Merge remote-tracking branch 'upstream/main'

.pre-commit-config.yaml

@@ -3,15 +3,15 @@ ci:
     autofix_prs: false
 repos:
 -   repo: https://github.com/python/black
-    rev: 23.12.1
+    rev: 24.1.1
     hooks:
     -   id: black
 -   repo: https://github.com/PyCQA/isort
     rev: 5.13.2
     hooks:
     -   id: isort
 -   repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.1.9
+    rev: v0.2.1
     hooks:
     -   id: ruff
         args: [

pandas-stubs/_libs/lib.pyi

@@ -5,7 +5,6 @@ from typing import (
 )
 
 import numpy as np
-from pandas import Interval
 from typing_extensions import (
     TypeAlias,
     TypeGuard,
@@ -21,7 +20,6 @@ def infer_dtype(value: object, skipna: bool = ...) -> str: ...
 def is_iterator(obj: object) -> bool: ...
 def is_scalar(val: object) -> bool: ...
 def is_list_like(obj: object, allow_sets: bool = ...) -> bool: ...
-def is_interval(val: object) -> TypeGuard[Interval]: ...
 def is_complex(val: object) -> TypeGuard[complex]: ...
 def is_bool(val: object) -> TypeGuard[bool | np.bool_]: ...
 def is_integer(val: object) -> TypeGuard[int | np.integer]: ...

pandas-stubs/_libs/properties.pyi

@@ -5,7 +5,10 @@ class CachedProperty:
     def __get__(self, obj, typ): ...
     def __set__(self, obj, value) -> None: ...
 
-cache_readonly: CachedProperty = ...
+# note: this is a lie to make type checkers happy (they special
+# case property). cache_readonly uses attribute names similar to
+# property (fget) but it does not provide fset and fdel.
+cache_readonly = property
 
 class AxisProperty:
     def __init__(self, axis: int = ..., doc: str = ...) -> None: ...

pandas-stubs/_libs/tslibs/nattype.pyi

@@ -5,9 +5,13 @@ from datetime import (
 )
 
 import numpy as np
-from typing_extensions import TypeAlias
+from typing_extensions import (
+    Self,
+    TypeAlias,
+)
 
 from pandas._libs.tslibs.period import Period
+from pandas._typing import TimeUnit
 
 NaT: NaTType
 iNaT: int
@@ -121,3 +125,6 @@ class NaTType:
     __le__: _NatComparison
     __gt__: _NatComparison
     __ge__: _NatComparison
+    @property
+    def unit(self) -> TimeUnit: ...
+    def as_unit(self, unit: TimeUnit, round_ok: bool = ...) -> Self: ...

pandas-stubs/_libs/tslibs/period.pyi

@@ -60,12 +60,9 @@ class PeriodMixin:
 class Period(PeriodMixin):
     def __init__(
         self,
-        value: Period
-        | str
-        | datetime.datetime
-        | datetime.date
-        | Timestamp
-        | None = ...,
+        value: (
+            Period | str | datetime.datetime | datetime.date | Timestamp | None
+        ) = ...,
         freq: str | BaseOffset | None = ...,
         ordinal: int | None = ...,
         year: int | None = ...,

pandas-stubs/_libs/tslibs/timedeltas.pyi

@@ -31,7 +31,10 @@ from pandas._libs.tslibs import (
 )
 from pandas._libs.tslibs.period import Period
 from pandas._libs.tslibs.timestamps import Timestamp
-from pandas._typing import npt
+from pandas._typing import (
+    TimeUnit,
+    npt,
+)
 
 class Components(NamedTuple):
     days: int
@@ -390,3 +393,6 @@ class Timedelta(timedelta):
     @property
     def components(self) -> Components: ...
     def view(self, dtype: npt.DTypeLike = ...) -> object: ...
+    @property
+    def unit(self) -> TimeUnit: ...
+    def as_unit(self, unit: TimeUnit, round_ok: bool = ...) -> Self: ...

