test_astype.py

import re

import numpy as np
import pytest

from pandas import (
    Categorical,
    CategoricalDtype,
    DataFrame,
    DatetimeTZDtype,
    Interval,
    IntervalDtype,
    NaT,
    Series,
    Timedelta,
    Timestamp,
    UInt64Index,
    concat,
    date_range,
    option_context,
)
import pandas._testing as tm
from pandas.core.arrays import integer_array


def _check_cast(df, v):
    """
    Check if all dtypes of df are equal to v
    """
    assert all(s.dtype.name == v for _, s in df.items())


class TestAstype:
    def test_astype_float(self, float_frame):
        casted = float_frame.astype(int)
        expected = DataFrame(
            float_frame.values.astype(int),
            index=float_frame.index,
            columns=float_frame.columns,
        )
        tm.assert_frame_equal(casted, expected)

        casted = float_frame.astype(np.int32)
        expected = DataFrame(
            float_frame.values.astype(np.int32),
            index=float_frame.index,
            columns=float_frame.columns,
        )
        tm.assert_frame_equal(casted, expected)

        float_frame["foo"] = "5"
        casted = float_frame.astype(int)
        expected = DataFrame(
            float_frame.values.astype(int),
            index=float_frame.index,
            columns=float_frame.columns,
        )
        tm.assert_frame_equal(casted, expected)

    def test_astype_mixed_float(self, mixed_float_frame):
        # mixed casting
        casted = mixed_float_frame.reindex(columns=["A", "B"]).astype("float32")
        _check_cast(casted, "float32")

        casted = mixed_float_frame.reindex(columns=["A", "B"]).astype("float16")
        _check_cast(casted, "float16")

    def test_astype_mixed_type(self, mixed_type_frame):
        # mixed casting
        mn = mixed_type_frame._get_numeric_data().copy()
        mn["little_float"] = np.array(12345.0, dtype="float16")
        mn["big_float"] = np.array(123456789101112.0, dtype="float64")

        casted = mn.astype("float64")
        _check_cast(casted, "float64")

        casted = mn.astype("int64")
        _check_cast(casted, "int64")

        casted = mn.reindex(columns=["little_float"]).astype("float16")
        _check_cast(casted, "float16")

        casted = mn.astype("float32")
        _check_cast(casted, "float32")

        casted = mn.astype("int32")
        _check_cast(casted, "int32")

        # to object
        casted = mn.astype("O")
        _check_cast(casted, "object")

    def test_astype_with_exclude_string(self, float_frame):
        df = float_frame.copy()
        expected = float_frame.astype(int)
        df["string"] = "foo"
        casted = df.astype(int, errors="ignore")

        expected["string"] = "foo"
        tm.assert_frame_equal(casted, expected)

        df = float_frame.copy()
        expected = float_frame.astype(np.int32)
        df["string"] = "foo"
        casted = df.astype(np.int32, errors="ignore")

        expected["string"] = "foo"
        tm.assert_frame_equal(casted, expected)

    def test_astype_with_view_float(self, float_frame):

        # this is the only real reason to do it this way
        tf = np.round(float_frame).astype(np.int32)
        casted = tf.astype(np.float32, copy=False)

        # TODO(wesm): verification?
        tf = float_frame.astype(np.float64)
        casted = tf.astype(np.int64, copy=False)  # noqa

    def test_astype_with_view_mixed_float(self, mixed_float_frame):

        tf = mixed_float_frame.reindex(columns=["A", "B", "C"])

        casted = tf.astype(np.int64)
        casted = tf.astype(np.float32)  # noqa

    @pytest.mark.parametrize("dtype", [np.int32, np.int64])
    @pytest.mark.parametrize("val", [np.nan, np.inf])
    def test_astype_cast_nan_inf_int(self, val, dtype):
        # see GH#14265
        #
        # Check NaN and inf --> raise error when converting to int.
        msg = "Cannot convert non-finite values \\(NA or inf\\) to integer"
        df = DataFrame([val])

