Help Pint-Pandas pass tests when HAS_UNCERTAINTIES is enabled #49181
Conversation
…abled. Signed-off-by: MichaelTiemann <72577720+MichaelTiemannOSC@users.noreply.github.com>
| @@ -1,6 +1,15 @@ | |||
| import numpy as np | |||
| import pytest | |||
|
|
|||
| try: | |||
| import uncertainties.unumpy as unp | |||
There was a problem hiding this comment.
this is not appropriate for the pandas test suite as this is not in any way tested in ci
There was a problem hiding this comment.
Thanks for the review! Do you have any suggestions as to what test cases I should look at to see how other extension packages deal with base tests that don't work as written? I've seen some examples where the hooks are there to create the correct sort of data (using fixtures) or hook the right functions (so that base test code calls extension functions). But am new to how I should deal with this case. If you guide, I can implement, test, and amend.
|
thanks for your pr - closing as per the above if there's a bug in pandas, please open an issue with a reproducible example and it can be addressed |
The Pint-Pandas testsuite relies on the Pandas test suite. In the process of enabling support for uncertainties in Pint-Pandas, I discovered places where the base Pandas test suite needs to follow along. I will extract the interesting test cases that triggered the need for these patches.
Signed-off-by: MichaelTiemann 72577720+MichaelTiemannOSC@users.noreply.github.com
doc/source/whatsnew/vX.X.X.rstfile if fixing a bug or adding a new feature.