From ea19b7c8be62b7e1d1fc2e35efa6221eb0d9e2dc Mon Sep 17 00:00:00 2001
From: Marco Gorelli <marcogorelli@protonmail.com>
Date: Tue, 9 Mar 2021 11:26:53 +0000
Subject: [PATCH 1/5] use pd_array in core

---
 .pre-commit-config.yaml                    |  7 ++
 pandas/core/strings/accessor.py            |  4 +-
 scripts/tests/test_use_pd_array_in_core.py | 26 ++++++++
 scripts/use_pd_array_in_core.py            | 78 ++++++++++++++++++++++
 4 files changed, 113 insertions(+), 2 deletions(-)
 create mode 100644 scripts/tests/test_use_pd_array_in_core.py
 create mode 100644 scripts/use_pd_array_in_core.py

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 99ba2feadbca7..9829d741ee8ed 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -227,3 +227,10 @@ repos:
             |\#\ type:\s?ignore(?!\[)
         language: pygrep
         types: [python]
+    -   id: use-pd_array-in-core
+        name: Import pandas.array as pd_array in core
+        language: python
+        entry: python scripts/use_pd_array_in_core.py
+        files: ^pandas/core/
+        exclude: ^pandas/core/api\.py$
+        types: [python]
diff --git a/pandas/core/strings/accessor.py b/pandas/core/strings/accessor.py
index 73a5ef9345fec..1c0538405dd71 100644
--- a/pandas/core/strings/accessor.py
+++ b/pandas/core/strings/accessor.py
@@ -3023,7 +3023,7 @@ def _str_extract_noexpand(arr, pat, flags=0):
     """
     from pandas import (
         DataFrame,
-        array,
+        array as pd_array,
     )
 
     regex = re.compile(pat, flags=flags)
@@ -3034,7 +3034,7 @@ def _str_extract_noexpand(arr, pat, flags=0):
         result = np.array([groups_or_na(val)[0] for val in arr], dtype=object)
         name = _get_single_group_name(regex)
         # not dispatching, so we have to reconstruct here.
-        result = array(result, dtype=result_dtype)
+        result = pd_array(result, dtype=result_dtype)
     else:
         if isinstance(arr, ABCIndex):
             raise ValueError("only one regex group is supported with Index")
diff --git a/scripts/tests/test_use_pd_array_in_core.py b/scripts/tests/test_use_pd_array_in_core.py
new file mode 100644
index 0000000000000..5c56634335798
--- /dev/null
+++ b/scripts/tests/test_use_pd_array_in_core.py
@@ -0,0 +1,26 @@
+import pytest
+
+from scripts.use_pd_array_in_core import use_pd_array
+
+BAD_FILE_0 = "import pandas as pd\npd.array"
+BAD_FILE_1 = "\nfrom pandas import array"
+GOOD_FILE_0 = "from pandas import array as pd_array"
+GOOD_FILE_1 = "from pandas.core.construction import array as pd_array"
+PATH = "t.py"
+
+
+@pytest.mark.parametrize("content", [BAD_FILE_0, BAD_FILE_1])
+def test_inconsistent_usage(content, capsys):
+    result_msg = (
+        r"t.py:2:0: Don't use pd.array in core, import array as pd_array instead\n"
+    )
+    with pytest.raises(SystemExit):
+        use_pd_array(content, PATH)
+    expected_msg, _ = capsys.readouterr()
+    assert result_msg == expected_msg
+
+
+@pytest.mark.parametrize("content", [GOOD_FILE_0, GOOD_FILE_1])
+def test_consistent_usage(content):
+    # should not raise
+    use_pd_array(content, PATH)
diff --git a/scripts/use_pd_array_in_core.py b/scripts/use_pd_array_in_core.py
new file mode 100644
index 0000000000000..8c03c602a3e55
--- /dev/null
+++ b/scripts/use_pd_array_in_core.py
@@ -0,0 +1,78 @@
+"""
+Check that pandas/core imports pandas.array as pd_array.
+
+This makes it easier to grep for usage of pandas array.
+
+This is meant to be run as a pre-commit hook - to run it manually, you can do:
+
+    pre-commit run use-pd_array-in-core --all-files
+
+"""
+
+import argparse
+import ast
+import sys
+from typing import (
+    Optional,
+    Sequence,
+)
+
+ERROR_MESSAGE = (
+    "{path}:{lineno}:{col_offset}: "
+    "Don't use pd.array in core, import array as pd_array instead\n"
+)
+
+
+class Visitor(ast.NodeVisitor):
+    def __init__(self, path: str) -> None:
+        self.path = path
+
+    def visit_ImportFrom(self, node: ast.ImportFrom) -> None:
+        # If array has been imported from somewhere in pandas,
+        # check it's aliased as pd_array.
+        if (
+            node.module is not None
+            and node.module.startswith("pandas")
+            and any(i.name == "array" and i.asname != "pd_array" for i in node.names)
+        ):
+            msg = ERROR_MESSAGE.format(
+                path=self.path, lineno=node.lineno, col_offset=node.col_offset
+            )
+            sys.stdout.write(msg)
+            sys.exit(1)
+        super().generic_visit(node)
+
+    def visit_Attribute(self, node: ast.Attribute) -> None:
+        if (
+            isinstance(node.value, ast.Name)
+            and node.value.id == "pd"
+            and node.attr == "array"
+        ):
+            msg = ERROR_MESSAGE.format(
+                path=self.path, lineno=node.lineno, col_offset=node.col_offset
+            )
+            sys.stdout.write(msg)
+            sys.exit(1)
+        super().generic_visit(node)
+
+
+def use_pd_array(content: str, path: str) -> None:
+    tree = ast.parse(content)
+    visitor = Visitor(path)
+    visitor.visit(tree)
+    return
+
+
+def main(argv: Optional[Sequence[str]] = None) -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("paths", nargs="*")
+    args = parser.parse_args(argv)
+
+    for path in args.paths:
+        with open(path, encoding="utf-8") as fd:
+            content = fd.read()
+        use_pd_array(content, path)
+
+
+if __name__ == "__main__":
+    main()

