From 7e461a18d9f6928132afec6f48ce968b3e989ba6 Mon Sep 17 00:00:00 2001
From: Kaiqi Dong <kaiqi@kth.se>
Date: Mon, 3 Dec 2018 17:43:52 +0100
Subject: [PATCH 01/28] remove \n from docstring

---
 pandas/core/arrays/datetimes.py  | 26 +++++++++++++-------------
 pandas/core/arrays/timedeltas.py | 16 ++++++++--------
 2 files changed, 21 insertions(+), 21 deletions(-)

diff --git a/pandas/core/arrays/datetimes.py b/pandas/core/arrays/datetimes.py
index cfe3afcf3730a..b3df505d56d78 100644
--- a/pandas/core/arrays/datetimes.py
+++ b/pandas/core/arrays/datetimes.py
@@ -82,7 +82,7 @@ def f(self):
         return result
 
     f.__name__ = name
-    f.__doc__ = docstring
+    f.__doc__ = "\n{}\n".format(docstring)
     return property(f)
 
 
@@ -1072,19 +1072,19 @@ def date(self):
 
         return tslib.ints_to_pydatetime(timestamps, box="date")
 
-    year = _field_accessor('year', 'Y', "\n The year of the datetime\n")
+    year = _field_accessor('year', 'Y', "The year of the datetime")
     month = _field_accessor('month', 'M',
-                            "\n The month as January=1, December=12 \n")
-    day = _field_accessor('day', 'D', "\nThe days of the datetime\n")
-    hour = _field_accessor('hour', 'h', "\nThe hours of the datetime\n")
-    minute = _field_accessor('minute', 'm', "\nThe minutes of the datetime\n")
-    second = _field_accessor('second', 's', "\nThe seconds of the datetime\n")
+                            "The month as January=1, December=12")
+    day = _field_accessor('day', 'D', "The days of the datetime")
+    hour = _field_accessor('hour', 'h', "The hours of the datetime")
+    minute = _field_accessor('minute', 'm', "The minutes of the datetime")
+    second = _field_accessor('second', 's', "The seconds of the datetime")
     microsecond = _field_accessor('microsecond', 'us',
-                                  "\nThe microseconds of the datetime\n")
+                                  "The microseconds of the datetime")
     nanosecond = _field_accessor('nanosecond', 'ns',
-                                 "\nThe nanoseconds of the datetime\n")
+                                 "The nanoseconds of the datetime")
     weekofyear = _field_accessor('weekofyear', 'woy',
-                                 "\nThe week ordinal of the year\n")
+                                 "The week ordinal of the year")
     week = weekofyear
     _dayofweek_doc = """
     The day of the week with Monday=0, Sunday=6.
@@ -1129,12 +1129,12 @@ def date(self):
         "The name of day in a week (ex: Friday)\n\n.. deprecated:: 0.23.0")
 
     dayofyear = _field_accessor('dayofyear', 'doy',
-                                "\nThe ordinal day of the year\n")
-    quarter = _field_accessor('quarter', 'q', "\nThe quarter of the date\n")
+                                "The ordinal day of the year")
+    quarter = _field_accessor('quarter', 'q', "The quarter of the date")
     days_in_month = _field_accessor(
         'days_in_month',
         'dim',
-        "\nThe number of days in the month\n")
+        "The number of days in the month")
     daysinmonth = days_in_month
     _is_month_doc = """
         Indicates whether the date is the {first_or_last} day of the month.
diff --git a/pandas/core/arrays/timedeltas.py b/pandas/core/arrays/timedeltas.py
index 830283d31a929..4afc9f5483c2a 100644
--- a/pandas/core/arrays/timedeltas.py
+++ b/pandas/core/arrays/timedeltas.py
@@ -59,7 +59,7 @@ def f(self):
         return result
 
     f.__name__ = name
-    f.__doc__ = docstring
+    f.__doc__ = "\n{}\n".format(docstring)
     return property(f)
 
 
@@ -684,16 +684,16 @@ def to_pytimedelta(self):
         return tslibs.ints_to_pytimedelta(self.asi8)
 
     days = _field_accessor("days", "days",
-                           "\nNumber of days for each element.\n")
+                           "Number of days for each element.")
     seconds = _field_accessor("seconds", "seconds",
-                              "\nNumber of seconds (>= 0 and less than 1 day) "
-                              "for each element.\n")
+                              "Number of seconds (>= 0 and less than 1 day) "
+                              "for each element.")
     microseconds = _field_accessor("microseconds", "microseconds",
-                                   "\nNumber of microseconds (>= 0 and less "
-                                   "than 1 second) for each element.\n")
+                                   "Number of microseconds (>= 0 and less "
+                                   "than 1 second) for each element.")
     nanoseconds = _field_accessor("nanoseconds", "nanoseconds",
-                                  "\nNumber of nanoseconds (>= 0 and less "
-                                  "than 1 microsecond) for each element.\n")
+                                  "Number of nanoseconds (>= 0 and less "
+                                  "than 1 microsecond) for each element.")
 
