REF: Move groupby.agg to apply (#39311)

Author: rhshadrach

.pre-commit-config.yaml

@@ -135,7 +135,7 @@ repos:
         files: ^pandas/tests/
     -   id: FrameOrSeriesUnion
         name: Check for use of Union[Series, DataFrame] instead of FrameOrSeriesUnion alias
-        entry: Union\[.*(Series.*DataFrame|DataFrame.*Series).*\]
+        entry: Union\[.*(Series,.*DataFrame|DataFrame,.*Series).*\]
         language: pygrep
         types: [python]
         exclude: ^pandas/_typing\.py$

pandas/core/apply.py

@@ -2,7 +2,18 @@
 
 import abc
 import inspect
-from typing import TYPE_CHECKING, Any, Dict, Iterator, List, Optional, Tuple, Type, cast
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Dict,
+    Iterator,
+    List,
+    Optional,
+    Tuple,
+    Type,
+    Union,
+    cast,
+)
 
 import numpy as np
 
@@ -13,6 +24,7 @@
     AggFuncType,
     AggFuncTypeBase,
     AggFuncTypeDict,
+    AggObjType,
     Axis,
     FrameOrSeriesUnion,
 )
@@ -34,6 +46,7 @@
 
 if TYPE_CHECKING:
     from pandas import DataFrame, Index, Series
+    from pandas.core.groupby import DataFrameGroupBy, SeriesGroupBy
 
 ResType = Dict[int, Any]
 
@@ -86,7 +99,7 @@ class Apply(metaclass=abc.ABCMeta):
 
     def __init__(
         self,
-        obj: FrameOrSeriesUnion,
+        obj: AggObjType,
         func,
         raw: bool,
         result_type: Optional[str],
@@ -646,3 +659,28 @@ def apply_standard(self) -> FrameOrSeriesUnion:
             return obj._constructor(mapped, index=obj.index).__finalize__(
                 obj, method="apply"
             )
+
+
+class GroupByApply(Apply):
+    obj: Union[SeriesGroupBy, DataFrameGroupBy]
+
+    def __init__(
+        self,
+        obj: Union[SeriesGroupBy, DataFrameGroupBy],
+        func: AggFuncType,
+        args,
+        kwds,
+    ):
+        kwds = kwds.copy()
+        self.axis = obj.obj._get_axis_number(kwds.get("axis", 0))
+        super().__init__(
+            obj,
+            func,
+            raw=False,
+            result_type=None,
+            args=args,
+            kwds=kwds,
+        )
+
+    def apply(self):
+        raise NotImplementedError

pandas/core/groupby/generic.py

@@ -57,11 +57,11 @@
 from pandas.core import algorithms, nanops
 from pandas.core.aggregation import (
     agg_list_like,
-    aggregate,
     maybe_mangle_lambdas,
     reconstruct_func,
     validate_func_kwargs,
 )
+from pandas.core.apply import GroupByApply
 from pandas.core.arrays import Categorical, ExtensionArray
 from pandas.core.base import DataError, SpecificationError
 import pandas.core.common as com
@@ -952,7 +952,8 @@ def aggregate(self, func=None, *args, engine=None, engine_kwargs=None, **kwargs)
         relabeling, func, columns, order = reconstruct_func(func, **kwargs)
         func = maybe_mangle_lambdas(func)
 
-        result, how = aggregate(self, func, *args, **kwargs)
+        op = GroupByApply(self, func, args, kwargs)
+        result, how = op.agg()
         if how is None:
             return result
 

pandas/tests/groupby/aggregate/test_aggregate.py

@@ -203,6 +203,18 @@ def test_aggregate_str_func(tsframe, groupbyfunc):
     tm.assert_frame_equal(result, expected)
 
 
+def test_agg_str_with_kwarg_axis_1_raises(df, reduction_func):
+    gb = df.groupby(level=0)
+    if reduction_func in ("idxmax", "idxmin"):
+        error = TypeError
+        msg = "reduction operation '.*' not allowed for this dtype"
+    else:
+        error = ValueError
+        msg = f"Operation {reduction_func} does not support axis=1"
+    with pytest.raises(error, match=msg):
+        gb.agg(reduction_func, axis=1)
+
+
 def test_aggregate_item_by_item(df):
     grouped = df.groupby("A")