textual-fastdatatable/stubs/pyarrow/__init__.pyi

from __future__ import annotations

from typing import Any, Iterable, Iterator, Literal, Mapping, Sequence, Type, TypeVar

import pandas as pd

from .compute import CastOptions

class DataType: ...
class Date32Type(DataType): ...
class Date64Type(DataType): ...
class TimestampType(DataType): ...

def string() -> DataType: ...
def null() -> DataType: ...
def bool_() -> DataType: ...
def int8() -> DataType: ...
def int16() -> DataType: ...
def int32() -> DataType: ...
def int64() -> DataType: ...
def uint8() -> DataType: ...
def uint16() -> DataType: ...
def uint32() -> DataType: ...
def uint64() -> DataType: ...
def float16() -> DataType: ...
def float32() -> DataType: ...
def float64() -> DataType: ...
def date32() -> DataType: ...
def date64() -> DataType: ...
def binary(length: int = -1) -> DataType: ...
def large_binary() -> DataType: ...
def large_string() -> DataType: ...
def month_day_nano_interval() -> DataType: ...
def time32(unit: Literal["s", "ms", "us", "ns"]) -> DataType: ...
def time64(unit: Literal["s", "ms", "us", "ns"]) -> DataType: ...
def timestamp(
    unit: Literal["s", "ms", "us", "ns"], tz: str | None = None
) -> DataType: ...
def duration(unit: Literal["s", "ms", "us", "ns"]) -> DataType: ...

class MemoryPool: ...
class Schema: ...
class Field: ...
class NativeFile: ...
class MonthDayNano: ...

class Scalar:
    def as_py(self) -> Any: ...
    @property
    def type(self) -> DataType: ...

A = TypeVar("A", bound="_PandasConvertible")

class _PandasConvertible:
    @property
    def type(self) -> DataType: ...  # noqa: A003
    def cast(
        self: A,
        target_type: DataType | None = None,
        safe: bool = True,
        options: CastOptions | None = None,
    ) -> A: ...
    def __getitem__(self, index: int) -> Scalar: ...
    def __iter__(self) -> Any: ...
    def to_pylist(self) -> list[Any]: ...
    def fill_null(self: A, fill_value: Any) -> A: ...
    def drop_null(self: A) -> A: ...

class Array(_PandasConvertible): ...
class ChunkedArray(_PandasConvertible): ...

class StructArray(Array):
    def flatten(self, memory_pool: MemoryPool | None = None) -> list[Array]: ...

T = TypeVar("T", bound="_Tabular")

class _Tabular:
    @classmethod
    def from_arrays(
        cls: Type[T],
        arrays: list[_PandasConvertible],
        names: list[str] | None = None,
        schema: Schema | None = None,
        metadata: Mapping | None = None,
    ) -> T: ...
    @classmethod
    def from_pydict(
        cls: Type[T],
        mapping: Mapping,
        schema: Schema | None = None,
        metadata: Mapping | None = None,
    ) -> T: ...
    def __getitem__(self, index: int) -> _PandasConvertible: ...
    def __len__(self) -> int: ...
    @property
    def column_names(self) -> list[str]: ...
    @property
    def columns(self) -> list[_PandasConvertible]: ...
    @property
    def num_rows(self) -> int: ...
    @property
    def num_columns(self) -> int: ...
    @property
    def schema(self) -> Schema: ...
    def append_column(
        self: T, field_: str | Field, column: Array | ChunkedArray
    ) -> T: ...
    def column(self, i: int | str) -> _PandasConvertible: ...
    def equals(self: T, other: T, check_metadata: bool = False) -> bool: ...
    def itercolumns(self) -> Iterator[_PandasConvertible]: ...
    def rename_columns(self: T, names: list[str]) -> T: ...
    def select(self: T, columns: Sequence[str | int]) -> T: ...
    def set_column(
        self: T, i: int, field_: str | Field, column: Array | ChunkedArray
    ) -> T: ...
    def slice(  # noqa: A003
        self: T,
        offset: int = 0,
        length: int | None = None,
    ) -> T: ...
    def sort_by(
        self: T,
        sorting: str | list[tuple[str, Literal["ascending", "descending"]]],
        **kwargs: Any,
    ) -> T: ...
    def to_pylist(self) -> list[dict[str, Any]]: ...

class RecordBatch(_Tabular): ...

class Table(_Tabular):
    @classmethod
    def from_batches(
        cls,
        batches: Iterable[RecordBatch],
        schema: Schema | None = None,
    ) -> "Table": ...
    def to_batches(self) -> list[RecordBatch]: ...

def scalar(value: Any, type: DataType) -> Scalar: ...  # noqa: A002
def array(
    obj: Iterable,
    type: DataType | None = None,  # noqa: A002
    mask: Array | None = None,
    size: int | None = None,
    from_pandas: bool | None = None,
    safe: bool = True,
    memory_pool: MemoryPool | None = None,
) -> Array | ChunkedArray: ...
def concat_arrays(
    arrays: Iterable[Array], memory_pool: MemoryPool | None = None
) -> Array: ...
def nulls(
    size: int,
    type: DataType | None = None,  # noqa: A002
    memory_pool: MemoryPool | None = None,
) -> Array: ...
def table(
    data: pd.DataFrame
    | Mapping[str, _PandasConvertible | list]
    | list[_PandasConvertible],
    names: list[str] | None = None,
    schema: Schema | None = None,
    metadata: Mapping | None = None,
    nthreads: int | None = None,
) -> Table: ...
def set_timezone_db_path(path: str) -> None: ...