test_series.py

from __future__ import annotations

import math
from datetime import date, datetime, time, timedelta
from decimal import Decimal
from typing import TYPE_CHECKING, Any, cast
from zoneinfo import ZoneInfo

import numpy as np
import pandas as pd
import pyarrow as pa
import pytest

import polars as pl
from polars._utils.construction import iterable_to_pyseries
from polars.datatypes import (
    Datetime,
    Field,
    Float64,
    Int32,
    Int64,
    Time,
    UInt32,
    UInt64,
    Unknown,
)
from polars.exceptions import (
    DuplicateError,
    InvalidOperationError,
    PolarsInefficientMapWarning,
    ShapeError,
)
from polars.testing import assert_frame_equal, assert_series_equal
from tests.unit.conftest import FLOAT_DTYPES, INTEGER_DTYPES
from tests.unit.utils.pycapsule_utils import PyCapsuleStreamHolder

if TYPE_CHECKING:
    from collections.abc import Iterator

    from polars._typing import EpochTimeUnit, PolarsDataType, TimeUnit
    from tests.conftest import PlMonkeyPatch


def test_cum_agg() -> None:
    # confirm that known series give expected results
    s = pl.Series("a", [1, 2, 3, 2])
    assert_series_equal(s.cum_sum(), pl.Series("a", [1, 3, 6, 8]))
    assert_series_equal(s.cum_min(), pl.Series("a", [1, 1, 1, 1]))
    assert_series_equal(s.cum_max(), pl.Series("a", [1, 2, 3, 3]))
    assert_series_equal(s.cum_prod(), pl.Series("a", [1, 2, 6, 12]))


def test_cum_agg_with_nulls() -> None:
    # confirm that known series give expected results
    s = pl.Series("a", [None, 2, None, 7, 8, None])
    assert_series_equal(s.cum_sum(), pl.Series("a", [None, 2, None, 9, 17, None]))
    assert_series_equal(s.cum_min(), pl.Series("a", [None, 2, None, 2, 2, None]))
    assert_series_equal(s.cum_max(), pl.Series("a", [None, 2, None, 7, 8, None]))
    assert_series_equal(s.cum_prod(), pl.Series("a", [None, 2, None, 14, 112, None]))


def test_cum_agg_with_infs() -> None:
    # confirm that inf values are handled correctly
    s = pl.Series([float("inf"), 0.0, 1.0])
    assert_series_equal(s.cum_min(), pl.Series([float("inf"), 0.0, 0.0]))

    s = pl.Series([float("-inf"), 0.0, 1.0])
    assert_series_equal(s.cum_max(), pl.Series([float("-inf"), 0.0, 1.0]))


def test_cum_min_max_bool() -> None:
    s = pl.Series("a", [None, True, True, None, False, None, True, False, False, None])
    assert_series_equal(s.cum_min().cast(pl.Int32), s.cast(pl.Int32).cum_min())
    assert_series_equal(s.cum_max().cast(pl.Int32), s.cast(pl.Int32).cum_max())
    assert_series_equal(
        s.cum_min(reverse=True).cast(pl.Int32), s.cast(pl.Int32).cum_min(reverse=True)
    )
    assert_series_equal(
        s.cum_max(reverse=True).cast(pl.Int32), s.cast(pl.Int32).cum_max(reverse=True)
    )


def test_init_inputs(plmonkeypatch: PlMonkeyPatch) -> None:
    nan = float("nan")
    # Good inputs
    pl.Series("a", [1, 2])
    pl.Series("a", values=[1, 2])
    pl.Series(name="a", values=[1, 2])
    pl.Series(values=[1, 2], name="a")

    assert pl.Series([1, 2]).dtype == pl.Int64
    assert pl.Series(values=[1, 2]).dtype == pl.Int64
    assert pl.Series("a").dtype == pl.Null  # Null dtype used in case of no data
    assert pl.Series().dtype == pl.Null
    assert pl.Series([]).dtype == pl.Null
    assert (
        pl.Series([None, None, None]).dtype == pl.Null
    )  # f32 type used for list with only None
    assert pl.Series(values=[True, False]).dtype == pl.Boolean
    assert pl.Series(values=np.array([True, False])).dtype == pl.Boolean
    assert pl.Series(values=np.array(["foo", "bar"])).dtype == pl.String
    assert pl.Series(values=["foo", "bar"]).dtype == pl.String
    assert pl.Series("a", [pl.Series([1, 2, 4]), pl.Series([3, 2, 1])]).dtype == pl.List
    assert pl.Series("a", [10000, 20000, 30000], dtype=pl.Time).dtype == pl.Time

    # 2d numpy array and/or list of 1d numpy arrays
    for res in (
        pl.Series(
            name="a",
            values=np.array([[1, 2], [3, nan]], dtype=np.float32),
            nan_to_null=True,
        ),
        pl.Series(
            name="a",
            values=[
                np.array([1, 2], dtype=np.float32),
                np.array([3, nan], dtype=np.float32),
            ],
            nan_to_null=True,
        ),
        pl.Series(
            name="a",
            values=(
                np.ndarray((2,), np.float32, np.array([1, 2], dtype=np.float32)),
                np.ndarray((2,), np.float32, np.array([3, nan], dtype=np.float32)),
            ),
            nan_to_null=True,
        ),
    ):
        assert res.dtype == pl.Array(pl.Float32, shape=2)
        assert res[0].to_list() == [1.0, 2.0]
        assert res[1].to_list() == [3.0, None]

    # numpy from arange, with/without dtype
    two_ints = np.arange(2, dtype=np.int64)
    three_ints = np.arange(3, dtype=np.int64)
    for res in (
        pl.Series("a", [two_ints, three_ints]),
        pl.Series("a", [two_ints, three_ints], dtype=pl.List(pl.Int64)),
    ):
        assert res.dtype == pl.List(pl.Int64)
        assert res.to_list() == [[0, 1], [0, 1, 2]]

    assert pl.Series(
        values=np.array([["foo", "bar"], ["foo2", "bar2"]])
    ).dtype == pl.Array(pl.String, shape=2)

    # lists
    assert pl.Series("a", [[1, 2], [3, 4]]).dtype == pl.List(pl.Int64)

    # conversion of Date to Datetime
    s = pl.Series([date(2023, 1, 1), date(2023, 1, 2)], dtype=pl.Datetime)
    assert s.to_list() == [datetime(2023, 1, 1), datetime(2023, 1, 2)]
    assert Datetime == s.dtype
    assert s.dtype.time_unit == "us"  # type: ignore[attr-defined]
    assert s.dtype.time_zone is None  # type: ignore[attr-defined]

    # conversion of Date to Datetime with specified timezone and units
    tu: TimeUnit = "ms"
    tz = "America/Argentina/Rio_Gallegos"
    s = pl.Series(
        [date(2023, 1, 1), date(2023, 1, 2)], dtype=pl.Datetime(tu)
    ).dt.replace_time_zone(tz)
    d1 = datetime(2023, 1, 1, 0, 0, 0, 0, ZoneInfo(tz))
    d2 = datetime(2023, 1, 2, 0, 0, 0, 0, ZoneInfo(tz))
    assert s.to_list() == [d1, d2]
    assert Datetime == s.dtype
    assert s.dtype.time_unit == tu  # type: ignore[attr-defined]
    assert s.dtype.time_zone == tz  # type: ignore[attr-defined]

    # datetime64: check timeunit (auto-detect, implicit/explicit) and NaT
    d64 = pd.date_range(date(2021, 8, 1), date(2021, 8, 3)).values.astype(
        "datetime64[ns]"
    )
    d64[1] = None

    expected = [datetime(2021, 8, 1, 0), None, datetime(2021, 8, 3, 0)]
    for dtype in (None, Datetime, Datetime("ns")):
        s = pl.Series("dates", d64, dtype)
        assert s.to_list() == expected
        assert Datetime == s.dtype
        assert s.dtype.time_unit == "ns"  # type: ignore[attr-defined]

    s = pl.Series(values=d64.astype("<M8[ms]"))
    assert s.dtype.time_unit == "ms"  # type: ignore[attr-defined]
    assert expected == s.to_list()

    # pandas
    assert pl.Series(pd.Series([1, 2])).dtype == pl.Int64

    # Bad inputs
    with pytest.raises(TypeError):
        pl.Series([1, 2, 3], [1, 2, 3])
    with pytest.raises(TypeError):
        pl.Series({"a": [1, 2, 3]})
    with pytest.raises(OverflowError):
        pl.Series("bigint", [2**128])

    # numpy not available
    plmonkeypatch.setattr(pl.series.series, "_check_for_numpy", lambda x: False)
    with pytest.raises(TypeError):
        pl.DataFrame(np.array([1, 2, 3]), schema=["a"])


def test_init_structured_objects() -> None:
    # validate init from dataclass, namedtuple, and pydantic model objects
    from typing import NamedTuple

    from polars._dependencies import dataclasses, pydantic

    @dataclasses.dataclass
    class TeaShipmentDC:
        exporter: str
        importer: str
        product: str
        tonnes: int | None

    class TeaShipmentNT(NamedTuple):
        exporter: str
        importer: str
        product: str
        tonnes: None | int

    class TeaShipmentPD(pydantic.BaseModel):
        exporter: str
        importer: str
        product: str
        tonnes: int

    for Tea in (TeaShipmentDC, TeaShipmentNT, TeaShipmentPD):
        t0 = Tea(exporter="Sri Lanka", importer="USA", product="Ceylon", tonnes=10)
        t1 = Tea(exporter="India", importer="UK", product="Darjeeling", tonnes=25)
        t2 = Tea(exporter="China", importer="UK", product="Keemum", tonnes=40)

        s = pl.Series("t", [t0, t1, t2])

        assert isinstance(s, pl.Series)
        assert s.dtype.fields == [  # type: ignore[attr-defined]
            Field("exporter", pl.String),
            Field("importer", pl.String),
            Field("product", pl.String),
            Field("tonnes", pl.Int64),
        ]
        assert s.to_list() == [
            {
                "exporter": "Sri Lanka",
                "importer": "USA",
                "product": "Ceylon",
                "tonnes": 10,
            },
            {
                "exporter": "India",
                "importer": "UK",
                "product": "Darjeeling",
                "tonnes": 25,
            },
            {
                "exporter": "China",
                "importer": "UK",
                "product": "Keemum",
                "tonnes": 40,
            },
        ]
        assert_frame_equal(s.to_frame(), pl.DataFrame({"t": [t0, t1, t2]}))


def test_to_frame() -> None:
    s1 = pl.Series([1, 2])
    s2 = pl.Series("s", [1, 2])

    df1 = s1.to_frame()
    df2 = s2.to_frame()
    df3 = s1.to_frame("xyz")
    df4 = s2.to_frame("xyz")

    for df, name in ((df1, ""), (df2, "s"), (df3, "xyz"), (df4, "xyz")):
        assert isinstance(df, pl.DataFrame)
        assert df.rows() == [(1,), (2,)]
        assert df.columns == [name]

    # note: the empty string IS technically a valid column name
    assert s2.to_frame("").columns == [""]
    assert s2.name == "s"


def test_bitwise_ops() -> None:
    a = pl.Series([True, False, True])
    b = pl.Series([False, True, True])
    assert_series_equal((a & b), pl.Series([False, False, True]))
    assert_series_equal((a | b), pl.Series([True, True, True]))
    assert_series_equal((a ^ b), pl.Series([True, True, False]))
    assert_series_equal((~a), pl.Series([False, True, False]))

