From 8370e3968478d50b538b12a0c9b665330456cb2f Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Thu, 5 Sep 2019 16:34:48 -0500
Subject: [PATCH 1/7] REF/ENH: Refactor NDFrame finalization

In preperation for https://github.com/pandas-dev/pandas/issues/27108
(disallowing duplicates), we need to enhance our metadata propagation.

*We need a way for a particiular attribute to deterimine how it's
propagated for a particular method*. Our current method of metadata
propagation lacked two features

1. It only copies an attribute from a source NDFrame to a new NDFrame.
   There is no way to propagate metadata from a collection of NDFrames
   (say from `pd.concat`) to a new NDFrame.
2. It only and always copies the attribute. This is not always
   appropriate when dealing with a collection of input NDFrames, as the
   source attributes may differ. The resolution of conflicts will differ
   by attribute (for `Series.name` we might throw away the name. For
   `Series.allow_duplicates`, any Series disallowing duplicates should
   mean the output disallows duplicates)
---
 pandas/core/_meta.py                  | 146 ++++++++++++++++++++++++++
 pandas/core/generic.py                |   7 +-
 pandas/tests/generic/test_metadata.py |  49 +++++++++
 3 files changed, 198 insertions(+), 4 deletions(-)
 create mode 100644 pandas/core/_meta.py
 create mode 100644 pandas/tests/generic/test_metadata.py

diff --git a/pandas/core/_meta.py b/pandas/core/_meta.py
new file mode 100644
index 0000000000000..07294a2e7d89f
--- /dev/null
+++ b/pandas/core/_meta.py
@@ -0,0 +1,146 @@
+"""
+Metadata propagation through pandas operations.
+
+This module contains the infrastructure for propagating ``NDFrame._metadata``
+through operations. We perform an operation (say :meth:`pandas.Series.copy`) that
+returns an ``NDFrame`` and would like to propagate the metadata (say ``Series.name``)
+from ``self`` to the new ``NDFrame``.
+
+.. note::
+
+   Currently, pandas doesn't provide a clean, documented API on
+
+   * which methods call finalize
+   * the types passed to finalize for each method
+
+   This is a known limitation we would like to address in the future.
+"""
+from collections import defaultdict
+from functools import wraps
+from typing import TYPE_CHECKING, Any, Callable, Union
+
+from pandas.core.dtypes.generic import ABCDataFrame, ABCSeries
+
+if TYPE_CHECKING:
+    from pandas.core.generic import NDFrame
+
+dispatch = defaultdict(dict)
+dispatch_method_type = Union[Callable[..., "NDFrame"], str]
+
+
+def key_of(method):
+    if isinstance(method, str):
+        # TODO: figure out if this is OK. May be necessary when we have
+        #   things like pd.merge and DataFrame.merge that hit the same finalize.
+        return method
+    elif method:
+        return method.__module__, method.__name__
+
+
+class PandasMetadata:
+    """
+    Dispatch metadata finalization for pandas metadata.
+
+    Users should instantiate a single `PandasMetadata` instance
+    for their piece of metadata and register finalizers for various
+    pandas methods using :meth:`PandsaMetadata.register`.
+
+    Parameters
+    ----------
+    name : str
+        The name of the attribute being finalized.
+
+    Examples
+    --------
+    >>> maxmeta = PandasMetadata("attr")
+
+    Register a finalizer for a given pandas method:
+
+    >>> @maxmeta.register(pd.concat)
+    ... def _(new, concatenator):
+    ...     new.attr = max(x.attr_meta for x in concatenator.objs)
+
+    >>> pd.DataFrame._metadata = ['attr']
+    >>> x = pd.DataFrame({"x"}); x.attr = 1
+    >>> y = pd.DataFrame({"y"}); y.attr = 2
+    >>> pd.concat([x, y]).attr
+    2
+    """
+
+    def __init__(self, name: str):
+        self.name = name
+
+    def register(self, pandas_method: dispatch_method_type):
+        """
+        A decorator to register a finalizer for a specific pandas method.
+
+        Parameters
+        ----------
+        pandas_method : callable or str
+            A pandas method, like :meth:`pandas.concat`, that this finalizer
+            should be used for. The function being decorated will be called
+            with the relevant arguments (typically the output and the source NDFrame).
+            When `NDFrame.__finalize__` is called as a result of `pandas_method`,
+            the registered finalizer will be called.
+        """
+
+        def decorate(func):
+            # TODO: warn of collisions?
+            dispatch[key_of(pandas_method)][self.name] = func
+
+            @wraps(func)
+            def wrapper(*args, **kwargs):
+                return func(*args, **kwargs)
+
+            return wrapper
+
+        return decorate
+
+
+def default_finalizer(new: "NDFrame", other: Any, *, name: str):
+    """
+    The default finalizer when this method, attribute hasn't been overridden.
+
+    This copies the ``_metadata`` attribute from ``other`` to ``self``, modifying
+    ``self`` inplace.
+
+    Parameters
+    ----------
+    new : NDFrame
+        The newly created NDFrame being finalized.
+    other : NDFrame
+        The source NDFrame attributes will be extracted from.
+    """
+    object.__setattr__(new, name, getattr(other, name, None))
+
+
+# ----------------------------------------------------------------------------
+# Pandas Internals.
+
+
+def ndframe_finalize(new: "NDFrame", other: Any, method: dispatch_method_type):
+    """
+    Finalize a new NDFrame.
+
+    The finalizer is looked up from finalizers registered with PandasMetadata.
+    `new` is modified inplace, and nothing is returned.
+
+    Parameters
+    ----------
+    new : NDFrame
+    other : NDFrame
+        Or a list of them? TBD
+    method : callable or str
+    """
+    # To avoid one isinstance per _metadata name, we check up front.
+    # Most of the time `other` is an ndframe, but in some cases (e.g. concat)
+    # it's `_Concatenator` object
+    other_is_ndframe = isinstance(other, (ABCSeries, ABCDataFrame))
+
+    for name in new._metadata:
+        finalizer = dispatch.get(key_of(method), {}).get(name)
+
+        if finalizer:
+            finalizer(new, other)
+        elif other_is_ndframe:
+            default_finalizer(new, other, name=name)
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index b427b1f0ac858..43c198d5707fe 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -62,6 +62,7 @@
 import pandas as pd
 from pandas._typing import Dtype, FilePathOrBuffer
 from pandas.core import missing, nanops
+from pandas.core._meta import ndframe_finalize
 import pandas.core.algorithms as algos
 from pandas.core.base import PandasObject, SelectionMixin
 import pandas.core.common as com
@@ -5175,9 +5176,7 @@ def __finalize__(self, other, method=None, **kwargs):
             types of propagation actions based on this
 
