From 109c0e7e46f68d13182e95007b02b5a86ebb52e2 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Wed, 9 Dec 2020 08:58:22 +0800
Subject: [PATCH 01/32] fix series.isin slow issue with Dtype IntegerArray

---
 pandas/core/series.py | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/pandas/core/series.py b/pandas/core/series.py
index 0e9476285c258..dd7f2fdfacb64 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -70,7 +70,7 @@
 from pandas.core import algorithms, base, generic, missing, nanops, ops
 from pandas.core.accessor import CachedAccessor
 from pandas.core.aggregation import aggregate, transform
-from pandas.core.arrays import ExtensionArray
+from pandas.core.arrays import ExtensionArray, IntegerArray
 from pandas.core.arrays.categorical import CategoricalAccessor
 from pandas.core.arrays.sparse import SparseAccessor
 import pandas.core.common as com
@@ -4630,7 +4630,10 @@ def isin(self, values) -> "Series":
         5    False
         Name: animal, dtype: bool
         """
-        result = algorithms.isin(self._values, values)
+        if isinstance(self._values, IntegerArray):
+            result = algorithms.isin(self._values._data, values)
+        else:
+            result = algorithms.isin(self._values, values)
         return self._constructor(result, index=self.index).__finalize__(
             self, method="isin"
         )

From e9f96ea2252bf2f46b5605a9d657a3561a6a5b2c Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Wed, 9 Dec 2020 15:09:03 +0800
Subject: [PATCH 02/32] Move isinstance(comps, IntegerArray) to algo.isin

---
 pandas/core/algorithms.py | 3 +++
 pandas/core/series.py     | 7 ++-----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 67a0e02fc2d4d..b2ed35fa4cab3 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -15,6 +15,7 @@
 from pandas._typing import AnyArrayLike, ArrayLike, DtypeObj, FrameOrSeriesUnion
 from pandas.util._decorators import doc
 
+from pandas.core.arrays import IntegerArray
 from pandas.core.dtypes.cast import (
     construct_1d_object_array_from_listlike,
     infer_dtype_from_array,
@@ -447,6 +448,8 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
     else:
         values = extract_array(values, extract_numpy=True)
 
+    if isinstance(comps, IntegerArray):
+        comps = comps._data
     comps = _ensure_arraylike(comps)
     comps = extract_array(comps, extract_numpy=True)
     if is_categorical_dtype(comps.dtype):
diff --git a/pandas/core/series.py b/pandas/core/series.py
index dd7f2fdfacb64..0e9476285c258 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -70,7 +70,7 @@
 from pandas.core import algorithms, base, generic, missing, nanops, ops
 from pandas.core.accessor import CachedAccessor
 from pandas.core.aggregation import aggregate, transform
-from pandas.core.arrays import ExtensionArray, IntegerArray
+from pandas.core.arrays import ExtensionArray
 from pandas.core.arrays.categorical import CategoricalAccessor
 from pandas.core.arrays.sparse import SparseAccessor
 import pandas.core.common as com
@@ -4630,10 +4630,7 @@ def isin(self, values) -> "Series":
         5    False
         Name: animal, dtype: bool
         """
-        if isinstance(self._values, IntegerArray):
-            result = algorithms.isin(self._values._data, values)
-        else:
-            result = algorithms.isin(self._values, values)
+        result = algorithms.isin(self._values, values)
         return self._constructor(result, index=self.index).__finalize__(
             self, method="isin"
         )

From a6be9c842195bf22723c81ba9bd5539adc7bf1ac Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Wed, 9 Dec 2020 15:58:23 +0800
Subject: [PATCH 03/32] cannot import IntegerArray due to circular import

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

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index b2ed35fa4cab3..3e74d11097b9e 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -15,7 +15,6 @@
 from pandas._typing import AnyArrayLike, ArrayLike, DtypeObj, FrameOrSeriesUnion
 from pandas.util._decorators import doc
 
-from pandas.core.arrays import IntegerArray
 from pandas.core.dtypes.cast import (
     construct_1d_object_array_from_listlike,
     infer_dtype_from_array,
@@ -448,7 +447,7 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
     else:
         values = extract_array(values, extract_numpy=True)
 
-    if isinstance(comps, IntegerArray):
+    if comps.__class__.__name__ is 'IntegerArray':
         comps = comps._data
     comps = _ensure_arraylike(comps)
     comps = extract_array(comps, extract_numpy=True)