    # rand/rxor/ror we trigger by casting the left hand to a list here in the test
    # Note that the type annotations only allow Series to be passed in, but there is
    # specific code to deal with non-Series inputs.
    assert_series_equal(
        (True & a),
        pl.Series([True, False, True]),
    )
    assert_series_equal(
        (True | a),
        pl.Series([True, True, True]),
    )
    assert_series_equal(
        (True ^ a),
        pl.Series([False, True, False]),
    )


def test_bitwise_floats_invert() -> None:
    s = pl.Series([2.0, 3.0, 0.0])

    with pytest.raises(InvalidOperationError):
        ~s


def test_equality() -> None:
    a = pl.Series("a", [1, 2])
    b = a

    cmp = a == b
    assert isinstance(cmp, pl.Series)
    assert cmp.sum() == 2
    assert (a != b).sum() == 0
    assert (a >= b).sum() == 2
    assert (a <= b).sum() == 2
    assert (a > b).sum() == 0
    assert (a < b).sum() == 0
    assert a.sum() == 3
    assert_series_equal(a, b)

    a = pl.Series("name", ["ham", "foo", "bar"])
    assert_series_equal((a == "ham"), pl.Series("name", [True, False, False]))

    a = pl.Series("name", [[1], [1, 2], [2, 3]])
    assert_series_equal((a == [1]), pl.Series("name", [True, False, False]))


def test_agg() -> None:
    series = pl.Series("a", [1, 2])
    assert series.mean() == 1.5
    assert series.min() == 1
    assert series.max() == 2


def test_date_agg() -> None:
    series = pl.Series(
        [
            date(2022, 8, 2),
            date(2096, 8, 1),
            date(9009, 9, 9),
        ],
        dtype=pl.Date,
    )
    assert series.min() == date(2022, 8, 2)
    assert series.max() == date(9009, 9, 9)


@pytest.mark.parametrize(
    ("s", "min", "max"),
    [
        (pl.Series(["c", "b", "a"], dtype=pl.Categorical()), "a", "c"),
        (pl.Series([None, "a", "c", "b"], dtype=pl.Categorical()), "a", "c"),
        (pl.Series([], dtype=pl.Categorical()), None, None),
        (pl.Series(["c", "b", "a"], dtype=pl.Enum(["c", "b", "a"])), "c", "a"),
        (pl.Series(["c", "b", "a"], dtype=pl.Enum(["c", "b", "a", "d"])), "c", "a"),
    ],
)
def test_categorical_agg(s: pl.Series, min: str | None, max: str | None) -> None:
    assert s.min() == min
    assert s.max() == max


def test_add_string() -> None:
    s = pl.Series(["hello", "weird"])
    result = s + " world"
    print(result)
    assert_series_equal(result, pl.Series(["hello world", "weird world"]))

    result = "pfx:" + s
    assert_series_equal(result, pl.Series("literal", ["pfx:hello", "pfx:weird"]))


@pytest.mark.parametrize(
    ("data", "expected_dtype"),
    [
        (100, pl.Int64),
        (8.5, pl.Float64),
        ("서울특별시", pl.String),
        (date.today(), pl.Date),
        (datetime.now(), pl.Datetime("us")),
        (time(23, 59, 59), pl.Time),
        (timedelta(hours=7, seconds=123), pl.Duration("us")),
    ],
)
def test_unknown_dtype(data: Any, expected_dtype: PolarsDataType) -> None:
    # if given 'Unknown', should be able to infer the correct dtype
    s = pl.Series([data], dtype=Unknown)
    assert s.dtype == expected_dtype
    assert s.to_list() == [data]


def test_various() -> None:
    a = pl.Series("a", [1, 2])
    assert a.is_null().sum() == 0
    assert a.name == "a"

    a = a.rename("b")
    assert a.name == "b"
    assert a.len() == 2
    assert len(a) == 2

    a.append(a.clone())
    assert_series_equal(a, pl.Series("b", [1, 2, 1, 2]))

    a = pl.Series("a", range(20))
    assert a.head(5).len() == 5
    assert a.tail(5).len() == 5
    assert (a.head(5) != a.tail(5)).all()

    a = pl.Series("a", [2, 1, 4])
    a.sort(in_place=True)
    assert_series_equal(a, pl.Series("a", [1, 2, 4]))
    a = pl.Series("a", [2, 1, 1, 4, 4, 4])
    assert_series_equal(
        a.arg_unique(), pl.Series("a", [0, 1, 3], dtype=pl.get_index_type())
    )

    assert_series_equal(a.gather([2, 3]), pl.Series("a", [1, 4]))


def test_series_dtype_is() -> None:
    s = pl.Series("s", [1, 2, 3])

    assert s.dtype.is_numeric()
    assert s.dtype.is_integer()
    assert s.dtype.is_signed_integer()
    assert not s.dtype.is_unsigned_integer()
    assert (s * 0.99).dtype.is_float()

    s = pl.Series("s", [1, 2, 3], dtype=pl.UInt8)
    assert s.dtype.is_numeric()
    assert s.dtype.is_integer()
    assert not s.dtype.is_signed_integer()
    assert s.dtype.is_unsigned_integer()

    s = pl.Series("bool", [True, None, False])
    assert not s.dtype.is_numeric()

    s = pl.Series("s", ["testing..."])
    assert s.dtype == pl.String
    assert s.dtype != pl.Boolean

    s = pl.Series("s", [], dtype=pl.Decimal(20, 15))
    assert not s.dtype.is_float()
    assert s.dtype.is_numeric()
    assert s.is_empty()

    s = pl.Series("s", [], dtype=pl.Datetime("ms", time_zone="UTC"))
    assert s.dtype.is_temporal()


def test_series_head_tail_limit() -> None:
    s = pl.Series(range(10))

    assert_series_equal(s.head(5), pl.Series(range(5)))
    assert_series_equal(s.limit(5), s.head(5))
    assert_series_equal(s.tail(5), pl.Series(range(5, 10)))

    # check if it doesn't fail when out of bounds
    assert s.head(100).len() == 10
    assert s.limit(100).len() == 10
    assert s.tail(100).len() == 10

    # negative values
    assert_series_equal(s.head(-7), pl.Series(range(3)))
    assert s.head(-2).len() == 8
    assert_series_equal(s.tail(-8), pl.Series(range(8, 10)))
    assert s.head(-6).len() == 4

    # negative values out of bounds
    assert s.head(-12).len() == 0
    assert s.limit(-12).len() == 0
    assert s.tail(-12).len() == 0


def test_filter_ops() -> None:
    a = pl.Series("a", range(20))
    assert a.filter(a > 1).len() == 18
    assert a.filter(a < 1).len() == 1
    assert a.filter(a <= 1).len() == 2
    assert a.filter(a >= 1).len() == 19
    assert a.filter(a == 1).len() == 1
    assert a.filter(a != 1).len() == 19


def test_cast() -> None:
    a = pl.Series("a", range(20))

    assert a.cast(pl.Float32).dtype == pl.Float32
    assert a.cast(pl.Float64).dtype == pl.Float64
    assert a.cast(pl.Int32).dtype == pl.Int32
    assert a.cast(pl.UInt32).dtype == pl.UInt32
    assert a.cast(pl.Datetime).dtype == pl.Datetime
    assert a.cast(pl.Date).dtype == pl.Date

    # display failed values, GH#4706
    with pytest.raises(InvalidOperationError, match="foobar"):
        pl.Series(["1", "2", "3", "4", "foobar"]).cast(int)


@pytest.mark.parametrize(
    "test_data",
    [
        [1, None, 2],
        ["abc", None, "xyz"],
        [None, datetime.now()],
        [[1, 2], [3, 4], None],
    ],
)
def test_to_pandas(test_data: list[Any]) -> None:
    a = pl.Series("s", test_data)
    b = a.to_pandas()

    assert a.name == b.name
    assert b.isnull().sum() == 1

    vals_b: list[Any]
    if a.dtype == pl.List:
        vals_b = [(None if x is None else x.tolist()) for x in b]
    else:
        vals_b = b.replace({np.nan: None}).values.tolist()  # type: ignore[dict-item]

    assert vals_b == test_data

    try:
        c = a.to_pandas(use_pyarrow_extension_array=True)
        assert a.name == c.name
        assert c.isnull().sum() == 1
        vals_c = [None if x is pd.NA else x for x in c.tolist()]
        assert vals_c == test_data
    except ModuleNotFoundError:
        # Skip test if pandas>=1.5.0 or Pyarrow>=8.0.0 is not installed.
        pass


def test_series_to_list() -> None:
    s = pl.Series("a", range(20))
    result = s.to_list()
    assert isinstance(result, list)
    assert len(result) == 20

    a = pl.Series("a", [1, None, 2])
    assert a.null_count() == 1
    assert a.to_list() == [1, None, 2]


@pytest.mark.may_fail_cloud  # reason: list.to_struct is a eager operation
def test_to_struct() -> None:
    s = pl.Series("nums", ["12 34", "56 78", "90 00"]).str.extract_all(r"\d+")

    assert s.list.to_struct().struct.fields == ["field_0", "field_1"]
    assert s.list.to_struct(fields=lambda idx: f"n{idx:02}").struct.fields == [
        "n00",
        "n01",
    ]
    assert_frame_equal(
        s.list.to_struct(fields=["one", "two"]).struct.unnest(),
        pl.DataFrame({"one": ["12", "56", "90"], "two": ["34", "78", "00"]}),
    )


def test_to_struct_empty() -> None:
    df = pl.DataFrame({"y": [[], [], []]}, schema={"y": pl.List(pl.Int64)})
    empty_df = df.select(pl.col("y").list.to_struct(fields=[]).struct.unnest())
    assert empty_df.shape == (0, 0)


def test_sort() -> None:
    a = pl.Series("a", [2, 1, 3])
    assert_series_equal(a.sort(), pl.Series("a", [1, 2, 3]))
    assert_series_equal(a.sort(descending=True), pl.Series("a", [3, 2, 1]))


def test_rechunk() -> None:
    a = pl.Series("a", [1, 2, 3])
    b = pl.Series("b", [4, 5, 6])
    a.append(b)
    assert a.n_chunks() == 2
    assert a.rechunk(in_place=False).n_chunks() == 1
    a.rechunk(in_place=True)
    assert a.n_chunks() == 1


def test_indexing() -> None:
    a = pl.Series("a", [1, 2, None])
    assert a[1] == 2
    assert a[2] is None
    b = pl.Series("b", [True, False])
    assert b[0]
    assert not b[1]
    a = pl.Series("a", ["a", None])
    assert a[0] == "a"
    assert a[1] is None
    a = pl.Series("a", [0.1, None])
    assert a[0] == 0.1
    assert a[1] is None


def test_arrow() -> None:
    a = pl.Series("a", [1, 2, 3, None])
    out = a.to_arrow()
    assert out == pa.array([1, 2, 3, None])

    b = pl.Series("b", [1.0, 2.0, 3.0, None])
    out = b.to_arrow()
    assert out == pa.array([1.0, 2.0, 3.0, None])

    c = pl.Series("c", ["A", "BB", "CCC", None])
    out = c.to_arrow()
    assert out == pa.array(["A", "BB", "CCC", None], type=pa.large_string())
    assert_series_equal(pl.from_arrow(out), c.rename(""))  # type: ignore[arg-type]

    out = c.to_frame().to_arrow()["c"]
    assert isinstance(out, (pa.Array, pa.ChunkedArray))
    assert_series_equal(pl.from_arrow(out), c)  # type: ignore[arg-type]
    assert_series_equal(pl.from_arrow(out, schema=["x"]), c.rename("x"))  # type: ignore[arg-type]

