From 7cc4d5375e57277f316c5a8bb17a58ab8003c5f1 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Wed, 12 Aug 2020 15:42:53 +0200
Subject: [PATCH 01/23] Fix GH-29442 DataFrame.groupby doesn't preserve
 _metadata

This bug is a regression in v1.1.0 and was introduced by the fix for GH-34214 in commit [6f065b].

Underlying cause is that the `*Splitter` classes do not use the `._constructor` property and do not call `__finalize__`.

Please note that the method name used for `__finalize__` calls was my best guess since documentation for the value has been hard to find.

[6f065b]: https://github.com/pandas-dev/pandas/commit/6f065b6d423ea211d803e8be93c27f547541c372
---
 pandas/core/groupby/ops.py                   |  5 +-
 pandas/tests/groupby/test_custom_metadata.py | 83 ++++++++++++++++++++
 2 files changed, 86 insertions(+), 2 deletions(-)
 create mode 100644 pandas/tests/groupby/test_custom_metadata.py

diff --git a/pandas/core/groupby/ops.py b/pandas/core/groupby/ops.py
index 64eb413fe78fa..b3baa39fe3f49 100644
--- a/pandas/core/groupby/ops.py
+++ b/pandas/core/groupby/ops.py
@@ -955,7 +955,8 @@ class SeriesSplitter(DataSplitter):
     def _chop(self, sdata: Series, slice_obj: slice) -> Series:
         # fastpath equivalent to `sdata.iloc[slice_obj]`
         mgr = sdata._mgr.get_slice(slice_obj)
-        return type(sdata)(mgr, name=sdata.name, fastpath=True)
+        return sdata._constructor(mgr, name=sdata.name, fastpath=True)\
+            .__finalize__(sdata, method='groupby')
 
 
 class FrameSplitter(DataSplitter):
@@ -971,7 +972,7 @@ def _chop(self, sdata: DataFrame, slice_obj: slice) -> DataFrame:
         # else:
         #     return sdata.iloc[:, slice_obj]
         mgr = sdata._mgr.get_slice(slice_obj, axis=1 - self.axis)
-        return type(sdata)(mgr)
+        return sdata._constructor(mgr).__finalize__(sdata, method='groupby')
 
 
 def get_splitter(
diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
new file mode 100644
index 0000000000000..f1b5d50d30b22
--- /dev/null
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -0,0 +1,83 @@
+"""
+Test metadata propagation in groupby
+
+The PandasTable class below is implemented according to the [guidelines], and as such would
+expect `__finalize__` to always be called so that the `_pandastable_metadata` is always populated. 
+
+[guidelines]: https://pandas.pydata.org/pandas-docs/stable/development/extending.html#override-constructor-properties
+"""
+
+import pytest
+import pandas as pd
+from warnings import warn
+from typing import List
+
+
+_TABLE_METADATA_FIELD_NAME = '_pandastable_metadata'
+
+
+def _combine_metadata(data: List[str]) -> str:
+    """
+    A mock implementation for testing
+    """
+    return '+'.join(data)
+
+
+class PandasTable(pd.DataFrame):
+    """
+    A pandas dataframe subclass with associated table metadata.
+    """
+
+    _metadata = [_TABLE_METADATA_FIELD_NAME]  # Register metadata fieldnames here
+
+    @property
+    def _constructor(self):
+        return PandasTable
+
+    def __finalize__(self, other, method=None, **kwargs):
+        """
+        This method is responsible for populating metadata when creating new Table-object.
+
+        The method argument is subject to change, and a robust handling of this is implemented
+        """
+        src = [other] #more logic here in actual implementation
+        metadata = _combine_metadata([d.get_metadata() for d in src if isinstance(d, PandasTable)])
+
+        if not metadata:
+            warn('__finalize__ unable to combine metadata for method "{method}", falling back to DataFrame')
+            return pd.DataFrame(self)
+        object.__setattr__(self, _TABLE_METADATA_FIELD_NAME, metadata)
+        return self
+
+    def get_metadata(self):
+        #return object.__getattribute__(self, _TABLE_METADATA_FIELD_NAME)
+        metadata = getattr(self, _TABLE_METADATA_FIELD_NAME, None)
+        if metadata is None:
+            warn('PandasTable object not correctly initialized: no metadata')
+        return metadata
+
+    @staticmethod
+    def from_table_data(df: pd.DataFrame, metadata) -> 'PandasTable':
+        df = PandasTable(df)
+        object.__setattr__(df, _TABLE_METADATA_FIELD_NAME, metadata)
+        return df
+
+
+@pytest.fixture
+def dft():
+    df = pd.DataFrame([[11, 12, 0], [21, 22, 0], [31, 32, 1]], columns={'a','b','g'})
+    return PandasTable.from_table_data(df, 'My metadata')    
+
+
+def test_initial_metadata(dft):
+    assert dft.get_metadata() == 'My metadata'
+
+
+def test_basic_propagation(dft):
+    gg = dft.loc[dft.g==0, :]
+    assert gg.get_metadata() == 'My metadata'
+
+
+def test_groupby(dft):
+    gg = [ab for g, ab in dft.groupby('g')]
+    assert gg[0].get_metadata() is not None

From 0ab766db189bd935557563e9638769dfc38c55bc Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Wed, 12 Aug 2020 16:01:04 +0200
Subject: [PATCH 02/23] Fix most PEP8 issues

I don't know how to truncate remaining long lines without loosing information.
---
 pandas/tests/groupby/test_custom_metadata.py | 23 +++++++++++---------
 1 file changed, 13 insertions(+), 10 deletions(-)

diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
index f1b5d50d30b22..4556566880617 100644
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -1,8 +1,9 @@
 """
 Test metadata propagation in groupby
 
-The PandasTable class below is implemented according to the [guidelines], and as such would
-expect `__finalize__` to always be called so that the `_pandastable_metadata` is always populated. 
+The PandasTable class below is implemented according to the [guidelines],
+and as such would expect `__finalize__` to always be called so that the
+`_pandastable_metadata` is always populated.
 
