Coverage for /pythoncovmergedfiles/medio/medio/usr/local/lib/python3.11/site-packages/numpy/_typing/_array_like.py: 96%

1import sys 

2from collections.abc import Callable, Collection, Sequence 

3from typing import TYPE_CHECKING, Any, Protocol, TypeAlias, TypeVar, runtime_checkable 

4 

5import numpy as np 

6from numpy import dtype 

7 

8from ._nbit_base import _32Bit, _64Bit 

9from ._nested_sequence import _NestedSequence 

10from ._shape import _AnyShape 

11 

12if TYPE_CHECKING: 

13 StringDType = np.dtypes.StringDType 

14else: 

15 # at runtime outside of type checking importing this from numpy.dtypes 

16 # would lead to a circular import 

17 from numpy._core.multiarray import StringDType 

18 

19_T = TypeVar("_T") 

20_ScalarT = TypeVar("_ScalarT", bound=np.generic) 

21_DTypeT = TypeVar("_DTypeT", bound=dtype[Any]) 

22_DTypeT_co = TypeVar("_DTypeT_co", covariant=True, bound=dtype[Any]) 

23 

24NDArray: TypeAlias = np.ndarray[_AnyShape, dtype[_ScalarT]] 

25 

26# The `_SupportsArray` protocol only cares about the default dtype 

27# (i.e. `dtype=None` or no `dtype` parameter at all) of the to-be returned 

28# array. 

29# Concrete implementations of the protocol are responsible for adding 

30# any and all remaining overloads 

31@runtime_checkable 

32class _SupportsArray(Protocol[_DTypeT_co]): 

33 def __array__(self) -> np.ndarray[Any, _DTypeT_co]: ... 

34 

35 

36@runtime_checkable 

37class _SupportsArrayFunc(Protocol): 

38 """A protocol class representing `~class.__array_function__`.""" 

39 def __array_function__( 

40 self, 

41 func: Callable[..., Any], 

42 types: Collection[type[Any]], 

43 args: tuple[Any, ...], 

44 kwargs: dict[str, Any], 

45 ) -> object: ... 

46 

47 

48# TODO: Wait until mypy supports recursive objects in combination with typevars 

49_FiniteNestedSequence: TypeAlias = ( 

50 _T 

51 | Sequence[_T] 

52 | Sequence[Sequence[_T]] 

53 | Sequence[Sequence[Sequence[_T]]] 

54 | Sequence[Sequence[Sequence[Sequence[_T]]]] 

55) 

56 

57# A subset of `npt.ArrayLike` that can be parametrized w.r.t. `np.generic` 

58_ArrayLike: TypeAlias = ( 

59 _SupportsArray[dtype[_ScalarT]] 

60 | _NestedSequence[_SupportsArray[dtype[_ScalarT]]] 

61) 

62 

63# A union representing array-like objects; consists of two typevars: 

64# One representing types that can be parametrized w.r.t. `np.dtype` 

65# and another one for the rest 

66_DualArrayLike: TypeAlias = ( 

67 _SupportsArray[_DTypeT] 

68 | _NestedSequence[_SupportsArray[_DTypeT]] 

69 | _T 

70 | _NestedSequence[_T] 

71) 

72 

73if sys.version_info >= (3, 12): 

74 from collections.abc import Buffer as _Buffer 

75else: 

76 @runtime_checkable 

77 class _Buffer(Protocol): 

78 def __buffer__(self, flags: int, /) -> memoryview: ... 

79 

80ArrayLike: TypeAlias = _Buffer | _DualArrayLike[dtype[Any], complex | bytes | str] 

81 

82# `ArrayLike<X>_co`: array-like objects that can be coerced into `X` 

83# given the casting rules `same_kind` 

84_ArrayLikeBool_co: TypeAlias = _DualArrayLike[dtype[np.bool], bool] 

85_ArrayLikeUInt_co: TypeAlias = _DualArrayLike[dtype[np.bool | np.unsignedinteger], bool] 

86_ArrayLikeInt_co: TypeAlias = _DualArrayLike[dtype[np.bool | np.integer], int] 

87_ArrayLikeFloat_co: TypeAlias = _DualArrayLike[dtype[np.bool | np.integer | np.floating], float] 

88_ArrayLikeComplex_co: TypeAlias = _DualArrayLike[dtype[np.bool | np.number], complex] 

89_ArrayLikeNumber_co: TypeAlias = _ArrayLikeComplex_co 

90_ArrayLikeTD64_co: TypeAlias = _DualArrayLike[dtype[np.bool | np.integer | np.timedelta64], int] 

91_ArrayLikeDT64_co: TypeAlias = _ArrayLike[np.datetime64] 

92_ArrayLikeObject_co: TypeAlias = _ArrayLike[np.object_] 

93 

94_ArrayLikeVoid_co: TypeAlias = _ArrayLike[np.void] 

95_ArrayLikeBytes_co: TypeAlias = _DualArrayLike[dtype[np.bytes_], bytes] 

96_ArrayLikeStr_co: TypeAlias = _DualArrayLike[dtype[np.str_], str] 

97_ArrayLikeString_co: TypeAlias = _DualArrayLike[StringDType, str] 

98_ArrayLikeAnyString_co: TypeAlias = _DualArrayLike[dtype[np.character] | StringDType, bytes | str] 

99 

100__Float64_co: TypeAlias = np.floating[_64Bit] | np.float32 | np.float16 | np.integer | np.bool 

101__Complex128_co: TypeAlias = np.number[_64Bit] | np.number[_32Bit] | np.float16 | np.integer | np.bool 

102_ArrayLikeFloat64_co: TypeAlias = _DualArrayLike[dtype[__Float64_co], float] 

103_ArrayLikeComplex128_co: TypeAlias = _DualArrayLike[dtype[__Complex128_co], complex] 

104 

105# NOTE: This includes `builtins.bool`, but not `numpy.bool`. 

106_ArrayLikeInt: TypeAlias = _DualArrayLike[dtype[np.integer], int]