    d = pl.Series("d", [None, None, None], pl.Null)
    out = d.to_arrow()
    assert out == pa.nulls(3)

    s = cast(
        "pl.Series",
        pl.from_arrow(pa.array([["foo"], ["foo", "bar"]], pa.list_(pa.utf8()))),
    )
    assert s.dtype == pl.List


def test_arrow_cat() -> None:
    # categorical dtype tests (including various forms of empty pyarrow array)
    arr0 = pa.array(["foo", "bar"], pa.dictionary(pa.int32(), pa.utf8()))
    assert_series_equal(
        pl.Series("arr", ["foo", "bar"], pl.Categorical), pl.Series("arr", arr0)
    )
    arr1 = pa.array(["xxx", "xxx", None, "yyy"]).dictionary_encode()
    arr2 = pa.chunked_array([], arr1.type)
    arr3 = pa.array([], arr1.type)
    arr4 = pa.array([]).dictionary_encode()

    assert_series_equal(
        pl.Series("arr", ["xxx", "xxx", None, "yyy"], dtype=pl.Categorical),
        pl.Series("arr", arr1),
    )
    for arr in (arr2, arr3):
        assert_series_equal(
            pl.Series("arr", [], dtype=pl.Categorical), pl.Series("arr", arr)
        )
    assert_series_equal(pl.Series("arr", [], dtype=pl.Null), pl.Series("arr", arr4))


def test_pycapsule_interface() -> None:
    a = pl.Series("a", [1, 2, 3, None])
    out = pa.chunked_array(PyCapsuleStreamHolder(a))
    out_arr = out.combine_chunks()
    assert out_arr == pa.array([1, 2, 3, None])


def test_get() -> None:
    a = pl.Series("a", [1, 2, 3])
    pos_idxs = pl.Series("idxs", [2, 0, 1, 0], dtype=pl.Int8)
    neg_and_pos_idxs = pl.Series(
        "neg_and_pos_idxs", [-2, 1, 0, -1, 2, -3], dtype=pl.Int8
    )
    empty_idxs = pl.Series("idxs", [], dtype=pl.Int8)
    empty_ints: list[int] = []
    assert a[0] == 1
    assert a[:2].to_list() == [1, 2]
    assert a[range(1)].to_list() == [1]
    assert a[range(0, 4, 2)].to_list() == [1, 3]
    assert a[:0].to_list() == []
    assert a[empty_ints].to_list() == []
    assert a[neg_and_pos_idxs.to_list()].to_list() == [2, 2, 1, 3, 3, 1]
    for dtype in (
        pl.UInt8,
        pl.UInt16,
        pl.UInt32,
        pl.UInt64,
        pl.Int8,
        pl.Int16,
        pl.Int32,
        pl.Int64,
    ):
        assert a[pos_idxs.cast(dtype)].to_list() == [3, 1, 2, 1]
        assert a[pos_idxs.cast(dtype).to_numpy()].to_list() == [3, 1, 2, 1]
        assert a[empty_idxs.cast(dtype)].to_list() == []
        assert a[empty_idxs.cast(dtype).to_numpy()].to_list() == []

    for dtype in (pl.Int8, pl.Int16, pl.Int32, pl.Int64):
        nps = a[neg_and_pos_idxs.cast(dtype).to_numpy()]
        assert nps.to_list() == [2, 2, 1, 3, 3, 1]


def test_set() -> None:
    a = pl.Series("a", [True, False, True])
    mask = pl.Series("msk", [True, False, True])
    a[mask] = False
    assert_series_equal(a, pl.Series("a", [False] * 3))


def test_set_value_as_list_fail() -> None:
    # only allowed for numerical physical types
    s = pl.Series("a", [1, 2, 3])
    s[[0, 2]] = [4, 5]
    assert s.to_list() == [4, 2, 5]

    # for other types it is not allowed
    s = pl.Series("a", ["a", "b", "c"])
    with pytest.raises(TypeError):
        s[[0, 1]] = ["d", "e"]

    s = pl.Series("a", [True, False, False])
    with pytest.raises(TypeError):
        s[[0, 1]] = [True, False]


@pytest.mark.parametrize("key", [True, False, 1.0])
def test_set_invalid_key(key: Any) -> None:
    s = pl.Series("a", [1, 2, 3])
    with pytest.raises(TypeError):
        s[key] = 1


@pytest.mark.parametrize(
    "key",
    [
        pl.Series([False, True, True]),
        pl.Series([1, 2], dtype=UInt32),
        pl.Series([1, 2], dtype=UInt64),
    ],
)
def test_set_key_series(key: pl.Series) -> None:
    """Only UInt32/UInt64/bool are allowed."""
    s = pl.Series("a", [1, 2, 3])
    s[key] = 4
    assert_series_equal(s, pl.Series("a", [1, 4, 4]))


def test_set_np_array_boolean_mask() -> None:
    a = pl.Series("a", [1, 2, 3])
    mask = np.array([True, False, True])
    a[mask] = 4
    assert_series_equal(a, pl.Series("a", [4, 2, 4]))


@pytest.mark.parametrize("dtype", [np.int32, np.int64, np.uint32, np.uint64])
def test_set_np_array(dtype: Any) -> None:
    a = pl.Series("a", [1, 2, 3])
    idx = np.array([0, 2], dtype=dtype)
    a[idx] = 4
    assert_series_equal(a, pl.Series("a", [4, 2, 4]))


@pytest.mark.parametrize("idx", [[0, 2], (0, 2)])
def test_set_list_and_tuple(idx: list[int] | tuple[int]) -> None:
    a = pl.Series("a", [1, 2, 3])
    a[idx] = 4
    assert_series_equal(a, pl.Series("a", [4, 2, 4]))


def test_init_nested_tuple() -> None:
    s1 = pl.Series("s", (1, 2, 3))
    assert s1.to_list() == [1, 2, 3]

    s2 = pl.Series("s", ((1, 2, 3),), dtype=pl.List(pl.UInt8))
    assert s2.to_list() == [[1, 2, 3]]
    assert s2.dtype == pl.List(pl.UInt8)

    s3 = pl.Series("s", ((1, 2, 3), (1, 2, 3)), dtype=pl.List(pl.Int32))
    assert s3.to_list() == [[1, 2, 3], [1, 2, 3]]
    assert s3.dtype == pl.List(pl.Int32)


def test_fill_null() -> None:
    s = pl.Series("a", [1, 2, None])
    assert_series_equal(s.fill_null(strategy="forward"), pl.Series("a", [1, 2, 2]))
    assert_series_equal(s.fill_null(14), pl.Series("a", [1, 2, 14], dtype=Int64))

    a = pl.Series("a", [0.0, 1.0, None, 2.0, None, 3.0])

    assert a.fill_null(0).to_list() == [0.0, 1.0, 0.0, 2.0, 0.0, 3.0]
    assert a.fill_null(strategy="zero").to_list() == [0.0, 1.0, 0.0, 2.0, 0.0, 3.0]
    assert a.fill_null(strategy="max").to_list() == [0.0, 1.0, 3.0, 2.0, 3.0, 3.0]
    assert a.fill_null(strategy="min").to_list() == [0.0, 1.0, 0.0, 2.0, 0.0, 3.0]
    assert a.fill_null(strategy="one").to_list() == [0.0, 1.0, 1.0, 2.0, 1.0, 3.0]
    assert a.fill_null(strategy="forward").to_list() == [0.0, 1.0, 1.0, 2.0, 2.0, 3.0]
    assert a.fill_null(strategy="backward").to_list() == [0.0, 1.0, 2.0, 2.0, 3.0, 3.0]
    assert a.fill_null(strategy="mean").to_list() == [0.0, 1.0, 1.5, 2.0, 1.5, 3.0]
    assert a.forward_fill().to_list() == [0.0, 1.0, 1.0, 2.0, 2.0, 3.0]
    assert a.backward_fill().to_list() == [0.0, 1.0, 2.0, 2.0, 3.0, 3.0]

    b = pl.Series("b", ["a", None, "c", None, "e"])
    assert b.fill_null(strategy="min").to_list() == ["a", "a", "c", "a", "e"]
    assert b.fill_null(strategy="max").to_list() == ["a", "e", "c", "e", "e"]
    assert b.fill_null(strategy="zero").to_list() == ["a", "", "c", "", "e"]
    assert b.fill_null(strategy="forward").to_list() == ["a", "a", "c", "c", "e"]
    assert b.fill_null(strategy="backward").to_list() == ["a", "c", "c", "e", "e"]

    c = pl.Series("c", [b"a", None, b"c", None, b"e"])
    assert c.fill_null(strategy="min").to_list() == [b"a", b"a", b"c", b"a", b"e"]
    assert c.fill_null(strategy="max").to_list() == [b"a", b"e", b"c", b"e", b"e"]
    assert c.fill_null(strategy="zero").to_list() == [b"a", b"", b"c", b"", b"e"]
    assert c.fill_null(strategy="forward").to_list() == [b"a", b"a", b"c", b"c", b"e"]
    assert c.fill_null(strategy="backward").to_list() == [b"a", b"c", b"c", b"e", b"e"]

    df = pl.DataFrame(
        [
            pl.Series("i32", [1, 2, None], dtype=pl.Int32),
            pl.Series("i64", [1, 2, None], dtype=pl.Int64),
            pl.Series("f32", [1, 2, None], dtype=pl.Float32),
            pl.Series("cat", ["a", "b", None], dtype=pl.Categorical),
            pl.Series("str", ["a", "b", None], dtype=pl.String),
            pl.Series("bool", [True, True, None], dtype=pl.Boolean),
        ]
    )

    assert df.fill_null(0, matches_supertype=False).fill_null("bar").fill_null(
        False
    ).to_dict(as_series=False) == {
        "i32": [1, 2, None],
        "i64": [1, 2, 0],
        "f32": [1.0, 2.0, None],
        "cat": ["a", "b", "bar"],
        "str": ["a", "b", "bar"],
        "bool": [True, True, False],
    }

    assert df.fill_null(0, matches_supertype=True).fill_null("bar").fill_null(
        False
    ).to_dict(as_series=False) == {
        "i32": [1, 2, 0],
        "i64": [1, 2, 0],
        "f32": [1.0, 2.0, 0.0],
        "cat": ["a", "b", "bar"],
        "str": ["a", "b", "bar"],
        "bool": [True, True, False],
    }
    df = pl.DataFrame({"a": [1, None, 2, None]})

    out = df.with_columns(
        pl.col("a").cast(pl.UInt8).alias("u8"),
        pl.col("a").cast(pl.UInt16).alias("u16"),
        pl.col("a").cast(pl.UInt32).alias("u32"),
        pl.col("a").cast(pl.UInt64).alias("u64"),
    ).fill_null(3)

    assert out.to_dict(as_series=False) == {
        "a": [1, 3, 2, 3],
        "u8": [1, 3, 2, 3],
        "u16": [1, 3, 2, 3],
        "u32": [1, 3, 2, 3],
        "u64": [1, 3, 2, 3],
    }
    assert out.dtypes == [pl.Int64, pl.UInt8, pl.UInt16, pl.UInt32, pl.UInt64]


def test_str_series_min_max_10674() -> None:
    str_series = pl.Series("b", ["a", None, "c", None, "e"], dtype=pl.String)
    assert str_series.min() == "a"
    assert str_series.max() == "e"
    assert str_series.sort(descending=False).min() == "a"
    assert str_series.sort(descending=True).max() == "e"


def test_fill_nan() -> None:
    nan = float("nan")
    a = pl.Series("a", [1.0, nan, 2.0, nan, 3.0])
    assert_series_equal(a.fill_nan(None), pl.Series("a", [1.0, None, 2.0, None, 3.0]))
    assert_series_equal(a.fill_nan(0), pl.Series("a", [1.0, 0.0, 2.0, 0.0, 3.0]))


def test_map_elements() -> None:
    a = pl.Series("a", [1, 2, None])
    with pytest.warns(PolarsInefficientMapWarning):
        b = a.map_elements(lambda x: x**2, return_dtype=pl.Int64)
    assert list(b) == [1, 4, None]

    a = pl.Series("a", ["foo", "bar", None])
    with pytest.warns(PolarsInefficientMapWarning):
        b = a.map_elements(lambda x: x + "py", return_dtype=pl.String)
    assert list(b) == ["foopy", "barpy", None]

    b = a.map_elements(lambda x: len(x), return_dtype=pl.Int32)
    assert list(b) == [3, 3, None]

    b = a.map_elements(lambda x: len(x))
    assert list(b) == [3, 3, None]