         """
-        if isinstance(other, NDFrame):
-            for name in self._metadata:
-                object.__setattr__(self, name, getattr(other, name, None))
+        ndframe_finalize(self, other, method)
         return self
 
     def __getattr__(self, name):
@@ -6016,7 +6015,7 @@ def copy(self, deep=True):
         dtype: object
         """
         data = self._data.copy(deep=deep)
-        return self._constructor(data).__finalize__(self)
+        return self._constructor(data).__finalize__(self, NDFrame.copy)
 
     def __copy__(self, deep=True):
         return self.copy(deep=deep)
diff --git a/pandas/tests/generic/test_metadata.py b/pandas/tests/generic/test_metadata.py
new file mode 100644
index 0000000000000..8912d7214e3fc
--- /dev/null
+++ b/pandas/tests/generic/test_metadata.py
@@ -0,0 +1,49 @@
+import pytest
+
+import pandas as pd
+from pandas.core._meta import PandasMetadata
+
+mymeta = PandasMetadata("attr")
+
+
+@mymeta.register(pd.core.generic.NDFrame.copy)
+def _(new, other):
+    new.attr = other.attr + 1
+
+
+@mymeta.register("concat")
+def _(new, other):
+    assert isinstance(other, pd.core.reshape.concat._Concatenator)
+    new.attr = sum(x.attr for x in other.objs)
+
+
+@pytest.fixture
+def custom_meta(monkeypatch):
+    original_metadata = []
+
+    for cls in [pd.Series, pd.DataFrame]:
+        original_metadata.append(cls._metadata)
+        custom_metadata = cls._metadata.copy()
+        custom_metadata.append("attr")
+
+        monkeypatch.setattr(cls, "_metadata", custom_metadata)
+
+
+def test_custom_finalizer(custom_meta):
+
+    df = pd.DataFrame({"A": [1, 2]})
+    df.attr = 0
+
+    result = df.copy()
+    assert result.attr == 1
+
+
+def test_concat(custom_meta):
+    df1 = pd.DataFrame({"A": [1, 2]})
+    df1.attr = 2
+
+    df2 = pd.DataFrame({"A": [1, 2]})
+    df2.attr = 3
+
+    result = pd.concat([df1, df2])
+    assert result.attr == 5

From d7bb99c022408b5c23cd780b82c677f566710725 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 9 Sep 2019 08:55:29 -0500
Subject: [PATCH 2/7] refactor

---
 pandas/core/_meta.py                  | 159 ++++++++++----------------
 pandas/core/generic.py                |  17 ++-
 pandas/tests/generic/test_metadata.py |  15 ++-
 3 files changed, 81 insertions(+), 110 deletions(-)

diff --git a/pandas/core/_meta.py b/pandas/core/_meta.py
index 07294a2e7d89f..55907af6c7083 100644
--- a/pandas/core/_meta.py
+++ b/pandas/core/_meta.py
@@ -5,46 +5,43 @@
 through operations. We perform an operation (say :meth:`pandas.Series.copy`) that
 returns an ``NDFrame`` and would like to propagate the metadata (say ``Series.name``)
 from ``self`` to the new ``NDFrame``.
+"""
+from typing import Any, Dict
 
-.. note::
+from pandas.core.dtypes.generic import ABCDataFrame, ABCSeries
 
-   Currently, pandas doesn't provide a clean, documented API on
+from pandas._typing import FrameOrSeries
 
-   * which methods call finalize
-   * the types passed to finalize for each method
 
-   This is a known limitation we would like to address in the future.
-"""
-from collections import defaultdict
-from functools import wraps
-from typing import TYPE_CHECKING, Any, Callable, Union
+class PandasMetadataType(type):
+    """
+    Metaclass controlling creation of metadata finalizers.
 
-from pandas.core.dtypes.generic import ABCDataFrame, ABCSeries
+    This ensures we have one finalizer instance per name, and
+    provides a place to look up finalizer per name.
+    """
 
-if TYPE_CHECKING:
-    from pandas.core.generic import NDFrame
+    # TODO(Py35): Replace metaclass with __subclass_init__
 
-dispatch = defaultdict(dict)
-dispatch_method_type = Union[Callable[..., "NDFrame"], str]
+    _instances = {}  # type: Dict[str, "PandasMetadata"]
 
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
 
-def key_of(method):
-    if isinstance(method, str):
-        # TODO: figure out if this is OK. May be necessary when we have
-        #   things like pd.merge and DataFrame.merge that hit the same finalize.
-        return method
-    elif method:
-        return method.__module__, method.__name__
+    def __call__(self, *args: Any, **kwds: Any) -> Any:
+        name = args[0]
+        if name in self._instances:
+            return self._instances[name]
+        else:
+            new = super().__call__(name, *args, **kwds)
+            self._instances[name] = new
+            return new
 
 
-class PandasMetadata:
+class PandasMetadata(metaclass=PandasMetadataType):
     """
     Dispatch metadata finalization for pandas metadata.
 