From 415b59008f90cec188adbbe93ba820db9872b4a9 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Wed, 9 Dec 2020 18:08:57 +0800
Subject: [PATCH 04/32] fix bug in pandas (Linux py38_np_dev)

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

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 3e74d11097b9e..dcef30a64f25b 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -447,7 +447,7 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
     else:
         values = extract_array(values, extract_numpy=True)
 
-    if comps.__class__.__name__ is 'IntegerArray':
+    if comps.__class__.__name__ == 'IntegerArray':
         comps = comps._data
     comps = _ensure_arraylike(comps)
     comps = extract_array(comps, extract_numpy=True)

From f3e5afb19a4cd03cad9caa50c8f599b20a1cc396 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Wed, 9 Dec 2020 18:18:44 +0800
Subject: [PATCH 05/32] fix pre commit issue.

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

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index dcef30a64f25b..09c63dd891db6 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -447,7 +447,7 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
     else:
         values = extract_array(values, extract_numpy=True)
 
-    if comps.__class__.__name__ == 'IntegerArray':
+    if comps.__class__.__name__ == "IntegerArray":
         comps = comps._data
     comps = _ensure_arraylike(comps)
     comps = extract_array(comps, extract_numpy=True)

From 14579fc83d257069c61c71ed41afb2f6e2d47f9f Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Wed, 9 Dec 2020 19:55:16 +0800
Subject: [PATCH 06/32] fix the code style issue.

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

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 09c63dd891db6..5e5c6153f5735 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -447,8 +447,8 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
     else:
         values = extract_array(values, extract_numpy=True)
 
-    if comps.__class__.__name__ == "IntegerArray":
-        comps = comps._data
+    if type(comps).__name__ == "IntegerArray":
+        comps = comps._data  # type: ignore[attr-defined, assignment]
     comps = _ensure_arraylike(comps)
     comps = extract_array(comps, extract_numpy=True)
     if is_categorical_dtype(comps.dtype):

From 562c91874cd7451a61f802b24d9140172f12c68c Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Fri, 11 Dec 2020 14:14:52 +0800
Subject: [PATCH 07/32] move the logic to elif block.

---
 pandas/core/algorithms.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 5e5c6153f5735..74511b57ce2b9 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -446,9 +446,7 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
         values = np.array(values)
     else:
         values = extract_array(values, extract_numpy=True)
-
-    if type(comps).__name__ == "IntegerArray":
-        comps = comps._data  # type: ignore[attr-defined, assignment]
+    
     comps = _ensure_arraylike(comps)
     comps = extract_array(comps, extract_numpy=True)
     if is_categorical_dtype(comps.dtype):
@@ -469,6 +467,8 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
     elif is_extension_array_dtype(comps.dtype) or is_extension_array_dtype(
         values.dtype
     ):
+        if type(comps).__name__ == "IntegerArray":
+            comps = comps._data  # type: ignore[attr-defined, assignment]
         return isin(np.asarray(comps), np.asarray(values))
 
     # GH16012

From 1449d3c06185d019894f130eb53a1f6e22bd6b7a Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Fri, 11 Dec 2020 14:17:22 +0800
Subject: [PATCH 08/32] remove blank line.

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

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 74511b57ce2b9..9e7362da0931a 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -446,7 +446,7 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
         values = np.array(values)
     else:
         values = extract_array(values, extract_numpy=True)
-    
+
     comps = _ensure_arraylike(comps)
     comps = extract_array(comps, extract_numpy=True)
     if is_categorical_dtype(comps.dtype):

From 3ccc917b22d47a10df6ab6dabd678d92f2eada6a Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sun, 27 Dec 2020 16:17:48 +0800
Subject: [PATCH 09/32] copy codes from #38422

---
 doc/source/reference/extensions.rst |  1 +
 pandas/core/algorithms.py           | 12 +++---------
 pandas/core/arrays/base.py          | 19 ++++++++++++++++++-
 pandas/core/arrays/masked.py        | 10 +++++++++-
 4 files changed, 31 insertions(+), 11 deletions(-)

diff --git a/doc/source/reference/extensions.rst b/doc/source/reference/extensions.rst
index fe4113d100abf..7b451ed3bf296 100644
--- a/doc/source/reference/extensions.rst
+++ b/doc/source/reference/extensions.rst
@@ -48,6 +48,7 @@ objects.
       api.extensions.ExtensionArray.equals
       api.extensions.ExtensionArray.factorize
       api.extensions.ExtensionArray.fillna
+      api.extensions.ExtensionArray.isin
       api.extensions.ExtensionArray.isna
       api.extensions.ExtensionArray.ravel
       api.extensions.ExtensionArray.repeat
diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 9e7362da0931a..fb299497df9fc 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -449,10 +449,8 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
 