 [guidelines]: https://pandas.pydata.org/pandas-docs/stable/development/extending.html#override-constructor-properties
 """
@@ -38,19 +39,21 @@ def __finalize__(self, other, method=None, **kwargs):
         """
         This method is responsible for populating metadata when creating new Table-object.
 
-        The method argument is subject to change, and a robust handling of this is implemented
+        The method argument is subject to change, and a robust handling of this
+        is implemented.
         """
-        src = [other] #more logic here in actual implementation
-        metadata = _combine_metadata([d.get_metadata() for d in src if isinstance(d, PandasTable)])
+        src = [other]  # more logic here in actual implementation
+        metadata = _combine_metadata(
+            [d.get_metadata() for d in src if isinstance(d, PandasTable)])
 
         if not metadata:
-            warn('__finalize__ unable to combine metadata for method "{method}", falling back to DataFrame')
+            warn('__finalize__ unable to combine metadata for method "{method}", '
+                 'falling back to DataFrame')
             return pd.DataFrame(self)
         object.__setattr__(self, _TABLE_METADATA_FIELD_NAME, metadata)
         return self
 
     def get_metadata(self):
-        #return object.__getattribute__(self, _TABLE_METADATA_FIELD_NAME)
         metadata = getattr(self, _TABLE_METADATA_FIELD_NAME, None)
         if metadata is None:
             warn('PandasTable object not correctly initialized: no metadata')
@@ -65,8 +68,8 @@ def from_table_data(df: pd.DataFrame, metadata) -> 'PandasTable':
 
 @pytest.fixture
 def dft():
-    df = pd.DataFrame([[11, 12, 0], [21, 22, 0], [31, 32, 1]], columns={'a','b','g'})
-    return PandasTable.from_table_data(df, 'My metadata')    
+    df = pd.DataFrame([[11, 12, 0], [21, 22, 0], [31, 32, 1]], columns={'a', 'b', 'g'})
+    return PandasTable.from_table_data(df, 'My metadata')
 
 
 def test_initial_metadata(dft):
@@ -74,7 +77,7 @@ def test_initial_metadata(dft):
 
 
 def test_basic_propagation(dft):
-    gg = dft.loc[dft.g==0, :]
+    gg = dft.loc[dft.g == 0, :]
     assert gg.get_metadata() == 'My metadata'
 
 

From d7b42e3d292b402293cb22925567857ef9cafa36 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Thu, 13 Aug 2020 09:10:43 +0200
Subject: [PATCH 03/23] Fix remaining PEP8 issues

---
 pandas/tests/groupby/test_custom_metadata.py | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
index 4556566880617..46176fcc15d3f 100644
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -5,7 +5,7 @@
 and as such would expect `__finalize__` to always be called so that the
 `_pandastable_metadata` is always populated.
 
-[guidelines]: https://pandas.pydata.org/pandas-docs/stable/development/extending.html#override-constructor-properties
+[guidelines]: https://pandas.pydata.org/pandas-docs/stable/development/extending.html#override-constructor-properties  # noqa
 """
 
 import pytest
@@ -37,10 +37,9 @@ def _constructor(self):
 
     def __finalize__(self, other, method=None, **kwargs):
         """
-        This method is responsible for populating metadata when creating new Table-object.
+        This method be called after constructor to populate metadata
 
-        The method argument is subject to change, and a robust handling of this
-        is implemented.
+        The "method" argument is subject to change and must be handled robustly.
         """
         src = [other]  # more logic here in actual implementation
         metadata = _combine_metadata(

From fbc602c6e183c22c340e7decbc0ef66732f287f0 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Thu, 13 Aug 2020 12:56:13 +0200
Subject: [PATCH 04/23] Apply "black" styling and fix typo

This should hopefully resolve remaining linter issues.
---
 pandas/core/groupby/ops.py                   |  7 +++--
 pandas/tests/groupby/test_custom_metadata.py | 29 +++++++++++---------
 2 files changed, 20 insertions(+), 16 deletions(-)

diff --git a/pandas/core/groupby/ops.py b/pandas/core/groupby/ops.py
index b3baa39fe3f49..4eef300eb6596 100644
--- a/pandas/core/groupby/ops.py
+++ b/pandas/core/groupby/ops.py
@@ -955,8 +955,9 @@ class SeriesSplitter(DataSplitter):
     def _chop(self, sdata: Series, slice_obj: slice) -> Series:
         # fastpath equivalent to `sdata.iloc[slice_obj]`
         mgr = sdata._mgr.get_slice(slice_obj)
-        return sdata._constructor(mgr, name=sdata.name, fastpath=True)\
-            .__finalize__(sdata, method='groupby')
+        return sdata._constructor(mgr, name=sdata.name, fastpath=True).__finalize__(
+            sdata, method="groupby"
+        )
 
 
 class FrameSplitter(DataSplitter):
@@ -972,7 +973,7 @@ def _chop(self, sdata: DataFrame, slice_obj: slice) -> DataFrame:
         # else:
         #     return sdata.iloc[:, slice_obj]
         mgr = sdata._mgr.get_slice(slice_obj, axis=1 - self.axis)
-        return sdata._constructor(mgr).__finalize__(sdata, method='groupby')
+        return sdata._constructor(mgr).__finalize__(sdata, method="groupby")
 
 
 def get_splitter(
diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
index 46176fcc15d3f..3e4d75fa51112 100644
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -14,14 +14,14 @@
 from typing import List
 
