BUG: Assignment of pyarrow arrays yield unexpected dtypes #58601
Conversation
…ssignment-unexpected-dtypes
…ssignment-unexpected-dtypes
pandas/core/frame.py
Outdated
| from pandas.compat._constants import REF_COUNT | ||
| from pandas.compat._optional import import_optional_dependency | ||
| from pandas.compat.numpy import function as nv | ||
|
|
||
| if not pa_version_under11p0: |
There was a problem hiding this comment.
What prevents this from working with pyarrow 10?
There was a problem hiding this comment.
Nothing prevents it ! :-)
It was a misanderstanding.
At the beginning I thought that the version 11.0.0 was the minimum version (because "Minimum Versions" tests in the CI were not working with pyarrow 10). But actually value in _sanitize_column is:
- A
pa.lib.ChunkedArrayin version10.0.1 - A
pa.lib.Arrayin versions above
So now I just test ifvalueis an instance of the parent classpa.lib._PandasConvertible
…ssignment-unexpected-dtypes
…s://github.com/droussea2001/pandas into BUG-56994/pyarrow-assignment-unexpected-dtypes
…ssignment-unexpected-dtypes
|
This pull request is stale because it has been open for thirty days with no activity. Please update and respond to this comment if you're still interested in working on this. |
I'm still working on it. |
|
@jorisvandenbossche : Hello Joris, hope you're doing well would you have any news about BUG: Assignment of pyarrow arrays yields unexpected dtypes ? Previously I was saying that "if a Series or a DataFrame is created or assigned with data dtype 'X' I would expect if it is possible, that it keeps the same 'X' dtype" but its only my humble opinion :-) and I would be equally happy to talk about it : maybe there is somewhere a discussion about how pyarrow types could be managed ? (thanks in advance !) |
|
Friendly ping @jorisvandenbossche |
doc/source/whatsnew/v3.0.0.rstfile if fixing a bug or adding a new feature.During a column assignment in a
DataFrame:valueis a pyarrow array (check tested withlib.is_pyarrow_array)sanitize_arraysetsdtypetoArrowDtype(value.type)sanitize_arrayprocessing are applied