1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324
|
from __future__ import annotations
import copy
from collections.abc import Callable, Sequence
from dataclasses import dataclass
from typing import TYPE_CHECKING, Generic, cast
import numpy as np
import pandas as pd
from packaging.version import Version
from pandas.api.extensions import ExtensionArray, ExtensionDtype
from pandas.api.types import is_scalar as pd_is_scalar
from xarray.core.types import DTypeLikeSave, T_ExtensionArray
from xarray.core.utils import (
NDArrayMixin,
is_allowed_extension_array,
is_allowed_extension_array_dtype,
)
HANDLED_EXTENSION_ARRAY_FUNCTIONS: dict[Callable, Callable] = {}
if TYPE_CHECKING:
from typing import Any
from pandas._typing import DtypeObj, Scalar
def is_scalar(value: object) -> bool:
"""Workaround: pandas is_scalar doesn't recognize Categorical nulls for some reason."""
return value is pd.CategoricalDtype.na_value or pd_is_scalar(value)
def implements(numpy_function_or_name: Callable | str) -> Callable:
"""Register an __array_function__ implementation.
Pass a function directly if it's guaranteed to exist in all supported numpy versions, or a
string to first check for its existence.
"""
def decorator(func):
if isinstance(numpy_function_or_name, str):
numpy_function = getattr(np, numpy_function_or_name, None)
else:
numpy_function = numpy_function_or_name
if numpy_function:
HANDLED_EXTENSION_ARRAY_FUNCTIONS[numpy_function] = func
return func
return decorator
@implements(np.issubdtype)
def __extension_duck_array__issubdtype(
extension_array_dtype: T_ExtensionArray, other_dtype: DTypeLikeSave
) -> bool:
return False # never want a function to think a pandas extension dtype is a subtype of numpy
@implements("astype") # np.astype was added in 2.1.0, but we only require >=1.24
def __extension_duck_array__astype(
array_or_scalar: T_ExtensionArray,
dtype: DTypeLikeSave,
order: str = "K",
casting: str = "unsafe",
subok: bool = True,
copy: bool = True,
device: str | None = None,
) -> ExtensionArray:
if (
not (
is_allowed_extension_array(array_or_scalar)
or is_allowed_extension_array_dtype(dtype)
)
or casting != "unsafe"
or not subok
or order != "K"
):
return NotImplemented
return as_extension_array(array_or_scalar, dtype, copy=copy)
@implements(np.asarray)
def __extension_duck_array__asarray(
array_or_scalar: np.typing.ArrayLike | T_ExtensionArray,
dtype: DTypeLikeSave | None = None,
) -> ExtensionArray:
if not is_allowed_extension_array(dtype):
return NotImplemented
return as_extension_array(array_or_scalar, dtype)
def as_extension_array(
array_or_scalar: np.typing.ArrayLike | T_ExtensionArray,
dtype: ExtensionDtype | DTypeLikeSave | None,
copy: bool = False,
) -> ExtensionArray:
if is_scalar(array_or_scalar):
return dtype.construct_array_type()._from_sequence( # type: ignore[union-attr]
[array_or_scalar], dtype=dtype
)
else:
return array_or_scalar.astype(dtype, copy=copy) # type: ignore[union-attr]
@implements(np.result_type)
def __extension_duck_array__result_type(
*arrays_and_dtypes: list[
np.typing.ArrayLike | np.typing.DTypeLike | ExtensionDtype | ExtensionArray
],
) -> DtypeObj:
extension_arrays_and_dtypes: list[ExtensionDtype | ExtensionArray] = [
cast(ExtensionDtype | ExtensionArray, x)
for x in arrays_and_dtypes
if is_allowed_extension_array(x) or is_allowed_extension_array_dtype(x)
]
if not extension_arrays_and_dtypes:
return NotImplemented
ea_dtypes: list[ExtensionDtype] = [
getattr(x, "dtype", cast(ExtensionDtype, x))
for x in extension_arrays_and_dtypes
]
scalars = [
x for x in arrays_and_dtypes if is_scalar(x) and x not in {pd.NA, np.nan}
]
# other_stuff could include:
# - arrays such as pd.ABCSeries, np.ndarray, or other array-api duck arrays
# - dtypes such as pd.DtypeObj, np.dtype, or other array-api duck dtypes
other_stuff = [
x
for x in arrays_and_dtypes
if not is_allowed_extension_array_dtype(x) and not is_scalar(x)
]