 
-_TABLE_METADATA_FIELD_NAME = '_pandastable_metadata'
+_TABLE_METADATA_FIELD_NAME = "_pandastable_metadata"
 
 
 def _combine_metadata(data: List[str]) -> str:
     """
     A mock implementation for testing
     """
-    return '+'.join(data)
+    return "+".join(data)
 
 
 class PandasTable(pd.DataFrame):
@@ -37,17 +37,20 @@ def _constructor(self):
 
     def __finalize__(self, other, method=None, **kwargs):
         """
-        This method be called after constructor to populate metadata
+        This method will be called after constructor to populate metadata
 
         The "method" argument is subject to change and must be handled robustly.
         """
         src = [other]  # more logic here in actual implementation
         metadata = _combine_metadata(
-            [d.get_metadata() for d in src if isinstance(d, PandasTable)])
+            [d.get_metadata() for d in src if isinstance(d, PandasTable)]
+        )
 
         if not metadata:
-            warn('__finalize__ unable to combine metadata for method "{method}", '
-                 'falling back to DataFrame')
+            warn(
+                '__finalize__ unable to combine metadata for method "{method}", '
+                "falling back to DataFrame"
+            )
             return pd.DataFrame(self)
         object.__setattr__(self, _TABLE_METADATA_FIELD_NAME, metadata)
         return self
@@ -55,11 +58,11 @@ def __finalize__(self, other, method=None, **kwargs):
     def get_metadata(self):
         metadata = getattr(self, _TABLE_METADATA_FIELD_NAME, None)
         if metadata is None:
-            warn('PandasTable object not correctly initialized: no metadata')
+            warn("PandasTable object not correctly initialized: no metadata")
         return metadata
 
     @staticmethod
-    def from_table_data(df: pd.DataFrame, metadata) -> 'PandasTable':
+    def from_table_data(df: pd.DataFrame, metadata) -> "PandasTable":
         df = PandasTable(df)
         object.__setattr__(df, _TABLE_METADATA_FIELD_NAME, metadata)
         return df
@@ -67,19 +70,19 @@ def from_table_data(df: pd.DataFrame, metadata) -> 'PandasTable':
 
 @pytest.fixture
 def dft():
-    df = pd.DataFrame([[11, 12, 0], [21, 22, 0], [31, 32, 1]], columns={'a', 'b', 'g'})
-    return PandasTable.from_table_data(df, 'My metadata')
+    df = pd.DataFrame([[11, 12, 0], [21, 22, 0], [31, 32, 1]], columns={"a", "b", "g"})
+    return PandasTable.from_table_data(df, "My metadata")
 
 
 def test_initial_metadata(dft):
-    assert dft.get_metadata() == 'My metadata'
+    assert dft.get_metadata() == "My metadata"
 
 
 def test_basic_propagation(dft):
     gg = dft.loc[dft.g == 0, :]
-    assert gg.get_metadata() == 'My metadata'
+    assert gg.get_metadata() == "My metadata"
 
 
 def test_groupby(dft):
-    gg = [ab for g, ab in dft.groupby('g')]
+    gg = [ab for g, ab in dft.groupby("g")]
     assert gg[0].get_metadata() is not None

From e9c64ac3b1aaefbe24985dc03106b9c1c7ae9b4f Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Thu, 13 Aug 2020 13:24:04 +0200
Subject: [PATCH 05/23] Fix import sorting issues reported by `isort`

---
 pandas/tests/groupby/test_custom_metadata.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
index 3e4d75fa51112..87742fdfecb1f 100644
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -8,11 +8,12 @@
 [guidelines]: https://pandas.pydata.org/pandas-docs/stable/development/extending.html#override-constructor-properties  # noqa
 """
 
-import pytest
-import pandas as pd
-from warnings import warn
 from typing import List
+from warnings import warn
 
+import pytest
+
+import pandas as pd
 
 _TABLE_METADATA_FIELD_NAME = "_pandastable_metadata"
 

From ec7fd00ffe5be5a2ea1eecf2bd4446e8bc9a3198 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Fri, 14 Aug 2020 10:58:13 +0200
Subject: [PATCH 06/23] Make testcase more concise

---
 pandas/tests/groupby/test_custom_metadata.py | 90 +++-----------------
 1 file changed, 13 insertions(+), 77 deletions(-)

diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
index 87742fdfecb1f..90d4c78d1f0c7 100644
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -1,89 +1,25 @@
-"""
-Test metadata propagation in groupby
+from pandas import DataFrame
 
-The PandasTable class below is implemented according to the [guidelines],
-and as such would expect `__finalize__` to always be called so that the
-`_pandastable_metadata` is always populated.
 
