Skip to content

REGR: Fix IntervalArray.nunique with infinity #63267

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

Merged
mroeschke merged 2 commits into from
Dec 5, 2025
+16 −3
Merged

REGR: Fix IntervalArray.nunique with infinity #63267

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fa44e36
REGR: Fix IntervalArray.nunique with infinity
rhshadrach Dec 4, 2025
d54d982
np.asarray
rhshadrach Dec 5, 2025
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
5 changes: 2 additions & 3 deletions pandas/core/arrays/interval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2128,9 +2128,8 @@ def _combined(self) -> IntervalSide:
)
comb = comb.view("complex128")[:, 0]
else:
comb = (np.array(left.ravel(), dtype="complex128")) + (
1j * np.array(right.ravel(), dtype="complex128")
)
comb = np.array(left.ravel(), dtype="complex128")
Copy link
Member

@mroeschke mroeschke Dec 4, 2025

Choose a reason for hiding this comment

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

Could we use np.asarray? (NumPy might copy anyways but would be nice if it didn't in this case)

Copy link
Member Author

@rhshadrach rhshadrach Dec 5, 2025

Choose a reason for hiding this comment

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

Updated, but I believe it always makes a copy when going float64 -> complex128. I also tried

comb = np.empty(left.shape, dtype="complex128")
comb.real = left
comb.imag = right

but that had slightly worse performance than the identical np.array and np.asarray.

comb.imag = right.ravel()
return comb

def _from_combined(self, combined: np.ndarray) -> IntervalArray:
Expand Down
14 changes: 14 additions & 0 deletions pandas/tests/arrays/interval/test_interval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -136,6 +136,20 @@ def test_unique_with_negatives(self):
)
tm.assert_index_equal(result, expected)

@pytest.mark.parametrize(
"data",
[
[Interval(-np.inf, 0), Interval(-np.inf, 1)],
[Interval(0, np.inf), Interval(1, np.inf)],
],
)
def test_unique_with_infinty(self, data):
# https://github.com/pandas-dev/pandas/issues/63218
s = pd.Series(data)
tm.assert_interval_array_equal(s.unique(), s.array)
assert s.nunique() == 2
tm.assert_series_equal(s.drop_duplicates(), s)


class TestSetitem:
def test_set_na(self, left_right_dtypes):
Expand Down
Loading