From 5fa51d847231f789d5445710f33a6b0dfa0d13fe Mon Sep 17 00:00:00 2001
From: Marco Gorelli <marcogorelli@protonmail.com>
Date: Tue, 9 Mar 2021 17:29:01 +0000
Subject: [PATCH 2/5] define pd_array in construction

---
 pandas/core/algorithms.py                  |  2 +-
 pandas/core/api.py                         |  2 +-
 pandas/core/apply.py                       |  2 +-
 pandas/core/arrays/categorical.py          |  2 +-
 pandas/core/arrays/datetimelike.py         |  2 +-
 pandas/core/arrays/interval.py             |  2 +-
 pandas/core/construction.py                |  4 ++--
 pandas/core/dtypes/concat.py               |  2 +-
 pandas/core/indexers.py                    |  2 +-
 pandas/core/indexing.py                    |  2 +-
 pandas/core/strings/accessor.py            |  6 ++----
 scripts/tests/test_use_pd_array_in_core.py | 15 +++++++--------
 scripts/use_pd_array_in_core.py            | 10 ++++++----
 13 files changed, 26 insertions(+), 27 deletions(-)

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index c3705fada724a..dd399184540e9 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -84,9 +84,9 @@
 
 from pandas.core.array_algos.take import take_nd
 from pandas.core.construction import (
-    array as pd_array,
     ensure_wrapped_if_datetimelike,
     extract_array,
+    pd_array,
 )
 from pandas.core.indexers import validate_indices
 
diff --git a/pandas/core/api.py b/pandas/core/api.py
index 2677530455b07..07f431d3b2c77 100644
--- a/pandas/core/api.py
+++ b/pandas/core/api.py
@@ -43,7 +43,7 @@
     UInt64Dtype,
 )
 from pandas.core.arrays.string_ import StringDtype
-from pandas.core.construction import array
+from pandas.core.construction import pd_array as array
 from pandas.core.flags import Flags
 from pandas.core.groupby import (
     Grouper,
diff --git a/pandas/core/apply.py b/pandas/core/apply.py
index 57147461284fb..384b5c389b962 100644
--- a/pandas/core/apply.py
+++ b/pandas/core/apply.py
@@ -52,8 +52,8 @@
 )
 import pandas.core.common as com
 from pandas.core.construction import (
-    array as pd_array,
     create_series_with_explicit_dtype,
+    pd_array,
 )
 
