Use flox for grouped first, last. (#10148)

* Optimize grouped first, last.

1. Use flox where possible.

Closes #9647

* simplify

* add whats-new

* typing

* more typing

* Fix

* docstrings and types

Author: dcherian

doc/whats-new.rst

@@ -41,6 +41,11 @@ New Features
 - Support reading to `GPU memory with Zarr <https://zarr.readthedocs.io/en/stable/user-guide/gpu.html>`_ (:pull:`10078`).
   By `Deepak Cherian <https://github.com/dcherian>`_.
 
+Performance
+~~~~~~~~~~~
+- :py:meth:`DatasetGroupBy.first` and :py:meth:`DatasetGroupBy.last` can now use ``flox`` if available. (:issue:`9647`)
+  By `Deepak Cherian <https://github.com/dcherian>`_.
+
 Breaking changes
 ~~~~~~~~~~~~~~~~
 - Rolled back code that would attempt to catch integer overflow when encoding
@@ -174,8 +179,6 @@ New Features
   :py:class:`pandas.DatetimeIndex` (:pull:`9965`). By `Spencer Clark
   <https://github.com/spencerkclark>`_ and `Kai Mühlbauer
   <https://github.com/kmuehlbauer>`_.
-- :py:meth:`DatasetGroupBy.first` and :py:meth:`DatasetGroupBy.last` can now use ``flox`` if available. (:issue:`9647`)
-  By `Deepak Cherian <https://github.com/dcherian>`_.
 
 Breaking changes
 ~~~~~~~~~~~~~~~~

xarray/core/groupby.py

@@ -993,7 +993,7 @@ def _flox_reduce(
         dim: Dims,
         keep_attrs: bool | None = None,
         **kwargs: Any,
-    ):
+    ) -> T_Xarray:
         """Adaptor function that translates our groupby API to that of flox."""
         import flox
         from flox.xarray import xarray_reduce
@@ -1116,6 +1116,8 @@ def _flox_reduce(
                     # flox always assigns an index so we must drop it here if we don't need it.
                     to_drop.append(grouper.name)
                     continue
+                # TODO: We can't simply use `self.encoded.coords` here because it corresponds to `unique_coord`,
+                # NOT `full_index`. We would need to construct a new Coordinates object, that corresponds to `full_index`.
                 new_coords.append(
                     # Using IndexVariable here ensures we reconstruct PandasMultiIndex with
                     # all associated levels properly.
@@ -1361,7 +1363,12 @@ def where(self, cond, other=dtypes.NA) -> T_Xarray:
         """
         return ops.where_method(self, cond, other)
 
-    def _first_or_last(self, op, skipna, keep_attrs):
+    def _first_or_last(
+        self,
+        op: Literal["first" | "last"],
+        skipna: bool | None,
+        keep_attrs: bool | None,
+    ):
         if all(
             isinstance(maybe_slice, slice)
             and (maybe_slice.stop == maybe_slice.start + 1)
@@ -1372,17 +1379,65 @@ def _first_or_last(self, op, skipna, keep_attrs):
             return self._obj
         if keep_attrs is None:
             keep_attrs = _get_keep_attrs(default=True)
-        return self.reduce(
-            op, dim=[self._group_dim], skipna=skipna, keep_attrs=keep_attrs
-        )
+        if (
+            module_available("flox", minversion="0.10.0")
+            and OPTIONS["use_flox"]
+            and contains_only_chunked_or_numpy(self._obj)
+        ):
+            result = self._flox_reduce(
+                dim=None, func=op, skipna=skipna, keep_attrs=keep_attrs
+            )
+        else:
+            result = self.reduce(
+                getattr(duck_array_ops, op),
+                dim=[self._group_dim],
+                skipna=skipna,
+                keep_attrs=keep_attrs,
+            )
+        return result
 
-    def first(self, skipna: bool | None = None, keep_attrs: bool | None = None):
-        """Return the first element of each group along the group dimension"""
-        return self._first_or_last(duck_array_ops.first, skipna, keep_attrs)
+    def first(
+        self, skipna: bool | None = None, keep_attrs: bool | None = None
+    ) -> T_Xarray:
+        """
+        Return the first element of each group along the group dimension
 
