BUG: Timestamp.unit now reflects changes in components after Timestamp.replace

[diogomsmiranda]

• closes BUG: Timestamp.unit should reflect changes in components after Timestamp.replace #57749
• [Tests added and passed]
• All [code checks passed]
• Added an entry in the latest doc/source/whatsnew/v3.0.0.rst

The logic behind this is having the Timestamp.unit set as the most precise unit changed on Timestamp.replace if one is more precise than the previous set.

[diogomsmiranda]

Hey, I've made the requested changes.

cc @jbrockmendel

[diogomsmiranda]

@jbrockmendel @mroeschke Hey!! Sorry for the ping but it's been a while since I last heard from you, is this change still of interest?

Let me know If you need anything from me :)

[diogomsmiranda]

@jbrockmendel I see you made a change on this files recently, does this still work as intended?
I haven't heard from you in a while. Is this still on track to be merged?

[diogomsmiranda]

Hey @mroeschke after speaking with him it is of my understanding that @jbrockmendel is taking some time mostly-off contributing for pandas, he indicated that you might be able to review this in the meantime? thanks in advance

[mroeschke]

Could you add a test like

ts = pd.Timestamp(year=2020, month=1, day=1, nanosecond=5)
result = ts.replace(nanosecond=0)
assert result.unit == "s"

[diogomsmiranda]

@mroeschke In fact that logic was not yet implemented, thank you for noticing! I have attempted to solve the issue by registering if a replacement with 0 was made and if so add an extra check to recalculate the replacement resolution. The test now passes successfully.

[diogomsmiranda]

@mroeschke Hey, I've added the test cases requested!! Had to change some things in the logic but I believe it's working as intended now.

[mroeschke]

Thanks for the pull request, but it appears to have gone stale. If interested in continuing, please merge in the main branch, address any review comments and/or failing tests, and we can reopen.