-    Users should instantiate a single `PandasMetadata` instance
-    for their piece of metadata and register finalizers for various
-    pandas methods using :meth:`PandsaMetadata.register`.
-
     Parameters
     ----------
     name : str
@@ -52,95 +49,61 @@ class PandasMetadata:
 
     Examples
     --------
-    >>> maxmeta = PandasMetadata("attr")
+    If you want the default resolution (copy from a source NDFrame
+    to a new NDFrame), you can just create an instance
 
-    Register a finalizer for a given pandas method:
+    >>> mymeta = PandasMetadata("mymeta")
 
-    >>> @maxmeta.register(pd.concat)
-    ... def _(new, concatenator):
-    ...     new.attr = max(x.attr_meta for x in concatenator.objs)
+    If you need custom metadata resolution, you'll need to subclass.
 
-    >>> pd.DataFrame._metadata = ['attr']
-    >>> x = pd.DataFrame({"x"}); x.attr = 1
-    >>> y = pd.DataFrame({"y"}); y.attr = 2
-    >>> pd.concat([x, y]).attr
-    2
+    >>> class IncrementMetadata:
+    ...     def finalize_default(self, new, other):
+    ...         setattr(new, self.attr, getattr(other, self.name, -1) + 1)
+
+    >>> increment_metadata = IncrementMetadata("attr")
     """
 
     def __init__(self, name: str):
         self.name = name
 
-    def register(self, pandas_method: dispatch_method_type):
+    def __repr__(self):
+        return "PandasMetadata(name='{}')".format(self.name)
+
+    def finalize_default(self, new: "FrameOrSeries", other: Any):
+        # TODO: tighten the type on Any. Union[NDFrame, _Concatenator, ...]
         """
-        A decorator to register a finalizer for a specific pandas method.
+        The default finalizer when this method, attribute hasn't been overridden.
+
+        This copies the ``_metadata`` attribute from ``other`` to ``self``, modifying
+        ``self`` inplace.
 
         Parameters
         ----------
-        pandas_method : callable or str
-            A pandas method, like :meth:`pandas.concat`, that this finalizer
-            should be used for. The function being decorated will be called
-            with the relevant arguments (typically the output and the source NDFrame).
-            When `NDFrame.__finalize__` is called as a result of `pandas_method`,
-            the registered finalizer will be called.
+        new : NDFrame
+            The newly created NDFrame being finalized.
+        other : Any
+            The source object attributes will be extracted from.
         """
+        # TODO: check perf on this isinstance.
+        if isinstance(other, (ABCSeries, ABCDataFrame)):
+            object.__setattr__(new, self.name, getattr(other, self.name, None))
 
-        def decorate(func):
-            # TODO: warn of collisions?
-            dispatch[key_of(pandas_method)][self.name] = func
+    def finalize_copy(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.finalize_default(new, other)
 
-            @wraps(func)
-            def wrapper(*args, **kwargs):
-                return func(*args, **kwargs)
+    def finalize_concat(self, new: "FrameOrSeries", other: Any):
+        # TODO: type other as _Concatenator
+        self.finalize_default(new, other)
 
-            return wrapper
 
-        return decorate
+class NameMetadata(PandasMetadata):
+    """Finalization for Series.name"""
 
 
-def default_finalizer(new: "NDFrame", other: Any, *, name: str):
-    """
-    The default finalizer when this method, attribute hasn't been overridden.
-
-    This copies the ``_metadata`` attribute from ``other`` to ``self``, modifying
-    ``self`` inplace.
+# TODO: having to create this here feels weird.
+name_metadata = NameMetadata("name")
 
-    Parameters
-    ----------
-    new : NDFrame
-        The newly created NDFrame being finalized.
-    other : NDFrame
-        The source NDFrame attributes will be extracted from.
-    """
-    object.__setattr__(new, name, getattr(other, name, None))
-
-
-# ----------------------------------------------------------------------------
-# Pandas Internals.
-
-
-def ndframe_finalize(new: "NDFrame", other: Any, method: dispatch_method_type):
-    """
-    Finalize a new NDFrame.
-
-    The finalizer is looked up from finalizers registered with PandasMetadata.
-    `new` is modified inplace, and nothing is returned.
-
-    Parameters
-    ----------
-    new : NDFrame
-    other : NDFrame
-        Or a list of them? TBD
-    method : callable or str
-    """
-    # To avoid one isinstance per _metadata name, we check up front.
-    # Most of the time `other` is an ndframe, but in some cases (e.g. concat)
-    # it's `_Concatenator` object
-    other_is_ndframe = isinstance(other, (ABCSeries, ABCDataFrame))
-
-    for name in new._metadata:
-        finalizer = dispatch.get(key_of(method), {}).get(name)
-
-        if finalizer:
-            finalizer(new, other)
-        elif other_is_ndframe:
-            default_finalizer(new, other, name=name)
+# For backwards compat. Do we care about this?
+# We can pretty easily deprecate, require subclasses to make their
+# own instance.
+default_finalizer = PandasMetadata("pandas")
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 43c198d5707fe..740ca9b2362a5 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -62,7 +62,7 @@
 import pandas as pd
 from pandas._typing import Dtype, FilePathOrBuffer
 from pandas.core import missing, nanops
-from pandas.core._meta import ndframe_finalize
+from pandas.core._meta import PandasMetadataType, default_finalizer
 import pandas.core.algorithms as algos
 from pandas.core.base import PandasObject, SelectionMixin
 import pandas.core.common as com
@@ -5164,7 +5164,7 @@ def pipe(self, func, *args, **kwargs):
     # ----------------------------------------------------------------------
     # Attribute access
 
-    def __finalize__(self, other, method=None, **kwargs):
+    def __finalize__(self, other, method=None):
         """
         Propagate metadata from other to self.
 
