Remove sel_points (#3261)

* remove .[i ]sel_points

* dangling decorator

* whatsnew

Author: max-sixty

doc/indexing.rst

@@ -392,14 +392,6 @@ These methods may also be applied to ``Dataset`` objects
   You may find increased performance by loading your data into memory first,
   e.g., with :py:meth:`~xarray.Dataset.load`.
 
-.. note::
-
-  Vectorized indexing is a new feature in v0.10.
-  In older versions of xarray, dimensions of indexers are ignored.
-  Dedicated methods for some advanced indexing use cases,
-  ``isel_points`` and ``sel_points`` are now deprecated.
-  See :ref:`more_advanced_indexing` for their alternative.
-
 .. note::
 
   If an indexer is a :py:meth:`~xarray.DataArray`, its coordinates should not

doc/whats-new.rst

@@ -21,6 +21,14 @@ v0.13.0 (unreleased)
 This release increases the minimum required Python version from 3.5.0 to 3.5.3
 (:issue:`3089`). By `Guido Imperiale <https://github.com/crusaderky>`_.
 
+Breaking changes
+~~~~~~~~~~~~~~~~
+
+  The ``isel_points`` and ``sel_points`` methods are removed, having been deprecated
+  since v0.10.0. These are redundant with the ``isel`` / ``sel`` methods.
+  See :ref:`vectorized_indexing` for the details 
+  By `Maximilian Roos <https://github.com/max-sixty>`_ 
+
 New functions/methods
 ~~~~~~~~~~~~~~~~~~~~~
 

xarray/core/dataarray.py

@@ -1026,32 +1026,6 @@ def sel(
         )
         return self._from_temp_dataset(ds)
 
-    def isel_points(self, dim="points", **indexers) -> "DataArray":
-        """Return a new DataArray whose data is given by pointwise integer
-        indexing along the specified dimension(s).
-
-        See Also
-        --------
-        Dataset.isel_points
-        """
-        ds = self._to_temp_dataset().isel_points(dim=dim, **indexers)
-        return self._from_temp_dataset(ds)
-
-    def sel_points(
-        self, dim="points", method=None, tolerance=None, **indexers
-    ) -> "DataArray":
-        """Return a new DataArray whose dataset is given by pointwise selection
-        of index labels along the specified dimension(s).
-
-        See Also
-        --------
-        Dataset.sel_points
-        """
-        ds = self._to_temp_dataset().sel_points(
-            dim=dim, method=method, tolerance=tolerance, **indexers
-        )
-        return self._from_temp_dataset(ds)
-
     def broadcast_like(
         self, other: Union["DataArray", Dataset], exclude: Iterable[Hashable] = None
     ) -> "DataArray":

xarray/core/dataset.py

@@ -40,7 +40,6 @@
     duck_array_ops,
     formatting,
     groupby,
-    indexing,
     ops,
     pdcompat,
     resample,
@@ -1999,214 +1998,6 @@ def sel(
         result = self.isel(indexers=pos_indexers, drop=drop)
         return result._overwrite_indexes(new_indexes)
 
-    def isel_points(self, dim: Any = "points", **indexers: Any) -> "Dataset":
-        """Returns a new dataset with each array indexed pointwise along the
-        specified dimension(s).
-
-        This method selects pointwise values from each array and is akin to
-        the NumPy indexing behavior of `arr[[0, 1], [0, 1]]`, except this
-        method does not require knowing the order of each array's dimensions.
-
-        Parameters
-        ----------
-        dim : hashable or DataArray or pandas.Index or other list-like object, 
-              optional
-            Name of the dimension to concatenate along. If dim is provided as a
-            hashable, it must be a new dimension name, in which case it is added
-            along axis=0. If dim is provided as a DataArray or Index or
-            list-like object, its name, which must not be present in the
-            dataset, is used as the dimension to concatenate along and the
-            values are added as a coordinate.
-        **indexers : {dim: indexer, ...}
-            Keyword arguments with names matching dimensions and values given
-            by array-like objects. All indexers must be the same length and
-            1 dimensional.
-
-        Returns
-        -------
-        obj : Dataset
-            A new Dataset with the same contents as this dataset, except each
-            array and dimension is indexed by the appropriate indexers. With
-            pointwise indexing, the new Dataset will always be a copy of the
-            original.
-
-        See Also
-        --------
-        Dataset.sel
-        Dataset.isel
-        Dataset.sel_points
-        DataArray.isel_points
-        """  # noqa
-        warnings.warn(
-            "Dataset.isel_points is deprecated: use Dataset.isel()" "instead.",
-            DeprecationWarning,
-            stacklevel=2,
-        )
-
-        indexer_dims = set(indexers)
-
-        def take(variable, slices):
-            # Note: remove helper function when once when numpy
-            # supports vindex https://github.com/numpy/numpy/pull/6075