     comps = _ensure_arraylike(comps)
     comps = extract_array(comps, extract_numpy=True)
-    if is_categorical_dtype(comps.dtype):
-        # TODO(extension)
-        # handle categoricals
-        return cast("Categorical", comps).isin(values)
+    if is_extension_array_dtype(comps.dtype):
+        return comps.isin(values)
 
     if needs_i8_conversion(comps.dtype):
         # Dispatch to DatetimeLikeArrayMixin.isin
@@ -464,11 +462,7 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
     elif needs_i8_conversion(values.dtype):
         return isin(comps, values.astype(object))
 
-    elif is_extension_array_dtype(comps.dtype) or is_extension_array_dtype(
-        values.dtype
-    ):
-        if type(comps).__name__ == "IntegerArray":
-            comps = comps._data  # type: ignore[attr-defined, assignment]
+    elif is_extension_array_dtype(values.dtype):
         return isin(np.asarray(comps), np.asarray(values))
 
     # GH16012
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 95470422f2ccd..089bb14772835 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -45,7 +45,7 @@
 from pandas.core.dtypes.missing import isna
 
 from pandas.core import ops
-from pandas.core.algorithms import factorize_array, unique
+from pandas.core.algorithms import factorize_array, isin, unique
 from pandas.core.missing import get_fill_func
 from pandas.core.sorting import nargminmax, nargsort
 
@@ -78,6 +78,7 @@ class ExtensionArray:
     factorize
     fillna
     equals
+    isin
     isna
     ravel
     repeat
@@ -833,6 +834,22 @@ def equals(self, other: object) -> bool:
             equal_na = self.isna() & other.isna()
             return bool((equal_values | equal_na).all())
 
+    def isin(self, values) -> np.ndarray:
+        """
+        Pointwise comparison for set containment in the given values.
+
+        Roughly equivalent to `np.array([x in values for x in self])`
+
+        Parameters
+        ----------
+        values : Sequence
+
+        Returns
+        -------
+        np.ndarray[bool]
+        """
+        return isin(np.asarray(self), values)
+
     def _values_for_factorize(self) -> Tuple[np.ndarray, Any]:
         """
         Return an array and missing value suitable for factorization.
diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index caed932cd7857..9b519152c7b75 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -19,7 +19,7 @@
 from pandas.core.dtypes.missing import isna, notna
 
 from pandas.core import nanops
-from pandas.core.algorithms import factorize_array, take
+from pandas.core.algorithms import factorize_array, isin, take
 from pandas.core.array_algos import masked_reductions
 from pandas.core.arraylike import OpsMixin
 from pandas.core.arrays import ExtensionArray
@@ -27,6 +27,7 @@
 
 if TYPE_CHECKING:
     from pandas import Series
+    from pandas.core.arrays import BooleanArray
 
 
 BaseMaskedArrayT = TypeVar("BaseMaskedArrayT", bound="BaseMaskedArray")
@@ -299,6 +300,13 @@ def take(
 
         return type(self)(result, mask, copy=False)
 
+    def isin(self, values) -> "BooleanArray":
+
+        from pandas.core.arrays import BooleanArray
+
+        result = isin(self._data, values)
+        return BooleanArray(result, self._mask.copy(), copy=False)
+
     def copy(self: BaseMaskedArrayT) -> BaseMaskedArrayT:
         data, mask = self._data, self._mask
         data = data.copy()

From 98a068304efed7abe802a64e0e65862e12c70f05 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sun, 27 Dec 2020 19:50:17 +0800
Subject: [PATCH 10/32] make `isin` correct for pd.NA

---
 pandas/core/arrays/masked.py | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 9b519152c7b75..06183becb6382 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -3,6 +3,7 @@
 from typing import TYPE_CHECKING, Any, Optional, Sequence, Tuple, Type, TypeVar, Union
 
 import numpy as np
+import pandas as pd
 
 from pandas._libs import lib, missing as libmissing
 from pandas._typing import Scalar
@@ -304,8 +305,11 @@ def isin(self, values) -> "BooleanArray":
 
         from pandas.core.arrays import BooleanArray
 
-        result = isin(self._data, values)
-        return BooleanArray(result, self._mask.copy(), copy=False)
+        result = isin(self._data, values) * np.invert(self._mask)
+        if any(x is pd.NA for x in values):
+            result += self._mask
+        mask = np.zeros_like(self, dtype=bool)
+        return BooleanArray(result, mask, copy=False)
 
     def copy(self: BaseMaskedArrayT) -> BaseMaskedArrayT:
         data, mask = self._data, self._mask