    # just check that it runs (somehow problem with conditional compilation)
    a = pl.Series("a", [2, 2, 3]).cast(pl.Datetime)
    a.map_elements(lambda x: x)
    a = pl.Series("a", [2, 2, 3]).cast(pl.Date)
    a.map_elements(lambda x: x)


def test_shape() -> None:
    s = pl.Series([1, 2, 3])
    assert s.shape == (3,)


@pytest.mark.parametrize("arrow_available", [True, False])
def test_create_list_series(
    arrow_available: bool, plmonkeypatch: PlMonkeyPatch
) -> None:
    plmonkeypatch.setattr(pl.series.series, "_PYARROW_AVAILABLE", arrow_available)
    a = [[1, 2], None, [None, 3]]
    s = pl.Series("", a)
    assert s.to_list() == a


def test_iter() -> None:
    s = pl.Series("", [1, 2, 3])

    itr = s.__iter__()
    assert itr.__next__() == 1
    assert itr.__next__() == 2
    assert itr.__next__() == 3
    assert sum(s) == 6


def test_empty() -> None:
    a = pl.Series(dtype=pl.Int8)
    assert a.dtype == pl.Int8
    assert a.is_empty()

    a = pl.Series()
    assert a.dtype == pl.Null
    assert a.is_empty()

    a = pl.Series("name", [])
    assert a.dtype == pl.Null
    assert a.is_empty()

    a = pl.Series(values=(), dtype=pl.Int8)
    assert a.dtype == pl.Int8
    assert a.is_empty()

    assert_series_equal(pl.Series(), pl.Series())
    assert_series_equal(
        pl.Series(dtype=pl.Int32), pl.Series(dtype=pl.Int64), check_dtypes=False
    )

    with pytest.raises(TypeError, match="ambiguous"):
        not pl.Series()


def test_round() -> None:
    a = pl.Series("f", [1.003, 2.003])
    b = a.round(2)
    assert b.to_list() == [1.00, 2.00]

    b = a.round()
    assert b.to_list() == [1.0, 2.0]


def test_round_int() -> None:
    s = pl.Series([1, 2, 3])
    assert_series_equal(s, s.round())


@pytest.mark.parametrize(
    ("series", "digits", "expected_result"),
    [
        pytest.param(pl.Series([1.234, 0.1234]), 2, pl.Series([1.2, 0.12]), id="f64"),
        pytest.param(
            pl.Series([1.234, 0.1234]).cast(pl.Float32),
            2,
            pl.Series([1.2, 0.12]).cast(pl.Float32),
            id="f32",
        ),
        pytest.param(pl.Series([123400, 1234]), 2, pl.Series([120000, 1200]), id="i64"),
        pytest.param(
            pl.Series([123400, 1234]).cast(pl.Int32),
            2,
            pl.Series([120000, 1200]).cast(pl.Int32),
            id="i32",
        ),
        pytest.param(
            pl.Series([0.0]), 2, pl.Series([0.0]), id="0 should remain the same"
        ),
    ],
)
def test_round_sig_figs(
    series: pl.Series, digits: int, expected_result: pl.Series
) -> None:
    result = series.round_sig_figs(digits=digits)
    assert_series_equal(result, expected_result)


def test_round_sig_figs_raises_exc() -> None:
    with pytest.raises(pl.exceptions.InvalidOperationError):
        pl.Series([1.234, 0.1234]).round_sig_figs(digits=0)


def test_apply_list_out() -> None:
    s = pl.Series("count", [3, 2, 2])
    out = s.map_elements(lambda val: pl.repeat(val, val, eager=True))
    assert out[0].to_list() == [3, 3, 3]
    assert out[1].to_list() == [2, 2]
    assert out[2].to_list() == [2, 2]


def test_reinterpret() -> None:
    s = pl.Series("a", [1, 1, 2], dtype=pl.UInt64)
    assert s.reinterpret(signed=True).dtype == pl.Int64
    df = pl.DataFrame([s])
    assert df.select([pl.col("a").reinterpret(signed=True)])["a"].dtype == pl.Int64


def test_mode() -> None:
    s = pl.Series("a", [1, 1, 2])
    assert s.mode().to_list() == [1]
    assert s.set_sorted().mode().to_list() == [1]

    df = pl.DataFrame([s])
    assert df.select([pl.col("a").mode()])["a"].to_list() == [1]
    assert (
        pl.Series(["foo", "bar", "buz", "bar"], dtype=pl.Categorical).mode().item()
        == "bar"
    )
    assert pl.Series([1.0, 2.0, 3.0, 2.0]).mode().item() == 2.0
    assert pl.Series(["a", "b", "c", "b"]).mode().item() == "b"

    # sorted data
    assert set(pl.int_range(0, 3, eager=True).mode().to_list()) == {0, 1, 2}


def test_diff() -> None:
    s = pl.Series("a", [1, 2, 3, 2, 2, 3, 0])

    assert_series_equal(
        s.diff(),
        pl.Series("a", [None, 1, 1, -1, 0, 1, -3]),
    )
    assert_series_equal(
        s.diff(null_behavior="drop"),
        pl.Series("a", [1, 1, -1, 0, 1, -3]),
    )


def test_diff_negative() -> None:
    s = pl.Series("a", [1, 2, 3, 2, 2, 3, 0])

    assert_series_equal(
        s.diff(-1),
        pl.Series("a", [-1, -1, 1, 0, -1, 3, None]),
    )
    assert_series_equal(
        s.diff(-1, null_behavior="drop"),
        pl.Series("a", [-1, -1, 1, 0, -1, 3]),
    )


def test_skew() -> None:
    s = pl.Series("a", [1, 2, 3, 2, 2, 3, 0])

    assert s.skew(bias=True) == pytest.approx(-0.5953924651018018)
    assert s.skew(bias=False) == pytest.approx(-0.7717168360221258)

    df = pl.DataFrame([s])
    assert np.isclose(
        df.select(pl.col("a").skew(bias=False))["a"][0], -0.7717168360221258
    )


def test_kurtosis() -> None:
    s = pl.Series("a", [1, 2, 3, 2, 2, 3, 0])
    expected = -0.6406250000000004

    assert s.kurtosis() == pytest.approx(expected)
    df = pl.DataFrame([s])
    assert np.isclose(df.select(pl.col("a").kurtosis())["a"][0], expected)


def test_sqrt() -> None:
    s = pl.Series("a", [1, 2])
    assert_series_equal(s.sqrt(), pl.Series("a", [1.0, np.sqrt(2)]))
    df = pl.DataFrame([s])
    assert_series_equal(
        df.select(pl.col("a").sqrt())["a"], pl.Series("a", [1.0, np.sqrt(2)])
    )


def test_cbrt() -> None:
    s = pl.Series("a", [1, 2])
    assert_series_equal(s.cbrt(), pl.Series("a", [1.0, np.cbrt(2)]))
    df = pl.DataFrame([s])
    assert_series_equal(
        df.select(pl.col("a").cbrt())["a"], pl.Series("a", [1.0, np.cbrt(2)])
    )


def test_range() -> None:
    s1 = pl.Series("a", [1, 2, 3, 2, 2, 3, 0])
    assert_series_equal(s1[2:5], s1[range(2, 5)])

    ranges = [range(-2, 1), range(3), range(2, 8, 2)]

    s2 = pl.Series("b", ranges, dtype=pl.List(pl.Int8))
    assert s2.to_list() == [[-2, -1, 0], [0, 1, 2], [2, 4, 6]]
    assert s2.dtype == pl.List(pl.Int8)
    assert s2.name == "b"

    s3 = pl.Series("c", (ranges for _ in range(3)))
    assert s3.to_list() == [
        [[-2, -1, 0], [0, 1, 2], [2, 4, 6]],
        [[-2, -1, 0], [0, 1, 2], [2, 4, 6]],
        [[-2, -1, 0], [0, 1, 2], [2, 4, 6]],
    ]
    assert s3.dtype == pl.List(pl.List(pl.Int64))

    df = pl.DataFrame([s1])
    assert_frame_equal(df[2:5], df[range(2, 5)])


def test_strict_cast() -> None:
    with pytest.raises(InvalidOperationError):
        pl.Series("a", [2**16]).cast(dtype=pl.Int16, strict=True)
    with pytest.raises(InvalidOperationError):
        pl.DataFrame({"a": [2**16]}).select([pl.col("a").cast(pl.Int16, strict=True)])


def test_floor_divide() -> None:
    s = pl.Series("a", [1, 2, 3])
    assert_series_equal(s // 2, pl.Series("a", [0, 1, 1]))
    assert_series_equal(
        pl.DataFrame([s]).select(pl.col("a") // 2)["a"], pl.Series("a", [0, 1, 1])
    )


def test_true_divide() -> None:
    s = pl.Series("a", [1, 2])
    assert_series_equal(s / 2, pl.Series("a", [0.5, 1.0]))
    assert_series_equal(
        pl.DataFrame([s]).select(pl.col("a") / 2)["a"], pl.Series("a", [0.5, 1.0])
    )

    # rtruediv
    assert_series_equal(
        pl.DataFrame([s]).select(2 / pl.col("a"))["literal"],
        pl.Series("literal", [2.0, 1.0]),
    )

    # https://github.com/pola-rs/polars/issues/1369
    vals = [3000000000, 2, 3]
    foo = pl.Series(vals)
    assert_series_equal(foo / 1, pl.Series(vals, dtype=Float64))
    assert_series_equal(
        pl.DataFrame({"a": vals}).select([pl.col("a") / 1])["a"],
        pl.Series("a", vals, dtype=Float64),
    )


def test_bitwise() -> None:
    a = pl.Series("a", [1, 2, 3])
    b = pl.Series("b", [3, 4, 5])
    assert_series_equal(a & b, pl.Series("a", [1, 0, 1]))
    assert_series_equal(a | b, pl.Series("a", [3, 6, 7]))
    assert_series_equal(a ^ b, pl.Series("a", [2, 6, 6]))

    df = pl.DataFrame([a, b])
    out = df.select(
        (pl.col("a") & pl.col("b")).alias("and"),
        (pl.col("a") | pl.col("b")).alias("or"),
        (pl.col("a") ^ pl.col("b")).alias("xor"),
    )
    assert_series_equal(out["and"], pl.Series("and", [1, 0, 1]))
    assert_series_equal(out["or"], pl.Series("or", [3, 6, 7]))
    assert_series_equal(out["xor"], pl.Series("xor", [2, 6, 6]))