 if TYPE_CHECKING:
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 0bf5e05786d4d..3ff3fda6ec768 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -97,8 +97,8 @@
 )
 import pandas.core.common as com
 from pandas.core.construction import (
-    array as pd_array,
     extract_array,
+    pd_array,
     sanitize_array,
 )
 from pandas.core.indexers import deprecate_ndim_indexing
diff --git a/pandas/core/arrays/datetimelike.py b/pandas/core/arrays/datetimelike.py
index bd5cc04659a06..36744bb5587d4 100644
--- a/pandas/core/arrays/datetimelike.py
+++ b/pandas/core/arrays/datetimelike.py
@@ -107,8 +107,8 @@
 )
 import pandas.core.common as com
 from pandas.core.construction import (
-    array as pd_array,
     extract_array,
+    pd_array,
 )
 from pandas.core.indexers import (
     check_array_indexer,
diff --git a/pandas/core/arrays/interval.py b/pandas/core/arrays/interval.py
index 1a626f327b97c..9ee0151792762 100644
--- a/pandas/core/arrays/interval.py
+++ b/pandas/core/arrays/interval.py
@@ -76,9 +76,9 @@
 from pandas.core.arrays.categorical import Categorical
 import pandas.core.common as com
 from pandas.core.construction import (
-    array as pd_array,
     ensure_wrapped_if_datetimelike,
     extract_array,
+    pd_array,
 )
 from pandas.core.indexers import check_array_indexer
 from pandas.core.indexes.base import ensure_index
diff --git a/pandas/core/construction.py b/pandas/core/construction.py
index 78a7f1890b5de..a2f04ebc0e586 100644
--- a/pandas/core/construction.py
+++ b/pandas/core/construction.py
@@ -74,7 +74,7 @@
     )
 
 
-def array(
+def pd_array(
     data: Union[Sequence[object], AnyArrayLike],
     dtype: Optional[Dtype] = None,
     copy: bool = True,
@@ -560,7 +560,7 @@ def sanitize_array(
         if is_object_dtype(subarr.dtype) and not is_object_or_str_dtype:
             inferred = lib.infer_dtype(subarr, skipna=False)
             if inferred in {"interval", "period"}:
-                subarr = array(subarr)
+                subarr = pd_array(subarr)
                 subarr = extract_array(subarr, extract_numpy=True)
 
     return subarr
diff --git a/pandas/core/dtypes/concat.py b/pandas/core/dtypes/concat.py
index 614a637f2d904..1d0e04cdc5a50 100644
--- a/pandas/core/dtypes/concat.py
+++ b/pandas/core/dtypes/concat.py
@@ -25,8 +25,8 @@
 from pandas.core.arrays import ExtensionArray
 from pandas.core.arrays.sparse import SparseArray
 from pandas.core.construction import (
-    array as pd_array,
     ensure_wrapped_if_datetimelike,
+    pd_array,
 )
 
 
diff --git a/pandas/core/indexers.py b/pandas/core/indexers.py
index 86d6b772fe2e4..34b2301d081cf 100644
--- a/pandas/core/indexers.py
+++ b/pandas/core/indexers.py
@@ -514,7 +514,7 @@ def check_array_indexer(array: AnyArrayLike, indexer: Any) -> Any:
     ...
     IndexError: arrays used as indices must be of integer or boolean type
     """
-    from pandas.core.construction import array as pd_array
+    from pandas.core.construction import pd_array
 
     # whatever is not an array-like is returned as-is (possible valid array
     # indexers that are not array-like: integer, slice, Ellipsis, None)
diff --git a/pandas/core/indexing.py b/pandas/core/indexing.py
index ae0f853db628e..c80496aee631d 100644
--- a/pandas/core/indexing.py
+++ b/pandas/core/indexing.py
@@ -48,7 +48,7 @@
 )
 
 import pandas.core.common as com
-from pandas.core.construction import array as pd_array
+from pandas.core.construction import pd_array
 from pandas.core.indexers import (
     check_array_indexer,
     is_exact_shape_match,
diff --git a/pandas/core/strings/accessor.py b/pandas/core/strings/accessor.py
index 1c0538405dd71..4021db7892db2 100644
--- a/pandas/core/strings/accessor.py
+++ b/pandas/core/strings/accessor.py
@@ -3021,10 +3021,8 @@ def _str_extract_noexpand(arr, pat, flags=0):
     Index.
 