From 6e2917e4015767d52a02ad817876eab9c131ba7f Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sun, 27 Dec 2020 20:21:43 +0800
Subject: [PATCH 11/32] sort imports

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

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 06183becb6382..36df0af38915e 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -3,7 +3,6 @@
 from typing import TYPE_CHECKING, Any, Optional, Sequence, Tuple, Type, TypeVar, Union
 
 import numpy as np
-import pandas as pd
 
 from pandas._libs import lib, missing as libmissing
 from pandas._typing import Scalar
@@ -19,6 +18,7 @@
 )
 from pandas.core.dtypes.missing import isna, notna
 
+import pandas as pd
 from pandas.core import nanops
 from pandas.core.algorithms import factorize_array, isin, take
 from pandas.core.array_algos import masked_reductions

From a4b6503741ab6cc0d23b7e1af61f61794fd35adc Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Thu, 31 Dec 2020 20:37:06 +0800
Subject: [PATCH 12/32] Avoiding import pandas as pd.

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

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 36df0af38915e..389cdc5b2219f 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -18,7 +18,7 @@
 )
 from pandas.core.dtypes.missing import isna, notna
 
-import pandas as pd
+from pandas import NA
 from pandas.core import nanops
 from pandas.core.algorithms import factorize_array, isin, take
 from pandas.core.array_algos import masked_reductions
@@ -306,7 +306,7 @@ def isin(self, values) -> "BooleanArray":
         from pandas.core.arrays import BooleanArray
 
         result = isin(self._data, values) * np.invert(self._mask)
-        if any(x is pd.NA for x in values):
+        if any(x is NA for x in values):
             result += self._mask
         mask = np.zeros_like(self, dtype=bool)
         return BooleanArray(result, mask, copy=False)

From f95fde94da7e45c401afe2ceccfbe0a0a4869a71 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Thu, 31 Dec 2020 20:50:12 +0800
Subject: [PATCH 13/32] fix cannot import NA issue.

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

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 389cdc5b2219f..df1f5f253848e 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -18,9 +18,9 @@
 )
 from pandas.core.dtypes.missing import isna, notna
 
-from pandas import NA
 from pandas.core import nanops
 from pandas.core.algorithms import factorize_array, isin, take
+from pandas.core.api import NA
 from pandas.core.array_algos import masked_reductions
 from pandas.core.arraylike import OpsMixin
 from pandas.core.arrays import ExtensionArray

From 13dc64f5aa0be208dc83b19d5c3221f7a6b284da Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sat, 9 Jan 2021 15:03:30 +0800
Subject: [PATCH 14/32] Adding Int64 and Float64 for benchmarks.

---
 asv_bench/benchmarks/series_methods.py | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/asv_bench/benchmarks/series_methods.py b/asv_bench/benchmarks/series_methods.py
index 3f4da8acf4db0..585102d8a476f 100644
--- a/asv_bench/benchmarks/series_methods.py
+++ b/asv_bench/benchmarks/series_methods.py
@@ -25,7 +25,7 @@ def time_constructor(self, data):
 
 class IsIn:
 
-    params = ["int64", "uint64", "object"]
+    params = ["int64", "uint64", "object", "Int64"]
     param_names = ["dtype"]
 
     def setup(self, dtype):
@@ -59,8 +59,12 @@ def time_isin_empty(self):
 
 
 class IsInFloat64:
-    def setup(self):
-        self.small = Series([1, 2], dtype=np.float64)
+
+    params = [np.float64, "Float64"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        self.small = Series([1, 2], dtype=dtype)
         self.many_different_values = np.arange(10 ** 6, dtype=np.float64)
         self.few_different_values = np.zeros(10 ** 7, dtype=np.float64)
         self.only_nans_values = np.full(10 ** 7, np.nan, dtype=np.float64)
@@ -114,7 +118,7 @@ def time_isin_long_series_long_values_floats(self):
 
 class IsInLongSeriesLookUpDominates:
     params = [
-        ["int64", "int32", "float64", "float32", "object"],
+        ["int64", "int32", "float64", "float32", "object", "Int64", "Float64"],
         [5, 1000],
         ["random_hits", "random_misses", "monotone_hits", "monotone_misses"],
     ]
@@ -141,7 +145,7 @@ def time_isin(self, dtypes, MaxNumber, series_type):
 
 class IsInLongSeriesValuesDominate:
     params = [
-        ["int64", "int32", "float64", "float32", "object"],
+        ["int64", "int32", "float64", "float32", "object", "Int64", "Float64"],
         ["random", "monotone"],
     ]
     param_names = ["dtype", "series_type"]