     @property
     def components(self):

From dea38f24c0067ae3fe9484b837c9649714213bba Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 14 Jan 2020 21:26:31 +0100
Subject: [PATCH 02/28] fix issue 17038

---
 pandas/core/reshape/pivot.py       |  4 +++-
 pandas/tests/reshape/test_pivot.py | 20 ++++++++++++++------
 2 files changed, 17 insertions(+), 7 deletions(-)

diff --git a/pandas/core/reshape/pivot.py b/pandas/core/reshape/pivot.py
index b443ba142369c..9743d90f4dd04 100644
--- a/pandas/core/reshape/pivot.py
+++ b/pandas/core/reshape/pivot.py
@@ -117,7 +117,9 @@ def pivot_table(
                 agged[v] = maybe_downcast_to_dtype(agged[v], data[v].dtype)
 
     table = agged
-    if table.index.nlevels > 1:
+
+    # GH 17038, this check should only happen if index is specified
+    if table.index.nlevels > 1 and index:
         # Related GH #17123
         # If index_names are integers, determine whether the integers refer
         # to the level position or name.
diff --git a/pandas/tests/reshape/test_pivot.py b/pandas/tests/reshape/test_pivot.py
index 743fc50c87e96..46a05123c9fdd 100644
--- a/pandas/tests/reshape/test_pivot.py
+++ b/pandas/tests/reshape/test_pivot.py
@@ -896,12 +896,6 @@ def _check_output(
             totals = table.loc[("All", ""), value_col]
             assert totals == self.data[value_col].mean()
 
-        # no rows
-        rtable = self.data.pivot_table(
-            columns=["AA", "BB"], margins=True, aggfunc=np.mean
-        )
-        assert isinstance(rtable, Series)
-
         table = self.data.pivot_table(index=["AA", "BB"], margins=True, aggfunc="mean")
         for item in ["DD", "EE", "FF"]:
             totals = table.loc[("All", ""), item]
@@ -972,6 +966,20 @@ def test_pivot_integer_columns(self):
 
         tm.assert_frame_equal(table, table2, check_names=False)
 
+    @pytest.mark.parametrize("cols", [(1, 2), ("a", "b"), (1, "b"), ("a", 1)])
+    def test_pivot_table_multiindex_only(self, cols):
+        # GH 17038
+        df2 = DataFrame({cols[0]: [1, 2, 3], cols[1]: [1, 2, 3], "v": [4, 5, 6]})
+
+        result = df2.pivot_table(values="v", columns=cols)
+        expected = DataFrame(
+            [[4, 5, 6]],
+            columns=MultiIndex.from_tuples([(1, 1), (2, 2), (3, 3)], names=cols),
+            index=Index(["v"]),
+        )
+
+        tm.assert_frame_equal(result, expected)
+
     def test_pivot_no_level_overlap(self):
         # GH #1181
 

From cd9e7ac3f31ffaf95cd628863df911dea9fa1248 Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 14 Jan 2020 21:29:43 +0100
Subject: [PATCH 03/28] revert change

---
 pandas/core/reshape/pivot.py       |  3 +--
 pandas/tests/reshape/test_pivot.py | 20 ++++++--------------
 2 files changed, 7 insertions(+), 16 deletions(-)

diff --git a/pandas/core/reshape/pivot.py b/pandas/core/reshape/pivot.py
index 9743d90f4dd04..a7cdbb0da7a4e 100644
--- a/pandas/core/reshape/pivot.py
+++ b/pandas/core/reshape/pivot.py
@@ -118,8 +118,7 @@ def pivot_table(
 
     table = agged
 
-    # GH 17038, this check should only happen if index is specified
-    if table.index.nlevels > 1 and index:
+    if table.index.nlevels > 1:
         # Related GH #17123
         # If index_names are integers, determine whether the integers refer
         # to the level position or name.
diff --git a/pandas/tests/reshape/test_pivot.py b/pandas/tests/reshape/test_pivot.py
index 46a05123c9fdd..743fc50c87e96 100644
--- a/pandas/tests/reshape/test_pivot.py
+++ b/pandas/tests/reshape/test_pivot.py
@@ -896,6 +896,12 @@ def _check_output(
             totals = table.loc[("All", ""), value_col]
             assert totals == self.data[value_col].mean()
 
+        # no rows
+        rtable = self.data.pivot_table(
+            columns=["AA", "BB"], margins=True, aggfunc=np.mean
+        )
+        assert isinstance(rtable, Series)
+
         table = self.data.pivot_table(index=["AA", "BB"], margins=True, aggfunc="mean")
         for item in ["DD", "EE", "FF"]:
             totals = table.loc[("All", ""), item]
@@ -966,20 +972,6 @@ def test_pivot_integer_columns(self):
 
         tm.assert_frame_equal(table, table2, check_names=False)
 
-    @pytest.mark.parametrize("cols", [(1, 2), ("a", "b"), (1, "b"), ("a", 1)])
-    def test_pivot_table_multiindex_only(self, cols):
-        # GH 17038
-        df2 = DataFrame({cols[0]: [1, 2, 3], cols[1]: [1, 2, 3], "v": [4, 5, 6]})
-
-        result = df2.pivot_table(values="v", columns=cols)
-        expected = DataFrame(
-            [[4, 5, 6]],
-            columns=MultiIndex.from_tuples([(1, 1), (2, 2), (3, 3)], names=cols),
-            index=Index(["v"]),
-        )
-
-        tm.assert_frame_equal(result, expected)
-
     def test_pivot_no_level_overlap(self):
         # GH #1181
 

From e5e912be0f596943067a7df812442764d311a086 Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 14 Jan 2020 21:30:16 +0100
Subject: [PATCH 04/28] revert change

---
 pandas/core/reshape/pivot.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/pandas/core/reshape/pivot.py b/pandas/core/reshape/pivot.py
index a7cdbb0da7a4e..b443ba142369c 100644
--- a/pandas/core/reshape/pivot.py
+++ b/pandas/core/reshape/pivot.py
@@ -117,7 +117,6 @@ def pivot_table(
                 agged[v] = maybe_downcast_to_dtype(agged[v], data[v].dtype)
 
     table = agged
-
     if table.index.nlevels > 1:
         # Related GH #17123
         # If index_names are integers, determine whether the integers refer