@@ -5176,7 +5176,16 @@ def __finalize__(self, other, method=None, **kwargs):
             types of propagation actions based on this
 
         """
-        ndframe_finalize(self, other, method)
+        for name in self._metadata:
+            finalizer = PandasMetadataType._instances.get(name, default_finalizer)
+            # TODO: measure perf of this vs. a dict.
+            if method == "copy":
+                finalizer.finalize_copy(self, other)
+            elif method == "concat":
+                finalizer.finalize_concat(self, other)
+            else:
+                finalizer.finalize_default(self, other)
+
         return self
 
     def __getattr__(self, name):
@@ -6015,7 +6024,7 @@ def copy(self, deep=True):
         dtype: object
         """
         data = self._data.copy(deep=deep)
-        return self._constructor(data).__finalize__(self, NDFrame.copy)
+        return self._constructor(data).__finalize__(self, method="copy")
 
     def __copy__(self, deep=True):
         return self.copy(deep=deep)
diff --git a/pandas/tests/generic/test_metadata.py b/pandas/tests/generic/test_metadata.py
index 8912d7214e3fc..9bee9a79b509c 100644
--- a/pandas/tests/generic/test_metadata.py
+++ b/pandas/tests/generic/test_metadata.py
@@ -3,18 +3,17 @@
 import pandas as pd
 from pandas.core._meta import PandasMetadata
 
-mymeta = PandasMetadata("attr")
 
+class MyMeta(PandasMetadata):
+    def finalize_copy(self, new, other):
+        new.attr = other.attr + 1
 
-@mymeta.register(pd.core.generic.NDFrame.copy)
-def _(new, other):
-    new.attr = other.attr + 1
+    def finalize_concat(self, new, other):
+        assert isinstance(other, pd.core.reshape.concat._Concatenator)
+        new.attr = sum(x.attr for x in other.objs)
 
 
-@mymeta.register("concat")
-def _(new, other):
-    assert isinstance(other, pd.core.reshape.concat._Concatenator)
-    new.attr = sum(x.attr for x in other.objs)
+mymeta = MyMeta("attr")
 
 
 @pytest.fixture

From b05782c0127dc0eceb59ff489680d1903f01d113 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 9 Sep 2019 09:08:16 -0500
Subject: [PATCH 3/7] move to meta

---
 pandas/core/generic.py                | 2 +-
 pandas/core/{_meta.py => meta.py}     | 0
 pandas/tests/generic/test_metadata.py | 2 +-
 3 files changed, 2 insertions(+), 2 deletions(-)
 rename pandas/core/{_meta.py => meta.py} (100%)

diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 58d2b7a7e5ba7..5ae08bb2267ee 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -62,7 +62,6 @@
 import pandas as pd
 from pandas._typing import Dtype, FilePathOrBuffer
 from pandas.core import missing, nanops
-from pandas.core._meta import PandasMetadataType, default_finalizer
 import pandas.core.algorithms as algos
 from pandas.core.base import PandasObject, SelectionMixin
 import pandas.core.common as com
@@ -77,6 +76,7 @@
 from pandas.core.indexes.period import Period, PeriodIndex
 import pandas.core.indexing as indexing
 from pandas.core.internals import BlockManager
+from pandas.core.meta import PandasMetadataType, default_finalizer
 from pandas.core.ops import _align_method_FRAME
 
 from pandas.io.formats import format as fmt
diff --git a/pandas/core/_meta.py b/pandas/core/meta.py
similarity index 100%
rename from pandas/core/_meta.py
rename to pandas/core/meta.py
diff --git a/pandas/tests/generic/test_metadata.py b/pandas/tests/generic/test_metadata.py
index 9bee9a79b509c..9899abf207a55 100644
--- a/pandas/tests/generic/test_metadata.py
+++ b/pandas/tests/generic/test_metadata.py
@@ -1,7 +1,7 @@
 import pytest
 
 import pandas as pd
-from pandas.core._meta import PandasMetadata
+from pandas.core.meta import PandasMetadata
 
 
 class MyMeta(PandasMetadata):

From 53576eb77151e24ddbf5edb37aaa3ad68f2e9379 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 9 Sep 2019 10:11:21 -0500
Subject: [PATCH 4/7] mypy

---
 pandas/core/meta.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/pandas/core/meta.py b/pandas/core/meta.py
index 55907af6c7083..f2bbbbdf9d492 100644
--- a/pandas/core/meta.py
+++ b/pandas/core/meta.py
@@ -28,8 +28,7 @@ class PandasMetadataType(type):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
 
-    def __call__(self, *args: Any, **kwds: Any) -> Any:
-        name = args[0]
+    def __call__(self, name, *args: Any, **kwds: Any) -> Any:  # type: ignore
         if name in self._instances:
             return self._instances[name]
         else:

From 3009732e5a437b5f95f543e583c521f31e61bf3a Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Mon, 9 Sep 2019 13:23:22 -0500
Subject: [PATCH 5/7] fixed subclass

