Merge branch 'main' into pr/9881

doc/whats-new.rst

@@ -37,14 +37,19 @@ Breaking changes
 
 Deprecations
 ~~~~~~~~~~~~
-
+- Finalize deprecation of ``closed`` parameters of :py:func:`cftime_range` and
+  :py:func:`date_range` (:pull:`9882`).
+  By `Kai Mühlbauer <https://github.com/kmuehlbauer>`_.
 
 Bug fixes
 ~~~~~~~~~
 - Fix type annotations for ``get_axis_num``. (:issue:`9822`, :pull:`9827`).
   By `Bruce Merry <https://github.com/bmerry>`_.
 - Fix unintended load on datasets when calling :py:meth:`DataArray.plot.scatter` (:pull:`9818`).
   By `Jimmy Westling <https://github.com/illviljan>`_.
+- Fix interpolation when non-numeric coordinate variables are present (:issue:`8099`, :issue:`9839`).
+  By `Deepak Cherian <https://github.com/dcherian>`_.
+
 
 Documentation
 ~~~~~~~~~~~~~

xarray/backends/zarr.py

@@ -1135,9 +1135,7 @@ def set_variables(self, variables, check_encoding_set, writer, unlimited_dims=No
                 else:
                     encoded_attrs[DIMENSION_KEY] = dims
 
-                encoding["exists_ok" if _zarr_v3() else "overwrite"] = (
-                    True if self._mode == "w" else False
-                )
+                encoding["overwrite"] = True if self._mode == "w" else False
 
                 zarr_array = self._create_new_array(
                     name=name,

xarray/coding/cftime_offsets.py

@@ -62,15 +62,13 @@
 )
 from xarray.core.common import _contains_datetime_like_objects, is_np_datetime_like
 from xarray.core.pdcompat import (
-    NoDefault,
     count_not_none,
     nanosecond_precision_timestamp,
-    no_default,
 )
 from xarray.core.utils import attempt_import, emit_user_level_warning
 
 if TYPE_CHECKING:
-    from xarray.core.types import InclusiveOptions, Self, SideOptions, TypeAlias
+    from xarray.core.types import InclusiveOptions, Self, TypeAlias
 
 
 DayOption: TypeAlias = Literal["start", "end"]
@@ -943,51 +941,14 @@ def _generate_range(start, end, periods, offset):
             current = next_date
 
 
-def _translate_closed_to_inclusive(closed):
-    """Follows code added in pandas #43504."""
-    emit_user_level_warning(
-        "Following pandas, the `closed` parameter is deprecated in "
-        "favor of the `inclusive` parameter, and will be removed in "
-        "a future version of xarray.",
-        FutureWarning,
-    )
-    if closed is None:
-        inclusive = "both"
-    elif closed in ("left", "right"):
-        inclusive = closed
-    else:
-        raise ValueError(
-            f"Argument `closed` must be either 'left', 'right', or None. "
-            f"Got {closed!r}."
-        )
-    return inclusive
-
-
-def _infer_inclusive(
-    closed: NoDefault | SideOptions, inclusive: InclusiveOptions | None
-) -> InclusiveOptions:
-    """Follows code added in pandas #43504."""
-    if closed is not no_default and inclusive is not None:
-        raise ValueError(
-            "Following pandas, deprecated argument `closed` cannot be "
-            "passed if argument `inclusive` is not None."
-        )
-    if closed is not no_default:
-        return _translate_closed_to_inclusive(closed)
-    if inclusive is None:
-        return "both"
-    return inclusive
-
-
 def cftime_range(
     start=None,
     end=None,
     periods=None,
     freq=None,
     normalize=False,
     name=None,
-    closed: NoDefault | SideOptions = no_default,
-    inclusive: None | InclusiveOptions = None,
+    inclusive: InclusiveOptions = "both",
     calendar="standard",
 ) -> CFTimeIndex:
     """Return a fixed frequency CFTimeIndex.
@@ -1006,16 +967,7 @@ def cftime_range(
         Normalize start/end dates to midnight before generating date range.
     name : str, default: None
         Name of the resulting index
-    closed : {None, "left", "right"}, default: "NO_DEFAULT"
-        Make the interval closed with respect to the given frequency to the
-        "left", "right", or both sides (None).
-
-        .. deprecated:: 2023.02.0
-            Following pandas, the ``closed`` parameter is deprecated in favor
-            of the ``inclusive`` parameter, and will be removed in a future
-            version of xarray.