        with pytest.raises(ValueError, match=msg):
            df.astype(dtype)

    def test_astype_str(self):
        # see GH#9757
        a = Series(date_range("2010-01-04", periods=5))
        b = Series(date_range("3/6/2012 00:00", periods=5, tz="US/Eastern"))
        c = Series([Timedelta(x, unit="d") for x in range(5)])
        d = Series(range(5))
        e = Series([0.0, 0.2, 0.4, 0.6, 0.8])

        df = DataFrame({"a": a, "b": b, "c": c, "d": d, "e": e})

        # Datetime-like
        result = df.astype(str)

        expected = DataFrame(
            {
                "a": list(map(str, map(lambda x: Timestamp(x)._date_repr, a._values))),
                "b": list(map(str, map(Timestamp, b._values))),
                "c": list(map(lambda x: Timedelta(x)._repr_base(), c._values)),
                "d": list(map(str, d._values)),
                "e": list(map(str, e._values)),
            }
        )

        tm.assert_frame_equal(result, expected)

    def test_astype_str_float(self):
        # see GH#11302
        result = DataFrame([np.NaN]).astype(str)
        expected = DataFrame(["nan"])

        tm.assert_frame_equal(result, expected)
        result = DataFrame([1.12345678901234567890]).astype(str)

        val = "1.1234567890123457"
        expected = DataFrame([val])
        tm.assert_frame_equal(result, expected)

    @pytest.mark.parametrize("dtype_class", [dict, Series])
    def test_astype_dict_like(self, dtype_class):
        # GH7271 & GH16717
        a = Series(date_range("2010-01-04", periods=5))
        b = Series(range(5))
        c = Series([0.0, 0.2, 0.4, 0.6, 0.8])
        d = Series(["1.0", "2", "3.14", "4", "5.4"])
        df = DataFrame({"a": a, "b": b, "c": c, "d": d})
        original = df.copy(deep=True)

        # change type of a subset of columns
        dt1 = dtype_class({"b": "str", "d": "float32"})
        result = df.astype(dt1)
        expected = DataFrame(
            {
                "a": a,
                "b": Series(["0", "1", "2", "3", "4"]),
                "c": c,
                "d": Series([1.0, 2.0, 3.14, 4.0, 5.4], dtype="float32"),
            }
        )
        tm.assert_frame_equal(result, expected)
        tm.assert_frame_equal(df, original)

        dt2 = dtype_class({"b": np.float32, "c": "float32", "d": np.float64})
        result = df.astype(dt2)
        expected = DataFrame(
            {
                "a": a,
                "b": Series([0.0, 1.0, 2.0, 3.0, 4.0], dtype="float32"),
                "c": Series([0.0, 0.2, 0.4, 0.6, 0.8], dtype="float32"),
                "d": Series([1.0, 2.0, 3.14, 4.0, 5.4], dtype="float64"),
            }
        )
        tm.assert_frame_equal(result, expected)
        tm.assert_frame_equal(df, original)

        # change all columns
        dt3 = dtype_class({"a": str, "b": str, "c": str, "d": str})
        tm.assert_frame_equal(df.astype(dt3), df.astype(str))
        tm.assert_frame_equal(df, original)

        # error should be raised when using something other than column labels
        # in the keys of the dtype dict
        dt4 = dtype_class({"b": str, 2: str})
        dt5 = dtype_class({"e": str})
        msg = "Only a column name can be used for the key in a dtype mappings argument"
        with pytest.raises(KeyError, match=msg):
            df.astype(dt4)
        with pytest.raises(KeyError, match=msg):
            df.astype(dt5)
        tm.assert_frame_equal(df, original)

        # if the dtypes provided are the same as the original dtypes, the
        # resulting DataFrame should be the same as the original DataFrame
        dt6 = dtype_class({col: df[col].dtype for col in df.columns})
        equiv = df.astype(dt6)
        tm.assert_frame_equal(df, equiv)
        tm.assert_frame_equal(df, original)