From 4e1abde5fae34a79c6912290e5cfbad1491093b7 Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Mon, 27 Jan 2020 20:31:28 +0100
Subject: [PATCH 05/28] try fix 31256

---
 pandas/core/groupby/groupby.py           |  2 +-
 pandas/tests/groupby/test_categorical.py | 15 +++++++++++++++
 2 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index a8c96840ff17b..f68471a037ce8 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -935,7 +935,7 @@ def _python_agg_general(self, func, *args, **kwargs):
             result, counts = self.grouper.agg_series(obj, f)
             assert result is not None
             key = base.OutputKey(label=name, position=idx)
-            output[key] = self._try_cast(result, obj, numeric_only=True)
+            output[key] = result
 
         if len(output) == 0:
             return self._python_apply_general(f)
diff --git a/pandas/tests/groupby/test_categorical.py b/pandas/tests/groupby/test_categorical.py
index 9323946581a0d..52e17f9a2594b 100644
--- a/pandas/tests/groupby/test_categorical.py
+++ b/pandas/tests/groupby/test_categorical.py
@@ -1342,3 +1342,18 @@ def test_series_groupby_categorical_aggregation_getitem():
     result = groups["foo"].agg("mean")
     expected = groups.agg("mean")["foo"]
     tm.assert_series_equal(result, expected)
+
+
+@pytest.mark.parametrize("func, expected_values", [
+    (pd.Series.nunique, [1, 1, 2]),
+    (pd.Series.count, [1, 2, 2])])
+def test_groupby_agg_categorical_columns(func, expected_values):
+    # 31256
+    df = pd.DataFrame({"id": [0, 1, 2, 3, 4],
+                       "groups": [0, 1, 1, 2, 2],
+                       "value": pd.Categorical([0, 0, 0, 0, 1])
+                       }).set_index('id')
+    result = df.groupby('groups').agg(func)
+
+    expected = pd.DataFrame({"value": expected_values}, index=pd.Index([0, 1, 2], name="groups"))
+    tm.assert_frame_equal(result, expected)

From 0b917c688218e7620dc1cf792d068810708de524 Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Mon, 27 Jan 2020 20:33:21 +0100
Subject: [PATCH 06/28] pep8

---
 pandas/tests/groupby/test_categorical.py | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

diff --git a/pandas/tests/groupby/test_categorical.py b/pandas/tests/groupby/test_categorical.py
index 52e17f9a2594b..c3b4b6deea68e 100644
--- a/pandas/tests/groupby/test_categorical.py
+++ b/pandas/tests/groupby/test_categorical.py
@@ -1344,16 +1344,22 @@ def test_series_groupby_categorical_aggregation_getitem():
     tm.assert_series_equal(result, expected)
 
 
-@pytest.mark.parametrize("func, expected_values", [
-    (pd.Series.nunique, [1, 1, 2]),
-    (pd.Series.count, [1, 2, 2])])
+@pytest.mark.parametrize(
+    "func, expected_values",
+    [(pd.Series.nunique, [1, 1, 2]), (pd.Series.count, [1, 2, 2])],
+)
 def test_groupby_agg_categorical_columns(func, expected_values):
     # 31256
-    df = pd.DataFrame({"id": [0, 1, 2, 3, 4],
-                       "groups": [0, 1, 1, 2, 2],
-                       "value": pd.Categorical([0, 0, 0, 0, 1])
-                       }).set_index('id')
-    result = df.groupby('groups').agg(func)
+    df = pd.DataFrame(
+        {
+            "id": [0, 1, 2, 3, 4],
+            "groups": [0, 1, 1, 2, 2],
+            "value": pd.Categorical([0, 0, 0, 0, 1]),
+        }
+    ).set_index("id")
+    result = df.groupby("groups").agg(func)
 
-    expected = pd.DataFrame({"value": expected_values}, index=pd.Index([0, 1, 2], name="groups"))
+    expected = pd.DataFrame(
+        {"value": expected_values}, index=pd.Index([0, 1, 2], name="groups")
+    )
     tm.assert_frame_equal(result, expected)

From e9cac5da3f0d23c79a9e7055f05b40652dd7bca9 Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Mon, 27 Jan 2020 21:40:00 +0100
Subject: [PATCH 07/28] fix test

---
 pandas/core/groupby/groupby.py | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index f68471a037ce8..ed5a82aa06aa0 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -807,7 +807,12 @@ def _try_cast(self, result, obj, numeric_only: bool = False):
             dtype = obj.dtype
 
         if not is_scalar(result):
-            if is_extension_array_dtype(dtype) and dtype.kind != "M":
+            # should only cast to ea if it is not a reduction
+            if (
+                is_extension_array_dtype(dtype)
+                and dtype.kind != "M"
+                and len(result) == len(obj)
+            ):
                 # The function can return something of any type, so check
                 #  if the type is compatible with the calling EA.
                 # datetime64tz is handled correctly in agg_series,
@@ -935,7 +940,7 @@ def _python_agg_general(self, func, *args, **kwargs):
             result, counts = self.grouper.agg_series(obj, f)
             assert result is not None
             key = base.OutputKey(label=name, position=idx)
-            output[key] = result
+            output[key] = self._try_cast(result, obj, numeric_only=True)
 
         if len(output) == 0:
             return self._python_apply_general(f)