From cc38088089644a5d4500239117581a974ffbeb2c Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sat, 9 Jan 2021 15:14:48 +0800
Subject: [PATCH 15/32] Adding isin benchmarks for Boolean array

---
 asv_bench/benchmarks/series_methods.py | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/asv_bench/benchmarks/series_methods.py b/asv_bench/benchmarks/series_methods.py
index 585102d8a476f..4d56042564192 100644
--- a/asv_bench/benchmarks/series_methods.py
+++ b/asv_bench/benchmarks/series_methods.py
@@ -36,6 +36,19 @@ def time_isin(self, dtypes):
         self.s.isin(self.values)
 
 
+class IsInBoolean:
+
+    params = ["boolean", "bool"]
+    param_names = ["dtype"]
+
+    def setup(self, dtype):
+        self.s = Series(np.random.randint(0, 2, 100000)).astype(dtype)
+        self.values = [True, False]
+
+    def time_isin(self, dtypes):
+        self.s.isin(self.values)
+
+
 class IsInDatetime64:
     def setup(self):
         dti = date_range(

From 94846cc16296480c17e8f5af0aed962623b91e3b Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sat, 9 Jan 2021 15:20:21 +0800
Subject: [PATCH 16/32] Adding what's new note.

---
 doc/source/whatsnew/v1.3.0.rst | 1 +
 1 file changed, 1 insertion(+)

