test: Use timezone aware datetime

[kesara]

This might prevent errors like https://github.com/ietf-tools/datatracker/actions/runs/10786037653/job/29912248540#step:6:1902

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 88.82%. Comparing base (c7f6bde) to head (7098833).
Report is 64 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #7918      +/-   ##
==========================================
+ Coverage   88.78%   88.82%   +0.04%     
==========================================
  Files         296      304       +8     
  Lines       41320    41495     +175     
==========================================
+ Hits        36687    36860     +173     
- Misses       4633     4635       +2     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[jennifer-richards]

Modulo date vs datetime, this should help.

There'll still be a very brief window around midnight UTC when this could fail. I wonder if we should write a test helper named something like date_or_wait(dt: TimeDelta) that will sleep if it's called within dt of a date change, then return the next day. I think this could protect us from random failures.

[kesara]

Modulo date vs datetime, this should help.

There'll still be a very brief window around midnight UTC when this could fail. I wonder if we should write a test helper named something like date_or_wait(dt: TimeDelta) that will sleep if it's called within dt of a date change, then return the next day. I think this could protect us from random failures.

We can use something like Freezgun to set a specific date too. The test I introduce follows the patterns of the existing test suite. The test for date is trivial & and go without it?

[jennifer-richards]

We can use something like Freezgun to set a specific date too. The test I introduce follows the patterns of the existing test suite. The test for date is trivial & and go without it?

Seems worth investigating, but what you've got here is fine.