From e03357e16599295e652faebfc9842721bbe17a9a Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 28 Jan 2020 21:59:26 +0100
Subject: [PATCH 08/28] fix

---
 pandas/core/arrays/base.py     | 2 +-
 pandas/core/groupby/groupby.py | 2 --
 2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index c3c91cea43f6b..caad19f4ee4f9 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -47,7 +47,7 @@ def try_cast_to_ea(cls_or_instance, obj, dtype=None):
     ExtensionArray or obj
     """
     try:
-        result = cls_or_instance._from_sequence(obj, dtype=dtype)
+        result = cls_or_instance._from_sequence(obj, dtype=dtype.name)
     except Exception:
         # We can't predict what downstream EA constructors may raise
         result = obj
diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index ed5a82aa06aa0..4d14c31955c16 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -807,11 +807,9 @@ def _try_cast(self, result, obj, numeric_only: bool = False):
             dtype = obj.dtype
 
         if not is_scalar(result):
-            # should only cast to ea if it is not a reduction
             if (
                 is_extension_array_dtype(dtype)
                 and dtype.kind != "M"
-                and len(result) == len(obj)
             ):
                 # The function can return something of any type, so check
                 #  if the type is compatible with the calling EA.

From a1df393492890f9bd6b6d5257a0649b6e9bfb6f3 Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 28 Jan 2020 22:08:00 +0100
Subject: [PATCH 09/28] fix up tests

---
 pandas/tests/groupby/test_categorical.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/pandas/tests/groupby/test_categorical.py b/pandas/tests/groupby/test_categorical.py
index c3b4b6deea68e..613f36ce4622f 100644
--- a/pandas/tests/groupby/test_categorical.py
+++ b/pandas/tests/groupby/test_categorical.py
@@ -738,7 +738,7 @@ def test_preserve_on_ordered_ops(func, values):
     g = df.groupby("payload")
     result = getattr(g, func)()
     expected = pd.DataFrame(
-        {"payload": [-2, -1], "col": pd.Series(values, dtype=c.dtype)}
+        {"payload": [-2, -1], "col": pd.Series(values, dtype="category")}
     ).set_index("payload")
     tm.assert_frame_equal(result, expected)
 
@@ -1360,6 +1360,7 @@ def test_groupby_agg_categorical_columns(func, expected_values):
     result = df.groupby("groups").agg(func)
 
     expected = pd.DataFrame(
-        {"value": expected_values}, index=pd.Index([0, 1, 2], name="groups")
+        {"value": pd.Index(expected_values, dtype="category")},
+        index=pd.Index([0, 1, 2], name="groups"),
     )
     tm.assert_frame_equal(result, expected)

From 8743c47e781f6219650f2f7cbb40c8e62407f7ac Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 28 Jan 2020 22:23:20 +0100
Subject: [PATCH 10/28] preserve order

---
 pandas/core/arrays/base.py               | 1 +
 pandas/tests/groupby/test_categorical.py | 3 ++-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index caad19f4ee4f9..cb5f1558eaf68 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -48,6 +48,7 @@ def try_cast_to_ea(cls_or_instance, obj, dtype=None):
     """
     try:
         result = cls_or_instance._from_sequence(obj, dtype=dtype.name)
+        result = result.set_ordered(dtype.ordered)
     except Exception:
         # We can't predict what downstream EA constructors may raise
         result = obj
diff --git a/pandas/tests/groupby/test_categorical.py b/pandas/tests/groupby/test_categorical.py
index 613f36ce4622f..e5fa2afc859a6 100644
--- a/pandas/tests/groupby/test_categorical.py
+++ b/pandas/tests/groupby/test_categorical.py
@@ -737,8 +737,9 @@ def test_preserve_on_ordered_ops(func, values):
     df = pd.DataFrame({"payload": [-1, -2, -1, -2], "col": c})
     g = df.groupby("payload")
     result = getattr(g, func)()
+    expected_col = pd.Categorical(values, ordered=True)
     expected = pd.DataFrame(
-        {"payload": [-2, -1], "col": pd.Series(values, dtype="category")}
+        {"payload": [-2, -1], "col": expected_col}
     ).set_index("payload")
     tm.assert_frame_equal(result, expected)
 

From 1e10d7179aa1f21829ee2c7020d5c4db6a43c965 Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 28 Jan 2020 22:33:51 +0100
Subject: [PATCH 11/28] add comment

---
 pandas/core/arrays/base.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index cb5f1558eaf68..1ea699f66bb65 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -48,7 +48,10 @@ def try_cast_to_ea(cls_or_instance, obj, dtype=None):
     """
     try:
         result = cls_or_instance._from_sequence(obj, dtype=dtype.name)
-        result = result.set_ordered(dtype.ordered)
+
+        # still preserve order for categorical
+        if hasattr(cls_or_instance, "ordered"):
+            result = result.set_ordered(dtype.ordered)
     except Exception:
         # We can't predict what downstream EA constructors may raise
         result = obj

From 905b3a5a4a6231ce1b1cb9633ce529f92808776b Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 28 Jan 2020 22:35:32 +0100
Subject: [PATCH 12/28] better test

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

diff --git a/pandas/tests/groupby/test_categorical.py b/pandas/tests/groupby/test_categorical.py
index e5fa2afc859a6..88fe5b113dc89 100644
--- a/pandas/tests/groupby/test_categorical.py
+++ b/pandas/tests/groupby/test_categorical.py
@@ -1361,7 +1361,7 @@ def test_groupby_agg_categorical_columns(func, expected_values):
     result = df.groupby("groups").agg(func)
 
     expected = pd.DataFrame(
-        {"value": pd.Index(expected_values, dtype="category")},
+        {"value": pd.Categorical(expected_values)},
         index=pd.Index([0, 1, 2], name="groups"),
     )
     tm.assert_frame_equal(result, expected)