diff --git a/doc/source/whatsnew/v1.3.0.rst b/doc/source/whatsnew/v1.3.0.rst
index 1d76c9d296255..7c02aa77f2feb 100644
--- a/doc/source/whatsnew/v1.3.0.rst
+++ b/doc/source/whatsnew/v1.3.0.rst
@@ -174,6 +174,7 @@ Performance improvements
 ~~~~~~~~~~~~~~~~~~~~~~~~
 - Performance improvement in :meth:`IntervalIndex.isin` (:issue:`38353`)
 - Performance improvement in :meth:`Series.mean` for nullable data types (:issue:`34814`)
+- Performance improvement in :meth:`Series.isin` for nullable data types (:issue:`38340`)
 -
 
 .. ---------------------------------------------------------------------------

From f4cb5ce2fb2f021fce0eae1b54fc2d550f051046 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sat, 9 Jan 2021 16:01:44 +0800
Subject: [PATCH 17/32] fix IsInFloat64 benchmarks

---
 asv_bench/benchmarks/series_methods.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/asv_bench/benchmarks/series_methods.py b/asv_bench/benchmarks/series_methods.py
index 4d56042564192..d1d9a0c15d396 100644
--- a/asv_bench/benchmarks/series_methods.py
+++ b/asv_bench/benchmarks/series_methods.py
@@ -82,15 +82,15 @@ def setup(self, dtype):
         self.few_different_values = np.zeros(10 ** 7, dtype=np.float64)
         self.only_nans_values = np.full(10 ** 7, np.nan, dtype=np.float64)
 
-    def time_isin_many_different(self):
+    def time_isin_many_different(self, dtypes):
         # runtime is dominated by creation of the lookup-table
         self.small.isin(self.many_different_values)
 
-    def time_isin_few_different(self):
+    def time_isin_few_different(self, dtypes):
         # runtime is dominated by creation of the lookup-table
         self.small.isin(self.few_different_values)
 
-    def time_isin_nan_values(self):
+    def time_isin_nan_values(self, dtypes):
         # runtime is dominated by creation of the lookup-table
         self.small.isin(self.few_different_values)
 

From 2ee8b05e024f00e3e73bb3b076f081ac727e46dd Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sat, 9 Jan 2021 16:46:46 +0800
Subject: [PATCH 18/32] always return false for null values.

---
 pandas/core/arrays/masked.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 6ed85a978af62..08218e001fbdb 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -335,8 +335,6 @@ def isin(self, values) -> "BooleanArray":
         from pandas.core.arrays import BooleanArray
 
         result = isin(self._data, values) * np.invert(self._mask)
-        if any(x is NA for x in values):
-            result += self._mask
         mask = np.zeros_like(self, dtype=bool)
         return BooleanArray(result, mask, copy=False)
 

From 7763b1828cfb18f7ef4170c81f713a45de16df7e Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sat, 9 Jan 2021 21:39:51 +0800
Subject: [PATCH 19/32] fix flake8 error.

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

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 08218e001fbdb..dee618933a459 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -22,7 +22,6 @@
 
 from pandas.core import nanops
 from pandas.core.algorithms import factorize_array, isin, take
-from pandas.core.api import NA
 from pandas.core.array_algos import masked_reductions
 from pandas.core.arraylike import OpsMixin
 from pandas.core.arrays import ExtensionArray

From 87dfff903651269f385434eaf732bbcd20e237f0 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sun, 10 Jan 2021 17:37:23 +0800
Subject: [PATCH 20/32] remove unused lines.

---
 pandas/core/algorithms.py | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 3e554e607174e..9ff49c3560d04 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -456,9 +456,6 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
     if is_extension_array_dtype(comps.dtype):
         return comps.isin(values)
 
-    elif is_interval_dtype(comps.dtype):
-        return cast("IntervalArray", comps).isin(values)
-
     elif needs_i8_conversion(comps.dtype):
         # Dispatch to DatetimeLikeArrayMixin.isin
         return array(comps).isin(values)

From d2d32d185eb56baae8c9f71668c591daecaef5d3 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sun, 10 Jan 2021 17:43:49 +0800
Subject: [PATCH 21/32] refactors for series benchmarks.

---
 asv_bench/benchmarks/series_methods.py | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/asv_bench/benchmarks/series_methods.py b/asv_bench/benchmarks/series_methods.py
index d1d9a0c15d396..e3ab8b243e25a 100644
--- a/asv_bench/benchmarks/series_methods.py
+++ b/asv_bench/benchmarks/series_methods.py
@@ -29,7 +29,8 @@ class IsIn:
     param_names = ["dtype"]
 
     def setup(self, dtype):
-        self.s = Series(np.random.randint(1, 10, 100000)).astype(dtype)
+        N = 10000
+        self.s = Series(np.random.randint(1, 10, N)).astype(dtype)
         self.values = [1, 2]
 
     def time_isin(self, dtypes):
@@ -42,7 +43,8 @@ class IsInBoolean:
     param_names = ["dtype"]
 
     def setup(self, dtype):
-        self.s = Series(np.random.randint(0, 2, 100000)).astype(dtype)
+        N = 10000
+        self.s = Series(np.random.randint(0, 2, N)).astype(dtype)
         self.values = [True, False]
 
     def time_isin(self, dtypes):
@@ -77,10 +79,12 @@ class IsInFloat64:
     param_names = ["dtype"]
 
     def setup(self, dtype):
+        N_many = 10 ** 5
+        N_few = 10 ** 6
         self.small = Series([1, 2], dtype=dtype)
-        self.many_different_values = np.arange(10 ** 6, dtype=np.float64)