     """
-    from pandas import (
-        DataFrame,
-        array as pd_array,
-    )
+    from pandas import DataFrame
+    from pandas.core.construction import pd_array
 
     regex = re.compile(pat, flags=flags)
     groups_or_na = _groups_or_na_fun(regex)
diff --git a/scripts/tests/test_use_pd_array_in_core.py b/scripts/tests/test_use_pd_array_in_core.py
index 5c56634335798..2f2c7012ae324 100644
--- a/scripts/tests/test_use_pd_array_in_core.py
+++ b/scripts/tests/test_use_pd_array_in_core.py
@@ -2,17 +2,17 @@
 
 from scripts.use_pd_array_in_core import use_pd_array
 
-BAD_FILE_0 = "import pandas as pd\npd.array"
-BAD_FILE_1 = "\nfrom pandas import array"
-GOOD_FILE_0 = "from pandas import array as pd_array"
-GOOD_FILE_1 = "from pandas.core.construction import array as pd_array"
+BAD_FILE_0 = "\nfrom pandas import array as pd_array"
+BAD_FILE_1 = "import pandas as pd\npd.array"
+GOOD_FILE = "from pandas.core.construction import pd_array"
 PATH = "t.py"
 
 
 @pytest.mark.parametrize("content", [BAD_FILE_0, BAD_FILE_1])
 def test_inconsistent_usage(content, capsys):
     result_msg = (
-        r"t.py:2:0: Don't use pd.array in core, import array as pd_array instead\n"
+        "t.py:2:0: Don't use pd.array in core, "
+        "instead use 'from pandas.core.construction import pd_array'\n"
     )
     with pytest.raises(SystemExit):
         use_pd_array(content, PATH)
@@ -20,7 +20,6 @@ def test_inconsistent_usage(content, capsys):
     assert result_msg == expected_msg
 
 
-@pytest.mark.parametrize("content", [GOOD_FILE_0, GOOD_FILE_1])
-def test_consistent_usage(content):
+def test_consistent_usage():
     # should not raise
-    use_pd_array(content, PATH)
+    use_pd_array(GOOD_FILE, PATH)
diff --git a/scripts/use_pd_array_in_core.py b/scripts/use_pd_array_in_core.py
index 8c03c602a3e55..4a2c4ab83fefc 100644
--- a/scripts/use_pd_array_in_core.py
+++ b/scripts/use_pd_array_in_core.py
@@ -1,7 +1,8 @@
 """
-Check that pandas/core imports pandas.array as pd_array.
+Check files use pd_array from pandas.core.construction rather than pd.array.
 
-This makes it easier to grep for usage of pandas array.
+This is only run within pandas/core and makes it easier to grep for usage of
+pandas array.
 
 This is meant to be run as a pre-commit hook - to run it manually, you can do:
 
@@ -19,7 +20,8 @@
 
 ERROR_MESSAGE = (
     "{path}:{lineno}:{col_offset}: "
-    "Don't use pd.array in core, import array as pd_array instead\n"
+    "Don't use pd.array in core, instead use "
+    "'from pandas.core.construction import pd_array'\n"
 )
 
 
@@ -33,7 +35,7 @@ def visit_ImportFrom(self, node: ast.ImportFrom) -> None:
         if (
             node.module is not None
             and node.module.startswith("pandas")
-            and any(i.name == "array" and i.asname != "pd_array" for i in node.names)
+            and any(i.name == "array" for i in node.names)
         ):
             msg = ERROR_MESSAGE.format(
                 path=self.path, lineno=node.lineno, col_offset=node.col_offset

From bec9a10918268b062626306bbb9012bcd49a3115 Mon Sep 17 00:00:00 2001
From: Marco Gorelli <marcogorelli@protonmail.com>
Date: Fri, 12 Mar 2021 10:36:02 +0000
Subject: [PATCH 3/5] import array as pd_array