    # ensure mistaken use of logical 'and'/'or' raises an exception
    with pytest.raises(TypeError, match="ambiguous"):
        a and b  # type: ignore[redundant-expr]

    with pytest.raises(TypeError, match="ambiguous"):
        a or b  # type: ignore[redundant-expr]


def test_from_generator_or_iterable() -> None:
    # generator function
    def gen(n: int) -> Iterator[int]:
        yield from range(n)

    # iterable object
    class Data:
        def __init__(self, n: int) -> None:
            self._n = n

        def __iter__(self) -> Iterator[int]:
            yield from gen(self._n)

    expected = pl.Series("s", range(10))
    assert expected.dtype == pl.Int64

    for generated_series in (
        pl.Series("s", values=gen(10)),
        pl.Series("s", values=Data(10)),
        pl.Series("s", values=(x for x in gen(10))),
    ):
        assert_series_equal(expected, generated_series)

    # test 'iterable_to_pyseries' directly to validate 'chunk_size' behaviour
    ps1 = iterable_to_pyseries("s", gen(10), dtype=pl.UInt8)
    ps2 = iterable_to_pyseries("s", gen(10), dtype=pl.UInt8, chunk_size=3)
    ps3 = iterable_to_pyseries("s", Data(10), dtype=pl.UInt8, chunk_size=6)

    expected = pl.Series("s", range(10), dtype=pl.UInt8)
    assert expected.dtype == pl.UInt8

    for ps in (ps1, ps2, ps3):
        generated_series = pl.Series("s")
        generated_series._s = ps
        assert_series_equal(expected, generated_series)

    # empty generator
    assert_series_equal(pl.Series("s", []), pl.Series("s", values=gen(0)))


def test_from_sequences(plmonkeypatch: PlMonkeyPatch) -> None:
    # test int, str, bool, flt
    values = [
        [[1], [None, 3]],
        [["foo"], [None, "bar"]],
        [[True], [None, False]],
        [[1.0], [None, 3.0]],
    ]

    for vals in values:
        plmonkeypatch.setattr(pl.series.series, "_PYARROW_AVAILABLE", False)
        a = pl.Series("a", vals)
        plmonkeypatch.setattr(pl.series.series, "_PYARROW_AVAILABLE", True)
        b = pl.Series("a", vals)
        assert_series_equal(a, b)
        assert a.to_list() == vals


def test_comparisons_int_series_to_float() -> None:
    srs_int = pl.Series([1, 2, 3, 4])

    assert_series_equal(srs_int - 1.0, pl.Series([0.0, 1.0, 2.0, 3.0]))
    assert_series_equal(srs_int + 1.0, pl.Series([2.0, 3.0, 4.0, 5.0]))
    assert_series_equal(srs_int * 2.0, pl.Series([2.0, 4.0, 6.0, 8.0]))
    assert_series_equal(srs_int / 2.0, pl.Series([0.5, 1.0, 1.5, 2.0]))
    assert_series_equal(srs_int % 2.0, pl.Series([1.0, 0.0, 1.0, 0.0]))
    assert_series_equal(4.0 % srs_int, pl.Series([0.0, 0.0, 1.0, 0.0]))

    assert_series_equal(srs_int // 2.0, pl.Series([0.0, 1.0, 1.0, 2.0]))
    assert_series_equal(srs_int < 3.0, pl.Series([True, True, False, False]))
    assert_series_equal(srs_int <= 3.0, pl.Series([True, True, True, False]))
    assert_series_equal(srs_int > 3.0, pl.Series([False, False, False, True]))
    assert_series_equal(srs_int >= 3.0, pl.Series([False, False, True, True]))
    assert_series_equal(srs_int == 3.0, pl.Series([False, False, True, False]))
    assert_series_equal(srs_int - True, pl.Series([0, 1, 2, 3]))


def test_comparisons_int_series_to_float_scalar() -> None:
    srs_int = pl.Series([1, 2, 3, 4])

    assert_series_equal(srs_int < 1.5, pl.Series([True, False, False, False]))
    assert_series_equal(srs_int > 1.5, pl.Series([False, True, True, True]))


def test_comparisons_datetime_series_to_date_scalar() -> None:
    srs_date = pl.Series([date(2023, 1, 1), date(2023, 1, 2), date(2023, 1, 3)])
    dt = datetime(2023, 1, 1, 12, 0, 0)

    assert_series_equal(srs_date < dt, pl.Series([True, False, False]))
    assert_series_equal(srs_date > dt, pl.Series([False, True, True]))


def test_comparisons_float_series_to_int() -> None:
    srs_float = pl.Series([1.0, 2.0, 3.0, 4.0])

    assert_series_equal(srs_float - 1, pl.Series([0.0, 1.0, 2.0, 3.0]))
    assert_series_equal(srs_float + 1, pl.Series([2.0, 3.0, 4.0, 5.0]))
    assert_series_equal(srs_float * 2, pl.Series([2.0, 4.0, 6.0, 8.0]))
    assert_series_equal(srs_float / 2, pl.Series([0.5, 1.0, 1.5, 2.0]))
    assert_series_equal(srs_float % 2, pl.Series([1.0, 0.0, 1.0, 0.0]))
    assert_series_equal(4 % srs_float, pl.Series([0.0, 0.0, 1.0, 0.0]))

    assert_series_equal(srs_float // 2, pl.Series([0.0, 1.0, 1.0, 2.0]))
    assert_series_equal(srs_float < 3, pl.Series([True, True, False, False]))
    assert_series_equal(srs_float <= 3, pl.Series([True, True, True, False]))
    assert_series_equal(srs_float > 3, pl.Series([False, False, False, True]))
    assert_series_equal(srs_float >= 3, pl.Series([False, False, True, True]))
    assert_series_equal(srs_float == 3, pl.Series([False, False, True, False]))
    assert_series_equal(srs_float - True, pl.Series([0.0, 1.0, 2.0, 3.0]))


def test_comparisons_bool_series_to_int() -> None:
    srs_bool = pl.Series([True, False])

    # (native bool comparison should work...)
    for t, f in ((True, False), (False, True)):
        assert list(srs_bool == t) == list(srs_bool != f) == [t, f]

    # TODO: do we want this to work?
    assert_series_equal(srs_bool / 1, pl.Series([True, False], dtype=Float64))
    match = (
        r"cannot do arithmetic with Series of dtype: Boolean"
        r" and argument of type: 'bool'"
    )
    with pytest.raises(TypeError, match=match):
        srs_bool - 1
    with pytest.raises(TypeError, match=match):
        srs_bool + 1
    match = (
        r"cannot do arithmetic with Series of dtype: Boolean"
        r" and argument of type: 'bool'"
    )
    with pytest.raises(TypeError, match=match):
        srs_bool % 2
    with pytest.raises(TypeError, match=match):
        srs_bool * 1

    from operator import ge, gt, le, lt

    for op in (ge, gt, le, lt):
        for scalar in (0, 1.0, True, False):
            op_str = op.__name__.replace("e", "t_eq")
            with pytest.raises(
                NotImplementedError,
                match=rf"Series of type Boolean does not have {op_str} operator",
            ):
                op(srs_bool, scalar)


@pytest.mark.parametrize(
    ("values", "compare_with", "compares_equal"),
    [
        (
            [date(1999, 12, 31), date(2021, 1, 31)],
            date(2021, 1, 31),
            [False, True],
        ),
        (
            [datetime(2021, 1, 1, 12, 0, 0), datetime(2021, 1, 2, 12, 0, 0)],
            datetime(2021, 1, 1, 12, 0, 0),
            [True, False],
        ),
        (
            [timedelta(days=1), timedelta(days=2)],
            timedelta(days=1),
            [True, False],
        ),
    ],
)
def test_temporal_comparison(
    values: list[Any], compare_with: Any, compares_equal: list[bool]
) -> None:
    assert_series_equal(
        pl.Series(values) == compare_with,
        pl.Series(compares_equal, dtype=pl.Boolean),
    )


@pytest.mark.parametrize(
    ("drop_nulls", "drop_first"),
    [
        (False, False),
        (False, True),
        (True, False),
        (True, True),
    ],
)
def test_to_dummies_with_nulls(drop_nulls: bool, drop_first: bool) -> None:
    s = pl.Series("s", [None, "a", "a", None, "b", "c"])
    expected = pl.DataFrame(
        {
            "s_a": [0, 1, 1, 0, 0, 0],
            "s_b": [0, 0, 0, 0, 1, 0],
            "s_c": [0, 0, 0, 0, 0, 1],
            "s_null": [1, 0, 0, 1, 0, 0],
        }
    ).cast(pl.UInt8)

    if drop_nulls:
        expected = expected.drop("s_null")
    if drop_first:
        expected = expected.drop("s_a")

    result = s.to_dummies(drop_nulls=drop_nulls, drop_first=drop_first)
    assert_frame_equal(result, expected)


@pytest.mark.parametrize(
    ("drop_nulls", "drop_first"),
    [
        (False, False),
        (False, True),
        (True, False),
        (True, True),
    ],
)
def test_to_dummies_no_nulls(drop_nulls: bool, drop_first: bool) -> None:
    s = pl.Series("s", ["a", "a", "b", "c"])
    expected = pl.DataFrame(
        {
            "s_a": [1, 1, 0, 0],
            "s_b": [0, 0, 1, 0],
            "s_c": [0, 0, 0, 1],
        }
    ).cast(pl.UInt8)

    if drop_first:
        expected = expected.drop("s_a")

    result = s.to_dummies(drop_nulls=drop_nulls, drop_first=drop_first)
    assert_frame_equal(result, expected)


def test_to_dummies_null_clash_19096() -> None:
    with pytest.raises(
        DuplicateError, match="column with name '_null' has more than one occurrence"
    ):
        pl.Series([None, "null"]).to_dummies()


def test_chunk_lengths() -> None:
    s = pl.Series("a", [1, 2, 2, 3])
    # this is a Series with one chunk, of length 4
    assert s.n_chunks() == 1
    assert s.chunk_lengths() == [4]


def test_limit() -> None:
    s = pl.Series("a", [1, 2, 3])
    assert_series_equal(s.limit(2), pl.Series("a", [1, 2]))


def test_filter() -> None:
    s = pl.Series("a", [1, 2, 3])
    mask = pl.Series("", [True, False, True])

    assert_series_equal(s.filter(mask), pl.Series("a", [1, 3]))
    assert_series_equal(s.filter([True, False, True]), pl.Series("a", [1, 3]))
    assert_series_equal(s.filter(np.array([True, False, True])), pl.Series("a", [1, 3]))

    with pytest.raises(RuntimeError, match="Expected a boolean mask"):
        s.filter(np.array([1, 0, 1]))


def test_gather_every() -> None:
    s = pl.Series("a", [1, 2, 3, 4])
    assert_series_equal(s.gather_every(2), pl.Series("a", [1, 3]))
    assert_series_equal(s.gather_every(2, offset=1), pl.Series("a", [2, 4]))


def test_arg_sort() -> None:
    s = pl.Series("a", [5, 3, 4, 1, 2])
    expected = pl.Series("a", [3, 4, 1, 2, 0], dtype=pl.get_index_type())

    assert_series_equal(s.arg_sort(), expected)

    expected_descending = pl.Series("a", [0, 2, 1, 4, 3], dtype=pl.get_index_type())
    assert_series_equal(s.arg_sort(descending=True), expected_descending)