pandas-stubs/_libs/tslibs/timestamps.pyi

@@ -38,6 +38,7 @@ from pandas._libs.tslibs import (
     Timedelta,
 )
 from pandas._typing import (
+    TimeUnit,
     np_ndarray_bool,
     npt,
 )
@@ -310,3 +311,6 @@ class Timestamp(datetime):
     def days_in_month(self) -> int: ...
     @property
     def daysinmonth(self) -> int: ...
+    @property
+    def unit(self) -> TimeUnit: ...
+    def as_unit(self, unit: TimeUnit, round_ok: bool = ...) -> Self: ...

pandas-stubs/_typing.pyi

@@ -48,6 +48,11 @@ from pandas.core.dtypes.dtypes import (
 
 from pandas.io.formats.format import EngFormatter
 
+# `Incomplete` is equivalent to `Any`. Use it to annotate symbols that you don't
+# know the type of yet and that should be changed in the future. Use `Any` only
+# where it is the only acceptable type.
+Incomplete: TypeAlias = Any
+
 ArrayLike: TypeAlias = ExtensionArray | np.ndarray
 AnyArrayLike: TypeAlias = Index | Series | np.ndarray
 PythonScalar: TypeAlias = str | bool | complex
@@ -80,6 +85,10 @@ class FulldatetimeDict(YearMonthDayDict, total=False):
     us: DatetimeDictArg
     ns: DatetimeDictArg
 
+CorrelationMethod: TypeAlias = (
+    Literal["pearson", "kendall", "spearman"]
+    | Callable[[np.ndarray, np.ndarray], float]
+)
 # dtypes
 NpDtype: TypeAlias = str | np.dtype[np.generic] | type[str | complex | bool | object]
 Dtype: TypeAlias = ExtensionDtype | NpDtype
@@ -444,6 +453,7 @@ class SequenceNotStr(Protocol[_T_co]):
 IndexLabel: TypeAlias = Hashable | Sequence[Hashable]
 Label: TypeAlias = Hashable | None
 Level: TypeAlias = Hashable | int
+Shape: TypeAlias = tuple[int, ...]
 Suffixes: TypeAlias = tuple[str | None, str | None]
 Ordered: TypeAlias = bool | None
 JSONSerializable: TypeAlias = PythonScalar | list | dict
@@ -469,8 +479,11 @@ AggFuncTypeSeriesToFrame: TypeAlias = list[AggFuncTypeBase] | AggFuncTypeDictSer
 AggFuncTypeFrame: TypeAlias = (
     AggFuncTypeBase | list[AggFuncTypeBase] | AggFuncTypeDictFrame
 )
+AggFuncTypeDict: TypeAlias = AggFuncTypeDictSeries | AggFuncTypeDictFrame
+AggFuncType: TypeAlias = AggFuncTypeBase | list[AggFuncTypeBase] | AggFuncTypeDict
 
 num: TypeAlias = complex
+AxisInt: TypeAlias = int
 AxisIndex: TypeAlias = Literal["index", 0]
 AxisColumn: TypeAlias = Literal["columns", 1]
 Axis: TypeAlias = AxisIndex | AxisColumn
@@ -563,9 +576,12 @@ IndexT = TypeVar("IndexT", bound=Index)
 IntervalT = TypeVar("IntervalT", bound=Interval)
 IntervalClosedType: TypeAlias = Literal["left", "right", "both", "neither"]
 
+ScalarIndexer: TypeAlias = int | np.integer
+SequenceIndexer: TypeAlias = slice | list[int] | np.ndarray
+PositionalIndexer: TypeAlias = ScalarIndexer | SequenceIndexer
 TakeIndexer: TypeAlias = Sequence[int] | Sequence[np.integer] | npt.NDArray[np.integer]
 
-IgnoreRaiseCoerce: TypeAlias = Literal["ignore", "raise", "coerce"]
+RaiseCoerce: TypeAlias = Literal["raise", "coerce"]
 