-[guidelines]: https://pandas.pydata.org/pandas-docs/stable/development/extending.html#override-constructor-properties  # noqa
-"""
-
-from typing import List
-from warnings import warn
-
-import pytest
-
-import pandas as pd
-
-_TABLE_METADATA_FIELD_NAME = "_pandastable_metadata"
-
-
-def _combine_metadata(data: List[str]) -> str:
-    """
-    A mock implementation for testing
+class CustomDataFrame(DataFrame):
     """
-    return "+".join(data)
-
+    Extension of DataFrame as described in [guidelines]
 
-class PandasTable(pd.DataFrame):
-    """
-    A pandas dataframe subclass with associated table metadata.
+    [guidelines]: https://pandas.pydata.org/pandas-docs/stable/development/extending.html#override-constructor-properties  # noqa
     """
 
-    _metadata = [_TABLE_METADATA_FIELD_NAME]  # Register metadata fieldnames here
+    _metadata = ["_custom_metadata"]
 
     @property
     def _constructor(self):
-        return PandasTable
-
-    def __finalize__(self, other, method=None, **kwargs):
-        """
-        This method will be called after constructor to populate metadata
-
-        The "method" argument is subject to change and must be handled robustly.
-        """
-        src = [other]  # more logic here in actual implementation
-        metadata = _combine_metadata(
-            [d.get_metadata() for d in src if isinstance(d, PandasTable)]
-        )
-
-        if not metadata:
-            warn(
-                '__finalize__ unable to combine metadata for method "{method}", '
-                "falling back to DataFrame"
-            )
-            return pd.DataFrame(self)
-        object.__setattr__(self, _TABLE_METADATA_FIELD_NAME, metadata)
-        return self
-
-    def get_metadata(self):
-        metadata = getattr(self, _TABLE_METADATA_FIELD_NAME, None)
-        if metadata is None:
-            warn("PandasTable object not correctly initialized: no metadata")
-        return metadata
-
-    @staticmethod
-    def from_table_data(df: pd.DataFrame, metadata) -> "PandasTable":
-        df = PandasTable(df)
-        object.__setattr__(df, _TABLE_METADATA_FIELD_NAME, metadata)
-        return df
-
-
-@pytest.fixture
-def dft():
-    df = pd.DataFrame([[11, 12, 0], [21, 22, 0], [31, 32, 1]], columns={"a", "b", "g"})
-    return PandasTable.from_table_data(df, "My metadata")
-
-
-def test_initial_metadata(dft):
-    assert dft.get_metadata() == "My metadata"
-
+        return CustomDataFrame
 
-def test_basic_propagation(dft):
-    gg = dft.loc[dft.g == 0, :]
-    assert gg.get_metadata() == "My metadata"
 
+def test_groupby_with_custom_metadata():
+    custom_df = CustomDataFrame(
+        [[11, 12, 0], [21, 22, 0], [31, 32, 1]], columns=["a", "b", "g"]
+    )
+    custom_df._custom_metadata = "Custom metadata"
 
-def test_groupby(dft):
-    gg = [ab for g, ab in dft.groupby("g")]
-    assert gg[0].get_metadata() is not None
+    for _, group_df in custom_df.groupby("g"):
+        assert group_df._custom_metadata == "Custom metadata"

From 14056678c7c51af5015bf5b9024615c599dd96d0 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Thu, 27 Aug 2020 09:28:10 +0200
Subject: [PATCH 07/23] Include test from original issue

Added testcase `test_groupby_sum_with_custom_metadata` for functionality exercised in the #GH-29442.
Testcase fails on current code.
---
 pandas/tests/groupby/test_custom_metadata.py | 36 +++++++++++++++-----
 1 file changed, 27 insertions(+), 9 deletions(-)

diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
index 90d4c78d1f0c7..65999e67be8d0 100644
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -1,25 +1,43 @@
-from pandas import DataFrame
+import pandas as pd
 
 
-class CustomDataFrame(DataFrame):
+class SubclassedDataFrame2(pd.DataFrame):
     """
     Extension of DataFrame as described in [guidelines]
 
     [guidelines]: https://pandas.pydata.org/pandas-docs/stable/development/extending.html#override-constructor-properties  # noqa
     """
 
-    _metadata = ["_custom_metadata"]
+    # temporary properties
+    _internal_names = pd.DataFrame._internal_names + ["internal_cache"]
+    _internal_names_set = set(_internal_names)
+
+    # normal properties
+    _metadata = ["added_property"]
 
     @property
     def _constructor(self):
-        return CustomDataFrame
+        return SubclassedDataFrame2
 
 
 def test_groupby_with_custom_metadata():
-    custom_df = CustomDataFrame(
+    custom_df = SubclassedDataFrame2(
         [[11, 12, 0], [21, 22, 0], [31, 32, 1]], columns=["a", "b", "g"]
     )
-    custom_df._custom_metadata = "Custom metadata"
-
-    for _, group_df in custom_df.groupby("g"):
-        assert group_df._custom_metadata == "Custom metadata"
+    custom_df.added_property = "hello_pandas"
+    grouped = custom_df.groupby("g")
+    for _, group_df in grouped:
+        assert group_df.added_property == "hello_pandas"
+
+
+def test_groupby_sum_with_custom_metadata():
+    my_data_as_dictionary = {
+        "mycategorical": [1, 1, 2],
+        "myfloat1": [1.0, 2.0, 3.0],
+        "myfloat2": [1.0, 2.0, 3.0],
+    }
+    sdf = SubclassedDataFrame2(my_data_as_dictionary)
+    sdf.added_property = "hello pandas"
+    grouped = sdf.groupby("mycategorical")[["myfloat1", "myfloat2"]]
+    df = grouped.sum()
+    assert df.added_property == "hello pandas"

From 8d3d896bcb20f7bcb0d3ac2b71098dfb05fff11e Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Thu, 27 Aug 2020 09:42:25 +0200
Subject: [PATCH 08/23] Fix metadata handling for `.groupby(...).sum()`

In order to propagate metadata fields, the `__finalize__` method must be
called for the resulting DataFrame with a reference to input. By
implementing this in `_GroupBy._agg_general`, this is performed as late
as possible for the `.sum()` (and similar) code-paths.