---
 pandas/core/algorithms.py                  |  2 +-
 pandas/core/api.py                         |  2 +-
 pandas/core/apply.py                       |  2 +-
 pandas/core/arrays/categorical.py          |  2 +-
 pandas/core/arrays/datetimelike.py         |  2 +-
 pandas/core/arrays/interval.py             |  2 +-
 pandas/core/construction.py                |  4 ++--
 pandas/core/dtypes/concat.py               |  2 +-
 pandas/core/indexers.py                    |  2 +-
 pandas/core/indexing.py                    |  2 +-
 pandas/core/strings/accessor.py            |  6 ++++--
 scripts/tests/test_use_pd_array_in_core.py | 15 ++++++++-------
 scripts/use_pd_array_in_core.py            | 10 ++++------
 13 files changed, 27 insertions(+), 26 deletions(-)

diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index dd399184540e9..c3705fada724a 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -84,9 +84,9 @@
 
 from pandas.core.array_algos.take import take_nd
 from pandas.core.construction import (
+    array as pd_array,
     ensure_wrapped_if_datetimelike,
     extract_array,
-    pd_array,
 )
 from pandas.core.indexers import validate_indices
 
diff --git a/pandas/core/api.py b/pandas/core/api.py
index 07f431d3b2c77..2677530455b07 100644
--- a/pandas/core/api.py
+++ b/pandas/core/api.py
@@ -43,7 +43,7 @@
     UInt64Dtype,
 )
 from pandas.core.arrays.string_ import StringDtype
-from pandas.core.construction import pd_array as array
+from pandas.core.construction import array
 from pandas.core.flags import Flags
 from pandas.core.groupby import (
     Grouper,
diff --git a/pandas/core/apply.py b/pandas/core/apply.py
index 384b5c389b962..57147461284fb 100644
--- a/pandas/core/apply.py
+++ b/pandas/core/apply.py
@@ -52,8 +52,8 @@
 )
 import pandas.core.common as com
 from pandas.core.construction import (
+    array as pd_array,
     create_series_with_explicit_dtype,
-    pd_array,
 )
 
 if TYPE_CHECKING:
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 3ff3fda6ec768..0bf5e05786d4d 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -97,8 +97,8 @@
 )
 import pandas.core.common as com
 from pandas.core.construction import (
+    array as pd_array,
     extract_array,
-    pd_array,
     sanitize_array,
 )
 from pandas.core.indexers import deprecate_ndim_indexing
diff --git a/pandas/core/arrays/datetimelike.py b/pandas/core/arrays/datetimelike.py
index 36744bb5587d4..bd5cc04659a06 100644
--- a/pandas/core/arrays/datetimelike.py
+++ b/pandas/core/arrays/datetimelike.py
@@ -107,8 +107,8 @@
 )
 import pandas.core.common as com
 from pandas.core.construction import (
+    array as pd_array,
     extract_array,
-    pd_array,
 )
 from pandas.core.indexers import (
     check_array_indexer,
diff --git a/pandas/core/arrays/interval.py b/pandas/core/arrays/interval.py
index 9ee0151792762..1a626f327b97c 100644
--- a/pandas/core/arrays/interval.py
+++ b/pandas/core/arrays/interval.py
@@ -76,9 +76,9 @@
 from pandas.core.arrays.categorical import Categorical
 import pandas.core.common as com
 from pandas.core.construction import (
+    array as pd_array,
     ensure_wrapped_if_datetimelike,
     extract_array,
-    pd_array,
 )
 from pandas.core.indexers import check_array_indexer
 from pandas.core.indexes.base import ensure_index
diff --git a/pandas/core/construction.py b/pandas/core/construction.py
index a2f04ebc0e586..78a7f1890b5de 100644
--- a/pandas/core/construction.py
+++ b/pandas/core/construction.py
@@ -74,7 +74,7 @@
     )
 