From c5d670bc0880202794b8478bdc0c7332eb18f007 Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 28 Jan 2020 23:11:02 +0100
Subject: [PATCH 13/28] fixup

---
 pandas/core/arrays/base.py     | 5 +----
 pandas/core/groupby/groupby.py | 6 +++++-
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 1ea699f66bb65..53cd6acdd613e 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -47,11 +47,8 @@ def try_cast_to_ea(cls_or_instance, obj, dtype=None):
     ExtensionArray or obj
     """
     try:
-        result = cls_or_instance._from_sequence(obj, dtype=dtype.name)
+        result = cls_or_instance._from_sequence(obj, dtype=dtype)
 
-        # still preserve order for categorical
-        if hasattr(cls_or_instance, "ordered"):
-            result = result.set_ordered(dtype.ordered)
     except Exception:
         # We can't predict what downstream EA constructors may raise
         result = obj
diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index 4d14c31955c16..6a931567bbccf 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -818,7 +818,11 @@ def _try_cast(self, result, obj, numeric_only: bool = False):
 
                 # return the same type (Series) as our caller
                 cls = dtype.construct_array_type()
-                result = try_cast_to_ea(cls, result, dtype=dtype)
+                result = try_cast_to_ea(cls, result, dtype=dtype.name)
+
+                # still preserve the order for categorical
+                if hasattr(result, "ordered"):
+                    result = result.set_ordered(dtype.ordered)
             elif numeric_only and is_numeric_dtype(dtype) or not numeric_only:
                 result = maybe_downcast_to_dtype(result, dtype)
 

From 916d9b27cac7bb5e7ac569e80547d387fcd407d0 Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 28 Jan 2020 23:11:37 +0100
Subject: [PATCH 14/28] remove blank line

---
 pandas/core/arrays/base.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 53cd6acdd613e..c3c91cea43f6b 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -48,7 +48,6 @@ def try_cast_to_ea(cls_or_instance, obj, dtype=None):
     """
     try:
         result = cls_or_instance._from_sequence(obj, dtype=dtype)
-
     except Exception:
         # We can't predict what downstream EA constructors may raise
         result = obj

From 2208fc26fdb9432b72cc678b71ece1ca7dc86229 Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 28 Jan 2020 23:34:50 +0100
Subject: [PATCH 15/28] fix test

---
 pandas/tests/groupby/test_nth.py | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/pandas/tests/groupby/test_nth.py b/pandas/tests/groupby/test_nth.py
index 0f850f2e94581..1c0c5819b7c8b 100644
--- a/pandas/tests/groupby/test_nth.py
+++ b/pandas/tests/groupby/test_nth.py
@@ -334,9 +334,7 @@ def test_first_last_tz_multi_column(method, ts, alpha):
     result = getattr(df.groupby("group"), method)()
     expected = pd.DataFrame(
         {
-            "category_string": pd.Categorical(
-                [alpha, "c"], dtype=category_string.dtype
-            ),
+            "category_string": pd.Categorical([alpha, "c"]),
             "datetimetz": [ts, Timestamp("2013-01-03", tz="US/Eastern")],
         },
         index=pd.Index([1, 2], name="group"),

From 86a254c9f116af1960678e2f3c0cf46e3ab9699b Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 28 Jan 2020 23:36:19 +0100
Subject: [PATCH 16/28] style

---
 pandas/core/groupby/groupby.py | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index 6a931567bbccf..d93681506bef8 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -807,10 +807,7 @@ def _try_cast(self, result, obj, numeric_only: bool = False):
             dtype = obj.dtype
 
         if not is_scalar(result):
-            if (
-                is_extension_array_dtype(dtype)
-                and dtype.kind != "M"
-            ):
+            if is_extension_array_dtype(dtype) and dtype.kind != "M":
                 # The function can return something of any type, so check
                 #  if the type is compatible with the calling EA.
                 # datetime64tz is handled correctly in agg_series,

From c36d97ba51ef003473c2ad581e4b0c3d4a2604bd Mon Sep 17 00:00:00 2001
From: Kaiqi <kaiqi@kth.se>
Date: Tue, 28 Jan 2020 23:55:12 +0100
Subject: [PATCH 17/28] linting

---
 pandas/tests/groupby/test_categorical.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/pandas/tests/groupby/test_categorical.py b/pandas/tests/groupby/test_categorical.py
index 88fe5b113dc89..cc772585c565f 100644
--- a/pandas/tests/groupby/test_categorical.py
+++ b/pandas/tests/groupby/test_categorical.py
@@ -738,9 +738,9 @@ def test_preserve_on_ordered_ops(func, values):
     g = df.groupby("payload")
     result = getattr(g, func)()
     expected_col = pd.Categorical(values, ordered=True)
-    expected = pd.DataFrame(
-        {"payload": [-2, -1], "col": expected_col}
-    ).set_index("payload")
+    expected = pd.DataFrame({"payload": [-2, -1], "col": expected_col}).set_index(
+        "payload"
+    )
     tm.assert_frame_equal(result, expected)
 
 

From 3f8ea8f8245831164daddbe7bd82590150145e98 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 07:32:12 -0600
Subject: [PATCH 18/28] wip

---
 pandas/core/groupby/groupby.py           | 9 +++------
 pandas/tests/groupby/test_categorical.py | 5 ++---
 2 files changed, 5 insertions(+), 9 deletions(-)

diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index d93681506bef8..8ab777768e39f 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -813,13 +813,10 @@ def _try_cast(self, result, obj, numeric_only: bool = False):
                 # datetime64tz is handled correctly in agg_series,
                 #  so is excluded here.
 
