TST: GH 32431 check print label indexing on nan #36635
Conversation
WillAyd
left a comment
WillAyd
left a comment
There was a problem hiding this comment.
Cool looks pretty good - thanks for the first contribution. Just a small change needed I think
| # GH 32431 | ||
| df = pd.Series([1, "{1,2}", 1, None]) | ||
| vc = df.value_counts(dropna=False) | ||
| print(vc.loc[np.nan]) |
There was a problem hiding this comment.
Rather than just printing you should construct the expected result and use tm.assert_series_equal(result, expected) - you should see this idiom throughout the test modules if you need a reference
There was a problem hiding this comment.
Thanks for the feedback! I constructed the expected result and expected an int. So I compared the results just with a normal assert, since I couldn't find a method on tm to assert ints. Please let me know if this satisfies your request
|
This pull request is stale because it has been open for thirty days with no activity. Please update or respond to this comment if you're still interested in working on this. |
|
thanks @wesleyboelrijk |
4 participants
black pandasgit diff upstream/master -u -- "*.py" | flake8 --diffThis is my first contribution to pandas. I added a simple test that was requested in one of the open issues.