Skip to content

Conversation

@jbrockmendel
Copy link
Member

Also fixes lookups/docstrings for class-level access to cache_readonly attributes.

Related: wesm/pandas2#27

@jreback
Copy link
Contributor

jreback commented Feb 15, 2018

this seems like quite a lot of extra code when you could simply override hasnans for Series

@jbrockmendel
Copy link
Member Author

this seems like quite a lot of extra code when you could simply override hasnans for Series

If it were just hasnans I would agree. The motivation comes from WIP to move arith/comparison ops from DatetimeIndexOpsMixin/DTI/TDI/PI to array classes, where there are currently a bunch of cache_readonlys that need to be flipped. This seemed like a reasonable scope to break off of what would otherwise be a giant diff (similar train of thought for #19672).

@jbrockmendel
Copy link
Member Author

Since this is failing tests anyway, I'll remove the new pieces and just fix hasnans and the docstrings.

@gfyoung gfyoung added Bug Missing-data np.nan, pd.NaT, pd.NA, dropna, isnull, interpolate labels Feb 15, 2018
@jbrockmendel
Copy link
Member Author

No idea what’s up with this test. Closing to narrow the queue

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Bug Missing-data np.nan, pd.NaT, pd.NA, dropna, isnull, interpolate

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Series.hasnans shouldn't be cached

3 participants