-        self.few_different_values = np.zeros(10 ** 7, dtype=np.float64)
-        self.only_nans_values = np.full(10 ** 7, np.nan, dtype=np.float64)
+        self.many_different_values = np.arange(N_many, dtype=np.float64)
+        self.few_different_values = np.zeros(N_few, dtype=np.float64)
+        self.only_nans_values = np.full(N_few, np.nan, dtype=np.float64)
 
     def time_isin_many_different(self, dtypes):
         # runtime is dominated by creation of the lookup-table

From 90ef57a21c6ec72b8a4201e2dac74866d5665be2 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sun, 10 Jan 2021 18:35:26 +0800
Subject: [PATCH 22/32] fix flake8 errors.

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

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 9ff49c3560d04..834b43d55a83e 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -36,7 +36,6 @@
     is_float_dtype,
     is_integer,
     is_integer_dtype,
-    is_interval_dtype,
     is_list_like,
     is_numeric_dtype,
     is_object_dtype,
@@ -68,7 +67,7 @@
 
 if TYPE_CHECKING:
     from pandas import Categorical, DataFrame, Index, Series
-    from pandas.core.arrays import DatetimeArray, IntervalArray, TimedeltaArray
+    from pandas.core.arrays import DatetimeArray, TimedeltaArray
 
 _shared_docs: Dict[str, str] = {}
 

From a48e00b8df7a063e843cb3eaa480005bf60085c0 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sun, 10 Jan 2021 21:18:07 +0800
Subject: [PATCH 23/32] Change back to see if can pass the tests.

---
 pandas/core/algorithms.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 834b43d55a83e..3e554e607174e 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -36,6 +36,7 @@
     is_float_dtype,
     is_integer,
     is_integer_dtype,
+    is_interval_dtype,
     is_list_like,
     is_numeric_dtype,
     is_object_dtype,
@@ -67,7 +68,7 @@
 
 if TYPE_CHECKING:
     from pandas import Categorical, DataFrame, Index, Series
-    from pandas.core.arrays import DatetimeArray, TimedeltaArray
+    from pandas.core.arrays import DatetimeArray, IntervalArray, TimedeltaArray
 
 _shared_docs: Dict[str, str] = {}
 
@@ -455,6 +456,9 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
     if is_extension_array_dtype(comps.dtype):
         return comps.isin(values)
 
+    elif is_interval_dtype(comps.dtype):
+        return cast("IntervalArray", comps).isin(values)
+
     elif needs_i8_conversion(comps.dtype):
         # Dispatch to DatetimeLikeArrayMixin.isin
         return array(comps).isin(values)

From c726d4a9e323f6693c5993022e9d10768af2e0e8 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Thu, 14 Jan 2021 23:20:52 +0800
Subject: [PATCH 24/32] makes NA isin [NA] return True.

---
 pandas/core/arrays/masked.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index dee618933a459..3227288de0757 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -22,6 +22,7 @@
 
 from pandas.core import nanops
 from pandas.core.algorithms import factorize_array, isin, take
+from pandas.core.api import NA
 from pandas.core.array_algos import masked_reductions
 from pandas.core.arraylike import OpsMixin
 from pandas.core.arrays import ExtensionArray
@@ -334,6 +335,8 @@ def isin(self, values) -> "BooleanArray":
         from pandas.core.arrays import BooleanArray
 
         result = isin(self._data, values) * np.invert(self._mask)
+        if NA in values:
+            result += self._mask
         mask = np.zeros_like(self, dtype=bool)
         return BooleanArray(result, mask, copy=False)
 

From 1134ad6cc302f7d973199e67c23cac479b5a3b83 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Thu, 14 Jan 2021 23:26:10 +0800
Subject: [PATCH 25/32] remove redundant codes.

---
 pandas/core/algorithms.py | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 7e3fafee90292..1c0bc0fc52b6c 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -456,9 +456,6 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
     if is_extension_array_dtype(comps.dtype):
         return comps.isin(values)
 
-    elif is_interval_dtype(comps.dtype):
-        return cast("IntervalArray", comps).isin(values)
-
     elif needs_i8_conversion(comps.dtype):
         # Dispatch to DatetimeLikeArrayMixin.isin
         return array(comps).isin(values)

From bce0e3e4c723650ef9119132cbc4ad83a7786a65 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Thu, 14 Jan 2021 23:34:22 +0800
Subject: [PATCH 26/32] makes performance better.

---
 pandas/core/arrays/masked.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 3227288de0757..1154480b25c22 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -12,6 +12,7 @@
 from pandas.core.dtypes.base import ExtensionDtype
 from pandas.core.dtypes.common import (
     is_dtype_equal,
+    is_extension_array_dtype,
     is_integer,
     is_object_dtype,
     is_scalar,
@@ -335,7 +336,8 @@ def isin(self, values) -> "BooleanArray":
         from pandas.core.arrays import BooleanArray
 
         result = isin(self._data, values) * np.invert(self._mask)
-        if NA in values:
+        if (is_extension_array_dtype(values.dtype) or is_object_dtype(values.dtype))\
+                and NA in values:
             result += self._mask
         mask = np.zeros_like(self, dtype=bool)
         return BooleanArray(result, mask, copy=False)

From bf788e55b2167214b1914d1d83fbbaf50039bc06 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Thu, 14 Jan 2021 23:43:21 +0800