-
-    inclusive : {None, "both", "neither", "left", "right"}, default None
+    inclusive : {"both", "neither", "left", "right"}, default "both"
         Include boundaries; whether to set each bound as closed or open.
 
         .. versionadded:: 2023.02.0
@@ -1193,8 +1145,6 @@ def cftime_range(
         offset = to_offset(freq)
         dates = np.array(list(_generate_range(start, end, periods, offset)))
 
-    inclusive = _infer_inclusive(closed, inclusive)
-
     if inclusive == "neither":
         left_closed = False
         right_closed = False
@@ -1229,8 +1179,7 @@ def date_range(
     tz=None,
     normalize=False,
     name=None,
-    closed: NoDefault | SideOptions = no_default,
-    inclusive: None | InclusiveOptions = None,
+    inclusive: InclusiveOptions = "both",
     calendar="standard",
     use_cftime=None,
 ):
@@ -1257,20 +1206,10 @@ def date_range(
         Normalize start/end dates to midnight before generating date range.
     name : str, default: None
         Name of the resulting index
-    closed : {None, "left", "right"}, default: "NO_DEFAULT"
-        Make the interval closed with respect to the given frequency to the
-        "left", "right", or both sides (None).
-
-        .. deprecated:: 2023.02.0
-            Following pandas, the `closed` parameter is deprecated in favor
-            of the `inclusive` parameter, and will be removed in a future
-            version of xarray.
-
-    inclusive : {None, "both", "neither", "left", "right"}, default: None
+    inclusive : {"both", "neither", "left", "right"}, default: "both"
         Include boundaries; whether to set each bound as closed or open.
 
         .. versionadded:: 2023.02.0
-
     calendar : str, default: "standard"
         Calendar type for the datetimes.
     use_cftime : boolean, optional
@@ -1294,8 +1233,6 @@ def date_range(
     if tz is not None:
         use_cftime = False
 
-    inclusive = _infer_inclusive(closed, inclusive)
-
     if _is_standard_calendar(calendar) and use_cftime is not True:
         try:
             return pd.date_range(

xarray/core/dataset.py

@@ -4167,7 +4167,7 @@ def _validate_interp_indexer(x, new_x):
             }
 
         variables: dict[Hashable, Variable] = {}
-        reindex: bool = False
+        reindex_vars: list[Hashable] = []
         for name, var in obj._variables.items():
             if name in indexers:
                 continue
@@ -4188,19 +4188,20 @@ def _validate_interp_indexer(x, new_x):
                 # booleans and objects and retains the dtype but inside
                 # this loop there might be some duplicate code that slows it
                 # down, therefore collect these signals and run it later:
-                reindex = True
+                reindex_vars.append(name)
             elif all(d not in indexers for d in var.dims):
                 # For anything else we can only keep variables if they
                 # are not dependent on any coords that are being
                 # interpolated along:
                 variables[name] = var
 
-        if reindex:
-            reindex_indexers = {
+        if reindex_vars and (
+            reindex_indexers := {
                 k: v for k, (_, v) in validated_indexers.items() if v.dims == (k,)
             }
+        ):
             reindexed = alignment.reindex(
-                obj,
+                obj[reindex_vars],
                 indexers=reindex_indexers,
                 method=method_non_numeric,
                 exclude_vars=variables.keys(),

xarray/tests/test_cftime_offsets.py

@@ -1057,15 +1057,6 @@ def test_rollback(calendar, offset, initial_date_args, partial_expected_date_arg
         False,
         [(1, 1, 2), (1, 1, 3)],
     ),
-    (
-        "0001-01-01",
-        "0001-01-04",
-        None,
-        "D",
-        None,
-        False,
-        [(1, 1, 1), (1, 1, 2), (1, 1, 3), (1, 1, 4)],
-    ),
     (
         "0001-01-01",
         "0001-01-04",
@@ -1294,13 +1285,6 @@ def test_invalid_cftime_range_inputs(