Fixes #GH-29442
---
 pandas/core/groupby/groupby.py | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index 4597afeeaddbf..95bfe35d9b22a 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -993,8 +993,9 @@ def _agg_general(
     ):
         with _group_selection_context(self):
             # try a cython aggregation if we can
+            result = None
             try:
-                return self._cython_agg_general(
+                result = self._cython_agg_general(
                     how=alias,
                     alt=npfunc,
                     numeric_only=numeric_only,
@@ -1009,10 +1010,10 @@ def _agg_general(
                     # raised in _get_cython_function, in some cases can
                     #  be trimmed by implementing cython funcs for more dtypes
                     pass
-
             # apply a non-cython aggregation
-            result = self.aggregate(lambda x: npfunc(x, axis=self.axis))
-            return result
+            if result is None:
+                result = self.aggregate(lambda x: npfunc(x, axis=self.axis))
+            return result.__finalize__(self.obj, method='groupby')
 
     def _cython_agg_general(
         self, how: str, alt=None, numeric_only: bool = True, min_count: int = -1

From 39e9f33d741b05905fcf5fac614881059940055d Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Thu, 27 Aug 2020 10:02:24 +0200
Subject: [PATCH 09/23] Apply black format

---
 pandas/core/groupby/groupby.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index 95bfe35d9b22a..b28a9c6ad0311 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -1013,7 +1013,7 @@ def _agg_general(
             # apply a non-cython aggregation
             if result is None:
                 result = self.aggregate(lambda x: npfunc(x, axis=self.axis))
-            return result.__finalize__(self.obj, method='groupby')
+            return result.__finalize__(self.obj, method="groupby")
 
     def _cython_agg_general(
         self, how: str, alt=None, numeric_only: bool = True, min_count: int = -1

From 7075ef28463cfb1c3db5eb9aa3659791066e84c7 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Wed, 23 Sep 2020 08:50:57 +0200
Subject: [PATCH 10/23] Remove xfail decorator for `sum` aggregation test

---
 pandas/tests/generic/test_finalize.py | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/pandas/tests/generic/test_finalize.py b/pandas/tests/generic/test_finalize.py
index 4d0f1a326225d..3cf45e61ee9c1 100644
--- a/pandas/tests/generic/test_finalize.py
+++ b/pandas/tests/generic/test_finalize.py
@@ -778,6 +778,20 @@ def test_categorical_accessor(method):
     [
         operator.methodcaller("sum"),
         lambda x: x.agg("sum"),
+    ],
+)
+def test_groupby_passing(obj, method):
+    obj.attrs = {"a": 1}
+    result = method(obj.groupby([0, 0]))
+    assert result.attrs == {"a": 1}
+
+
+@pytest.mark.parametrize(
+    "obj", [pd.Series([0, 0]), pd.DataFrame({"A": [0, 1], "B": [1, 2]})]
+)
+@pytest.mark.parametrize(
+    "method",
+    [
         lambda x: x.agg(["sum", "count"]),
         lambda x: x.transform(lambda y: y),
         lambda x: x.apply(lambda y: y),

From 4720c979dea17b2942694ec223fe2db487c299e6 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Fri, 2 Oct 2020 08:33:37 +0200
Subject: [PATCH 11/23] Move `__finalize__` outsize context handler

---
 pandas/core/groupby/groupby.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index a3d2efb8a8e6a..cc25ac1c40f8c 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -1017,7 +1017,7 @@ def _agg_general(
             # apply a non-cython aggregation
             if result is None:
                 result = self.aggregate(lambda x: npfunc(x, axis=self.axis))
-            return result.__finalize__(self.obj, method="groupby")
+        return result.__finalize__(self.obj, method="groupby")
 
     def _cython_agg_general(
         self, how: str, alt=None, numeric_only: bool = True, min_count: int = -1

From 6b290e8eceada67a65f2b685409bd90676e25e54 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Fri, 2 Oct 2020 09:15:22 +0200
Subject: [PATCH 12/23] Revert __finalize__ call in DataSplitter

The call was an attempt to ensure that `__finalize__` was called for DataFrames returned when iterating over grouping result, but it caused unreasonable performance degradation.
The relevant test has been marked with `xfail` in order to preserve it. When a solution to the issue is implemented, this test should be reactivated.
---
 pandas/core/groupby/ops.py                   |  6 ++----
 pandas/tests/groupby/test_custom_metadata.py | 10 ++++++----
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/pandas/core/groupby/ops.py b/pandas/core/groupby/ops.py
index b0b1217143b4f..84efde93c222c 100644
--- a/pandas/core/groupby/ops.py
+++ b/pandas/core/groupby/ops.py
@@ -909,9 +909,7 @@ class SeriesSplitter(DataSplitter):
     def _chop(self, sdata: Series, slice_obj: slice) -> Series:
         # fastpath equivalent to `sdata.iloc[slice_obj]`
         mgr = sdata._mgr.get_slice(slice_obj)
-        return sdata._constructor(mgr, name=sdata.name, fastpath=True).__finalize__(
-            sdata, method="groupby"
-        )
+        return sdata._constructor(mgr, name=sdata.name, fastpath=True)
 
 
 class FrameSplitter(DataSplitter):
@@ -927,7 +925,7 @@ def _chop(self, sdata: DataFrame, slice_obj: slice) -> DataFrame:
         # else:
         #     return sdata.iloc[:, slice_obj]
         mgr = sdata._mgr.get_slice(slice_obj, axis=1 - self.axis)
-        return sdata._constructor(mgr).__finalize__(sdata, method="groupby")
+        return sdata._constructor(mgr)
 
 
 def get_splitter(
diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
index 65999e67be8d0..1f95d17472f9e 100644
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -1,3 +1,5 @@
+import pytest
+
 import pandas as pd
 
