Fix missing type stubs error from pyright (#1404)

* fix reportMissingTypeStubs error

* fix isort issues

* fix python 3.10 issue

* add pyarrow-stubs. add comment about check_array_indexer

Author: Dr-Irv

pandas-stubs/__init__.pyi

@@ -94,7 +94,6 @@ from pandas._config import (
 )
 
 from pandas.util._print_versions import show_versions as show_versions
-from pandas.util._tester import test as test
 
 from pandas.io.api import (
     ExcelFile as ExcelFile,

pandas-stubs/core/arrays/arrow/__init__.pyi

@@ -0,0 +1,2 @@
+from pandas.core.indexers.objects import BaseIndexer as BaseIndexer
+from pandas.core.indexers.utils import check_array_indexer as check_array_indexer

pandas-stubs/core/indexers/__init__.pyi

@@ -0,0 +1,31 @@
+from typing import Any
+
+import numpy as np
+from pandas import DatetimeIndex
+
+from pandas._libs.tslibs import BaseOffset
+
+class BaseIndexer:
+    def __init__(
+        self, index_array: np.ndarray | None = None, window_size: int = 0, **kwargs: Any
+    ) -> None: ...
+    def get_window_bounds(
+        self,
+        num_values: int,
+        min_periods: int | None,
+        center: bool | None,
+        closed: str | None = None,
+        step: int | None = None,
+    ) -> tuple[np.ndarray, np.ndarray]: ...
+
+class FixedForwardWindowIndexer(BaseIndexer): ...
+
+class VariableOffsetWindowIndexer(BaseIndexer):
+    def __init__(
+        self,
+        index_array: np.ndarray | None = None,
+        window_size: int = 0,
+        index: DatetimeIndex | None = None,
+        offset: BaseOffset | None = None,
+        **kwargs: Any,
+    ) -> None: ...

pandas-stubs/core/indexers/objects.pyi

@@ -0,0 +1,15 @@
+from typing import overload
+
+from pandas._typing import (
+    AnyArrayLike,
+    ListLike,
+    np_ndarray_int,
+)
+
+# Docs may not be right. See pandas-dev/pandas#62562
+@overload
+def check_array_indexer(array: AnyArrayLike, indexer: int) -> int: ...
+@overload
+def check_array_indexer(array: AnyArrayLike, indexer: slice) -> slice: ...
+@overload
+def check_array_indexer(array: AnyArrayLike, indexer: ListLike) -> np_ndarray_int: ...

pandas-stubs/core/indexers/utils.pyi

@@ -5,7 +5,6 @@ from pandas.io.excel import (
     read_excel as read_excel,
 )
 from pandas.io.feather_format import read_feather as read_feather
-from pandas.io.gbq import read_gbq as read_gbq
 from pandas.io.html import read_html as read_html
 from pandas.io.json import read_json as read_json
 from pandas.io.orc import read_orc as read_orc

pandas-stubs/io/api.pyi

@@ -12,10 +12,10 @@ from typing import (
     overload,
 )
 
-from odf.opendocument import OpenDocument
+from odf.opendocument import OpenDocument  # pyright: ignore[reportMissingTypeStubs]
 from openpyxl.workbook.workbook import Workbook
 from pandas.core.frame import DataFrame
-import pyxlsb.workbook
+import pyxlsb.workbook  # pyright: ignore[reportMissingTypeStubs]
 from typing_extensions import Self
 from xlrd.book import Book
 

pandas-stubs/io/excel/_base.pyi

@@ -34,8 +34,6 @@ from pandas._typing import (
     UsecolsArgType,
 )
 
-from pandas.io.common import IOHandles
-
 @overload
 def read_csv(
     filepath_or_buffer: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str],
@@ -472,7 +470,6 @@ class TextFileReader(abc.Iterator):
     chunksize: int | None
     nrows: int | None
     squeeze: bool
-    handles: IOHandles | None
     def __init__(
         self,
         f: FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str] | list,

pandas-stubs/io/parsers/readers.pyi

@@ -68,6 +68,7 @@ types-python-dateutil = ">=2.8.19"
 beautifulsoup4 = ">=4.14.2"
 html5lib = ">=1.1"
 python-calamine = ">=0.2.0"
+pyarrow-stubs = { version = ">=20.0.0.20250928", python = "<4" }
 
 [build-system]
 requires = ["poetry-core>=1.0.0"]
@@ -240,7 +241,6 @@ enableTypeIgnoreComments = false    # use pyright-specific ignores
 # disable subset of strict
 reportMissingParameterType = false
 reportMissingTypeArgument = false
-reportMissingTypeStubs = false
 reportUnnecessaryTypeIgnoreComment = true
 reportUnknownArgumentType = false
 reportUnknownLambdaType = false

pyproject.toml

@@ -6,7 +6,6 @@
 import sys
 
 import numpy as np
-import pandas as pd
 from pandas.api.extensions import (
     no_default,
     register_extension_dtype,
@@ -165,7 +164,7 @@ def __getitem__(self, item):
             return self._data[item]
         else:
             # array, slice.
-            item = pd.api.indexers.check_array_indexer(self, item)
+            item = check_array_indexer(self, item)
             return type(self)(self._data[item])
 
     def take(