Skip to content

BUG: Refactor construction process for MaskedArrays in series constructors #58900

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from
Closed

BUG: Refactor construction process for MaskedArrays in series constructors #58900

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 9 additions & 0 deletions pandas/core/construction.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,6 +36,8 @@
)
from pandas.core.dtypes.common import (
ensure_object,
is_float_dtype,
is_integer_dtype,
is_list_like,
is_object_dtype,
pandas_dtype,
Expand Down Expand Up @@ -510,11 +512,18 @@ def sanitize_masked_array(data: ma.MaskedArray) -> np.ndarray:
Convert numpy MaskedArray to ensure mask is softened.
"""
mask = ma.getmaskarray(data)
original = data
original_dtype = data.dtype
if mask.any():
dtype, fill_value = maybe_promote(data.dtype, np.nan)
dtype = cast(np.dtype, dtype)
data = ma.asarray(data.astype(dtype, copy=True))
data.soften_mask() # set hardmask False if it was True
if not mask.all():
idx = np.unravel_index(np.nanargmax(data, axis=None), data.shape)
if not mask[idx] and int(data[idx]) != original[idx]:
if is_integer_dtype(original_dtype) and is_float_dtype(data.dtype):
data = ma.asarray(original, "object")
Comment on lines +522 to +526
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not sure if this is the appropriate solution. Converting to object should be avoided

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you please expand on your idea? What could I use instead of an object? Thanks for your help!

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I imagine the dtype output from maybe_promote isn't entirely appropriate given the original issue. Casting to object would cover up the root issue. This issue I suspect is fairly non-trivial.

data[mask] = fill_value
else:
data = data.copy()
Expand Down
25 changes: 25 additions & 0 deletions pandas/tests/series/test_constructors.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2151,6 +2151,31 @@ def test_inference_on_pandas_objects(self):
result = Series(idx)
assert result.dtype != np.object_

def test_series_constructor_maskedarray_int_overflow(self):
# GH#56566
mx = ma.masked_array(
[
4873214862074861312,
4875446630161458944,
4824652147895424384,
0,
3526420114272476800,
],
mask=[0, 0, 0, 1, 0],
)
result = Series(mx, dtype="Int64")
expected = np.array(
[
4873214862074861312,
4875446630161458944,
4824652147895424384,
3526420114272476800,
],
dtype="int64",
)
result = np.array(result.dropna(ignore_index=True).values)
assert np.all(expected == result)


class TestSeriesConstructorIndexCoercion:
def test_series_constructor_datetimelike_index_coercion(self):
Expand Down
Loading