REGR: Fix conversion of mixed dtype DataFrame to numpy str #35473
Conversation
pandas/core/internals/managers.py
Outdated
| @@ -847,7 +847,7 @@ def _interleave(self, dtype=None, na_value=lib.no_default) -> np.ndarray: | |||
| # Give EAs some input on what happens here. Sparse needs this. | |||
| if isinstance(dtype, SparseDtype): | |||
| dtype = dtype.subtype | |||
| elif is_extension_array_dtype(dtype): | |||
| elif is_extension_array_dtype(dtype) or is_dtype_equal(dtype, str): | |||
There was a problem hiding this comment.
are you sure this is actually hit?
There was a problem hiding this comment.
Yes, this is to avoid initializing result as np.empty(dtype=str) which was creating an array with dtype "<U1" and then breaking things.
Before the change that caused this regression the dtype was always being set to the inferred dtype from _interleaved_dtype (object in this case), so here I'm trying to make sure that this still happens.
There was a problem hiding this comment.
make this a separate elif as it is very confusing here the way it is written
simonjayhawkins
added
Compat
pandas/core/internals/managers.py
Outdated
| @@ -847,7 +847,7 @@ def _interleave(self, dtype=None, na_value=lib.no_default) -> np.ndarray: | |||
| # Give EAs some input on what happens here. Sparse needs this. | |||
| if isinstance(dtype, SparseDtype): | |||
| dtype = dtype.subtype | |||
| elif is_extension_array_dtype(dtype): | |||
| elif is_extension_array_dtype(dtype) or is_dtype_equal(dtype, str): | |||
There was a problem hiding this comment.
make this a separate elif as it is very confusing here the way it is written
|
lgtm. ping on green. |
@jreback Green, thanks for reviewing |
|
thanks! |
…aFrame to numpy str
…numpy str (#35617) Co-authored-by: Daniel Saxton <[email protected]>
black pandasgit diff upstream/master -u -- "*.py" | flake8 --diff