 
-def pd_array(
+def array(
     data: Union[Sequence[object], AnyArrayLike],
     dtype: Optional[Dtype] = None,
     copy: bool = True,
@@ -560,7 +560,7 @@ def sanitize_array(
         if is_object_dtype(subarr.dtype) and not is_object_or_str_dtype:
             inferred = lib.infer_dtype(subarr, skipna=False)
             if inferred in {"interval", "period"}:
-                subarr = pd_array(subarr)
+                subarr = array(subarr)
                 subarr = extract_array(subarr, extract_numpy=True)
 
     return subarr
diff --git a/pandas/core/dtypes/concat.py b/pandas/core/dtypes/concat.py
index 1d0e04cdc5a50..614a637f2d904 100644
--- a/pandas/core/dtypes/concat.py
+++ b/pandas/core/dtypes/concat.py
@@ -25,8 +25,8 @@
 from pandas.core.arrays import ExtensionArray
 from pandas.core.arrays.sparse import SparseArray
 from pandas.core.construction import (
+    array as pd_array,
     ensure_wrapped_if_datetimelike,
-    pd_array,
 )
 
 
diff --git a/pandas/core/indexers.py b/pandas/core/indexers.py
index 34b2301d081cf..86d6b772fe2e4 100644
--- a/pandas/core/indexers.py
+++ b/pandas/core/indexers.py
@@ -514,7 +514,7 @@ def check_array_indexer(array: AnyArrayLike, indexer: Any) -> Any:
     ...
     IndexError: arrays used as indices must be of integer or boolean type
     """
-    from pandas.core.construction import pd_array
+    from pandas.core.construction import array as pd_array
 
     # whatever is not an array-like is returned as-is (possible valid array
     # indexers that are not array-like: integer, slice, Ellipsis, None)
diff --git a/pandas/core/indexing.py b/pandas/core/indexing.py
index c80496aee631d..ae0f853db628e 100644
--- a/pandas/core/indexing.py
+++ b/pandas/core/indexing.py
@@ -48,7 +48,7 @@
 )
 
 import pandas.core.common as com
-from pandas.core.construction import pd_array
+from pandas.core.construction import array as pd_array
 from pandas.core.indexers import (
     check_array_indexer,
     is_exact_shape_match,
diff --git a/pandas/core/strings/accessor.py b/pandas/core/strings/accessor.py
index 4021db7892db2..1c0538405dd71 100644
--- a/pandas/core/strings/accessor.py
+++ b/pandas/core/strings/accessor.py
@@ -3021,8 +3021,10 @@ def _str_extract_noexpand(arr, pat, flags=0):
     Index.
 
     """
-    from pandas import DataFrame
-    from pandas.core.construction import pd_array
+    from pandas import (
+        DataFrame,
+        array as pd_array,
+    )
 
     regex = re.compile(pat, flags=flags)
     groups_or_na = _groups_or_na_fun(regex)
diff --git a/scripts/tests/test_use_pd_array_in_core.py b/scripts/tests/test_use_pd_array_in_core.py
index 2f2c7012ae324..5c56634335798 100644
--- a/scripts/tests/test_use_pd_array_in_core.py
+++ b/scripts/tests/test_use_pd_array_in_core.py
@@ -2,17 +2,17 @@
 
 from scripts.use_pd_array_in_core import use_pd_array
 
-BAD_FILE_0 = "\nfrom pandas import array as pd_array"
-BAD_FILE_1 = "import pandas as pd\npd.array"
-GOOD_FILE = "from pandas.core.construction import pd_array"
+BAD_FILE_0 = "import pandas as pd\npd.array"
+BAD_FILE_1 = "\nfrom pandas import array"
+GOOD_FILE_0 = "from pandas import array as pd_array"
+GOOD_FILE_1 = "from pandas.core.construction import array as pd_array"
 PATH = "t.py"
 
 
 @pytest.mark.parametrize("content", [BAD_FILE_0, BAD_FILE_1])
 def test_inconsistent_usage(content, capsys):
     result_msg = (
-        "t.py:2:0: Don't use pd.array in core, "
-        "instead use 'from pandas.core.construction import pd_array'\n"
+        r"t.py:2:0: Don't use pd.array in core, import array as pd_array instead\n"
     )
     with pytest.raises(SystemExit):
         use_pd_array(content, PATH)
@@ -20,6 +20,7 @@ def test_inconsistent_usage(content, capsys):
     assert result_msg == expected_msg
 
