Closed tlambert03 closed 11 months ago
Attention: 3 lines
in your changes are missing coverage. Please review.
Comparison is base (
aa38630
) 87.75% compared to head (6f41ae5
) 87.73%.
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
Missing a couple of test coverage lines, but looks good to me. I basically agree with your assessment. I wouldn't look to the pandas API but the draft DataFrame API, which bizarrely has drop_columns
but not drop_rows
— perhaps they are leaking implementation details into their API choices? 😂 I also couldn't find an issue in their repo. But anyway, drop_columns
appears to be in-place (returns Self), so maybe drop_rows
would be a good analogous method name? And maybe we should make an issue in their repo, though I don't feel super strongly about it.
Thanks!
Btw: I would not assume that it's in place because of -> Self
. That means it returns "type of self" (not actual self)
See this type of usage for example: https://peps.python.org/pep-0673/#use-in-classmethod-signatures
Fascinating! Thanks for the info!
cc @jni
would appreciate your input on the API. I considered using
Table.drop
like the pandas API, but I thought it might be confusing that this one is in-place and that one is not. I'm also wary of the confusion between row indices and row headers (both of which could conceivably be integers), so I forced the usage of a kwarg.