TYP: tighten Axis (#48612)

* TYP: tighten Axis

* allow 'rows'

pandas/_typing.py

@@ -106,7 +106,8 @@
 
 NumpyIndexT = TypeVar("NumpyIndexT", np.ndarray, "Index")
 
-Axis = Union[str, int]
+AxisInt = int
+Axis = Union[AxisInt, Literal["index", "columns", "rows"]]
 IndexLabel = Union[Hashable, Sequence[Hashable]]
 Level = Hashable
 Shape = Tuple[int, ...]

pandas/compat/numpy/function.py

@@ -29,7 +29,10 @@
     is_bool,
     is_integer,
 )
-from pandas._typing import Axis
+from pandas._typing import (
+    Axis,
+    AxisInt,
+)
 from pandas.errors import UnsupportedFunctionCall
 from pandas.util._validators import (
     validate_args,
@@ -413,7 +416,7 @@ def validate_resampler_func(method: str, args, kwargs) -> None:
             raise TypeError("too many arguments passed in")
 
 
-def validate_minmax_axis(axis: int | None, ndim: int = 1) -> None:
+def validate_minmax_axis(axis: AxisInt | None, ndim: int = 1) -> None:
     """
     Ensure that the axis argument passed to min, max, argmin, or argmax is zero
     or None, as otherwise it will be incorrectly ignored.

pandas/core/algorithms.py

@@ -29,6 +29,7 @@
 from pandas._typing import (
     AnyArrayLike,
     ArrayLike,
+    AxisInt,
     DtypeObj,
     IndexLabel,
     TakeIndexer,
@@ -1105,7 +1106,7 @@ def mode(
 
 def rank(
     values: ArrayLike,
-    axis: int = 0,
+    axis: AxisInt = 0,
     method: str = "average",
     na_option: str = "keep",
     ascending: bool = True,
@@ -1483,7 +1484,7 @@ def get_indexer(current_indexer, other_indexer):
 def take(
     arr,
     indices: TakeIndexer,
-    axis: int = 0,
+    axis: AxisInt = 0,
     allow_fill: bool = False,
     fill_value=None,
 ):
@@ -1675,7 +1676,7 @@ def searchsorted(
 _diff_special = {"float64", "float32", "int64", "int32", "int16", "int8"}
 
 
-def diff(arr, n: int, axis: int = 0):
+def diff(arr, n: int, axis: AxisInt = 0):
     """
     difference of n between self,
     analogous to s-s.shift(n)

pandas/core/apply.py

@@ -31,6 +31,7 @@
     AggFuncTypeDict,
     AggObjType,
     Axis,
+    AxisInt,
     NDFrameT,
     npt,
 )
@@ -104,7 +105,7 @@ def frame_apply(
 
 
 class Apply(metaclass=abc.ABCMeta):
-    axis: int
+    axis: AxisInt
 
     def __init__(
         self,

pandas/core/array_algos/masked_reductions.py

@@ -9,7 +9,10 @@
 import numpy as np
 
 from pandas._libs import missing as libmissing
-from pandas._typing import npt
+from pandas._typing import (
+    AxisInt,
+    npt,
+)
 
 from pandas.core.nanops import check_below_min_count
 
@@ -21,7 +24,7 @@ def _reductions(
     *,
     skipna: bool = True,
     min_count: int = 0,
-    axis: int | None = None,
+    axis: AxisInt | None = None,
     **kwargs,
 ):
     """
@@ -62,7 +65,7 @@ def sum(
     *,
     skipna: bool = True,
     min_count: int = 0,
-    axis: int | None = None,
+    axis: AxisInt | None = None,
 ):
     return _reductions(
         np.sum, values=values, mask=mask, skipna=skipna, min_count=min_count, axis=axis
@@ -75,7 +78,7 @@ def prod(
     *,
     skipna: bool = True,
     min_count: int = 0,
-    axis: int | None = None,
+    axis: AxisInt | None = None,
 ):
     return _reductions(
         np.prod, values=values, mask=mask, skipna=skipna, min_count=min_count, axis=axis
@@ -88,7 +91,7 @@ def _minmax(
     mask: npt.NDArray[np.bool_],
     *,
     skipna: bool = True,
-    axis: int | None = None,
+    axis: AxisInt | None = None,
 ):
     """
     Reduction for 1D masked array.
@@ -125,7 +128,7 @@ def min(
     mask: npt.NDArray[np.bool_],
     *,
     skipna: bool = True,
-    axis: int | None = None,
+    axis: AxisInt | None = None,
 ):
     return _minmax(np.min, values=values, mask=mask, skipna=skipna, axis=axis)
 
