CLN: avoid values_from_object in reshape.merge #32537
Conversation
pandas/core/reshape/merge.py
Outdated
| and is_extension_array_dtype(rk.dtype) | ||
| and lk.dtype == rk.dtype | ||
| ): | ||
| elif is_extension_array_dtype(lk.dtype) and lk.dtype == rk.dtype: |
There was a problem hiding this comment.
Is this guaranteed to be equivalent?
There was a problem hiding this comment.
I'm pretty sure. How could you have this condition but not is_extension_array_dtype(rk.dtype)?
There was a problem hiding this comment.
should change to lk.is_dtype_equal(rk) (technically the lk.dtype == rk.dtype is unsafe and we should never use this)
| lk = getattr(lk, "_values", lk)._data | ||
| rk = getattr(rk, "_values", rk)._data | ||
| # Extract the ndarray (UTC-localized) values | ||
| # Note: we dont need the dtypes to match, as these can still be compared |
There was a problem hiding this comment.
That does happen somewhere else?
There was a problem hiding this comment.
yes, in the EA-dtype check below we require matching dtypes
…-values_from_object-merge
pandas/core/reshape/merge.py
Outdated
| and is_extension_array_dtype(rk.dtype) | ||
| and lk.dtype == rk.dtype | ||
| ): | ||
| elif is_extension_array_dtype(lk.dtype) and lk.dtype == rk.dtype: |
There was a problem hiding this comment.
should change to lk.is_dtype_equal(rk) (technically the lk.dtype == rk.dtype is unsafe and we should never use this)
not sure i follow. can you give an example? |
…-values_from_object-merge
are there any merge cases that work that didn't previously? |
there are some dtype combinations that arent tested that looked sketchy. there's a TODO comment about adding tests for them |
kk |
|
lgtm. merge o ngreen. |
|
hmm, something broke |
…-values_from_object-merge
|
updated+green |
|
Thanks @jbrockmendel |
No description provided.