Skip to content

fix: don't destroy dtypes in to_dataframe #10705

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
merged 3 commits into from
Sep 5, 2025
Merged

fix: don't destroy dtypes in to_dataframe #10705

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
cb2c244
fix: don't destroy dtypes in to_dataframe
ianhi Sep 5, 2025
9cb6d80
add a used-ignore
ianhi Sep 5, 2025
f039120
put type ignore in correct spot
ianhi Sep 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
6 changes: 4 additions & 2 deletions xarray/core/coordinates.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -180,11 +180,13 @@ def to_index(self, ordered_dims: Sequence[Hashable] | None = None) -> pd.Index:
np.tile(np.repeat(code, repeat_counts[i]), tile_counts[i])
for code in codes
]
level_list += [list(level) for level in levels]
level_list += levels
names += index.names

return pd.MultiIndex(
levels=level_list, codes=[list(c) for c in code_list], names=names
levels=level_list, # type: ignore[arg-type,unused-ignore]
Copy link
Contributor Author

Choose a reason for hiding this comment

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

I think that the pandas stubs might be wrong here. They want a list of hashables, but a list of dtypes does work fine

codes=[list(c) for c in code_list],
names=names,
)


Expand Down
28 changes: 28 additions & 0 deletions xarray/tests/test_dataarray.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3528,6 +3528,34 @@ def test_to_dataframe_0length(self) -> None:
assert len(actual) == 0
assert_array_equal(actual.index.names, list("ABC"))

@pytest.mark.parametrize(
"x_dtype,y_dtype,v_dtype",
[
(np.uint32, np.float32, np.uint32),
(np.int16, np.float64, np.int64),
(np.uint8, np.float32, np.uint16),
(np.int32, np.float32, np.int8),
],
)
def test_to_dataframe_coord_dtypes_2d(self, x_dtype, y_dtype, v_dtype) -> None:
x = np.array([1], dtype=x_dtype)
y = np.array([1.0], dtype=y_dtype)
v = np.array([[42]], dtype=v_dtype)

da = DataArray(v, dims=["x", "y"], coords={"x": x, "y": y})
df = da.to_dataframe(name="v").reset_index()

# Check that coordinate dtypes are preserved
assert df["x"].dtype == np.dtype(x_dtype), (
f"x coord: expected {x_dtype}, got {df['x'].dtype}"
)
assert df["y"].dtype == np.dtype(y_dtype), (
f"y coord: expected {y_dtype}, got {df['y'].dtype}"
)
assert df["v"].dtype == np.dtype(v_dtype), (
f"v data: expected {v_dtype}, got {df['v'].dtype}"
)

@requires_dask_expr
@requires_dask
@pytest.mark.xfail(not has_dask_ge_2025_1_0, reason="dask-expr is broken")
Expand Down
Loading