Microsecond-precision test failures

[nathany-copia]

Steps to reproduce

Running mix test with Elixir 1.18 after updating dependencies so that they will compile (see #772).

Description of issue

• What are the expected results? All tests pass in current versions of Elixir.

I'm not entirely sure what the desired behaviour is for these tests, so I'm hoping someone more familiar with this code base can take a look or provide some direction.

 1) test shift to zero (ShiftTests)
     test/shift_test.exs:175
     Assertion with === failed
     code:  assert expected === result
     left:  ~N[0000-01-01 00:00:00]
     right: ~N[0000-01-01 00:00:00.000000]
     stacktrace:
       test/shift_test.exs:178: (test)



  2) test format RFC3339 (DateFormatTest.FormatDefault)
     test/format_default_test.exs:338
     Assertion with == failed
     code:  assert format(date, "{RFC3339}") == {:ok, "0010-09-26T07:00:02+00:50"}
     left:  {:ok, "0010-09-26T07:00:02+00:53"}
     right: {:ok, "0010-09-26T07:00:02+00:50"}
     stacktrace:
       test/format_default_test.exs:360: (test)



  3) test add (TimexTests)
     test/timex_test.exs:14
     Assertion with === failed
     code:  assert expected === result
     left:  ~U[2015-07-02 14:27:52Z]
     right: ~U[2015-07-02 14:27:52.000000Z]
     stacktrace:
       test/timex_test.exs:18: (test)



  4) test subtract milliseconds (TimexTests)
     test/timex_test.exs:35
     Assertion with === failed
     code:  assert subtracted.microsecond === {900_000, 2}
     left:  {900000, 6}
     right: {900000, 2}
     stacktrace:
       test/timex_test.exs:39: (test)



  5) test subtract (TimexTests)
     test/timex_test.exs:28
     Assertion with === failed
     code:  assert expected === result
     left:  ~U[2015-06-16 14:27:52Z]
     right: ~U[2015-06-16 14:27:52.000000Z]
     stacktrace:
       test/timex_test.exs:32: (test)

[nathany-copia]

These appear to be different than the test failures happening on main. 🤔

https://github.com/bitwalker/timex/actions

[nathany-copia]

For some of these we could just truncate the .000000 from the time -- but do we want to update the tests, or should the code be modified? 🤔

[nathany-copia]

@bitwalker What should be done here? I'm willing to do the work of opening a PR, but I would appreciate some direction.