BUG: agg order for list is not maintained (pandas-dev#41017)

doc/source/whatsnew/v1.3.0.rst

@@ -973,6 +973,7 @@ Other
 - Bug in :meth:`DataFrame.equals`, :meth:`Series.equals`, :meth:`Index.equals` with object-dtype containing ``np.datetime64("NaT")`` or ``np.timedelta64("NaT")`` (:issue:`39650`)
 - Bug in :func:`pandas.util.show_versions` where console JSON output was not proper JSON (:issue:`39701`)
 - Bug in :meth:`DataFrame.convert_dtypes` incorrectly raised ValueError when called on an empty DataFrame (:issue:`40393`)
+- Bug in :meth:`DataFrame.agg()` not sorting the aggregated axis in the order of the provided aggragation functions when one or more aggregation function fails to produce results (:issue:`33634`)
 - Bug in :meth:`DataFrame.clip` not interpreting missing values as no threshold (:issue:`40420`)
 - Bug in :class:`Series` backed by :class:`DatetimeArray` or :class:`TimedeltaArray` sometimes failing to set the array's ``freq`` to ``None`` (:issue:`41425`)
 

pandas/core/apply.py

@@ -376,12 +376,10 @@ def agg_list_like(self) -> FrameOrSeriesUnion:
             raise ValueError("no results")
 
         try:
-            return concat(results, keys=keys, axis=1, sort=False)
+            concatenated = concat(results, keys=keys, axis=1, sort=False)
         except TypeError as err:
-
             # we are concatting non-NDFrame objects,
             # e.g. a list of scalars
-
             from pandas import Series
 
             result = Series(results, index=keys, name=obj.name)
@@ -390,6 +388,16 @@ def agg_list_like(self) -> FrameOrSeriesUnion:
                     "cannot combine transform and aggregation operations"
                 ) from err
             return result
+        else:
+            # Concat uses the first index to determine the final indexing order.
+            # The union of a shorter first index with the other indices causes
+            # the index sorting to be different from the order of the aggregating
+            # functions. Reindex if this is the case.
+            index_size = concatenated.index.size
+            full_ordered_index = next(
+                result.index for result in results if result.index.size == index_size
+            )
+            return concatenated.reindex(full_ordered_index, copy=False)
 
     def agg_dict_like(self) -> FrameOrSeriesUnion:
         """

pandas/tests/apply/test_frame_apply.py

@@ -1110,10 +1110,9 @@ def test_agg_multiple_mixed_no_warning():
     with tm.assert_produces_warning(None):
         result = mdf[["D", "C", "B", "A"]].agg(["sum", "min"])
 
-    # For backwards compatibility, the result's index is
-    # still sorted by function name, so it's ['min', 'sum']
-    # not ['sum', 'min'].
-    expected = expected[["D", "C", "B", "A"]]
+    # GH40420: the result of .agg should have an index that is sorted
+    # according to the arguments provided to agg.
+    expected = expected[["D", "C", "B", "A"]].reindex(["sum", "min"])
     tm.assert_frame_equal(result, expected)
 
 
@@ -1521,6 +1520,38 @@ def test_apply_np_reducer(float_frame, op, how):
     tm.assert_series_equal(result, expected)
 
 
+def test_aggregation_func_column_order():
+    # GH40420: the result of .agg should have an index that is sorted
+    # according to the arguments provided to agg.
+    df = DataFrame(
+        [
+            ("1", 1, 0, 0),
+            ("2", 2, 0, 0),
+            ("3", 3, 0, 0),
+            ("4", 4, 5, 4),
+            ("5", 5, 6, 6),
+            ("6", 6, 7, 7),
+        ],
+        columns=("item", "att1", "att2", "att3"),
+    )
+
+    def foo(s):
+        return s.sum() / 2
+
+    aggs = ["sum", foo, "count", "min"]
+    result = df.agg(aggs)
+    expected = DataFrame(
+        {
+            "item": ["123456", np.nan, 6, "1"],
+            "att1": [21.0, 10.5, 6.0, 1.0],
+            "att2": [18.0, 9.0, 6.0, 0.0],
+            "att3": [17.0, 8.5, 6.0, 0.0],
+        },
+        index=["sum", "foo", "count", "min"],
+    )
+    tm.assert_frame_equal(result, expected)
+
+
 def test_apply_getitem_axis_1():
     # GH 13427
     df = DataFrame({"a": [0, 1, 2], "b": [1, 2, 3]})