 
-def test_consistent_usage():
+@pytest.mark.parametrize("content", [GOOD_FILE_0, GOOD_FILE_1])
+def test_consistent_usage(content):
     # should not raise
-    use_pd_array(GOOD_FILE, PATH)
+    use_pd_array(content, PATH)
diff --git a/scripts/use_pd_array_in_core.py b/scripts/use_pd_array_in_core.py
index 4a2c4ab83fefc..8c03c602a3e55 100644
--- a/scripts/use_pd_array_in_core.py
+++ b/scripts/use_pd_array_in_core.py
@@ -1,8 +1,7 @@
 """
-Check files use pd_array from pandas.core.construction rather than pd.array.
+Check that pandas/core imports pandas.array as pd_array.
 
-This is only run within pandas/core and makes it easier to grep for usage of
-pandas array.
+This makes it easier to grep for usage of pandas array.
 
 This is meant to be run as a pre-commit hook - to run it manually, you can do:
 
@@ -20,8 +19,7 @@
 
 ERROR_MESSAGE = (
     "{path}:{lineno}:{col_offset}: "
-    "Don't use pd.array in core, instead use "
-    "'from pandas.core.construction import pd_array'\n"
+    "Don't use pd.array in core, import array as pd_array instead\n"
 )
 
 
@@ -35,7 +33,7 @@ def visit_ImportFrom(self, node: ast.ImportFrom) -> None:
         if (
             node.module is not None
             and node.module.startswith("pandas")
-            and any(i.name == "array" for i in node.names)
+            and any(i.name == "array" and i.asname != "pd_array" for i in node.names)
         ):
             msg = ERROR_MESSAGE.format(
                 path=self.path, lineno=node.lineno, col_offset=node.col_offset

From 96e9851925b0482314c5847d65820e2cf58be499 Mon Sep 17 00:00:00 2001
From: Marco Gorelli <marcogorelli@protonmail.com>
Date: Fri, 12 Mar 2021 15:00:32 +0000
Subject: [PATCH 4/5] fixup test

---
 scripts/tests/test_use_pd_array_in_core.py | 2 +-
 scripts/use_pd_array_in_core.py            | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/scripts/tests/test_use_pd_array_in_core.py b/scripts/tests/test_use_pd_array_in_core.py
index 5c56634335798..9c66199a82846 100644
--- a/scripts/tests/test_use_pd_array_in_core.py
+++ b/scripts/tests/test_use_pd_array_in_core.py
@@ -12,7 +12,7 @@
 @pytest.mark.parametrize("content", [BAD_FILE_0, BAD_FILE_1])
 def test_inconsistent_usage(content, capsys):
     result_msg = (
-        r"t.py:2:0: Don't use pd.array in core, import array as pd_array instead\n"
+        "t.py:2:0: Don't use pd.array in core, import array as pd_array instead\n"
     )
     with pytest.raises(SystemExit):
         use_pd_array(content, PATH)
diff --git a/scripts/use_pd_array_in_core.py b/scripts/use_pd_array_in_core.py
index 8c03c602a3e55..531084683bdb1 100644
--- a/scripts/use_pd_array_in_core.py
+++ b/scripts/use_pd_array_in_core.py
@@ -60,7 +60,6 @@ def use_pd_array(content: str, path: str) -> None:
     tree = ast.parse(content)
     visitor = Visitor(path)
     visitor.visit(tree)
-    return
 
 
 def main(argv: Optional[Sequence[str]] = None) -> None:

From 6eaee7d95abbd7cb011e2859ba2ad1b38499d36e Mon Sep 17 00:00:00 2001
From: Marco Gorelli <marcogorelli@protonmail.com>
Date: Fri, 2 Apr 2021 10:45:24 +0100
Subject: [PATCH 5/5] add source for coverage

---
 setup.cfg | 1 +
 1 file changed, 1 insertion(+)

diff --git a/setup.cfg b/setup.cfg
index 7165fc2275dc0..2aaafa0391531 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -140,6 +140,7 @@ omit =
     pandas/_typing.py
     pandas/_version.py
 plugins = Cython.Coverage
+source = pandas
 
 [coverage:report]
 ignore_errors = False