-            if hasattr(variable.data, "vindex"):
-                # Special case for dask backed arrays to use vectorised list
-                # indexing
-                sel = variable.data.vindex[slices]
-            else:
-                # Otherwise assume backend is numpy array with 'fancy' indexing
-                sel = variable.data[slices]
-            return sel
-
-        def relevant_keys(mapping):
-            return [
-                k for k, v in mapping.items() if any(d in indexer_dims for d in v.dims)
-            ]
-
-        coords = relevant_keys(self.coords)
-        indexers = {k: np.asarray(v) for k, v in indexers.items()}
-        non_indexed_dims = set(self.dims) - indexer_dims
-        non_indexed_coords = set(self.coords) - set(coords)
-
-        # All the indexers should be iterables
-        # Check that indexers are valid dims, integers, and 1D
-        for k, v in indexers.items():
-            if k not in self.dims:
-                raise ValueError("dimension %s does not exist" % k)
-            if v.dtype.kind != "i":  # type: ignore
-                raise TypeError("Indexers must be integers")
-            if v.ndim != 1:  # type: ignore
-                raise ValueError("Indexers must be 1 dimensional")
-
-        # all the indexers should have the same length
-        lengths = {len(v) for k, v in indexers.items()}
-        if len(lengths) > 1:
-            raise ValueError("All indexers must be the same length")
-
-        # Existing dimensions are not valid choices for the dim argument
-        if isinstance(dim, str):
-            if dim in self.dims:
-                # dim is an invalid string
-                raise ValueError(
-                    "Existing dimension names are not valid "
-                    "choices for the dim argument in sel_points"
-                )
-
-        elif hasattr(dim, "dims"):
-            # dim is a DataArray or Coordinate
-            if dim.name in self.dims:
-                # dim already exists
-                raise ValueError(
-                    "Existing dimensions are not valid choices "
-                    "for the dim argument in sel_points"
-                )
-
-        # Set the new dim_name, and optionally the new dim coordinate
-        # dim is either an array-like or a string
-        if not utils.is_scalar(dim):
-            # dim is array like get name or assign 'points', get as variable
-            dim_name = "points" if not hasattr(dim, "name") else dim.name
-            dim_coord = as_variable(dim, name=dim_name)
-        else:
-            # dim is a string
-            dim_name = dim
-            dim_coord = None  # type: ignore
-
-        reordered = self.transpose(*list(indexer_dims), *list(non_indexed_dims))
-
-        variables = OrderedDict()  # type: ignore
-
-        for name, var in reordered.variables.items():
-            if name in indexers or any(d in indexer_dims for d in var.dims):
-                # slice if var is an indexer or depends on an indexed dim
-                slc = [indexers.get(k, slice(None)) for k in var.dims]
-
-                var_dims = [dim_name] + [d for d in var.dims if d in non_indexed_dims]
-                selection = take(var, tuple(slc))
-                var_subset = type(var)(var_dims, selection, var.attrs)
-                variables[name] = var_subset
-            else:
-                # If not indexed just add it back to variables or coordinates
-                variables[name] = var
-
-        coord_names = (set(coords) & set(variables)) | non_indexed_coords
-
-        dset = self._replace_vars_and_dims(variables, coord_names=coord_names)
-        # Add the dim coord to the new dset. Must be done after creation
-        # because_replace_vars_and_dims can only access existing coords,
-        # not add new ones
-        if dim_coord is not None:
-            dset.coords[dim_name] = dim_coord
-        return dset
-
-    def sel_points(
-        self,
-        dim: Any = "points",
-        method: str = None,
-        tolerance: Number = None,
-        **indexers: Any
-    ):
-        """Returns a new dataset with each array indexed pointwise by tick
-        labels along the specified dimension(s).
-
-        In contrast to `Dataset.isel_points`, indexers for this method should
-        use labels instead of integers.
-
-        In contrast to `Dataset.sel`, this method selects points along the
-        diagonal of multi-dimensional arrays, not the intersection.
-
-        Parameters
-        ----------
-        dim : hashable or DataArray or pandas.Index or other list-like object, 
-              optional
-            Name of the dimension to concatenate along. If dim is provided as a
-            hashable, it must be a new dimension name, in which case it is added
-            along axis=0. If dim is provided as a DataArray or Index or