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

diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 5ae08bb2267ee..3538e4c3ba852 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -76,7 +76,7 @@
 from pandas.core.indexes.period import Period, PeriodIndex
 import pandas.core.indexing as indexing
 from pandas.core.internals import BlockManager
-from pandas.core.meta import PandasMetadataType, default_finalizer
+from pandas.core.meta import PandasMetadata
 from pandas.core.ops import _align_method_FRAME
 
 from pandas.io.formats import format as fmt
@@ -5177,7 +5177,7 @@ def __finalize__(self, other, method=None):
 
         """
         for name in self._metadata:
-            finalizer = PandasMetadataType._instances.get(name, default_finalizer)
+            finalizer = PandasMetadata(name)
             # TODO: measure perf of this vs. a dict.
             if method == "copy":
                 finalizer.finalize_copy(self, other)

From 710d73a07d6b875e59ac641ea7b575736ef0e1ec Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Tue, 10 Sep 2019 11:58:35 -0500
Subject: [PATCH 6/7] define API

---
 pandas/_typing.py            |   8 ++
 pandas/core/frame.py         |   2 +-
 pandas/core/generic.py       |   5 +
 pandas/core/indexes/multi.py |   4 +-
 pandas/core/meta.py          | 233 +++++++++++++++++++++++++++++++++--
 pandas/core/series.py        |   1 +
 6 files changed, 241 insertions(+), 12 deletions(-)

diff --git a/pandas/_typing.py b/pandas/_typing.py
index 37a5d7945955d..8cffe3c9de0ad 100644
--- a/pandas/_typing.py
+++ b/pandas/_typing.py
@@ -15,6 +15,14 @@
     from pandas.core.sparse.series import SparseSeries  # noqa: F401
     from pandas.core.generic import NDFrame  # noqa: F401
 
+    # TODO: Are we OK with these becoming semi-public?
+    from pandas.core.reshape.concat import _Concatenator as Concatenator  # noqa: F401
+    from pandas.core.reshape.merge import _OrderedMerge as OrderedMerge  # noqa: F401
+    from pandas.core.reshape.merge import _AsOfMerge as AsOfMerge  # noqa: F401
+    from pandas.core.reshape.merge import (  # noqa: F401
+        _MergeOperation as MergeOperation,
+    )
+
 
 AnyArrayLike = TypeVar(
     "AnyArrayLike", "ExtensionArray", "Index", "Series", "SparseSeries", np.ndarray
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 16fece1c7eb8b..34cafe2b6c5df 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -2920,7 +2920,7 @@ def _getitem_multilevel(self, key):
                 result = self._constructor(
                     new_values, index=self.index, columns=result_columns
                 )
-                result = result.__finalize__(self)
+                result = result.__finalize__(self, method="geitem_multilevel")
 
             # If there is only one column being returned, and its name is
             # either an empty string, or a tuple with an empty string as its
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 3538e4c3ba852..b59a9e2e833b4 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -5183,6 +5183,11 @@ def __finalize__(self, other, method=None):
                 finalizer.finalize_copy(self, other)
             elif method == "concat":
                 finalizer.finalize_concat(self, other)
+            elif method == "getitem_multilevel":
+                finalizer.getitem_multilevel(self, other)
+            elif method == "sort_index":
+                finalizer.getitem_sort_index(self, other)
+
             else:
                 finalizer.finalize_default(self, other)
 
diff --git a/pandas/core/indexes/multi.py b/pandas/core/indexes/multi.py
index 761862b9f30e9..122cf721d934c 100644
--- a/pandas/core/indexes/multi.py
+++ b/pandas/core/indexes/multi.py
@@ -1490,7 +1490,9 @@ def _try_mi(k):
             new_index = maybe_droplevels(new_index, k)
             return series._constructor(
                 new_values, index=new_index, name=series.name
-            ).__finalize__(self)
+            ).__finalize__(
+                self
+            )  # TODO: figure out finalize API
 
         try:
             return self._engine.get_value(s, k)
diff --git a/pandas/core/meta.py b/pandas/core/meta.py
index f2bbbbdf9d492..0c5ee307c87d4 100644
--- a/pandas/core/meta.py
+++ b/pandas/core/meta.py
@@ -6,11 +6,19 @@
 returns an ``NDFrame`` and would like to propagate the metadata (say ``Series.name``)
 from ``self`` to the new ``NDFrame``.
 """
-from typing import Any, Dict
+from typing import TYPE_CHECKING, Any, Dict, Union
 
 from pandas.core.dtypes.generic import ABCDataFrame, ABCSeries
 
-from pandas._typing import FrameOrSeries
+if TYPE_CHECKING:
+    from pandas._typing import (
+        FrameOrSeries,
+        Series,
+        Concatenator,
+        OrderedMerge,
+        MergeOperation,
+        AsOfMerge,
+    )
 
 
 class PandasMetadataType(type):
@@ -56,7 +64,7 @@ class PandasMetadata(metaclass=PandasMetadataType):
     If you need custom metadata resolution, you'll need to subclass.
 
     >>> class IncrementMetadata:
-    ...     def finalize_default(self, new, other):
+    ...     def default(self, new, other):
     ...         setattr(new, self.attr, getattr(other, self.name, -1) + 1)
 
     >>> increment_metadata = IncrementMetadata("attr")
@@ -68,8 +76,9 @@ def __init__(self, name: str):
     def __repr__(self):
         return "PandasMetadata(name='{}')".format(self.name)
 
-    def finalize_default(self, new: "FrameOrSeries", other: Any):
-        # TODO: tighten the type on Any. Union[NDFrame, _Concatenator, ...]
+    def default(
+        self, new: "FrameOrSeries", other: Union["FrameOrSeries", "Concatenator"]
+    ):
         """
         The default finalizer when this method, attribute hasn't been overridden.
 