# We implement one special case: when possible, preserve Categoricals (avoid promoting
# to object) by merging the categories of all given Categoricals + scalars + NA.
# Ideally this could be upstreamed into pandas find_result_type / find_common_type.
if not other_stuff and all(
isinstance(x, pd.CategoricalDtype) and not x.ordered for x in ea_dtypes
):
return union_unordered_categorical_and_scalar(
cast(list[pd.CategoricalDtype], ea_dtypes),
scalars, # type: ignore[arg-type]
)
if not other_stuff and all(
isinstance(x, type(ea_type := ea_dtypes[0])) for x in ea_dtypes
):
return ea_type
raise ValueError(
f"Cannot cast values to shared type, found values: {arrays_and_dtypes}"
)
def union_unordered_categorical_and_scalar(
categorical_dtypes: list[pd.CategoricalDtype], scalars: list[Scalar]
) -> pd.CategoricalDtype:
scalars = [x for x in scalars if x is not pd.CategoricalDtype.na_value]
all_categories = set().union(*(x.categories for x in categorical_dtypes))
all_categories = all_categories.union(scalars)
return pd.CategoricalDtype(categories=list(all_categories))
@implements(np.broadcast_to)
def __extension_duck_array__broadcast(arr: T_ExtensionArray, shape: tuple):
if shape[0] == len(arr) and len(shape) == 1:
return arr
raise NotImplementedError("Cannot broadcast 1d-only pandas extension array.")
@implements(np.stack)
def __extension_duck_array__stack(arr: T_ExtensionArray, axis: int):
raise NotImplementedError("Cannot stack 1d-only pandas extension array.")
@implements(np.concatenate)
def __extension_duck_array__concatenate(
arrays: Sequence[T_ExtensionArray], axis: int = 0, out=None
) -> T_ExtensionArray:
return type(arrays[0])._concat_same_type(arrays) # type: ignore[attr-defined]
@implements(np.where)
def __extension_duck_array__where(
condition: T_ExtensionArray | np.typing.ArrayLike,
x: T_ExtensionArray,
y: T_ExtensionArray | np.typing.ArrayLike,
) -> T_ExtensionArray:
# pd.where won't broadcast 0-dim arrays across a scalar-like series; scalar y's must be preserved
if hasattr(y, "shape") and len(y.shape) == 1 and y.shape[0] == 1:
y = y[0] # type: ignore[index]
return cast(T_ExtensionArray, pd.Series(x).where(condition, y).array) # type: ignore[arg-type]
def _replace_duck(args, replacer: Callable[[PandasExtensionArray], list]) -> list:
args_as_list = list(args)
for index, value in enumerate(args_as_list):
if isinstance(value, PandasExtensionArray):
args_as_list[index] = replacer(value)
elif isinstance(value, tuple): # should handle more than just tuple? iterable?
args_as_list[index] = tuple(_replace_duck(value, replacer))
elif isinstance(value, list):
args_as_list[index] = _replace_duck(value, replacer)
return args_as_list
def replace_duck_with_extension_array(args) -> tuple:
return tuple(_replace_duck(args, lambda duck: duck.array))
def replace_duck_with_series(args) -> tuple:
return tuple(_replace_duck(args, lambda duck: pd.Series(duck.array)))
@implements(np.ndim)
def __extension_duck_array__ndim(x: PandasExtensionArray) -> int:
return x.ndim
@implements(np.reshape)
def __extension_duck_array__reshape(
arr: T_ExtensionArray, shape: tuple
) -> T_ExtensionArray:
if (shape[0] == len(arr) and len(shape) == 1) or shape == (-1,):
return arr
raise NotImplementedError(
f"Cannot reshape 1d-only pandas extension array to: {shape}"
)
@dataclass(frozen=True)
class PandasExtensionArray(NDArrayMixin, Generic[T_ExtensionArray]):
"""NEP-18 compliant wrapper for pandas extension arrays.