 
@@ -8,10 +10,6 @@ class SubclassedDataFrame2(pd.DataFrame):
     [guidelines]: https://pandas.pydata.org/pandas-docs/stable/development/extending.html#override-constructor-properties  # noqa
     """
 
-    # temporary properties
-    _internal_names = pd.DataFrame._internal_names + ["internal_cache"]
-    _internal_names_set = set(_internal_names)
-
     # normal properties
     _metadata = ["added_property"]
 
@@ -20,6 +18,10 @@ def _constructor(self):
         return SubclassedDataFrame2
 
 
+@pytest.mark.xfail(
+    reason="Missing high-performance implementation of metadata support for groupby. "
+    "__finalize__ is not called for grouped dataframes"
+)
 def test_groupby_with_custom_metadata():
     custom_df = SubclassedDataFrame2(
         [[11, 12, 0], [21, 22, 0], [31, 32, 1]], columns=["a", "b", "g"]

From a2bdc3d3dfe80018bb6cabce3583fdf9df6cdca9 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Fri, 2 Oct 2020 09:19:27 +0200
Subject: [PATCH 13/23] Add performance test counting __finalize__ calls

---
 pandas/tests/groupby/test_custom_metadata.py | 36 ++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
index 1f95d17472f9e..9b4c69f575d0a 100644
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -1,3 +1,4 @@
+import numpy as np
 import pytest
 
 import pandas as pd
@@ -43,3 +44,38 @@ def test_groupby_sum_with_custom_metadata():
     grouped = sdf.groupby("mycategorical")[["myfloat1", "myfloat2"]]
     df = grouped.sum()
     assert df.added_property == "hello pandas"
+
+
+def test_groupby_apply_performance_with_custom_metadata():
+    """
+    Check that __finalize__ is not called for each group during .apply
+    """
+    counter = {"value": 0}
+
+    class SubclassedDataFrame3(pd.DataFrame):
+        # normal properties
+        _metadata = ["added_property"]
+
+        @property
+        def _constructor(self):
+            return SubclassedDataFrame3
+
+        def __finalize__(self, *args, **kwargs):
+            counter["value"] += 1
+            return super().__finalize__(*args, **kwargs)
+
+    N = 10 ** 4
+    labels = np.random.randint(0, N // 5, size=N)
+    labels2 = np.random.randint(0, 3, size=N)
+    df = SubclassedDataFrame3(
+        {
+            "key": labels,
+            "key2": labels2,
+            "value1": np.random.randn(N),
+            "value2": ["foo", "bar", "baz", "qux"] * (N // 4),
+        }
+    )
+    df.index = pd.CategoricalIndex(df.key)
+    df.groupby(level="key").apply(lambda x: 1)
+    finalize_call_count = counter["value"]
+    assert finalize_call_count < 42  # Random number << len(labels2)

From 2af844716d4b1ade43cf0cebffd751fa4b8317d7 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Fri, 2 Oct 2020 09:58:04 +0200
Subject: [PATCH 14/23] Add whatsnew entry

---
 doc/source/whatsnew/v1.1.3.rst | 1 +
 1 file changed, 1 insertion(+)

diff --git a/doc/source/whatsnew/v1.1.3.rst b/doc/source/whatsnew/v1.1.3.rst
index 15777abcb8084..87748735ec9ae 100644
--- a/doc/source/whatsnew/v1.1.3.rst
+++ b/doc/source/whatsnew/v1.1.3.rst
@@ -58,6 +58,7 @@ Bug fixes
 - Bug in :meth:`Series.astype` showing too much precision when casting from ``np.float32`` to string dtype (:issue:`36451`)
 - Bug in :meth:`Series.isin` and :meth:`DataFrame.isin` when using ``NaN`` and a row length above 1,000,000 (:issue:`22205`)
 - Bug in :func:`cut` raising a ``ValueError`` when passed a :class:`Series` of labels with ``ordered=False`` (:issue:`36603`)
+- Bug in :class:`BaseGroupedBy` causing ``groupby(...).sum()` and similar to not preserve metadata (:issue:`29442`)
 
 .. ---------------------------------------------------------------------------
 

From ec3eeb154972de907b5ab8a227066657b6418db0 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Fri, 2 Oct 2020 10:37:57 +0200
Subject: [PATCH 15/23] Move __finalize__ inside context manager

In attempt to address test failure in pandas\tests\plotting\test_groupby.py that I cannot reproduce locally.
---
 pandas/core/groupby/groupby.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index cc25ac1c40f8c..a3d2efb8a8e6a 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -1017,7 +1017,7 @@ def _agg_general(
             # apply a non-cython aggregation
             if result is None:
                 result = self.aggregate(lambda x: npfunc(x, axis=self.axis))
-        return result.__finalize__(self.obj, method="groupby")
+            return result.__finalize__(self.obj, method="groupby")
 
     def _cython_agg_general(
         self, how: str, alt=None, numeric_only: bool = True, min_count: int = -1

From 54875637014830710a779a97e627e646a46f0740 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Fri, 2 Oct 2020 12:01:05 +0200
Subject: [PATCH 16/23] Fix typo in whatsnew entry