@@ -87,12 +96,216 @@ def finalize_default(self, new: "FrameOrSeries", other: Any):
         if isinstance(other, (ABCSeries, ABCDataFrame)):
             object.__setattr__(new, self.name, getattr(other, self.name, None))
 
-    def finalize_copy(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.finalize_default(new, other)
+    def sort_index(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        self.default(new, other)
 
-    def finalize_concat(self, new: "FrameOrSeries", other: Any):
-        # TODO: type other as _Concatenator
-        self.finalize_default(new, other)
+    def sort_values(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        self.default(new, other)
+
+    def copy(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def rename(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    # TODO: these are private methods.
+    def getitem_multilevel(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        self.default(new, other)
+
+    def simple_replace(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def array_wrap(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def align_frame(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        # TODO: Check the types here. We end up calling __finalize__ twice.
+        #   left.__finalize__(self)
+        #   right.__finalize__(other)
+        #  I think the type on right is Frame->Frame.
+        return self.default(new, other)
+
+    def align_series(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        # TODO: Check the types here. We end up calling __finalize__ twice.
+        #   left.__finalize__(self)
+        #   right.__finalize__(other)
+        #  I think the type on right is Series->Series.
+        return self.default(new, other)
+
+    def consolidate(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def convert(self, new: "FrameOrSeries", other: "FrameOrSeries"):  # XXX: private
+        return self.default(new, other)
+
+    def get_bool_data(
+        self, new: "FrameOrSeries", other: "FrameOrSeries"
+    ):  # XXX: private
+        return self.default(new, other)
+
+    def get_numeric_data(
+        self, new: "FrameOrSeries", other: "FrameOrSeries"
+    ):  # XXX: private
+        return self.default(new, other)
+
+    def reindex_with_indexers(
+        self, new: "FrameOrSeries", other: "FrameOrSeries"
+    ):  # XXX: private
+        return self.default(new, other)
+
+    def slice(self, new: "FrameOrSeries", other: "FrameOrSeries"):  # XXX: private
+        return self.default(new, other)
+
+    def to_dict_of_blocks(
+        self, new: "FrameOrSeries", other: "FrameOrSeries"
+    ):  # XXX: private
+        return self.default(new, other)
+
+    def where(self, new: "FrameOrSeries", other: "FrameOrSeries"):  # XXX: private
+        return self.default(new, other)
+
+    def astype(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def fillna(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def infer_objects(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def interpolate(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def rank(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def replace(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def shift(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def slice_shift(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def swapaxes(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def take(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def transpose(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def tshift(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def tz_convert(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def tz_localize(self, new: "FrameOrSeries", other: "FrameOrSeries"):
+        return self.default(new, other)
+
+    def cum_func(self, new: "FrameOrSeries", other: "FrameOrSeries"):  # XXX: private
+        return self.default(new, other)
+
+    # ops
+    # TODO: constrcut_result, wrapper
+    # def construct_result(self):
+    def flex_wrapper(self, new: "FrameOrSeries", other: Any):
+        return self.default(new, other)
+
+    # ----------------------------------------------------------------------------------
+    # Series
+    # TODO: How should we namespace these?
+    def series_getitem(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_get_values(self, new: "Series", other: "Series"):  # XXX: private
+        return self.default(new, other)
+
+    def series_get_values_tuple(self, new: "Series", other: "Series"):  # XXX: private
+        return self.default(new, other)
+
+    def series_apply(self, new: "Series", other: "Series"):
+        # TODO: We don't call __finalize__ when expanding to a DataFrame. May need to
+        # update the type
+        return self.default(new, other)
+
+    def series_argsort(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_count(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_diff(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_dot(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_isin(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_map(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_repeat(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_reset_index(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_round(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_sort_index(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_sort_values(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_swaplevel(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_take(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_to_period(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_to_sparse(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_to_timestamp(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_view(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    def series_duplicated(self, new: "Series", other: "Series"):
+        return self.default(new, other)
+
+    # Other finalizers
+    def concat(self, new: "FrameOrSeries", other: "Concatenator"):
+        self.default(new, other)
+
+    def merge(
+        self, new: "FrameOrSeries", other: "MergeOperation"
+    ):  # todo: specialize? merge_asof?
+        self.default(new, other)
+
+    def ordered_merge(
+        self, new: "FrameOrSeries", other: "OrderedMerge"
+    ):  # todo: specialize? merge_asof?
+        self.default(new, other)
+
+    def asof_merge(
+        self, new: "FrameOrSeries", other: "AsOfMerge"
+    ):  # todo: specialize? merge_asof?
+        self.default(new, other)
+
+    # TODO: Sparse? Just ignoring for now I think.
 
 
 class NameMetadata(PandasMetadata):
diff --git a/pandas/core/series.py b/pandas/core/series.py
index 10d50e89ca92e..bcfd1884172e2 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -3999,6 +3999,7 @@ def f(x):
         if len(mapped) and isinstance(mapped[0], Series):
             # GH 25959 use pd.array instead of tolist
             # so extension arrays can be used
+            # TODO: would like to apply finalize here.
             return self._constructor_expanddim(pd.array(mapped), index=self.index)
         else:
             return self._constructor(mapped, index=self.index).__finalize__(self)

From ecf3989ae5d6e4d73b4f250b07b8dfa11531e60c Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Tue, 10 Sep 2019 12:13:59 -0500
Subject: [PATCH 7/7] simplify