        # GH#16717
        # if dtypes provided is empty, the resulting DataFrame
        # should be the same as the original DataFrame
        dt7 = dtype_class({}) if dtype_class is dict else dtype_class({}, dtype=object)
        equiv = df.astype(dt7)
        tm.assert_frame_equal(df, equiv)
        tm.assert_frame_equal(df, original)

    def test_astype_duplicate_col(self):
        a1 = Series([1, 2, 3, 4, 5], name="a")
        b = Series([0.1, 0.2, 0.4, 0.6, 0.8], name="b")
        a2 = Series([0, 1, 2, 3, 4], name="a")
        df = concat([a1, b, a2], axis=1)

        result = df.astype(str)
        a1_str = Series(["1", "2", "3", "4", "5"], dtype="str", name="a")
        b_str = Series(["0.1", "0.2", "0.4", "0.6", "0.8"], dtype=str, name="b")
        a2_str = Series(["0", "1", "2", "3", "4"], dtype="str", name="a")
        expected = concat([a1_str, b_str, a2_str], axis=1)
        tm.assert_frame_equal(result, expected)

        result = df.astype({"a": "str"})
        expected = concat([a1_str, b, a2_str], axis=1)
        tm.assert_frame_equal(result, expected)

    @pytest.mark.parametrize(
        "dtype",
        [
            "category",
            CategoricalDtype(),
            CategoricalDtype(ordered=True),
            CategoricalDtype(ordered=False),
            CategoricalDtype(categories=list("abcdef")),
            CategoricalDtype(categories=list("edba"), ordered=False),
            CategoricalDtype(categories=list("edcb"), ordered=True),
        ],
        ids=repr,
    )
    def test_astype_categorical(self, dtype):
        # GH#18099
        d = {"A": list("abbc"), "B": list("bccd"), "C": list("cdde")}
        df = DataFrame(d)
        result = df.astype(dtype)
        expected = DataFrame({k: Categorical(d[k], dtype=dtype) for k in d})
        tm.assert_frame_equal(result, expected)

    @pytest.mark.parametrize("cls", [CategoricalDtype, DatetimeTZDtype, IntervalDtype])
    def test_astype_categoricaldtype_class_raises(self, cls):
        df = DataFrame({"A": ["a", "a", "b", "c"]})
        xpr = f"Expected an instance of {cls.__name__}"
        with pytest.raises(TypeError, match=xpr):
            df.astype({"A": cls})

        with pytest.raises(TypeError, match=xpr):
            df["A"].astype(cls)

    @pytest.mark.parametrize("dtype", ["Int64", "Int32", "Int16"])
    def test_astype_extension_dtypes(self, dtype):
        # GH#22578
        df = DataFrame([[1.0, 2.0], [3.0, 4.0], [5.0, 6.0]], columns=["a", "b"])

        expected1 = DataFrame(
            {
                "a": integer_array([1, 3, 5], dtype=dtype),
                "b": integer_array([2, 4, 6], dtype=dtype),
            }
        )
        tm.assert_frame_equal(df.astype(dtype), expected1)
        tm.assert_frame_equal(df.astype("int64").astype(dtype), expected1)
        tm.assert_frame_equal(df.astype(dtype).astype("float64"), df)

        df = DataFrame([[1.0, 2.0], [3.0, 4.0], [5.0, 6.0]], columns=["a", "b"])
        df["b"] = df["b"].astype(dtype)
        expected2 = DataFrame(
            {"a": [1.0, 3.0, 5.0], "b": integer_array([2, 4, 6], dtype=dtype)}
        )
        tm.assert_frame_equal(df, expected2)

        tm.assert_frame_equal(df.astype(dtype), expected1)
        tm.assert_frame_equal(df.astype("int64").astype(dtype), expected1)