 # Shared by functions such as drop and astype
 IgnoreRaise: TypeAlias = Literal["ignore", "raise"]
@@ -758,5 +774,10 @@ RandomState: TypeAlias = (
     | np.random.BitGenerator
     | np.random.RandomState
 )
+Frequency: TypeAlias = str | BaseOffset
+TimeUnit: TypeAlias = Literal["s", "ms", "us", "ns"]
+TimeGrouperOrigin: TypeAlias = (
+    Timestamp | Literal["epoch", "start", "start_day", "end", "end_day"]
+)
 
 __all__ = ["npt", "type_t"]

pandas-stubs/api/types/__init__.pyi

@@ -19,7 +19,6 @@ from pandas.core.dtypes.api import (
     is_hashable as is_hashable,
     is_integer as is_integer,
     is_integer_dtype as is_integer_dtype,
-    is_interval as is_interval,
     is_iterator as is_iterator,
     is_list_like as is_list_like,
     is_named_tuple as is_named_tuple,

pandas-stubs/core/arrays/datetimelike.pyi

@@ -5,16 +5,21 @@ from pandas.core.arrays.base import (
     ExtensionArray,
     ExtensionOpsMixin,
 )
+from typing_extensions import Self
 
 from pandas._libs import (
     NaT as NaT,
     NaTType as NaTType,
 )
+from pandas._typing import TimeUnit
 
 class DatelikeOps:
     def strftime(self, date_format): ...
 
 class TimelikeOps:
+    @property
+    def unit(self) -> TimeUnit: ...
+    def as_unit(self, unit: TimeUnit) -> Self: ...
     def round(self, freq, ambiguous: str = ..., nonexistent: str = ...): ...
     def floor(self, freq, ambiguous: str = ..., nonexistent: str = ...): ...
     def ceil(self, freq, ambiguous: str = ..., nonexistent: str = ...): ...

pandas-stubs/core/base.pyi

@@ -1,7 +1,12 @@
-from collections.abc import Iterator
+from collections.abc import (
+    Hashable,
+    Iterator,
+)
 from typing import (
+    Any,
     Generic,
     Literal,
+    final,
 )
 
 import numpy as np
@@ -19,13 +24,19 @@ from pandas._typing import (
     Scalar,
     npt,
 )
+from pandas.util._decorators import cache_readonly
 
 class NoNewAttributesMixin:
-    def __setattr__(self, key, value) -> None: ...
+    def __setattr__(self, key: str, value: Any) -> None: ...
 
 class SelectionMixin(Generic[NDFrameT]):
+    obj: NDFrameT
+    exclusions: frozenset[Hashable]
+    @final
+    @cache_readonly
     def ndim(self) -> int: ...
     def __getitem__(self, key): ...
+    def aggregate(self, func, *args, **kwargs): ...
 
 class IndexOpsMixin(OpsMixin, Generic[S1]):
     __array_priority__: int = ...

pandas-stubs/core/dtypes/api.pyi

@@ -17,7 +17,6 @@ from pandas.core.dtypes.common import (
     is_hashable as is_hashable,
     is_integer as is_integer,
     is_integer_dtype as is_integer_dtype,
-    is_interval as is_interval,
     is_iterator as is_iterator,
     is_list_like as is_list_like,
     is_named_tuple as is_named_tuple,

pandas-stubs/core/dtypes/common.pyi

@@ -17,7 +17,6 @@ from pandas.core.dtypes.inference import (
     is_float as is_float,
     is_hashable as is_hashable,
     is_integer as is_integer,
-    is_interval as is_interval,
     is_iterator as is_iterator,
     is_list_like as is_list_like,
     is_named_tuple as is_named_tuple,

pandas-stubs/core/dtypes/inference.pyi

@@ -5,7 +5,6 @@ is_integer = lib.is_integer
 is_float = lib.is_float
 is_complex = lib.is_complex
 is_scalar = lib.is_scalar
-is_interval = lib.is_interval
 is_list_like = lib.is_list_like
 is_iterator = lib.is_iterator