BUG: Fix to_datetime not respecting dayfirst

[Aloqeely]

• closes BUG: to_datetime behaves differently depending of the format of the string provided #58859
• Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• Added type annotations to new arguments/methods/functions.
• Added an entry in the latest doc/source/whatsnew/vX.X.X.rst file if fixing a bug or adding a new feature.

[github-actions]

This pull request is stale because it has been open for thirty days with no activity. Please update and respond to this comment if you're still interested in working on this.

[MarcoGorelli]

thanks for your PR!

Are you sure that dateutil doesn't take dayfirst into account? it's an argument that's passed in, and it seems to still be respected in cases when the dateutil path is used:

In [11]: pd.to_datetime(['foo', '10.03.2000'], dayfirst=True, errors='coerce')
<ipython-input-11-69ca28852760>:1: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(['foo', '10.03.2000'], dayfirst=True, errors='coerce')
Out[11]: DatetimeIndex(['NaT', '2000-03-10'], dtype='datetime64[s]', freq=None)

In [12]: pd.to_datetime(['foo', '10.03.2000'], errors='coerce')
<ipython-input-12-8697ff36715a>:1: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(['foo', '10.03.2000'], errors='coerce')
Out[12]: DatetimeIndex(['NaT', '2000-10-03'], dtype='datetime64[s]', freq=None)

but it's been a while since I looked at this code too closely, so I might be missing something

[Aloqeely]

Ah, I guess the dateutil path was not being used here.

[Aloqeely]

The formatting makes it seem like I made a lot of changes but essentially I just added an if not dayfirst check before using string_to_dts per your suggestion.

[MarcoGorelli]

thanks @Aloqeely , this looks better!

the test you've added hits the conversion.pyx code you've modified

however, I don't think it hits the line you've modified in parsing.pyx? in which case, could you add such a test please?

[Aloqeely]

The test fails if I revert the changes in parsing.pyx, I believe this is the line of code in the test that hits the changes:

pandas/pandas/tests/tools/test_to_datetime.py

Lines 3044 to 3046 in 4c27530

	result1, _ = parsing.parse_datetime_string_with_reso(
	date_str, dayfirst=dayfirst, yearfirst=yearfirst
	)

[MarcoGorelli]

ah nice, so the test already covers both paths

looks good, just a merge conflict to resolve

[mroeschke]

Thanks @Aloqeely