    @pytest.mark.parametrize("dtype", ["Int64", "Int32", "Int16"])
    def test_astype_extension_dtypes_1d(self, dtype):
        # GH#22578
        df = DataFrame({"a": [1.0, 2.0, 3.0]})

        expected1 = DataFrame({"a": integer_array([1, 2, 3], dtype=dtype)})
        tm.assert_frame_equal(df.astype(dtype), expected1)
        tm.assert_frame_equal(df.astype("int64").astype(dtype), expected1)

        df = DataFrame({"a": [1.0, 2.0, 3.0]})
        df["a"] = df["a"].astype(dtype)
        expected2 = DataFrame({"a": integer_array([1, 2, 3], dtype=dtype)})
        tm.assert_frame_equal(df, expected2)

        tm.assert_frame_equal(df.astype(dtype), expected1)
        tm.assert_frame_equal(df.astype("int64").astype(dtype), expected1)

    @pytest.mark.parametrize("dtype", ["category", "Int64"])
    def test_astype_extension_dtypes_duplicate_col(self, dtype):
        # GH#24704
        a1 = Series([0, np.nan, 4], name="a")
        a2 = Series([np.nan, 3, 5], name="a")
        df = concat([a1, a2], axis=1)

        result = df.astype(dtype)
        expected = concat([a1.astype(dtype), a2.astype(dtype)], axis=1)
        tm.assert_frame_equal(result, expected)

    @pytest.mark.parametrize(
        "dtype", [{100: "float64", 200: "uint64"}, "category", "float64"]
    )
    def test_astype_column_metadata(self, dtype):
        # GH#19920
        columns = UInt64Index([100, 200, 300], name="foo")
        df = DataFrame(np.arange(15).reshape(5, 3), columns=columns)
        df = df.astype(dtype)
        tm.assert_index_equal(df.columns, columns)

    @pytest.mark.parametrize("dtype", ["M8", "m8"])
    @pytest.mark.parametrize("unit", ["ns", "us", "ms", "s", "h", "m", "D"])
    def test_astype_from_datetimelike_to_object(self, dtype, unit):
        # tests astype to object dtype
        # GH#19223 / GH#12425
        dtype = f"{dtype}[{unit}]"
        arr = np.array([[1, 2, 3]], dtype=dtype)
        df = DataFrame(arr)
        result = df.astype(object)
        assert (result.dtypes == object).all()

        if dtype.startswith("M8"):
            assert result.iloc[0, 0] == Timestamp(1, unit=unit)
        else:
            assert result.iloc[0, 0] == Timedelta(1, unit=unit)

    @pytest.mark.parametrize("arr_dtype", [np.int64, np.float64])
    @pytest.mark.parametrize("dtype", ["M8", "m8"])
    @pytest.mark.parametrize("unit", ["ns", "us", "ms", "s", "h", "m", "D"])
    def test_astype_to_datetimelike_unit(self, arr_dtype, dtype, unit):
        # tests all units from numeric origination
        # GH#19223 / GH#12425
        dtype = f"{dtype}[{unit}]"
        arr = np.array([[1, 2, 3]], dtype=arr_dtype)
        df = DataFrame(arr)
        result = df.astype(dtype)
        expected = DataFrame(arr.astype(dtype))

        tm.assert_frame_equal(result, expected)

    @pytest.mark.parametrize("unit", ["ns", "us", "ms", "s", "h", "m", "D"])
    def test_astype_to_datetime_unit(self, unit):
        # tests all units from datetime origination
        # GH#19223
        dtype = f"M8[{unit}]"
        arr = np.array([[1, 2, 3]], dtype=dtype)
        df = DataFrame(arr)
        result = df.astype(dtype)
        expected = DataFrame(arr.astype(dtype))

        tm.assert_frame_equal(result, expected)