@pytest.mark.parametrize(
    ("series", "argmin", "argmax"),
    [
        # Numeric
        (pl.Series([5, 3, 4, 1, 2]), 3, 0),
        (pl.Series([None, 5, 1]), 2, 1),
        (pl.Series([float("nan"), 3.0, 5.0]), 1, 2),
        (pl.Series([None, float("nan"), 3.0, 5.0]), 2, 3),
        # Boolean
        (pl.Series([True, False]), 1, 0),
        (pl.Series([True, True]), 0, 0),
        (pl.Series([False, False]), 0, 0),
        (pl.Series([None, True, False, True]), 2, 1),
        (pl.Series([None, True, True]), 1, 1),
        (pl.Series([None, False, False]), 1, 1),
        # String
        (pl.Series(["a", "c", "b"]), 0, 1),
        (pl.Series([None, "a", None, "b"]), 1, 3),
        # Binary
        (pl.Series([b"a", b"c", b"b"]), 0, 1),
        (pl.Series([None, b"a", None, b"b"]), 1, 3),
        # Decimal
        (pl.Series([Decimal("1.1"), Decimal("2.2"), Decimal("0.5")]), 2, 1),
        (pl.Series([None, Decimal("1.1"), None, Decimal("2.2")]), 1, 3),
        # Categorical
        (pl.Series(["c", "b", "a"], dtype=pl.Categorical()), 2, 0),
        (pl.Series("s", [None, "c", "b", None, "a"], pl.Categorical()), 4, 1),
    ],
)
def test_arg_min_arg_max(series: pl.Series, argmin: int, argmax: int) -> None:
    assert series.arg_min() == argmin, (
        f"values: {series.to_list()}, expected {argmin} got {series.arg_min()}"
    )
    assert series.arg_max() == argmax, (
        f"values: {series.to_list()}, expected {argmax} got {series.arg_max()}"
    )


@pytest.mark.parametrize(
    ("series"),
    [
        # All nulls
        pl.Series([None, None], dtype=pl.Int32),
        pl.Series([None, None], dtype=pl.Boolean),
        pl.Series([None, None], dtype=pl.String),
        pl.Series([None, None], dtype=pl.Categorical),
        pl.Series([None, None], dtype=pl.Categorical()),
        # Empty Series
        pl.Series([], dtype=pl.Int32),
        pl.Series([], dtype=pl.Boolean),
        pl.Series([], dtype=pl.String),
        pl.Series([], dtype=pl.Categorical),
    ],
)
def test_arg_min_arg_max_all_nulls_or_empty(series: pl.Series) -> None:
    assert series.arg_min() is None
    assert series.arg_max() is None


def test_arg_min_and_arg_max_sorted() -> None:
    # test ascending and descending numerical series
    s = pl.Series([None, 1, 2, 3, 4, 5])
    s.sort(in_place=True)  # set ascending sorted flag
    assert s.flags == {"SORTED_ASC": True, "SORTED_DESC": False}
    assert s.arg_min() == 1
    assert s.arg_max() == 5
    s = pl.Series([None, 5, 4, 3, 2, 1])
    s.sort(descending=True, in_place=True)  # set descing sorted flag
    assert s.flags == {"SORTED_ASC": False, "SORTED_DESC": True}
    assert s.arg_min() == 5
    assert s.arg_max() == 1

    # test ascending and descending str series
    s = pl.Series([None, "a", "b", "c", "d", "e"])
    s.sort(in_place=True)  # set ascending sorted flag
    assert s.flags == {"SORTED_ASC": True, "SORTED_DESC": False}
    assert s.arg_min() == 1
    assert s.arg_max() == 5
    s = pl.Series([None, "e", "d", "c", "b", "a"])
    s.sort(descending=True, in_place=True)  # set descing sorted flag
    assert s.flags == {"SORTED_ASC": False, "SORTED_DESC": True}
    assert s.arg_min() == 5
    assert s.arg_max() == 1


def test_is_null_is_not_null() -> None:
    s = pl.Series("a", [1.0, 2.0, 3.0, None])
    assert_series_equal(s.is_null(), pl.Series("a", [False, False, False, True]))
    assert_series_equal(s.is_not_null(), pl.Series("a", [True, True, True, False]))


def test_is_finite_is_infinite() -> None:
    s = pl.Series("a", [1.0, 2.0, np.inf])
    assert_series_equal(s.is_finite(), pl.Series("a", [True, True, False]))
    assert_series_equal(s.is_infinite(), pl.Series("a", [False, False, True]))


@pytest.mark.parametrize("float_type", [pl.Float32, pl.Float64])
def test_is_nan_is_not_nan(float_type: PolarsDataType) -> None:
    s = pl.Series([1.0, np.nan, None], dtype=float_type)

    assert_series_equal(s.is_nan(), pl.Series([False, True, None]))
    assert_series_equal(s.is_not_nan(), pl.Series([True, False, None]))
    assert_series_equal(s.fill_nan(2.0), pl.Series([1.0, 2.0, None], dtype=float_type))
    assert_series_equal(s.drop_nans(), pl.Series([1.0, None], dtype=float_type))


def test_float_methods_on_ints() -> None:
    # these float-specific methods work on non-float numeric types
    s = pl.Series([1, None], dtype=pl.Int32)
    assert_series_equal(s.is_finite(), pl.Series([True, None]))
    assert_series_equal(s.is_infinite(), pl.Series([False, None]))
    assert_series_equal(s.is_nan(), pl.Series([False, None]))
    assert_series_equal(s.is_not_nan(), pl.Series([True, None]))


def test_dot() -> None:
    s1 = pl.Series("a", [1, 2, 3])
    s2 = pl.Series("b", [4.0, 5.0, 6.0])

    assert np.array([1, 2, 3]) @ np.array([4, 5, 6]) == 32

    for dot_result in (
        s1.dot(s2),
        s1 @ s2,
        [1, 2, 3] @ s2,
        s1 @ np.array([4, 5, 6]),
    ):
        assert dot_result == 32

    with pytest.raises(ShapeError, match="length mismatch"):
        s1 @ [4, 5, 6, 7, 8]


@pytest.mark.parametrize(
    ("dtype"),
    [pl.Int8, pl.Int16, pl.Int32, pl.Float32, pl.Float64],
)
def test_peak_max_peak_min(dtype: pl.DataType) -> None:
    s = pl.Series("a", [4, 1, 3, 2, 5], dtype=dtype)

    result = s.peak_min()
    expected = pl.Series("a", [False, True, False, True, False])
    assert_series_equal(result, expected)

    result = s.peak_max()
    expected = pl.Series("a", [True, False, True, False, True])
    assert_series_equal(result, expected)


def test_peak_max_peak_min_bool() -> None:
    s = pl.Series("a", [False, True, False, True, True, False], dtype=pl.Boolean)
    result = s.peak_min()
    expected = pl.Series("a", [False, False, True, False, False, False])
    assert_series_equal(result, expected)

    result = s.peak_max()
    expected = pl.Series("a", [False, True, False, False, False, False])
    assert_series_equal(result, expected)


def test_shrink_to_fit() -> None:
    s = pl.Series("a", [4, 1, 3, 2, 5])
    sf = s.shrink_to_fit(in_place=True)
    assert sf is s

    s = pl.Series("a", [4, 1, 3, 2, 5])
    sf = s.shrink_to_fit(in_place=False)
    assert s is not sf


@pytest.mark.parametrize("unit", ["ns", "us", "ms"])
def test_cast_datetime_to_time(unit: TimeUnit) -> None:
    a = pl.Series(
        "a",
        [
            datetime(2022, 9, 7, 0, 0),
            datetime(2022, 9, 6, 12, 0),
            datetime(2022, 9, 7, 23, 59, 59),
            datetime(2022, 9, 7, 23, 59, 59, 201),
        ],
        dtype=Datetime(unit),
    )
    if unit == "ms":
        # NOTE: microseconds are lost for `unit=ms`
        expected_values = [time(0, 0), time(12, 0), time(23, 59, 59), time(23, 59, 59)]
    else:
        expected_values = [
            time(0, 0),
            time(12, 0),
            time(23, 59, 59),
            time(23, 59, 59, 201),
        ]
    expected = pl.Series("a", expected_values)
    assert_series_equal(a.cast(Time), expected)


def test_init_categorical() -> None:
    for values in [[None], ["foo", "bar"], [None, "foo", "bar"]]:
        expected = pl.Series("a", values, dtype=pl.String).cast(pl.Categorical)
        a = pl.Series("a", values, dtype=pl.Categorical)
        assert_series_equal(a, expected)


def test_iter_nested_list() -> None:
    elems = list(pl.Series("s", [[1, 2], [3, 4]]))
    assert_series_equal(elems[0], pl.Series([1, 2]))
    assert_series_equal(elems[1], pl.Series([3, 4]))

    rev_elems = list(reversed(pl.Series("s", [[1, 2], [3, 4]])))
    assert_series_equal(rev_elems[0], pl.Series([3, 4]))
    assert_series_equal(rev_elems[1], pl.Series([1, 2]))


def test_iter_nested_struct() -> None:
    # note: this feels inconsistent with the above test for nested list, but
    # let's ensure the behaviour is codified before potentially modifying...
    elems = list(pl.Series("s", [{"a": 1, "b": 2}, {"a": 3, "b": 4}]))
    assert elems[0] == {"a": 1, "b": 2}
    assert elems[1] == {"a": 3, "b": 4}

    rev_elems = list(reversed(pl.Series("s", [{"a": 1, "b": 2}, {"a": 3, "b": 4}])))
    assert rev_elems[0] == {"a": 3, "b": 4}
    assert rev_elems[1] == {"a": 1, "b": 2}


@pytest.mark.parametrize(
    "dtype",
    [
        pl.UInt8,
        pl.Float32,
        pl.Int32,
        pl.Boolean,
        pl.List(pl.String),
        pl.Struct([pl.Field("a", pl.Int64), pl.Field("b", pl.Boolean)]),
    ],
)
def test_nested_list_types_preserved(dtype: pl.DataType) -> None:
    srs = pl.Series([pl.Series([], dtype=dtype) for _ in range(5)])
    for srs_nested in srs:
        assert srs_nested.dtype == dtype


def test_to_physical() -> None:
    # casting an int result in an int
    s = pl.Series("a", [1, 2, 3])
    assert_series_equal(s.to_physical(), s)

    # casting a date results in an Int32
    s = pl.Series("a", [date(2020, 1, 1)] * 3)
    expected = pl.Series("a", [18262] * 3, dtype=Int32)
    assert_series_equal(s.to_physical(), expected)

    # casting a categorical results in a UInt32
    s = pl.Series(["cat1"]).cast(pl.Categorical)
    assert s.to_physical().dtype == pl.UInt32

    # casting a small enum results in a UInt8
    s = pl.Series(["cat1"]).cast(pl.Enum(["cat1"]))
    assert s.to_physical().dtype == pl.UInt8

    # casting a List(Categorical) results in a List(UInt32)
    s = pl.Series([["cat1"]]).cast(pl.List(pl.Categorical))
    assert s.to_physical().dtype == pl.List(pl.UInt32)