Subject: [PATCH 27/32] fix flake8 errors.

---
 pandas/core/algorithms.py | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 1c0bc0fc52b6c..be65adeb34139 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -36,7 +36,6 @@
     is_float_dtype,
     is_integer,
     is_integer_dtype,
-    is_interval_dtype,
     is_list_like,
     is_numeric_dtype,
     is_object_dtype,
@@ -68,7 +67,7 @@
 
 if TYPE_CHECKING:
     from pandas import Categorical, DataFrame, Index, Series
-    from pandas.core.arrays import DatetimeArray, IntervalArray, TimedeltaArray
+    from pandas.core.arrays import DatetimeArray, TimedeltaArray
 
 _shared_docs: Dict[str, str] = {}
 
@@ -477,7 +476,7 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
         # If the values include nan we need to check for nan explicitly
         # since np.nan it not equal to np.nan
         if isna(values).any():
-            f = lambda c, v: np.logical_or(np.in1d(c, v), np.isnan(c))
+            def f(c, v): return np.logical_or(np.in1d(c, v), np.isnan(c))
         else:
             f = np.in1d
 

From 40950bef647807a509ff8e7f9cddccd46f336b10 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sat, 16 Jan 2021 16:19:58 +0800
Subject: [PATCH 28/32] polish codes

---
 pandas/core/arrays/masked.py | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 1154480b25c22..0b0786b8d2ff7 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -12,7 +12,6 @@
 from pandas.core.dtypes.base import ExtensionDtype
 from pandas.core.dtypes.common import (
     is_dtype_equal,
-    is_extension_array_dtype,
     is_integer,
     is_object_dtype,
     is_scalar,
@@ -335,10 +334,12 @@ def isin(self, values) -> "BooleanArray":
 
         from pandas.core.arrays import BooleanArray
 
-        result = isin(self._data, values) * np.invert(self._mask)
-        if (is_extension_array_dtype(values.dtype) or is_object_dtype(values.dtype))\
-                and NA in values:
-            result += self._mask
+        result = isin(self._data, values)
+        if self._hasna:
+            if NA in values:
+                result += self._mask
+            else:
+                result *= np.invert(self._mask)
         mask = np.zeros_like(self, dtype=bool)
         return BooleanArray(result, mask, copy=False)
 

From 570d640c167ee52e98408516f3362d322d983cd8 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sat, 16 Jan 2021 16:21:11 +0800
Subject: [PATCH 29/32] not import NA

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

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 0b0786b8d2ff7..6b5d3e1230a23 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -22,7 +22,6 @@
 
 from pandas.core import nanops
 from pandas.core.algorithms import factorize_array, isin, take
-from pandas.core.api import NA
 from pandas.core.array_algos import masked_reductions
 from pandas.core.arraylike import OpsMixin
 from pandas.core.arrays import ExtensionArray
@@ -336,7 +335,7 @@ def isin(self, values) -> "BooleanArray":
 
         result = isin(self._data, values)
         if self._hasna:
-            if NA in values:
+            if libmissing.NA in values:
                 result += self._mask
             else:
                 result *= np.invert(self._mask)

From 0f89578f02a69b5a6c9d817f0ca7c1f254c1343c Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sat, 16 Jan 2021 16:21:37 +0800
Subject: [PATCH 30/32] fix code style

---
 pandas/core/algorithms.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index be65adeb34139..8105500071a70 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -476,7 +476,8 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
         # If the values include nan we need to check for nan explicitly
         # since np.nan it not equal to np.nan
         if isna(values).any():
-            def f(c, v): return np.logical_or(np.in1d(c, v), np.isnan(c))
+            def f(c, v):
+                return np.logical_or(np.in1d(c, v), np.isnan(c))
         else:
             f = np.in1d
 

From 199c11c478f36292e7bd84afa7964e9dc5d03671 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sat, 16 Jan 2021 22:08:28 +0800
Subject: [PATCH 31/32] fix black error.

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

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index 8105500071a70..9f938a95d070e 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -476,8 +476,10 @@ def isin(comps: AnyArrayLike, values: AnyArrayLike) -> np.ndarray:
         # If the values include nan we need to check for nan explicitly
         # since np.nan it not equal to np.nan
         if isna(values).any():
+
             def f(c, v):
                 return np.logical_or(np.in1d(c, v), np.isnan(c))
+
         else:
             f = np.in1d
 

From 9f35b5b674c2b656b99973c0901e235bfabbdb81 Mon Sep 17 00:00:00 2001
From: tushushu <m18591915960@163.com>
Date: Sun, 17 Jan 2021 11:58:23 +0800
Subject: [PATCH 32/32] fix CI

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

diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 6b5d3e1230a23..d062d114abb6b 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -329,7 +329,7 @@ def take(
 
         return type(self)(result, mask, copy=False)
 
-    def isin(self, values) -> "BooleanArray":
+    def isin(self, values) -> BooleanArray:
 
         from pandas.core.arrays import BooleanArray