Parameters
----------
array : T_ExtensionArray
The array to be wrapped upon e.g,. :py:class:`xarray.Variable` creation.
```
"""
array: T_ExtensionArray
def __post_init__(self):
if not isinstance(self.array, pd.api.extensions.ExtensionArray):
raise TypeError(f"{self.array} is not a pandas ExtensionArray.")
# This does not use the UNSUPPORTED_EXTENSION_ARRAY_TYPES whitelist because
# we do support extension arrays from datetime, for example, that need
# duck array support internally via this class. These can appear from `DatetimeIndex`
# wrapped by `PandasIndex` internally, for example.
if not is_allowed_extension_array(self.array):
raise TypeError(
f"{self.array.dtype!r} should be converted to a numpy array in `xarray` internally."
)
def __array_function__(self, func, types, args, kwargs):
if func not in HANDLED_EXTENSION_ARRAY_FUNCTIONS:
raise KeyError("Function not registered for pandas extension arrays.")
args = replace_duck_with_extension_array(args)
res = HANDLED_EXTENSION_ARRAY_FUNCTIONS[func](*args, **kwargs)
if isinstance(res, ExtensionArray):
return PandasExtensionArray(res)
return res
def __array_ufunc__(self, ufunc, method, *inputs, **kwargs):
return ufunc(*inputs, **kwargs)
def __getitem__(self, key) -> PandasExtensionArray[T_ExtensionArray]:
if (
isinstance(key, tuple) and len(key) == 1
): # pyarrow type arrays can't handle single-length tuples
(key,) = key
item = self.array[key]
if is_allowed_extension_array(item):
return PandasExtensionArray(item)
if is_scalar(item) or isinstance(key, int):
return PandasExtensionArray(type(self.array)._from_sequence([item])) # type: ignore[call-arg,attr-defined,unused-ignore]
return PandasExtensionArray(item)
def __setitem__(self, key, val):
self.array[key] = val
def __len__(self):
return len(self.array)
def __eq__(self, other):
if isinstance(other, PandasExtensionArray):
return self.array == other.array
return self.array == other
def __ne__(self, other):
return ~(self == other)
@property
def ndim(self) -> int:
return 1
def __array__(
self, dtype: np.typing.DTypeLike | None = None, /, *, copy: bool | None = None
) -> np.ndarray:
if Version(np.__version__) >= Version("2.0.0"):
return np.asarray(self.array, dtype=dtype, copy=copy)
else:
return np.asarray(self.array, dtype=dtype)
def __getattr__(self, attr: str) -> Any:
# with __deepcopy__ or __copy__, the object is first constructed and then the sub-objects are attached (see https://docs.python.org/3/library/copy.html)
# Thus, if we didn't have `super().__getattribute__("array")` this method would call `self.array` (i.e., `getattr(self, "array")`) again while looking for `__setstate__`
# (which is apparently the first thing sought in copy.copy from the under-construction copied object),
# which would cause a recursion error since `array` is not present on the object when it is being constructed during `__{deep}copy__`.
# Even though we have defined these two methods now below due to `test_extension_array_copy_arrow_type` (cause unknown)
# we leave this here as it more robust than self.array
return getattr(super().__getattribute__("array"), attr)
def __copy__(self) -> PandasExtensionArray[T_ExtensionArray]:
return PandasExtensionArray(copy.copy(self.array))
def __deepcopy__(
self, memo: dict[int, Any] | None = None
) -> PandasExtensionArray[T_ExtensionArray]:
return PandasExtensionArray(copy.deepcopy(self.array, memo=memo))
|