---
 pandas/_typing.py                     |   8 -
 pandas/core/frame.py                  |   2 +-
 pandas/core/generic.py                |  15 +-
 pandas/core/indexes/multi.py          |   4 +-
 pandas/core/meta.py                   | 255 ++++----------------------
 pandas/tests/generic/test_metadata.py |  16 +-
 6 files changed, 49 insertions(+), 251 deletions(-)

diff --git a/pandas/_typing.py b/pandas/_typing.py
index 8cffe3c9de0ad..37a5d7945955d 100644
--- a/pandas/_typing.py
+++ b/pandas/_typing.py
@@ -15,14 +15,6 @@
     from pandas.core.sparse.series import SparseSeries  # noqa: F401
     from pandas.core.generic import NDFrame  # noqa: F401
 
-    # TODO: Are we OK with these becoming semi-public?
-    from pandas.core.reshape.concat import _Concatenator as Concatenator  # noqa: F401
-    from pandas.core.reshape.merge import _OrderedMerge as OrderedMerge  # noqa: F401
-    from pandas.core.reshape.merge import _AsOfMerge as AsOfMerge  # noqa: F401
-    from pandas.core.reshape.merge import (  # noqa: F401
-        _MergeOperation as MergeOperation,
-    )
-
 
 AnyArrayLike = TypeVar(
     "AnyArrayLike", "ExtensionArray", "Index", "Series", "SparseSeries", np.ndarray
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 34cafe2b6c5df..16fece1c7eb8b 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -2920,7 +2920,7 @@ def _getitem_multilevel(self, key):
                 result = self._constructor(
                     new_values, index=self.index, columns=result_columns
                 )
-                result = result.__finalize__(self, method="geitem_multilevel")
+                result = result.__finalize__(self)
 
             # If there is only one column being returned, and its name is
             # either an empty string, or a tuple with an empty string as its
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index b59a9e2e833b4..3eabeafe296ae 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -5178,18 +5178,7 @@ def __finalize__(self, other, method=None):
         """
         for name in self._metadata:
             finalizer = PandasMetadata(name)
-            # TODO: measure perf of this vs. a dict.
-            if method == "copy":
-                finalizer.finalize_copy(self, other)
-            elif method == "concat":
-                finalizer.finalize_concat(self, other)
-            elif method == "getitem_multilevel":
-                finalizer.getitem_multilevel(self, other)
-            elif method == "sort_index":
-                finalizer.getitem_sort_index(self, other)
-
-            else:
-                finalizer.finalize_default(self, other)
+            finalizer.finalize(self, other, method)
 
         return self
 
@@ -6029,7 +6018,7 @@ def copy(self, deep=True):
         dtype: object
         """
         data = self._data.copy(deep=deep)
-        return self._constructor(data).__finalize__(self, method="copy")
+        return self._constructor(data).__finalize__(self)
 
     def __copy__(self, deep=True):
         return self.copy(deep=deep)
diff --git a/pandas/core/indexes/multi.py b/pandas/core/indexes/multi.py
index 122cf721d934c..761862b9f30e9 100644
--- a/pandas/core/indexes/multi.py
+++ b/pandas/core/indexes/multi.py
@@ -1490,9 +1490,7 @@ def _try_mi(k):
             new_index = maybe_droplevels(new_index, k)
             return series._constructor(
                 new_values, index=new_index, name=series.name
-            ).__finalize__(
-                self
-            )  # TODO: figure out finalize API
+            ).__finalize__(self)
 
         try:
             return self._engine.get_value(s, k)
diff --git a/pandas/core/meta.py b/pandas/core/meta.py
index 0c5ee307c87d4..bf2617b446a39 100644
--- a/pandas/core/meta.py
+++ b/pandas/core/meta.py
@@ -6,19 +6,12 @@
 returns an ``NDFrame`` and would like to propagate the metadata (say ``Series.name``)
 from ``self`` to the new ``NDFrame``.
 """
-from typing import TYPE_CHECKING, Any, Dict, Union
+from typing import TYPE_CHECKING, Any, Dict
 
 from pandas.core.dtypes.generic import ABCDataFrame, ABCSeries
 
 if TYPE_CHECKING:
-    from pandas._typing import (
-        FrameOrSeries,
-        Series,
-        Concatenator,
-        OrderedMerge,
-        MergeOperation,
-        AsOfMerge,
-    )
+    from pandas._typing import FrameOrSeries
 
 
 class PandasMetadataType(type):
@@ -76,9 +69,36 @@ def __init__(self, name: str):
     def __repr__(self):
         return "PandasMetadata(name='{}')".format(self.name)
 
-    def default(
-        self, new: "FrameOrSeries", other: Union["FrameOrSeries", "Concatenator"]
-    ):
+    def finalize(self, new: "FrameOrSeries", other: Any, method):
+        """
+        Run the finalization for `method`.
+
+        Parameters
+        ----------
+        new : DataFrame or Series
+        other : Any
+            One of the following types
+
+            * DataFrame
+            * Series
+            * Concatenator
+            * MergeOperation
+
+        method : str
+            The source method.
+
+        Returns
+        -------
+        None
+            Expected to operate inplace.
+
+        Notes
+        -----
+        The default implementation simply calls ``.default``, ignoring `method`.
+        """
+        self.default(new, other)
+
+    def default(self, new: "FrameOrSeries", other: Any):
         """
         The default finalizer when this method, attribute hasn't been overridden.
 