-                # return the same type (Series) as our caller
-                cls = dtype.construct_array_type()
-                result = try_cast_to_ea(cls, result, dtype=dtype.name)
+                if len(result) and isinstance(result[0], dtype.type):
+                    cls = dtype.construct_array_type()
+                    result = try_cast_to_ea(cls, result, dtype=dtype)
 
-                # still preserve the order for categorical
-                if hasattr(result, "ordered"):
-                    result = result.set_ordered(dtype.ordered)
             elif numeric_only and is_numeric_dtype(dtype) or not numeric_only:
                 result = maybe_downcast_to_dtype(result, dtype)
 
diff --git a/pandas/tests/groupby/test_categorical.py b/pandas/tests/groupby/test_categorical.py
index cc772585c565f..8a3b306a9e40b 100644
--- a/pandas/tests/groupby/test_categorical.py
+++ b/pandas/tests/groupby/test_categorical.py
@@ -737,7 +737,7 @@ def test_preserve_on_ordered_ops(func, values):
     df = pd.DataFrame({"payload": [-1, -2, -1, -2], "col": c})
     g = df.groupby("payload")
     result = getattr(g, func)()
-    expected_col = pd.Categorical(values, ordered=True)
+    expected_col = pd.Categorical(values, dtype=c.dtype)
     expected = pd.DataFrame({"payload": [-2, -1], "col": expected_col}).set_index(
         "payload"
     )
@@ -1361,7 +1361,6 @@ def test_groupby_agg_categorical_columns(func, expected_values):
     result = df.groupby("groups").agg(func)
 
     expected = pd.DataFrame(
-        {"value": pd.Categorical(expected_values)},
-        index=pd.Index([0, 1, 2], name="groups"),
+        {"value": expected_values}, index=pd.Index([0, 1, 2], name="groups"),
     )
     tm.assert_frame_equal(result, expected)

From a6ad1a2bfb41db77a83da4b0f13ffd33f30a8a40 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 07:37:36 -0600
Subject: [PATCH 19/28] alternative

---
 pandas/tests/groupby/test_nth.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/pandas/tests/groupby/test_nth.py b/pandas/tests/groupby/test_nth.py
index 1c0c5819b7c8b..0f850f2e94581 100644
--- a/pandas/tests/groupby/test_nth.py
+++ b/pandas/tests/groupby/test_nth.py
@@ -334,7 +334,9 @@ def test_first_last_tz_multi_column(method, ts, alpha):
     result = getattr(df.groupby("group"), method)()
     expected = pd.DataFrame(
         {
-            "category_string": pd.Categorical([alpha, "c"]),
+            "category_string": pd.Categorical(
+                [alpha, "c"], dtype=category_string.dtype
+            ),
             "datetimetz": [ts, Timestamp("2013-01-03", tz="US/Eastern")],
         },
         index=pd.Index([1, 2], name="group"),

From c4ebfa960b11dad3e755b56569ba2e555e54a039 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 09:21:36 -0600
Subject: [PATCH 20/28] Fixups

* Ensure int for resample().sum()
* Ensure float for resample().mean()
---
 pandas/core/groupby/ops.py                   | 11 +++++++++++
 pandas/tests/resample/test_datetime_index.py |  4 +++-
 pandas/tests/resample/test_timedelta.py      |  2 +-
 3 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/pandas/core/groupby/ops.py b/pandas/core/groupby/ops.py
index 679d3668523c2..2e95daa392976 100644
--- a/pandas/core/groupby/ops.py
+++ b/pandas/core/groupby/ops.py
@@ -543,6 +543,17 @@ def _cython_operation(
             if mask.any():
                 result = result.astype("float64")
                 result[mask] = np.nan
+        elif (
+            how == "add"
+            and is_integer_dtype(orig_values.dtype)
+            and is_extension_array_dtype(orig_values.dtype)
+        ):
+            # We need this to ensure that Series[Int64Dtype].resample().sum()
+            # remains int64 dtype.
+            # Two options for avoiding this special case
+            # 1. mask-aware ops and avoid casting to float with NaN above
+            # 2. specify the result dtype when calling this method
+            result = result.astype("int64")
 
         if kind == "aggregate" and self._filter_empty_groups and not counts.all():
             assert result.ndim != 2
diff --git a/pandas/tests/resample/test_datetime_index.py b/pandas/tests/resample/test_datetime_index.py
index 4860329718f54..3ad82b9e075a8 100644
--- a/pandas/tests/resample/test_datetime_index.py
+++ b/pandas/tests/resample/test_datetime_index.py
@@ -122,7 +122,9 @@ def test_resample_integerarray():
 
     result = ts.resample("3T").mean()
     expected = Series(
-        [1, 4, 7], index=pd.date_range("1/1/2000", periods=3, freq="3T"), dtype="Int64"
+        [1, 4, 7],
+        index=pd.date_range("1/1/2000", periods=3, freq="3T"),
+        dtype="float64",
     )
     tm.assert_series_equal(result, expected)
 
diff --git a/pandas/tests/resample/test_timedelta.py b/pandas/tests/resample/test_timedelta.py
index d1bcdc55cb509..a4d14f127b80e 100644
--- a/pandas/tests/resample/test_timedelta.py
+++ b/pandas/tests/resample/test_timedelta.py
@@ -105,7 +105,7 @@ def test_resample_categorical_data_with_timedeltaindex():
         index=pd.to_timedelta([0, 10], unit="s"),
     )
     expected = expected.reindex(["Group_obj", "Group"], axis=1)
-    expected["Group"] = expected["Group_obj"].astype("category")
+    expected["Group"] = expected["Group_obj"]
     tm.assert_frame_equal(result, expected)
 