    @pytest.mark.parametrize("unit", ["ns"])
    def test_astype_to_timedelta_unit_ns(self, unit):
        # preserver the timedelta conversion
        # GH#19223
        dtype = f"m8[{unit}]"
        arr = np.array([[1, 2, 3]], dtype=dtype)
        df = DataFrame(arr)
        result = df.astype(dtype)
        expected = DataFrame(arr.astype(dtype))

        tm.assert_frame_equal(result, expected)

    @pytest.mark.parametrize("unit", ["us", "ms", "s", "h", "m", "D"])
    def test_astype_to_timedelta_unit(self, unit):
        # coerce to float
        # GH#19223
        dtype = f"m8[{unit}]"
        arr = np.array([[1, 2, 3]], dtype=dtype)
        df = DataFrame(arr)
        result = df.astype(dtype)
        expected = DataFrame(df.values.astype(dtype).astype(float))

        tm.assert_frame_equal(result, expected)

    @pytest.mark.parametrize("unit", ["ns", "us", "ms", "s", "h", "m", "D"])
    def test_astype_to_incorrect_datetimelike(self, unit):
        # trying to astype a m to a M, or vice-versa
        # GH#19224
        dtype = f"M8[{unit}]"
        other = f"m8[{unit}]"

        df = DataFrame(np.array([[1, 2, 3]], dtype=dtype))
        msg = (
            fr"cannot astype a datetimelike from \[datetime64\[ns\]\] to "
            fr"\[timedelta64\[{unit}\]\]"
        )
        with pytest.raises(TypeError, match=msg):
            df.astype(other)

        msg = (
            fr"cannot astype a timedelta from \[timedelta64\[ns\]\] to "
            fr"\[datetime64\[{unit}\]\]"
        )
        df = DataFrame(np.array([[1, 2, 3]], dtype=other))
        with pytest.raises(TypeError, match=msg):
            df.astype(dtype)

    def test_astype_arg_for_errors(self):
        # GH#14878

        df = DataFrame([1, 2, 3])

        msg = (
            "Expected value of kwarg 'errors' to be one of "
            "['raise', 'ignore']. Supplied value is 'True'"
        )
        with pytest.raises(ValueError, match=re.escape(msg)):
            df.astype(np.float64, errors=True)

        df.astype(np.int8, errors="ignore")

    def test_astype_arg_for_errors_dictlist(self):
        # GH#25905
        df = DataFrame(
            [
                {"a": "1", "b": "16.5%", "c": "test"},
                {"a": "2.2", "b": "15.3", "c": "another_test"},
            ]
        )
        expected = DataFrame(
            [
                {"a": 1.0, "b": "16.5%", "c": "test"},
                {"a": 2.2, "b": "15.3", "c": "another_test"},
            ]
        )
        type_dict = {"a": "float64", "b": "float64", "c": "object"}

        result = df.astype(dtype=type_dict, errors="ignore")

        tm.assert_frame_equal(result, expected)

    def test_astype_dt64tz(self, timezone_frame):
        # astype
        expected = np.array(
            [
                [
                    Timestamp("2013-01-01 00:00:00"),
                    Timestamp("2013-01-02 00:00:00"),
                    Timestamp("2013-01-03 00:00:00"),
                ],
                [
                    Timestamp("2013-01-01 00:00:00-0500", tz="US/Eastern"),
                    NaT,
                    Timestamp("2013-01-03 00:00:00-0500", tz="US/Eastern"),
                ],
                [
                    Timestamp("2013-01-01 00:00:00+0100", tz="CET"),
                    NaT,
                    Timestamp("2013-01-03 00:00:00+0100", tz="CET"),
                ],
            ],
            dtype=object,
        ).T
        expected = DataFrame(
            expected,
            index=timezone_frame.index,
            columns=timezone_frame.columns,
            dtype=object,
        )
        result = timezone_frame.astype(object)
        tm.assert_frame_equal(result, expected)