-            list-like object, its name, which must not be present in the
-            dataset, is used as the dimension to concatenate along and the
-            values are added as a coordinate.
-        method : {None, 'nearest', 'pad'/'ffill', 'backfill'/'bfill'}, optional
-            Method to use for inexact matches (requires pandas>=0.16):
-
-            * None (default): only exact matches
-            * pad / ffill: propagate last valid index value forward
-            * backfill / bfill: propagate next valid index value backward
-            * nearest: use nearest valid index value
-        tolerance : optional
-            Maximum distance between original and new labels for inexact
-            matches. The values of the index at the matching locations must
-            satisfy the equation ``abs(index[indexer] - target) <= tolerance``.
-            Requires pandas>=0.17.
-        **indexers : {dim: indexer, ...}
-            Keyword arguments with names matching dimensions and values given
-            by array-like objects. All indexers must be the same length and
-            1 dimensional.
-
-        Returns
-        -------
-        obj : Dataset
-            A new Dataset with the same contents as this dataset, except each
-            array and dimension is indexed by the appropriate indexers. With
-            pointwise indexing, the new Dataset will always be a copy of the
-            original.
-
-        See Also
-        --------
-        Dataset.sel
-        Dataset.isel
-        Dataset.isel_points
-        DataArray.sel_points
-        """  # noqa
-        warnings.warn(
-            "Dataset.sel_points is deprecated: use Dataset.sel()" "instead.",
-            DeprecationWarning,
-            stacklevel=2,
-        )
-
-        pos_indexers, _ = indexing.remap_label_indexers(
-            self, indexers, method=method, tolerance=tolerance
-        )
-        return self.isel_points(dim=dim, **pos_indexers)
-
     def broadcast_like(
         self, other: Union["Dataset", "DataArray"], exclude: Iterable[Hashable] = None
     ) -> "Dataset":

xarray/tests/test_dataarray.py

@@ -1001,64 +1001,6 @@ def test_isel_drop(self):
         selected = data.isel(x=0, drop=False)
         assert_identical(expected, selected)
 
-    @pytest.mark.filterwarnings("ignore:Dataset.isel_points")
-    def test_isel_points(self):
-        shape = (10, 5, 6)
-        np_array = np.random.random(shape)
-        da = DataArray(
-            np_array, dims=["time", "y", "x"], coords={"time": np.arange(0, 100, 10)}
-        )
-        y = [1, 3]
-        x = [3, 0]
-
-        expected = da.values[:, y, x]
-
-        actual = da.isel_points(y=y, x=x, dim="test_coord")
-        assert actual.coords["test_coord"].shape == (len(y),)
-        assert list(actual.coords) == ["time"]
-        assert actual.dims == ("test_coord", "time")
-
-        actual = da.isel_points(y=y, x=x)
-        assert "points" in actual.dims
-        # Note that because xarray always concatenates along the first
-        # dimension, We must transpose the result to match the numpy style of
-        # concatenation.
-        np.testing.assert_equal(actual.T, expected)
-
-        # a few corner cases
-        da.isel_points(time=[1, 2], x=[2, 2], y=[3, 4])
-        np.testing.assert_allclose(
-            da.isel_points(time=[1], x=[2], y=[4]).values.squeeze(),
-            np_array[1, 4, 2].squeeze(),
-        )
-        da.isel_points(time=[1, 2])
-        y = [-1, 0]
-        x = [-2, 2]
-        expected = da.values[:, y, x]
-        actual = da.isel_points(x=x, y=y).values
-        np.testing.assert_equal(actual.T, expected)
-
-        # test that the order of the indexers doesn't matter
-        assert_identical(da.isel_points(y=y, x=x), da.isel_points(x=x, y=y))
-
-        # make sure we're raising errors in the right places
-        with raises_regex(ValueError, "All indexers must be the same length"):
-            da.isel_points(y=[1, 2], x=[1, 2, 3])
-        with raises_regex(ValueError, "dimension bad_key does not exist"):
-            da.isel_points(bad_key=[1, 2])
-        with raises_regex(TypeError, "Indexers must be integers"):
-            da.isel_points(y=[1.5, 2.2])
-        with raises_regex(TypeError, "Indexers must be integers"):
-            da.isel_points(x=[1, 2, 3], y=slice(3))
-        with raises_regex(ValueError, "Indexers must be 1 dimensional"):
-            da.isel_points(y=1, x=2)
-        with raises_regex(ValueError, "Existing dimension names are not"):
-            da.isel_points(y=[1, 2], x=[1, 2], dim="x")
-
-        # using non string dims
-        actual = da.isel_points(y=[1, 2], x=[1, 2], dim=["A", "B"])
-        assert "points" in actual.coords
-
     def test_loc(self):
         self.ds["x"] = ("x", np.array(list("abcdefghij")))
         da = self.ds["foo"]