@@ -135,7 +138,7 @@ def max(
     mask: npt.NDArray[np.bool_],
     *,
     skipna: bool = True,
-    axis: int | None = None,
+    axis: AxisInt | None = None,
 ):
     return _minmax(np.max, values=values, mask=mask, skipna=skipna, axis=axis)
 
@@ -145,7 +148,7 @@ def mean(
     mask: npt.NDArray[np.bool_],
     *,
     skipna: bool = True,
-    axis: int | None = None,
+    axis: AxisInt | None = None,
 ):
     if not values.size or mask.all():
         return libmissing.NA
@@ -157,7 +160,7 @@ def var(
     mask: npt.NDArray[np.bool_],
     *,
     skipna: bool = True,
-    axis: int | None = None,
+    axis: AxisInt | None = None,
     ddof: int = 1,
 ):
     if not values.size or mask.all():

pandas/core/array_algos/take.py

@@ -15,6 +15,7 @@
 )
 from pandas._typing import (
     ArrayLike,
+    AxisInt,
     npt,
 )
 
@@ -36,7 +37,7 @@
 def take_nd(
     arr: np.ndarray,
     indexer,
-    axis: int = ...,
+    axis: AxisInt = ...,
     fill_value=...,
     allow_fill: bool = ...,
 ) -> np.ndarray:
@@ -47,7 +48,7 @@ def take_nd(
 def take_nd(
     arr: ExtensionArray,
     indexer,
-    axis: int = ...,
+    axis: AxisInt = ...,
     fill_value=...,
     allow_fill: bool = ...,
 ) -> ArrayLike:
@@ -57,7 +58,7 @@ def take_nd(
 def take_nd(
     arr: ArrayLike,
     indexer,
-    axis: int = 0,
+    axis: AxisInt = 0,
     fill_value=lib.no_default,
     allow_fill: bool = True,
 ) -> ArrayLike:
@@ -120,7 +121,7 @@ def take_nd(
 def _take_nd_ndarray(
     arr: np.ndarray,
     indexer: npt.NDArray[np.intp] | None,
-    axis: int,
+    axis: AxisInt,
     fill_value,
     allow_fill: bool,
 ) -> np.ndarray:
@@ -287,7 +288,7 @@ def take_2d_multi(
 
 @functools.lru_cache(maxsize=128)
 def _get_take_nd_function_cached(
-    ndim: int, arr_dtype: np.dtype, out_dtype: np.dtype, axis: int
+    ndim: int, arr_dtype: np.dtype, out_dtype: np.dtype, axis: AxisInt
 ):
     """
     Part of _get_take_nd_function below that doesn't need `mask_info` and thus
@@ -324,7 +325,11 @@ def _get_take_nd_function_cached(
 
 
 def _get_take_nd_function(
-    ndim: int, arr_dtype: np.dtype, out_dtype: np.dtype, axis: int = 0, mask_info=None
+    ndim: int,
+    arr_dtype: np.dtype,
+    out_dtype: np.dtype,
+    axis: AxisInt = 0,
+    mask_info=None,
 ):
     """
     Get the appropriate "take" implementation for the given dimension, axis
@@ -503,7 +508,7 @@ def _take_nd_object(
     arr: np.ndarray,
     indexer: npt.NDArray[np.intp],
     out: np.ndarray,
-    axis: int,
+    axis: AxisInt,
     fill_value,
     mask_info,
 ) -> None:

pandas/core/array_algos/transforms.py

@@ -6,8 +6,10 @@
 
 import numpy as np
 
+from pandas._typing import AxisInt
 
-def shift(values: np.ndarray, periods: int, axis: int, fill_value) -> np.ndarray:
+
+def shift(values: np.ndarray, periods: int, axis: AxisInt, fill_value) -> np.ndarray:
     new_values = values
 
     if periods == 0 or values.size == 0:

pandas/core/arrays/_mixins.py

@@ -17,6 +17,7 @@
 from pandas._libs.arrays import NDArrayBacked
 from pandas._typing import (
     ArrayLike,
+    AxisInt,
     Dtype,
     F,
     PositionalIndexer2D,
@@ -157,7 +158,7 @@ def take(
         *,
         allow_fill: bool = False,
         fill_value: Any = None,
-        axis: int = 0,
+        axis: AxisInt = 0,
     ) -> NDArrayBackedExtensionArrayT:
         if allow_fill:
             fill_value = self._validate_scalar(fill_value)
@@ -192,15 +193,15 @@ def _values_for_factorize(self):
         return self._ndarray, self._internal_fill_value
 