---
 doc/source/whatsnew/v1.1.3.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/source/whatsnew/v1.1.3.rst b/doc/source/whatsnew/v1.1.3.rst
index 87748735ec9ae..501fe8efd10cd 100644
--- a/doc/source/whatsnew/v1.1.3.rst
+++ b/doc/source/whatsnew/v1.1.3.rst
@@ -58,7 +58,7 @@ Bug fixes
 - Bug in :meth:`Series.astype` showing too much precision when casting from ``np.float32`` to string dtype (:issue:`36451`)
 - Bug in :meth:`Series.isin` and :meth:`DataFrame.isin` when using ``NaN`` and a row length above 1,000,000 (:issue:`22205`)
 - Bug in :func:`cut` raising a ``ValueError`` when passed a :class:`Series` of labels with ``ordered=False`` (:issue:`36603`)
-- Bug in :class:`BaseGroupedBy` causing ``groupby(...).sum()` and similar to not preserve metadata (:issue:`29442`)
+- Bug in :class:`BaseGroupedBy` causing ``groupby(...).sum()`` and similar to not preserve metadata (:issue:`29442`)
 
 .. ---------------------------------------------------------------------------
 

From 8bfd08d75e36a30886b7536b3ee26126a50344ab Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Tue, 6 Oct 2020 10:08:20 +0200
Subject: [PATCH 17/23] Revert 1.1.3 whatsnew entry

---
 doc/source/whatsnew/v1.1.3.rst | 1 -
 1 file changed, 1 deletion(-)

diff --git a/doc/source/whatsnew/v1.1.3.rst b/doc/source/whatsnew/v1.1.3.rst
index 501fe8efd10cd..15777abcb8084 100644
--- a/doc/source/whatsnew/v1.1.3.rst
+++ b/doc/source/whatsnew/v1.1.3.rst
@@ -58,7 +58,6 @@ Bug fixes
 - Bug in :meth:`Series.astype` showing too much precision when casting from ``np.float32`` to string dtype (:issue:`36451`)
 - Bug in :meth:`Series.isin` and :meth:`DataFrame.isin` when using ``NaN`` and a row length above 1,000,000 (:issue:`22205`)
 - Bug in :func:`cut` raising a ``ValueError`` when passed a :class:`Series` of labels with ``ordered=False`` (:issue:`36603`)
-- Bug in :class:`BaseGroupedBy` causing ``groupby(...).sum()`` and similar to not preserve metadata (:issue:`29442`)
 
 .. ---------------------------------------------------------------------------
 

From 7878041149bee9389b623d46afd0650f8c58c62a Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Tue, 6 Oct 2020 10:10:02 +0200
Subject: [PATCH 18/23] Add 1.1.4 whatsnew entry

---
 doc/source/whatsnew/v1.1.4.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/source/whatsnew/v1.1.4.rst b/doc/source/whatsnew/v1.1.4.rst
index e63912ebc8fee..7617d8d36c07d 100644
--- a/doc/source/whatsnew/v1.1.4.rst
+++ b/doc/source/whatsnew/v1.1.4.rst
@@ -22,7 +22,7 @@ Fixed regressions
 