        result = timezone_frame.astype("datetime64[ns]")
        expected = DataFrame(
            {
                "A": date_range("20130101", periods=3),
                "B": (
                    date_range("20130101", periods=3, tz="US/Eastern")
                    .tz_convert("UTC")
                    .tz_localize(None)
                ),
                "C": (
                    date_range("20130101", periods=3, tz="CET")
                    .tz_convert("UTC")
                    .tz_localize(None)
                ),
            }
        )
        expected.iloc[1, 1] = NaT
        expected.iloc[1, 2] = NaT
        tm.assert_frame_equal(result, expected)

    def test_astype_dt64tz_to_str(self, timezone_frame):
        # str formatting
        result = timezone_frame.astype(str)
        expected = DataFrame(
            [
                [
                    "2013-01-01",
                    "2013-01-01 00:00:00-05:00",
                    "2013-01-01 00:00:00+01:00",
                ],
                ["2013-01-02", "NaT", "NaT"],
                [
                    "2013-01-03",
                    "2013-01-03 00:00:00-05:00",
                    "2013-01-03 00:00:00+01:00",
                ],
            ],
            columns=timezone_frame.columns,
        )
        tm.assert_frame_equal(result, expected)

        with option_context("display.max_columns", 20):
            result = str(timezone_frame)
            assert (
                "0 2013-01-01 2013-01-01 00:00:00-05:00 2013-01-01 00:00:00+01:00"
            ) in result
            assert (
                "1 2013-01-02                       NaT                       NaT"
            ) in result
            assert (
                "2 2013-01-03 2013-01-03 00:00:00-05:00 2013-01-03 00:00:00+01:00"
            ) in result

    def test_astype_empty_dtype_dict(self):
        # issue mentioned further down in the following issue's thread
        # https://github.com/pandas-dev/pandas/issues/33113
        df = DataFrame()
        result = df.astype(dict())
        tm.assert_frame_equal(result, df)
        assert result is not df

    @pytest.mark.parametrize(
        "df",
        [
            DataFrame(Series(["x", "y", "z"], dtype="string")),
            DataFrame(Series(["x", "y", "z"], dtype="category")),
            DataFrame(Series(3 * [Timestamp("2020-01-01", tz="UTC")])),
            DataFrame(Series(3 * [Interval(0, 1)])),
        ],
    )
    @pytest.mark.parametrize("errors", ["raise", "ignore"])
    def test_astype_ignores_errors_for_extension_dtypes(self, df, errors):
        # https://github.com/pandas-dev/pandas/issues/35471
        if errors == "ignore":
            expected = df
            result = df.astype(float, errors=errors)
            tm.assert_frame_equal(result, expected)
        else:
            msg = "(Cannot cast)|(could not convert)"
            with pytest.raises((ValueError, TypeError), match=msg):
                df.astype(float, errors=errors)

    def test_astype_tz_conversion(self):
        # GH 35973
        val = {"tz": date_range("2020-08-30", freq="d", periods=2, tz="Europe/London")}
        df = DataFrame(val)
        result = df.astype({"tz": "datetime64[ns, Europe/Berlin]"})

        expected = df
        expected["tz"] = expected["tz"].dt.tz_convert("Europe/Berlin")
        tm.assert_frame_equal(result, expected)

    @pytest.mark.parametrize("tz", ["UTC", "Europe/Berlin"])
    def test_astype_tz_object_conversion(self, tz):
        # GH 35973
        val = {"tz": date_range("2020-08-30", freq="d", periods=2, tz="Europe/London")}
        expected = DataFrame(val)

        # convert expected to object dtype from other tz str (independently tested)
        result = expected.astype({"tz": f"datetime64[ns, {tz}]"})
        result = result.astype({"tz": "object"})

        # do real test: object dtype to a specified tz, different from construction tz.
        result = result.astype({"tz": "datetime64[ns, Europe/London]"})
        tm.assert_frame_equal(result, expected)