-    def last(self, skipna: bool | None = None, keep_attrs: bool | None = None):
-        """Return the last element of each group along the group dimension"""
-        return self._first_or_last(duck_array_ops.last, skipna, keep_attrs)
+        Parameters
+        ----------
+        skipna : bool or None, optional
+            If True, skip missing values (as marked by NaN). By default, only
+            skips missing values for float dtypes; other dtypes either do not
+            have a sentinel missing value (int) or ``skipna=True`` has not been
+            implemented (object, datetime64 or timedelta64).
+        keep_attrs : bool or None, optional
+            If True, ``attrs`` will be copied from the original
+            object to the new one.  If False, the new object will be
+            returned without attributes.
+
+        """
+        return self._first_or_last("first", skipna, keep_attrs)
+
+    def last(
+        self, skipna: bool | None = None, keep_attrs: bool | None = None
+    ) -> T_Xarray:
+        """
+        Return the last element of each group along the group dimension
+
+        Parameters
+        ----------
+        skipna : bool or None, optional
+            If True, skip missing values (as marked by NaN). By default, only
+            skips missing values for float dtypes; other dtypes either do not
+            have a sentinel missing value (int) or ``skipna=True`` has not been
+            implemented (object, datetime64 or timedelta64).
+        keep_attrs : bool or None, optional
+            If True, ``attrs`` will be copied from the original
+            object to the new one.  If False, the new object will be
+            returned without attributes.
+
+
+        """
+        return self._first_or_last("last", skipna, keep_attrs)
 
     def assign_coords(self, coords=None, **coords_kwargs):
         """Assign coordinates by group.

xarray/core/resample.py

@@ -2,7 +2,7 @@
 
 import warnings
 from collections.abc import Callable, Hashable, Iterable, Sequence
-from typing import TYPE_CHECKING, Any
+from typing import TYPE_CHECKING, Any, Literal
 
 from xarray.core._aggregations import (
     DataArrayResampleAggregations,
@@ -55,8 +55,11 @@ def _flox_reduce(
         keep_attrs: bool | None = None,
         **kwargs,
     ) -> T_Xarray:
-        result = super()._flox_reduce(dim=dim, keep_attrs=keep_attrs, **kwargs)
-        result = result.rename({RESAMPLE_DIM: self._group_dim})
+        result: T_Xarray = (
+            super()
+            ._flox_reduce(dim=dim, keep_attrs=keep_attrs, **kwargs)
+            .rename({RESAMPLE_DIM: self._group_dim})  # type: ignore[assignment]
+        )
         return result
 
     def shuffle_to_chunks(self, chunks: T_Chunks = None):
@@ -103,6 +106,21 @@ def shuffle_to_chunks(self, chunks: T_Chunks = None):
         (grouper,) = self.groupers
         return self._shuffle_obj(chunks).drop_vars(RESAMPLE_DIM)
 
+    def _first_or_last(
+        self, op: Literal["first", "last"], skipna: bool | None, keep_attrs: bool | None
+    ) -> T_Xarray:
+        from xarray.core.dataset import Dataset
+
+        result = super()._first_or_last(op=op, skipna=skipna, keep_attrs=keep_attrs)
+        if isinstance(result, Dataset):
+            # Can't do this in the base class because group_dim is RESAMPLE_DIM
+            # which is not present in the original object
+            for var in result.data_vars:
+                result._variables[var] = result._variables[var].transpose(
+                    *self._obj._variables[var].dims
+                )
+        return result
+
     def _drop_coords(self) -> T_Xarray:
         """Drop non-dimension coordinates along the resampled dimension."""
         obj = self._obj

xarray/groupers.py

@@ -242,7 +242,11 @@ def _factorize_unique(self) -> EncodedGroups:
         unique_coord = Variable(
             dims=codes.name, data=unique_values, attrs=self.group.attrs
         )
-        full_index = pd.Index(unique_values)
+        full_index = (
+            unique_values
+            if isinstance(unique_values, pd.MultiIndex)
+            else pd.Index(unique_values)
+        )
 
         return EncodedGroups(
             codes=codes,

xarray/tests/test_groupby.py

@@ -1618,6 +1618,8 @@ def test_groupby_first_and_last(self) -> None:
         expected = array  # should be a no-op
         assert_identical(expected, actual)
 
+        # TODO: groupby_bins too
+
     def make_groupby_multidim_example_array(self) -> DataArray:
         return DataArray(
             [[[0, 1], [2, 3]], [[5, 10], [15, 20]]],
@@ -2374,13 +2376,13 @@ def test_resample_and_first(self) -> None:
         # upsampling
         expected_time = pd.date_range("2000-01-01", freq="3h", periods=19)
         expected = ds.reindex(time=expected_time)
-        actual = ds.resample(time="3h")
+        rs = ds.resample(time="3h")
         for how in ["mean", "sum", "first", "last"]:
-            method = getattr(actual, how)
+            method = getattr(rs, how)
             result = method()
             assert_equal(expected, result)
         for method in [np.mean]:
-            result = actual.reduce(method)
+            result = rs.reduce(method)
             assert_equal(expected, result)
 
     def test_resample_min_count(self) -> None: