the tests need the pytest parameterized decorator to run, and this doesn't work in the class-based test modules that unittest requires. I ran the existing tests with pytest and it seems to run them all fine, so we could change to using pytest to run the test without needing to convert all the tests from unittest to pytest.

I'd be happy with a change to pytest personally; however I wonder if that should be a larger PR that does a wholesale conversion rather than have multiple testing frameworks in the test. You could use a unittest.TestCase with self.subTest to test over each of the cases; for example here:

pystac/tests/validation/test_validate.py

Lines 39 to 40 in 0426bd1

	for example in TestCases.get_examples_info():
	with self.subTest(example.path):

As it stands there's not a change to scripts/test to use pytest, so these unit tests wouldn't be run as part of CI.

I made some changes, so I need someone else to review.

[pjhartzell]

Looks good to me.

[philvarner]

Got this merged in just under a year!