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

Fix Dataset.where with drop=True and mixed dims #6690

Merged
merged 3 commits into from
Jun 12, 2022
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
3 changes: 3 additions & 0 deletions doc/whats-new.rst
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,9 @@ Bug fixes
allowing the ``encoding`` and ``unlimited_dims`` options with ``save_mfdataset``.
(:issue:`6684`)
By `Travis A. O'Brien <https://github.com/taobrienlbl>`_.
- :py:meth:`Dataset.where` with ``drop=True`` now behaves correctly with mixed dimensions.
(:issue:`6227`, :pull:`6690`)
By `Michael Niklas <https://github.com/headtr1ck>`_.

Documentation
~~~~~~~~~~~~~
Expand Down
23 changes: 14 additions & 9 deletions xarray/core/common.py
Original file line number Diff line number Diff line change
Expand Up @@ -1362,18 +1362,23 @@ def where(
f"cond argument is {cond!r} but must be a {Dataset!r} or {DataArray!r}"
)

# align so we can use integer indexing
self, cond = align(self, cond) # type: ignore[assignment]

# get cond with the minimal size needed for the Dataset
if isinstance(cond, Dataset):
clipcond = cond.to_array().any("variable")
else:
clipcond = cond
def _dataarray_indexer(dim: Hashable) -> DataArray:
return cond.any(dim=(d for d in cond.dims if d != dim))

def _dataset_indexer(dim: Hashable) -> DataArray:
cond_wdim = cond.drop(var for var in cond if dim not in cond[var].dims)
keepany = cond_wdim.any(dim=(d for d in cond.dims.keys() if d != dim))
return keepany.to_array().any("variable")

_get_indexer = (
_dataarray_indexer if isinstance(cond, DataArray) else _dataset_indexer
)

# clip the data corresponding to coordinate dims that are not used
nonzeros = zip(clipcond.dims, np.nonzero(clipcond.values))
indexers = {k: np.unique(v) for k, v in nonzeros}
indexers = {}
for dim in cond.sizes.keys():
indexers[dim] = _get_indexer(dim)

self = self.isel(**indexers)
cond = cond.isel(**indexers)
Expand Down
19 changes: 19 additions & 0 deletions xarray/tests/test_dataset.py
Original file line number Diff line number Diff line change
Expand Up @@ -4715,6 +4715,25 @@ def test_where_drop(self) -> None:
actual8 = ds.where(ds > 0, drop=True)
assert_identical(expected8, actual8)

# mixed dimensions: PR#6690, Issue#6227
ds = xr.Dataset(
{
"a": ("x", [1, 2, 3]),
"b": ("y", [2, 3, 4]),
"c": (("x", "y"), np.arange(9).reshape((3, 3))),
}
)
expected9 = xr.Dataset(
{
"a": ("x", [np.nan, 3]),
"b": ("y", [np.nan, 3, 4]),
"c": (("x", "y"), np.arange(3.0, 9.0).reshape((2, 3))),
}
)
actual9 = ds.where(ds > 2, drop=True)
assert actual9.sizes["x"] == 2
assert_identical(expected9, actual9)

def test_where_drop_empty(self) -> None:
# regression test for GH1341
array = DataArray(np.random.rand(100, 10), dims=["nCells", "nVertLevels"])
Expand Down