@@ -96,217 +116,6 @@ def default(
         if isinstance(other, (ABCSeries, ABCDataFrame)):
             object.__setattr__(new, self.name, getattr(other, self.name, None))
 
-    def sort_index(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        self.default(new, other)
-
-    def sort_values(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        self.default(new, other)
-
-    def copy(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def rename(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    # TODO: these are private methods.
-    def getitem_multilevel(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        self.default(new, other)
-
-    def simple_replace(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def array_wrap(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def align_frame(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        # TODO: Check the types here. We end up calling __finalize__ twice.
-        #   left.__finalize__(self)
-        #   right.__finalize__(other)
-        #  I think the type on right is Frame->Frame.
-        return self.default(new, other)
-
-    def align_series(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        # TODO: Check the types here. We end up calling __finalize__ twice.
-        #   left.__finalize__(self)
-        #   right.__finalize__(other)
-        #  I think the type on right is Series->Series.
-        return self.default(new, other)
-
-    def consolidate(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def convert(self, new: "FrameOrSeries", other: "FrameOrSeries"):  # XXX: private
-        return self.default(new, other)
-
-    def get_bool_data(
-        self, new: "FrameOrSeries", other: "FrameOrSeries"
-    ):  # XXX: private
-        return self.default(new, other)
-
-    def get_numeric_data(
-        self, new: "FrameOrSeries", other: "FrameOrSeries"
-    ):  # XXX: private
-        return self.default(new, other)
-
-    def reindex_with_indexers(
-        self, new: "FrameOrSeries", other: "FrameOrSeries"
-    ):  # XXX: private
-        return self.default(new, other)
-
-    def slice(self, new: "FrameOrSeries", other: "FrameOrSeries"):  # XXX: private
-        return self.default(new, other)
-
-    def to_dict_of_blocks(
-        self, new: "FrameOrSeries", other: "FrameOrSeries"
-    ):  # XXX: private
-        return self.default(new, other)
-
-    def where(self, new: "FrameOrSeries", other: "FrameOrSeries"):  # XXX: private
-        return self.default(new, other)
-
-    def astype(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def fillna(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def infer_objects(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def interpolate(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def rank(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def replace(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def shift(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def slice_shift(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def swapaxes(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def take(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def transpose(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def tshift(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def tz_convert(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def tz_localize(self, new: "FrameOrSeries", other: "FrameOrSeries"):
-        return self.default(new, other)
-
-    def cum_func(self, new: "FrameOrSeries", other: "FrameOrSeries"):  # XXX: private
-        return self.default(new, other)
-
-    # ops
-    # TODO: constrcut_result, wrapper
-    # def construct_result(self):
-    def flex_wrapper(self, new: "FrameOrSeries", other: Any):
-        return self.default(new, other)
-
-    # ----------------------------------------------------------------------------------
-    # Series
-    # TODO: How should we namespace these?
-    def series_getitem(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_get_values(self, new: "Series", other: "Series"):  # XXX: private
-        return self.default(new, other)
-
-    def series_get_values_tuple(self, new: "Series", other: "Series"):  # XXX: private
-        return self.default(new, other)
-
-    def series_apply(self, new: "Series", other: "Series"):
-        # TODO: We don't call __finalize__ when expanding to a DataFrame. May need to
-        # update the type
-        return self.default(new, other)
-
-    def series_argsort(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_count(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_diff(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_dot(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_isin(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_map(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_repeat(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_reset_index(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_round(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_sort_index(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_sort_values(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_swaplevel(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_take(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_to_period(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_to_sparse(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_to_timestamp(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_view(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    def series_duplicated(self, new: "Series", other: "Series"):
-        return self.default(new, other)
-
-    # Other finalizers
-    def concat(self, new: "FrameOrSeries", other: "Concatenator"):
-        self.default(new, other)
-
-    def merge(
-        self, new: "FrameOrSeries", other: "MergeOperation"
-    ):  # todo: specialize? merge_asof?
-        self.default(new, other)
-
-    def ordered_merge(
-        self, new: "FrameOrSeries", other: "OrderedMerge"
-    ):  # todo: specialize? merge_asof?
-        self.default(new, other)
-
-    def asof_merge(
-        self, new: "FrameOrSeries", other: "AsOfMerge"
-    ):  # todo: specialize? merge_asof?
-        self.default(new, other)
-
-    # TODO: Sparse? Just ignoring for now I think.
-
 
 class NameMetadata(PandasMetadata):
     """Finalization for Series.name"""
diff --git a/pandas/tests/generic/test_metadata.py b/pandas/tests/generic/test_metadata.py
index 9899abf207a55..0a7ca78f11f99 100644
--- a/pandas/tests/generic/test_metadata.py
+++ b/pandas/tests/generic/test_metadata.py
@@ -5,7 +5,15 @@
 
 
 class MyMeta(PandasMetadata):
-    def finalize_copy(self, new, other):
+    def finalize(self, new, other, method):
+        if method == "concat":
+            self.finalize_concat(new, other)
+        elif method == "copy":
+            self.finalize_copy(new, other)
+        else:
+            super().finalize(new, other, method)
+
+    def default(self, new, other):
         new.attr = other.attr + 1
 
     def finalize_concat(self, new, other):
@@ -28,7 +36,8 @@ def custom_meta(monkeypatch):
         monkeypatch.setattr(cls, "_metadata", custom_metadata)
 
 
-def test_custom_finalizer(custom_meta):
+@pytest.mark.usefixtures("custom_meta")
+def test_custom_finalizer():
 
     df = pd.DataFrame({"A": [1, 2]})
     df.attr = 0
@@ -37,7 +46,8 @@ def test_custom_finalizer(custom_meta):
     assert result.attr == 1
 
 
-def test_concat(custom_meta):
+@pytest.mark.usefixtures("custom_meta")
+def test_concat():
     df1 = pd.DataFrame({"A": [1, 2]})
     df1.attr = 2