 Bug fixes
 ~~~~~~~~~
--
+- Bug in :class:`BaseGroupedBy` causing ``groupby(...).sum()`` and similar to not preserve metadata (:issue:`29442`)
 
 .. ---------------------------------------------------------------------------
 

From 554869960ea9439950f470bf5142465a74033828 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Thu, 8 Oct 2020 15:34:33 +0200
Subject: [PATCH 19/23] Revert all changes not strictly related to issue

---
 pandas/core/groupby/ops.py                   |  4 ++--
 pandas/tests/groupby/test_custom_metadata.py | 14 --------------
 2 files changed, 2 insertions(+), 16 deletions(-)

diff --git a/pandas/core/groupby/ops.py b/pandas/core/groupby/ops.py
index 84efde93c222c..6051aa3022da1 100644
--- a/pandas/core/groupby/ops.py
+++ b/pandas/core/groupby/ops.py
@@ -909,7 +909,7 @@ class SeriesSplitter(DataSplitter):
     def _chop(self, sdata: Series, slice_obj: slice) -> Series:
         # fastpath equivalent to `sdata.iloc[slice_obj]`
         mgr = sdata._mgr.get_slice(slice_obj)
-        return sdata._constructor(mgr, name=sdata.name, fastpath=True)
+        return type(sdata)(mgr, name=sdata.name, fastpath=True)
 
 
 class FrameSplitter(DataSplitter):
@@ -925,7 +925,7 @@ def _chop(self, sdata: DataFrame, slice_obj: slice) -> DataFrame:
         # else:
         #     return sdata.iloc[:, slice_obj]
         mgr = sdata._mgr.get_slice(slice_obj, axis=1 - self.axis)
-        return sdata._constructor(mgr)
+        return type(sdata)(mgr)
 
 
 def get_splitter(
diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
index 9b4c69f575d0a..5f951ac588dee 100644
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -19,20 +19,6 @@ def _constructor(self):
         return SubclassedDataFrame2
 
 
-@pytest.mark.xfail(
-    reason="Missing high-performance implementation of metadata support for groupby. "
-    "__finalize__ is not called for grouped dataframes"
-)
-def test_groupby_with_custom_metadata():
-    custom_df = SubclassedDataFrame2(
-        [[11, 12, 0], [21, 22, 0], [31, 32, 1]], columns=["a", "b", "g"]
-    )
-    custom_df.added_property = "hello_pandas"
-    grouped = custom_df.groupby("g")
-    for _, group_df in grouped:
-        assert group_df.added_property == "hello_pandas"
-
-
 def test_groupby_sum_with_custom_metadata():
     my_data_as_dictionary = {
         "mycategorical": [1, 1, 2],

From 18e8ff5f434e3db1c1d057ab6db48c75173f4877 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Thu, 8 Oct 2020 15:40:07 +0200
Subject: [PATCH 20/23] Use descriptive names for unit tests

---
 pandas/tests/generic/test_finalize.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/pandas/tests/generic/test_finalize.py b/pandas/tests/generic/test_finalize.py
index 4056fa2227324..5507e98d974c1 100644
--- a/pandas/tests/generic/test_finalize.py
+++ b/pandas/tests/generic/test_finalize.py
@@ -774,7 +774,7 @@ def test_categorical_accessor(method):
         lambda x: x.agg("sum"),
     ],
 )
-def test_groupby_passing(obj, method):
+def test_groupby_finalize(obj, method):
     obj.attrs = {"a": 1}
     result = method(obj.groupby([0, 0]))
     assert result.attrs == {"a": 1}
@@ -792,7 +792,7 @@ def test_groupby_passing(obj, method):
     ],
 )
 @not_implemented_mark
-def test_groupby(obj, method):
+def test_groupby_finalize_not_implemented(obj, method):
     obj.attrs = {"a": 1}
     result = method(obj.groupby([0, 0]))
     assert result.attrs == {"a": 1}

From 57694ae6d1724c01f5a5a0c6e5016bfdfc899aa9 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Thu, 8 Oct 2020 16:02:13 +0200
Subject: [PATCH 21/23] Remove extraneous imports

---
 pandas/tests/groupby/test_custom_metadata.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
index 5f951ac588dee..28c2ad2729fc0 100644
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ b/pandas/tests/groupby/test_custom_metadata.py
@@ -1,5 +1,4 @@
 import numpy as np
-import pytest
 
 import pandas as pd
 

From d86ae78b3fbcd902a2b0757eb6b5e90a0160aeca Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Fri, 9 Oct 2020 14:40:36 +0200
Subject: [PATCH 22/23] Delete tests for custom metadata

---
 pandas/tests/groupby/test_custom_metadata.py | 66 --------------------
 1 file changed, 66 deletions(-)
 delete mode 100644 pandas/tests/groupby/test_custom_metadata.py

diff --git a/pandas/tests/groupby/test_custom_metadata.py b/pandas/tests/groupby/test_custom_metadata.py
deleted file mode 100644
index 28c2ad2729fc0..0000000000000
--- a/pandas/tests/groupby/test_custom_metadata.py
+++ /dev/null
@@ -1,66 +0,0 @@
-import numpy as np
-
-import pandas as pd
-
-
-class SubclassedDataFrame2(pd.DataFrame):
-    """
-    Extension of DataFrame as described in [guidelines]
-
-    [guidelines]: https://pandas.pydata.org/pandas-docs/stable/development/extending.html#override-constructor-properties  # noqa
-    """
-
-    # normal properties
-    _metadata = ["added_property"]
-
-    @property
-    def _constructor(self):
-        return SubclassedDataFrame2
-
-
-def test_groupby_sum_with_custom_metadata():
-    my_data_as_dictionary = {
-        "mycategorical": [1, 1, 2],
-        "myfloat1": [1.0, 2.0, 3.0],
-        "myfloat2": [1.0, 2.0, 3.0],
-    }
-    sdf = SubclassedDataFrame2(my_data_as_dictionary)
-    sdf.added_property = "hello pandas"
-    grouped = sdf.groupby("mycategorical")[["myfloat1", "myfloat2"]]
-    df = grouped.sum()
-    assert df.added_property == "hello pandas"
-
-
-def test_groupby_apply_performance_with_custom_metadata():
-    """
-    Check that __finalize__ is not called for each group during .apply
-    """
-    counter = {"value": 0}
-
-    class SubclassedDataFrame3(pd.DataFrame):
-        # normal properties
-        _metadata = ["added_property"]
-
-        @property
-        def _constructor(self):
-            return SubclassedDataFrame3
-
-        def __finalize__(self, *args, **kwargs):
-            counter["value"] += 1
-            return super().__finalize__(*args, **kwargs)
-
-    N = 10 ** 4
-    labels = np.random.randint(0, N // 5, size=N)
-    labels2 = np.random.randint(0, 3, size=N)
-    df = SubclassedDataFrame3(
-        {
-            "key": labels,
-            "key2": labels2,
-            "value1": np.random.randn(N),
-            "value2": ["foo", "bar", "baz", "qux"] * (N // 4),
-        }
-    )
-    df.index = pd.CategoricalIndex(df.key)
-    df.groupby(level="key").apply(lambda x: 1)
-    finalize_call_count = counter["value"]
-    assert finalize_call_count < 42  # Random number << len(labels2)

From 6227894372829d4e92231401906bbe229c9d1908 Mon Sep 17 00:00:00 2001
From: janus <janus@insignificancegalore.net>
Date: Fri, 9 Oct 2020 14:41:54 +0200
Subject: [PATCH 23/23] Remove reference to internal class from whatsnew

---
 doc/source/whatsnew/v1.1.4.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/source/whatsnew/v1.1.4.rst b/doc/source/whatsnew/v1.1.4.rst
index 7617d8d36c07d..c9dde182ab831 100644
--- a/doc/source/whatsnew/v1.1.4.rst
+++ b/doc/source/whatsnew/v1.1.4.rst
@@ -22,7 +22,7 @@ Fixed regressions
 
 Bug fixes
 ~~~~~~~~~
-- Bug in :class:`BaseGroupedBy` causing ``groupby(...).sum()`` and similar to not preserve metadata (:issue:`29442`)
+- Bug causing ``groupby(...).sum()`` and similar to not preserve metadata (:issue:`29442`)
 
 .. ---------------------------------------------------------------------------