Skip to content
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.

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

BUG: Creation of UInt64 column with 18446744073709551615 throws RuntimeWarning #60214

Open
wants to merge 16 commits into
base: main
Choose a base branch
from
17 changes: 11 additions & 6 deletions pandas/core/construction.py
Original file line number Diff line number Diff line change
Expand Up @@ -509,15 +509,20 @@ def ensure_wrapped_if_datetimelike(arr):

def sanitize_masked_array(data: ma.MaskedArray) -> np.ndarray:
"""
Convert numpy MaskedArray to ensure mask is softened.
Convert numpy MaskedArray to ensure mask is softened,

"""
mask = ma.getmaskarray(data)
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
data[mask] = fill_value
dtype = cast(np.dtype, data.dtype)
if isinstance(dtype, ExtensionDtype) and dtype.name.startswith("Masked"):
Copy link
Member

Choose a reason for hiding this comment

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

I haven't checked thoroughly, but I think you want dtype._can_hold_na instead of having logic depend on the name of the dtype.

data = ma.asarray(data.astype(dtype, copy=True))
data.soften_mask() # If the data is a Masked EA, directly soften the mask.
else:
dtype, fill_value = maybe_promote(data.dtype, np.nan)
data = ma.asarray(data.astype(dtype, copy=True))
data.soften_mask() # set hardmask False if it was True
data[mask] = fill_value
else:
data = data.copy()
return data
Expand Down
35 changes: 35 additions & 0 deletions pandas/tests/dtypes/cast/test_construct_ndarray.py
Original file line number Diff line number Diff line change
Expand Up @@ -34,3 +34,38 @@ def test_construct_1d_ndarray_preserving_na_datetimelike(dtype):

result = sanitize_array(arr, index=None, dtype=np.dtype(object))
tm.assert_numpy_array_equal(result, expected)


@pytest.mark.parametrize(
"values, dtype, expected",
[
(
np.ma.masked_array([1, 2, 3], mask=[False, True, False]),
"int64",
np.array([1, 2, 3], dtype=np.int64),
),
(
np.ma.masked_array([1, 2, 3], mask=[False, True, False]),
"float64",
np.array([1, 2, 3], dtype=np.float64),
),
(
np.ma.masked_array([1, 2, 3], mask=[False, True, False]),
"UInt64",
np.array([1, 2, 3], dtype=np.uint64),
),
(
np.ma.masked_array([1.0, 2.0, 3.0], mask=[False, True, False]),
"float64",
np.array([1.0, 2.0, 3.0], dtype=np.float64),
),
(
np.ma.masked_array([1.0, 2.0, 3.0], mask=[False, True, False]),
"Int64",
np.array([1, 2, 3], dtype=np.int64),
),
],
)
def test_sanitize_masked_array_with_masked_ea(values, dtype, expected):
Copy link
Member

Choose a reason for hiding this comment

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

Can you add a case that tests for a large number as in an issue.

Also start the test with a reference to the GitHub issue:

# GH#60050

result = sanitize_array(values, index=None, dtype=dtype)
tm.assert_masked_array_equal(result, expected)
Loading