    # casting a List(Enum) with a small enum results in a List(UInt8)
    s = pl.Series(["cat1"]).cast(pl.List(pl.Enum(["cat1"])))
    assert s.to_physical().dtype == pl.List(pl.UInt8)


def test_to_physical_rechunked_21285() -> None:
    # A series with multiple chunks, dtype is array or list of structs with a
    # null field (causes rechunking) and a field with a different physical and
    # logical repr (causes the full body of `to_physical_repr` to run).
    arr_dtype = pl.Array(pl.Struct({"f0": pl.Time, "f1": pl.Null}), shape=(1,))
    s = pl.Series("a", [None], arr_dtype)  # content doesn't matter
    s = s.append(s)
    expected_arr_dtype = pl.Array(pl.Struct({"f0": Int64, "f1": pl.Null}), shape=(1,))
    expected = pl.Series("a", [None, None], expected_arr_dtype)
    assert_series_equal(s.to_physical(), expected)

    list_dtype = pl.List(pl.Struct({"f0": pl.Time, "f1": pl.Null}))
    s = pl.Series("a", [None], list_dtype)  # content doesn't matter
    s = s.append(s)
    expected_list_dtype = pl.List(pl.Struct({"f0": Int64, "f1": pl.Null}))
    expected = pl.Series("a", [None, None], expected_list_dtype)
    assert_series_equal(s.to_physical(), expected)


def test_is_between_datetime() -> None:
    s = pl.Series("a", [datetime(2020, 1, 1, 10, 0, 0), datetime(2020, 1, 1, 20, 0, 0)])
    start = datetime(2020, 1, 1, 12, 0, 0)
    end = datetime(2020, 1, 1, 23, 0, 0)
    expected = pl.Series("a", [False, True])

    # only on the expression api
    result = s.to_frame().with_columns(pl.col("*").is_between(start, end)).to_series()
    assert_series_equal(result, expected)


@pytest.mark.parametrize(
    "f",
    [
        "sin",
        "cos",
        "tan",
        "arcsin",
        "arccos",
        "arctan",
        "sinh",
        "cosh",
        "tanh",
        "arcsinh",
        "arccosh",
        "arctanh",
    ],
)
@pytest.mark.filterwarnings("ignore:invalid value encountered:RuntimeWarning")
def test_trigonometric(f: str) -> None:
    s = pl.Series("a", [0.0, math.pi, None, math.nan])
    expected = (
        pl.Series("a", getattr(np, f)(s.to_numpy()))
        .to_frame()
        .with_columns(pl.when(s.is_null()).then(None).otherwise(pl.col("a")).alias("a"))
        .to_series()
    )
    result = getattr(s, f)()
    assert_series_equal(result, expected)


@pytest.mark.filterwarnings("ignore:invalid value encountered:RuntimeWarning")
def test_trigonometric_cot() -> None:
    # cotangent is not available in numpy...
    s = pl.Series("a", [0.0, math.pi, None, math.nan])
    expected = pl.Series("a", [math.inf, -8.1656e15, None, math.nan])
    assert_series_equal(s.cot(), expected)


def test_trigonometric_invalid_input() -> None:
    # String
    s = pl.Series("a", ["1", "2", "3"])
    with pytest.raises(InvalidOperationError):
        s.sin()

    # Date
    s = pl.Series("a", [date(1990, 2, 28), date(2022, 7, 26)])
    with pytest.raises(InvalidOperationError):
        s.cosh()


@pytest.mark.parametrize("dtype", INTEGER_DTYPES)
def test_product_ints(dtype: PolarsDataType) -> None:
    a = pl.Series("a", [1, 2, 3], dtype=dtype)
    out = a.product()
    assert out == 6
    a = pl.Series("a", [1, 2, None], dtype=dtype)
    out = a.product()
    assert out == 2
    a = pl.Series("a", [None, 2, 3], dtype=dtype)
    out = a.product()
    assert out == 6


@pytest.mark.parametrize("dtype", FLOAT_DTYPES)
def test_product_floats(dtype: PolarsDataType) -> None:
    a = pl.Series("a", [], dtype=dtype)
    out = a.product()
    assert out == 1
    a = pl.Series("a", [None, None], dtype=dtype)
    out = a.product()
    assert out == 1
    a = pl.Series("a", [3.0, None, float("nan")], dtype=dtype)
    out = a.product()
    assert math.isnan(out)


def test_ceil() -> None:
    s = pl.Series([1.8, 1.2, 3.0])
    expected = pl.Series([2.0, 2.0, 3.0])
    assert_series_equal(s.ceil(), expected)


def test_duration_arithmetic() -> None:
    # apply some basic duration math to series
    s = pl.Series([datetime(2022, 1, 1, 10, 20, 30), datetime(2022, 1, 2, 20, 40, 50)])
    d1 = pl.duration(days=5, microseconds=123456)
    d2 = timedelta(days=5, microseconds=123456)

    expected_values = [
        datetime(2022, 1, 6, 10, 20, 30, 123456),
        datetime(2022, 1, 7, 20, 40, 50, 123456),
    ]
    for d in (d1, d2):
        df1 = pl.select((s + d).alias("d_offset"))
        df2 = pl.select((d + s).alias("d_offset"))
        assert df1["d_offset"].to_list() == expected_values
        assert_series_equal(df1["d_offset"], df2["d_offset"])


def test_mean_overflow() -> None:
    arr = np.array([255] * (1 << 17), dtype="int16")
    assert arr.mean() == 255.0


def test_sign() -> None:
    # Integers
    a = pl.Series("a", [-9, -0, 0, 4, None])
    expected = pl.Series("a", [-1, 0, 0, 1, None])
    assert_series_equal(a.sign(), expected)

    # Floats
    a = pl.Series("a", [-9.0, -0.0, 0.0, 4.0, float("nan"), None])
    expected = pl.Series("a", [-1.0, 0.0, 0.0, 1.0, float("nan"), None])
    assert_series_equal(a.sign(), expected)

    # Decimal
    s = pl.Series("a", [1, -1, 10, -10])
    for scale in [0, 1, 2, 3, 7, 16, 20, 30]:
        dtype = pl.Decimal(scale=scale)
        assert_series_equal(s.sign().cast(dtype), s.cast(dtype).sign())

        s = pl.Series("a", ["1.00", "20.00", "-1", "0", "-7"], dtype)
        assert_series_equal(
            s.sign(), pl.Series("a", ["1", "1", "-1", "0", "-1"], dtype)
        )

    # Invalid input
    a = pl.Series("a", [date(1950, 2, 1), date(1970, 1, 1), date(2022, 12, 12), None])
    with pytest.raises(InvalidOperationError):
        a.sign()


def test_exp() -> None:
    s = pl.Series("a", [0.1, 0.01, None])
    expected = pl.Series("a", [1.1051709180756477, 1.010050167084168, None])
    assert_series_equal(s.exp(), expected)
    # test if we can run on empty series as well.
    assert s[:0].exp().to_list() == []


def test_cumulative_eval() -> None:
    s = pl.Series("values", [1, 2, 3, 4, 5])

    # evaluate expressions individually
    expr1 = pl.element().first()
    expr2 = pl.element().last() ** 2

    expected1 = pl.Series("values", [1, 1, 1, 1, 1])
    expected2 = pl.Series("values", [1, 4, 9, 16, 25])
    assert_series_equal(s.cumulative_eval(expr1), expected1)
    assert_series_equal(s.cumulative_eval(expr2), expected2)

    # evaluate combined expressions and validate
    expr3 = expr1 - expr2
    expected3 = pl.Series("values", [0, -3, -8, -15, -24])
    assert_series_equal(s.cumulative_eval(expr3), expected3)


def test_first_last() -> None:
    # Ensure multiple chunks
    s1 = pl.Series("a", [None, None], dtype=pl.Int32)
    s2 = pl.Series("a", [None, 3, 4, None], dtype=pl.Int32)
    s3 = pl.Series("a", [None, None], dtype=pl.Int32)
    s = s1.append(s2).append(s3)
    assert s.first() is None
    assert s.first(ignore_nulls=True) == 3
    assert s.last() is None
    assert s.last(ignore_nulls=True) == 4


def test_clip() -> None:
    s = pl.Series("foo", [-50, 5, None, 50])
    assert s.clip(1, 10).to_list() == [1, 5, None, 10]


def test_repr() -> None:
    s = pl.Series("ints", [1001, 2002, 3003])
    s_repr = repr(s)

    assert "shape: (3,)" in s_repr
    assert "Series: 'ints' [i64]" in s_repr
    for n in s.to_list():
        assert str(n) in s_repr

    class XSeries(pl.Series):
        """Custom Series class."""

    # check custom class name reflected in repr output
    x = XSeries("ints", [1001, 2002, 3003])
    x_repr = repr(x)

    assert "shape: (3,)" in x_repr
    assert "XSeries: 'ints' [i64]" in x_repr
    assert "1001" in x_repr
    for n in x.to_list():
        assert str(n) in x_repr


def test_repr_html(df: pl.DataFrame) -> None:
    # check it does not panic/error, and appears to contain a table
    html = pl.Series("misc", [123, 456, 789])._repr_html_()
    assert "<table" in html


@pytest.mark.parametrize(
    ("value", "time_unit", "exp", "exp_type"),
    [
        (
            13285,
            "d",
            date(2006, 5, 17),
            pl.Date,
        ),
        (
            1147880044,
            "s",
            datetime(2006, 5, 17, 15, 34, 4),
            pl.Datetime("us"),
        ),
        (
            1147880044 * 1_000,
            "ms",
            datetime(2006, 5, 17, 15, 34, 4),
            pl.Datetime("us"),
        ),
        (
            1147880044 * 1_000_000,
            "us",
            datetime(2006, 5, 17, 15, 34, 4),
            pl.Datetime("us"),
        ),
        (
            1147880044 * 1_000_000_000,
            "ns",
            datetime(2006, 5, 17, 15, 34, 4),
            pl.Datetime("ns"),
        ),
    ],
)
def test_from_epoch_expr(
    value: int,
    time_unit: EpochTimeUnit,
    exp: date | datetime,
    exp_type: PolarsDataType,
) -> None:
    s = pl.Series("timestamp", [value, None])
    result = pl.from_epoch(s, time_unit=time_unit)

    expected = pl.Series("timestamp", [exp, None]).cast(exp_type)
    assert_series_equal(result, expected)


def test_get_chunks() -> None:
    a = pl.Series("a", [1, 2])
    b = pl.Series("a", [3, 4])
    chunks = pl.concat([a, b], rechunk=False).get_chunks()
    assert_series_equal(chunks[0], a)
    assert_series_equal(chunks[1], b)


def test_null_comparisons() -> None:
    s = pl.Series("s", [None, "str", "a"])
    assert (s.shift() == s).null_count() == 2
    assert (s.shift() != s).null_count() == 2


def test_min_max_agg_on_str() -> None:
    strings = ["b", "a", "x"]
    s = pl.Series(strings)
    assert (s.min(), s.max()) == ("a", "x")


def test_min_max_full_nan_15058() -> None:
    s = pl.Series([float("nan")] * 2)
    assert all(x != x for x in [s.min(), s.max()])


def test_is_between() -> None:
    s = pl.Series("num", [1, 2, None, 4, 5])
    assert s.is_between(2, 4).to_list() == [False, True, None, True, False]

    s = pl.Series("num", [1, 2, None, 4, 5])
    assert s.is_between(2, 4, closed="left").to_list() == [
        False,
        True,
        None,
        False,
        False,
    ]

    s = pl.Series("num", [1, 2, None, 4, 5])
    assert s.is_between(2, 4, closed="right").to_list() == [
        False,
        False,
        None,
        True,
        False,
    ]

    s = pl.Series("num", [1, 2, None, 4, 5])
    assert s.is_between(pl.lit(2) / 2, pl.lit(4) * 2, closed="both").to_list() == [
        True,
        True,
        None,
        True,
        True,
    ]

    s = pl.Series("s", ["a", "b", "c", "d", "e"])
    assert s.is_between("b", "d").to_list() == [
        False,
        True,
        True,
        True,
        False,
    ]