         cftime_range(start, end, periods, freq, inclusive=inclusive)  # type: ignore[arg-type]
 
 
-def test_invalid_cftime_arg() -> None:
-    with pytest.warns(
-        FutureWarning, match="Following pandas, the `closed` parameter is deprecated"
-    ):
-        cftime_range("2000", "2001", None, "YE", closed="left")
-
-
 _CALENDAR_SPECIFIC_MONTH_END_TESTS = [
     ("noleap", [(2, 28), (4, 30), (6, 30), (8, 31), (10, 31), (12, 31)]),
     ("all_leap", [(2, 29), (4, 30), (6, 30), (8, 31), (10, 31), (12, 31)]),
@@ -1534,15 +1518,6 @@ def as_timedelta_not_implemented_error():
         tick.as_timedelta()
 
 
-@pytest.mark.parametrize("function", [cftime_range, date_range])
-def test_cftime_or_date_range_closed_and_inclusive_error(function: Callable) -> None:
-    if function == cftime_range and not has_cftime:
-        pytest.skip("requires cftime")
-
-    with pytest.raises(ValueError, match="Following pandas, deprecated"):
-        function("2000", periods=3, closed=None, inclusive="right")
-
-
 @pytest.mark.parametrize("function", [cftime_range, date_range])
 def test_cftime_or_date_range_invalid_inclusive_value(function: Callable) -> None:
     if function == cftime_range and not has_cftime:
@@ -1552,29 +1527,6 @@ def test_cftime_or_date_range_invalid_inclusive_value(function: Callable) -> Non
         function("2000", periods=3, inclusive="foo")
 
 
-@pytest.mark.parametrize(
-    "function",
-    [
-        pytest.param(cftime_range, id="cftime", marks=requires_cftime),
-        pytest.param(date_range, id="date"),
-    ],
-)
-@pytest.mark.parametrize(
-    ("closed", "inclusive"), [(None, "both"), ("left", "left"), ("right", "right")]
-)
-def test_cftime_or_date_range_closed(
-    function: Callable,
-    closed: Literal["left", "right", None],
-    inclusive: Literal["left", "right", "both"],
-) -> None:
-    with pytest.warns(FutureWarning, match="Following pandas"):
-        result_closed = function("2000-01-01", "2000-01-04", freq="D", closed=closed)
-        result_inclusive = function(
-            "2000-01-01", "2000-01-04", freq="D", inclusive=inclusive
-        )
-        np.testing.assert_equal(result_closed.values, result_inclusive.values)
-
-
 @pytest.mark.parametrize("function", [cftime_range, date_range])
 def test_cftime_or_date_range_inclusive_None(function) -> None:
     if function == cftime_range and not has_cftime:

xarray/tests/test_interp.py

@@ -1065,6 +1065,44 @@ def test_interp1d_complex_out_of_bounds() -> None:
     assert_identical(actual, expected)
 
 
+@requires_scipy
+def test_interp_non_numeric_1d() -> None:
+    ds = xr.Dataset(
+        {
+            "numeric": ("time", 1 + np.arange(0, 4, 1)),
+            "non_numeric": ("time", np.array(["a", "b", "c", "d"])),
+        },
+        coords={"time": (np.arange(0, 4, 1))},
+    )
+    actual = ds.interp(time=np.linspace(0, 3, 7))
+
+    expected = xr.Dataset(
+        {
+            "numeric": ("time", 1 + np.linspace(0, 3, 7)),
+            "non_numeric": ("time", np.array(["a", "b", "b", "c", "c", "d", "d"])),
+        },
+        coords={"time": np.linspace(0, 3, 7)},
+    )
+    xr.testing.assert_identical(actual, expected)
+
+
+@requires_scipy
+def test_interp_non_numeric_nd() -> None:
+    # regression test for GH8099, GH9839
+    ds = xr.Dataset({"x": ("a", np.arange(4))}, coords={"a": (np.arange(4) - 1.5)})
+    t = xr.DataArray(
+        np.random.randn(6).reshape((2, 3)) * 0.5,
+        dims=["r", "s"],
+        coords={"r": np.arange(2) - 0.5, "s": np.arange(3) - 1},
+    )
+    ds["m"] = ds.x > 1
+
+    actual = ds.interp(a=t, method="linear")
+    # with numeric only
+    expected = ds[["x"]].interp(a=t, method="linear")
+    assert_identical(actual[["x"]], expected)
+
+
 @requires_dask
 @requires_scipy
 def test_interp_vectorized_dask():
@@ -1128,4 +1166,4 @@ def test_interp_vectorized_shared_dims(chunk):
         dims=("t", "u"),
         coords={"u": [45, 55], "t": [10, 12], "x": dx, "y": dy},
     )
-    assert_identical(actual, expected)
+    assert_identical(actual, expected)