     # Signature of "argmin" incompatible with supertype "ExtensionArray"
-    def argmin(self, axis: int = 0, skipna: bool = True):  # type: ignore[override]
+    def argmin(self, axis: AxisInt = 0, skipna: bool = True):  # type: ignore[override]
         # override base class by adding axis keyword
         validate_bool_kwarg(skipna, "skipna")
         if not skipna and self._hasna:
             raise NotImplementedError
         return nargminmax(self, "argmin", axis=axis)
 
     # Signature of "argmax" incompatible with supertype "ExtensionArray"
-    def argmax(self, axis: int = 0, skipna: bool = True):  # type: ignore[override]
+    def argmax(self, axis: AxisInt = 0, skipna: bool = True):  # type: ignore[override]
         # override base class by adding axis keyword
         validate_bool_kwarg(skipna, "skipna")
         if not skipna and self._hasna:
@@ -216,7 +217,7 @@ def unique(self: NDArrayBackedExtensionArrayT) -> NDArrayBackedExtensionArrayT:
     def _concat_same_type(
         cls: type[NDArrayBackedExtensionArrayT],
         to_concat: Sequence[NDArrayBackedExtensionArrayT],
-        axis: int = 0,
+        axis: AxisInt = 0,
     ) -> NDArrayBackedExtensionArrayT:
         dtypes = {str(x.dtype) for x in to_concat}
         if len(dtypes) != 1:
@@ -351,7 +352,7 @@ def fillna(
     # ------------------------------------------------------------------------
     # Reductions
 
-    def _wrap_reduction_result(self, axis: int | None, result):
+    def _wrap_reduction_result(self, axis: AxisInt | None, result):
         if axis is None or self.ndim == 1:
             return self._box_func(result)
         return self._from_backing_data(result)

pandas/core/arrays/base.py

@@ -30,6 +30,7 @@
 from pandas._typing import (
     ArrayLike,
     AstypeArg,
+    AxisInt,
     Dtype,
     FillnaOptions,
     PositionalIndexer,
@@ -1137,7 +1138,7 @@ def factorize(
     @Substitution(klass="ExtensionArray")
     @Appender(_extension_array_shared_docs["repeat"])
     def repeat(
-        self: ExtensionArrayT, repeats: int | Sequence[int], axis: int | None = None
+        self: ExtensionArrayT, repeats: int | Sequence[int], axis: AxisInt | None = None
     ) -> ExtensionArrayT:
         nv.validate_repeat((), {"axis": axis})
         ind = np.arange(len(self)).repeat(repeats)
@@ -1567,7 +1568,7 @@ def _fill_mask_inplace(
     def _rank(
         self,
         *,
-        axis: int = 0,
+        axis: AxisInt = 0,
         method: str = "average",
         na_option: str = "keep",
         ascending: bool = True,

pandas/core/arrays/categorical.py

@@ -39,6 +39,7 @@
 from pandas._typing import (
     ArrayLike,
     AstypeArg,
+    AxisInt,
     Dtype,
     NpDtype,
     Ordered,
@@ -1988,7 +1989,7 @@ def sort_values(
     def _rank(
         self,
         *,
-        axis: int = 0,
+        axis: AxisInt = 0,
         method: str = "average",
         na_option: str = "keep",
         ascending: bool = True,
@@ -2464,7 +2465,7 @@ def equals(self, other: object) -> bool:
 
     @classmethod
     def _concat_same_type(
-        cls: type[CategoricalT], to_concat: Sequence[CategoricalT], axis: int = 0
+        cls: type[CategoricalT], to_concat: Sequence[CategoricalT], axis: AxisInt = 0
     ) -> CategoricalT:
         from pandas.core.dtypes.concat import union_categoricals