@pytest.mark.parametrize(
    ("dtype", "lower", "upper"),
    [
        (pl.Int8, -128, 127),
        (pl.UInt8, 0, 255),
        (pl.Int16, -32768, 32767),
        (pl.UInt16, 0, 65535),
        (pl.Int32, -2147483648, 2147483647),
        (pl.UInt32, 0, 4294967295),
        (pl.Int64, -9223372036854775808, 9223372036854775807),
        (pl.UInt64, 0, 18446744073709551615),
        (pl.Float32, float("-inf"), float("inf")),
        (pl.Float64, float("-inf"), float("inf")),
    ],
)
def test_upper_lower_bounds(
    dtype: PolarsDataType, upper: int | float, lower: int | float
) -> None:
    s = pl.Series("s", dtype=dtype)
    assert s.lower_bound().item() == lower
    assert s.upper_bound().item() == upper


def test_numpy_series_arithmetic() -> None:
    sx = pl.Series(values=[1, 2])
    y = np.array([3.0, 4.0])

    result_add1 = y + sx
    result_add2 = sx + y
    expected_add = pl.Series([4.0, 6.0], dtype=pl.Float64)
    assert_series_equal(result_add1, expected_add)  # type: ignore[arg-type]
    assert_series_equal(result_add2, expected_add)

    result_sub1 = cast("pl.Series", y - sx)  # py37 is different vs py311 on this one
    expected = pl.Series([2.0, 2.0], dtype=pl.Float64)
    assert_series_equal(result_sub1, expected)
    result_sub2 = sx - y
    expected = pl.Series([-2.0, -2.0], dtype=pl.Float64)
    assert_series_equal(result_sub2, expected)

    result_mul1 = y * sx
    result_mul2 = sx * y
    expected = pl.Series([3.0, 8.0], dtype=pl.Float64)
    assert_series_equal(result_mul1, expected)  # type: ignore[arg-type]
    assert_series_equal(result_mul2, expected)

    result_div1 = y / sx
    expected = pl.Series([3.0, 2.0], dtype=pl.Float64)
    assert_series_equal(result_div1, expected)  # type: ignore[arg-type]
    result_div2 = sx / y
    expected = pl.Series([1 / 3, 0.5], dtype=pl.Float64)
    assert_series_equal(result_div2, expected)

    result_pow1 = y**sx
    expected = pl.Series([3.0, 16.0], dtype=pl.Float64)
    assert_series_equal(result_pow1, expected)  # type: ignore[arg-type]
    result_pow2 = sx**y
    expected = pl.Series([1.0, 16.0], dtype=pl.Float64)
    assert_series_equal(result_pow2, expected)  # type: ignore[arg-type]


def test_from_epoch_seq_input() -> None:
    seq_input = [1147880044]
    expected = pl.Series([datetime(2006, 5, 17, 15, 34, 4)])
    result = pl.from_epoch(seq_input)
    assert_series_equal(result, expected)


def test_symmetry_for_max_in_names() -> None:
    # int
    a = pl.Series("a", [1])
    assert (a - a.max()).name == (a.max() - a).name == a.name  # type: ignore[union-attr]
    # float
    a = pl.Series("a", [1.0])
    assert (a - a.max()).name == (a.max() - a).name == a.name  # type: ignore[union-attr]
    # duration
    a = pl.Series("a", [1], dtype=pl.Duration("ns"))
    assert (a - a.max()).name == (a.max() - a).name == a.name  # type: ignore[union-attr]
    # datetime
    a = pl.Series("a", [1], dtype=pl.Datetime("ns"))
    assert (a - a.max()).name == (a.max() - a).name == a.name  # type: ignore[union-attr]

    # TODO: time arithmetic support?
    # a = pl.Series("a", [1], dtype=pl.Time)
    # assert (a - a.max()).name == (a.max() - a).name == a.name


def test_series_getitem_out_of_bounds_positive() -> None:
    s = pl.Series([1, 2])
    with pytest.raises(
        IndexError, match="index 10 is out of bounds for sequence of length 2"
    ):
        s[10]


def test_series_getitem_out_of_bounds_negative() -> None:
    s = pl.Series([1, 2])
    with pytest.raises(
        IndexError, match="index -10 is out of bounds for sequence of length 2"
    ):
        s[-10]


def test_series_cmp_fast_paths() -> None:
    assert (
        pl.Series([None], dtype=pl.Int32) != pl.Series([1, 2], dtype=pl.Int32)
    ).to_list() == [None, None]
    assert (
        pl.Series([None], dtype=pl.Int32) == pl.Series([1, 2], dtype=pl.Int32)
    ).to_list() == [None, None]

    assert (
        pl.Series([None], dtype=pl.String) != pl.Series(["a", "b"], dtype=pl.String)
    ).to_list() == [None, None]
    assert (
        pl.Series([None], dtype=pl.String) == pl.Series(["a", "b"], dtype=pl.String)
    ).to_list() == [None, None]

    assert (
        pl.Series([None], dtype=pl.Boolean)
        != pl.Series([True, False], dtype=pl.Boolean)
    ).to_list() == [None, None]
    assert (
        pl.Series([None], dtype=pl.Boolean)
        == pl.Series([False, False], dtype=pl.Boolean)
    ).to_list() == [None, None]


def test_comp_series_with_str_13123() -> None:
    s = pl.Series(["1", "2", None])
    assert_series_equal(s != "1", pl.Series([False, True, None]))
    assert_series_equal(s == "1", pl.Series([True, False, None]))
    assert_series_equal(s.eq_missing("1"), pl.Series([True, False, False]))
    assert_series_equal(s.ne_missing("1"), pl.Series([False, True, True]))


@pytest.mark.parametrize(
    ("data", "single", "multiple", "single_expected", "multiple_expected"),
    [
        ([1, 2, 3], 1, [2, 4], 0, [1, 3]),
        (["a", "b", "c"], "d", ["a", "d"], 3, [0, 3]),
        ([b"a", b"b", b"c"], b"d", [b"a", b"d"], 3, [0, 3]),
        (
            [date(2022, 1, 2), date(2023, 4, 1)],
            date(2022, 1, 1),
            [date(1999, 10, 1), date(2024, 1, 1)],
            0,
            [0, 2],
        ),
        ([1, 2, 3], 1, np.array([2, 4]), 0, [1, 3]),  # test np array.
    ],
)
def test_search_sorted(
    data: list[Any],
    single: Any,
    multiple: list[Any],
    single_expected: Any,
    multiple_expected: list[Any],
) -> None:
    s = pl.Series(data)
    single_s = s.search_sorted(single)
    assert single_s == single_expected

    multiple_s = s.search_sorted(multiple)
    assert_series_equal(
        multiple_s, pl.Series(multiple_expected, dtype=pl.get_index_type())
    )


def test_series_from_pandas_with_dtype() -> None:
    expected = pl.Series("foo", [1, 2, 3], dtype=pl.Int8)
    s = pl.Series("foo", pd.Series([1, 2, 3]), pl.Int8)
    assert_series_equal(s, expected)
    s = pl.Series("foo", pd.Series([1, 2, 3], dtype="Int16"), pl.Int8)
    assert_series_equal(s, expected)

    with pytest.raises(InvalidOperationError, match="conversion from"):
        pl.Series("foo", pd.Series([-1, 2, 3]), pl.UInt8)
    s = pl.Series("foo", pd.Series([-1, 2, 3]), pl.UInt8, strict=False)
    assert s.to_list() == [None, 2, 3]
    assert s.dtype == pl.UInt8

    with pytest.raises(InvalidOperationError, match="conversion from"):
        pl.Series("foo", pd.Series([-1, 2, 3], dtype="Int8"), pl.UInt8)
    s = pl.Series("foo", pd.Series([-1, 2, 3], dtype="Int8"), pl.UInt8, strict=False)
    assert s.to_list() == [None, 2, 3]
    assert s.dtype == pl.UInt8


def test_series_from_pyarrow_with_dtype() -> None:
    s = pl.Series("foo", pa.array([-1, 2, 3]), pl.Int8)
    assert_series_equal(s, pl.Series("foo", [-1, 2, 3], dtype=pl.Int8))

    with pytest.raises(InvalidOperationError, match="conversion from"):
        pl.Series("foo", pa.array([-1, 2, 3]), pl.UInt8)

    s = pl.Series("foo", pa.array([-1, 2, 3]), dtype=pl.UInt8, strict=False)
    assert s.to_list() == [None, 2, 3]
    assert s.dtype == pl.UInt8


def test_series_from_numpy_with_dtype() -> None:
    s = pl.Series("foo", np.array([-1, 2, 3]), pl.Int8)
    assert_series_equal(s, pl.Series("foo", [-1, 2, 3], dtype=pl.Int8))

    with pytest.raises(InvalidOperationError, match="conversion from"):
        pl.Series("foo", np.array([-1, 2, 3]), pl.UInt8)

    s = pl.Series("foo", np.array([-1, 2, 3]), dtype=pl.UInt8, strict=False)
    assert s.to_list() == [None, 2, 3]
    assert s.dtype == pl.UInt8


def test_raise_invalid_is_between() -> None:
    with pytest.raises(pl.exceptions.InvalidOperationError):
        pl.select(pl.lit(2).is_between(pl.lit("11"), pl.lit("33")))


def test_construction_large_nested_u64_17231() -> None:
    import polars as pl

    values = [{"f0": [9223372036854775808]}]
    dtype = pl.Struct({"f0": pl.List(pl.UInt64)})
    assert pl.Series(values, dtype=dtype).to_list() == values


def test_repeat_by() -> None:
    calculated = pl.select(a=pl.Series("a", [1, 2]).repeat_by(2))
    expected = pl.select(a=pl.Series("a", [[1, 1], [2, 2]]))
    assert calculated.equals(expected)


def test_is_close() -> None:
    a = pl.Series(
        "a",
        [
            1.0,
            1.0,
            float("-inf"),
            float("inf"),
            float("inf"),
            float("inf"),
            float("nan"),
        ],
    )
    b = pl.Series(
        "b", [1.3, 1.7, float("-inf"), float("inf"), float("-inf"), 1.0, float("nan")]
    )
    assert a.is_close(b, abs_tol=0.5).to_list() == [
        True,
        False,
        True,
        True,
        False,
        False,
        False,
    ]


def test_is_close_literal() -> None:
    a = pl.Series("a", [1.1, 1.2, 1.3, 1.4, float("inf"), float("nan")])
    assert a.is_close(1.2).to_list() == [False, True, False, False, False, False]


def test_is_close_nans_equal() -> None:
    a = pl.Series("a", [1.0, float("nan")])
    b = pl.Series("b", [2.0, float("nan")])
    assert a.is_close(b, nans_equal=True).to_list() == [False, True]


def test_is_close_invalid_abs_tol() -> None:
    with pytest.raises(pl.exceptions.ComputeError):
        pl.select(pl.lit(1.0).is_close(1, abs_tol=-1.0))


def test_is_close_invalid_rel_tol() -> None:
    with pytest.raises(pl.exceptions.ComputeError):
        pl.select(pl.lit(1.0).is_close(1, rel_tol=-1.0))


def test_comparisons_structs_raise() -> None:
    s = pl.Series([{"x": 1}, {"x": 2}, {"x": 3}])
    rhss = ["", " ", 5, {"x": 1}]
    for rhs in rhss:
        with pytest.raises(
            NotImplementedError,
            match=r"Series of type Struct\(\{'x': Int64\}\) does not have eq operator",
        ):
            s == rhs  # noqa: B015