 

From bbad886420c1f1e65943d52cb7a8da9978d8cffd Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 09:29:52 -0600
Subject: [PATCH 21/28] revert extranesou

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

diff --git a/pandas/tests/groupby/test_categorical.py b/pandas/tests/groupby/test_categorical.py
index 8a3b306a9e40b..718ba14b57497 100644
--- a/pandas/tests/groupby/test_categorical.py
+++ b/pandas/tests/groupby/test_categorical.py
@@ -737,10 +737,9 @@ def test_preserve_on_ordered_ops(func, values):
     df = pd.DataFrame({"payload": [-1, -2, -1, -2], "col": c})
     g = df.groupby("payload")
     result = getattr(g, func)()
-    expected_col = pd.Categorical(values, dtype=c.dtype)
-    expected = pd.DataFrame({"payload": [-2, -1], "col": expected_col}).set_index(
-        "payload"
-    )
+    expected = pd.DataFrame(
+        {"payload": [-2, -1], "col": pd.Series(values, dtype=c.dtype)}
+    ).set_index("payload")
     tm.assert_frame_equal(result, expected)
 
 

From a6a498efc58432f25d7cfea0744e0231c659f7a9 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 09:32:46 -0600
Subject: [PATCH 22/28] non-numeric

---
 pandas/tests/groupby/test_categorical.py | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/pandas/tests/groupby/test_categorical.py b/pandas/tests/groupby/test_categorical.py
index 718ba14b57497..6beb43a39383b 100644
--- a/pandas/tests/groupby/test_categorical.py
+++ b/pandas/tests/groupby/test_categorical.py
@@ -1363,3 +1363,12 @@ def test_groupby_agg_categorical_columns(func, expected_values):
         {"value": expected_values}, index=pd.Index([0, 1, 2], name="groups"),
     )
     tm.assert_frame_equal(result, expected)
+
+
+def test_groupby_agg_non_numeric():
+    df = pd.DataFrame(
+        {"A": pd.Categorical(["a", "a", "b"], categories=["a", "b", "c"])}
+    )
+    result = df.groupby([1, 2, 1]).nunique()
+    expected = pd.DataFrame({"A": [2, 1]}, index=[1, 2])
+    tm.assert_frame_equal(result, expected)

From 2a3f5a20b5ae9affd282f806989974e92de7bd8e Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 09:43:42 -0600
Subject: [PATCH 23/28] xfailing test

---
 .../tests/groupby/aggregate/test_aggregate.py | 21 +++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/pandas/tests/groupby/aggregate/test_aggregate.py b/pandas/tests/groupby/aggregate/test_aggregate.py
index 67bdcc246579e..2d31996a8a964 100644
--- a/pandas/tests/groupby/aggregate/test_aggregate.py
+++ b/pandas/tests/groupby/aggregate/test_aggregate.py
@@ -663,6 +663,27 @@ def test_aggregate_mixed_types():
     tm.assert_frame_equal(result, expected)
 
 
+@pytest.mark.xfail(reason="Not implemented.")
+def test_aggregate_udf_na_extension_type():
+    # https://github.com/pandas-dev/pandas/pull/31359
+    # This is currently failing to cast back to Int64Dtype.
+    # The presence of the NA causes two problems
+    # 1. NA is not an instance of Int64Dtype.type (numpy.int64)
+    # 2. The presence of an NA forces object type, so the non-NA values is
+    #    a Python int rather than a NumPy int64. Python ints aren't
+    #    instances of numpy.int64.
+    def aggfunc(x):
+        if all(x > 2):
+            return 1
+        else:
+            return pd.NA
+
+    df = pd.DataFrame({"A": pd.array([1, 2, 3])})
+    result = df.groupby([1, 1, 2]).agg(aggfunc)
+    expected = pd.DataFrame({"A": pd.array([1, pd.NA], dtype="Int64")}, index=[1, 2])
+    tm.assert_frame_equal(result, expected)
+
+
 class TestLambdaMangling:
     def test_basic(self):
         df = pd.DataFrame({"A": [0, 0, 1, 1], "B": [1, 2, 3, 4]})

From ceef95e9303b2022f5f8733e08575ee485f4363c Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 11:24:40 -0600
Subject: [PATCH 24/28] release note

---
 doc/source/whatsnew/v1.0.0.rst | 45 ++++++++++++++++++++++++++++++++++
 1 file changed, 45 insertions(+)

diff --git a/doc/source/whatsnew/v1.0.0.rst b/doc/source/whatsnew/v1.0.0.rst
index 2abe85f042af1..ce474f0b149d4 100755
--- a/doc/source/whatsnew/v1.0.0.rst
+++ b/doc/source/whatsnew/v1.0.0.rst
@@ -626,6 +626,51 @@ consistent with the behaviour of :class:`DataFrame` and :class:`Index`.
    DeprecationWarning: The default dtype for empty Series will be 'object' instead of 'float64' in a future version. Specify a dtype explicitly to silence this warning.
    Series([], dtype: float64)
 
+Result dtype inference changes for resample operations
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The rules for the result dtype in :meth:`DataFrame.resample` aggregations have change for extension types (:issue:`31359`).
+Previously, pandas would attempt to convert the result back to the original dtype, falling back to the usual
+inference rules if that was not possible. Now, pandas will only return a result of the original dtype if the
+scalar values in the result are instances of the extension dtype's scalar type.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"A": ['a', 'b']}, dtype='category', index=pd.date_range('2000', periods=2))
+   df
+
+
+*pandas 0.25.x*
+
+.. code-block:: python
+
+   >>> df.resample("2D").agg(lambda x: 'a').A.dtype
+   CategoricalDtype(categories=['a', 'b'], ordered=False)
+
+*pandas 1.0.0*
+
+.. ipython:: python
+
+   df.resample("2D").agg(lambda x: 'a').A.dtype
+
+This fixes an inconsistency between ``resample`` and ``groupby``.
+This also fixes a potential bug, where the **values** of the result might change
+depending on how the results are cast back to the original dtype.
+
+*pandas 0.25.x*
+
+.. code-block:: python
+
+   df.resample("2D").agg(lambda x: 'c')
+
+        A
+   0  NaN
+
+.. ipython:: python
+
+   df.resample("2D").agg(lambda x: 'c')
+
+
 .. _whatsnew_100.api_breaking.python:
 
 Increased minimum version for Python

From ed91cc166ef7164598291a83b528542da80a208e Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 11:55:08 -0600
Subject: [PATCH 25/28] fixup

---
 doc/source/whatsnew/v1.0.0.rst | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/doc/source/whatsnew/v1.0.0.rst b/doc/source/whatsnew/v1.0.0.rst
index ce474f0b149d4..3e9920b422814 100755
--- a/doc/source/whatsnew/v1.0.0.rst
+++ b/doc/source/whatsnew/v1.0.0.rst
@@ -636,7 +636,8 @@ scalar values in the result are instances of the extension dtype's scalar type.
 
 .. ipython:: python
 
-   df = pd.DataFrame({"A": ['a', 'b']}, dtype='category', index=pd.date_range('2000', periods=2))
+   df = pd.DataFrame({"A": ['a', 'b']}, dtype='category',
+                     index=pd.date_range('2000', periods=2))
    df
 
 
@@ -661,7 +662,7 @@ depending on how the results are cast back to the original dtype.
 
 .. code-block:: python
 
-   df.resample("2D").agg(lambda x: 'c')
+   >>> df.resample("2D").agg(lambda x: 'c')
 
         A
    0  NaN

From 9c7af0fe6a85ea30c85ec2116a1c435323604c63 Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 13:05:05 -0600
Subject: [PATCH 26/28] fixup

---
 doc/source/whatsnew/v1.0.0.rst | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/doc/source/whatsnew/v1.0.0.rst b/doc/source/whatsnew/v1.0.0.rst
index 3e9920b422814..70ab4e62f51ba 100755
--- a/doc/source/whatsnew/v1.0.0.rst
+++ b/doc/source/whatsnew/v1.0.0.rst
@@ -667,6 +667,8 @@ depending on how the results are cast back to the original dtype.
         A
    0  NaN
 
+*pandas 1.0.0*
+
 .. ipython:: python
 
    df.resample("2D").agg(lambda x: 'c')

From ca3564867b62feb0f70c6b9ae406c0c182e91c4f Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 13:06:51 -0600
Subject: [PATCH 27/28] fixup

---
 pandas/tests/groupby/test_categorical.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/pandas/tests/groupby/test_categorical.py b/pandas/tests/groupby/test_categorical.py
index 6beb43a39383b..1c2de8c8c223f 100644
--- a/pandas/tests/groupby/test_categorical.py
+++ b/pandas/tests/groupby/test_categorical.py
@@ -1369,6 +1369,10 @@ def test_groupby_agg_non_numeric():
     df = pd.DataFrame(
         {"A": pd.Categorical(["a", "a", "b"], categories=["a", "b", "c"])}
     )
-    result = df.groupby([1, 2, 1]).nunique()
     expected = pd.DataFrame({"A": [2, 1]}, index=[1, 2])
+
+    result = df.groupby([1, 2, 1]).agg(pd.Series.nunique)
+    tm.assert_frame_equal(result, expected)
+
+    result = df.groupby([1, 2, 1]).nunique()
     tm.assert_frame_equal(result, expected)

From 1b826bb710f4fb6f480127042b3c4e26ebb00f7e Mon Sep 17 00:00:00 2001
From: Tom Augspurger <tom.w.augspurger@gmail.com>
Date: Wed, 29 Jan 2020 13:08:00 -0600
Subject: [PATCH 28/28] fixup

---
 doc/source/whatsnew/v1.0.0.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/source/whatsnew/v1.0.0.rst b/doc/source/whatsnew/v1.0.0.rst
index 70ab4e62f51ba..8025f7762f110 100755
--- a/doc/source/whatsnew/v1.0.0.rst
+++ b/doc/source/whatsnew/v1.0.0.rst
@@ -629,7 +629,7 @@ consistent with the behaviour of :class:`DataFrame` and :class:`Index`.
 Result dtype inference changes for resample operations
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-The rules for the result dtype in :meth:`DataFrame.resample` aggregations have change for extension types (:issue:`31359`).
+The rules for the result dtype in :meth:`DataFrame.resample` aggregations have changed for extension types (:issue:`31359`).
 Previously, pandas would attempt to convert the result back to the original dtype, falling back to the usual
 inference rules if that was not possible. Now, pandas will only return a result of the original dtype if the
 scalar values in the result are instances of the extension dtype's scalar type.