maxframe 0.1.0b5__cp39-cp39-macosx_11_0_arm64.whl

maxframe/dataframe/datasource/tests/test_datasource.py

@@ -0,0 +1,401 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+from collections import OrderedDict
+
+import numpy as np
+import pandas as pd
+import pytest
+from odps import ODPS
+
+from .... import tensor as mt
+from ....tests.utils import tn
+from ....utils import lazy_import
+from ... import read_odps_query, read_odps_table
+from ...core import DatetimeIndex, Float64Index, IndexValue, Int64Index, MultiIndex
+from ..dataframe import from_pandas as from_pandas_df
+from ..date_range import date_range
+from ..from_tensor import (
+    dataframe_from_1d_tileables,
+    dataframe_from_tensor,
+    series_from_tensor,
+)
+from ..index import from_pandas as from_pandas_index
+from ..index import from_tileable
+from ..read_odps_query import ColumnSchema, _resolve_task_sector
+from ..series import from_pandas as from_pandas_series
+
+ray = lazy_import("ray")
+
+
+def test_from_pandas_dataframe():
+    data = pd.DataFrame(
+        np.random.rand(10, 10), columns=["c" + str(i) for i in range(10)]
+    )
+    df = from_pandas_df(data, chunk_size=4)
+
+    pd.testing.assert_series_equal(df.op.dtypes, data.dtypes)
+    assert isinstance(df.index_value._index_value, IndexValue.RangeIndex)
+    assert df.index_value._index_value._slice == slice(0, 10, 1)
+    assert df.index_value.is_monotonic_increasing is True
+    assert df.index_value.is_monotonic_decreasing is False
+    assert df.index_value.is_unique is True
+    assert df.index_value.min_val == 0
+    assert df.index_value.max_val == 9
+    np.testing.assert_equal(df.columns_value._index_value._data, data.columns.values)
+
+    data2 = data[::2]
+    df2 = from_pandas_df(data2, chunk_size=4)
+
+    pd.testing.assert_series_equal(df.op.dtypes, data2.dtypes)
+    assert isinstance(df2.index_value._index_value, IndexValue.RangeIndex)
+    assert df2.index_value._index_value._slice == slice(0, 10, 2)
+
+
+def test_from_pandas_dataframe_with_multi_index():
+    index = pd.MultiIndex.from_tuples([("k1", "v1")], names=["X", "Y"])
+    data = np.random.randint(0, 100, size=(1, 3))
+    pdf = pd.DataFrame(data, columns=["A", "B", "C"], index=index)
+    df = from_pandas_df(pdf, chunk_size=4)
+    assert isinstance(df.index, MultiIndex)
+    assert df.index.names == ["X", "Y"]
+    assert df.index.name is None
+
+
+def test_from_pandas_series():
+    data = pd.Series(np.random.rand(10), name="a")
+    series = from_pandas_series(data, chunk_size=4)
+
+    assert series.name == data.name
+    assert isinstance(series.index_value._index_value, IndexValue.RangeIndex)
+    assert series.index_value._index_value._slice == slice(0, 10, 1)
+    assert series.index_value.is_monotonic_increasing is True
+    assert series.index_value.is_monotonic_decreasing is False
+    assert series.index_value.is_unique is True
+    assert series.index_value.min_val == 0
+    assert series.index_value.max_val == 9
+
+
+def test_from_pandas_index():
+    data = pd.date_range("2020-1-1", periods=10, name="date")
+    index = from_pandas_index(data, chunk_size=4)
+
+    assert isinstance(index, DatetimeIndex)
+    assert index.name == data.name
+    assert index.dtype == data.dtype
+    assert isinstance(index.index_value.value, IndexValue.DatetimeIndex)
+
+
+def test_from_tileable_index():
+    t = mt.random.rand(10, 4)
+
+    with pytest.raises(ValueError):
+        from_tileable(t)
+
+    pd_df = pd.DataFrame(
+        np.random.rand(10, 4), index=np.arange(10, 0, -1).astype(np.int64)
+    )
+    pd_df.index.name = "ind"
+    df = from_pandas_df(pd_df, chunk_size=6)
+
+    for o in [df, df[0]]:
+        index = o.index
+        assert isinstance(index, Int64Index)
+        assert index.dtype == np.int64
+        assert index.name == pd_df.index.name
+        assert isinstance(index.index_value.value, IndexValue.Int64Index)
+
+    t = mt.random.rand(10, chunk_size=6)
+    index = from_tileable(t, name="new_name")
+
+    assert isinstance(index, Float64Index)
+    assert index.dtype == np.float64
+    assert index.name == "new_name"
+    assert isinstance(index.index_value.value, IndexValue.Float64Index)
+
+
+def test_from_tensor():
+    tensor = mt.random.rand(10, 10, chunk_size=5)
+    df = dataframe_from_tensor(tensor)
+    assert isinstance(df.index_value._index_value, IndexValue.RangeIndex)
+    assert df.dtypes[0] == tensor.dtype
+
+    # test converted from scalar
+    scalar = mt.array(1)
+    np.testing.assert_equal(scalar.ndim, 0)
+    with pytest.raises(TypeError):
+        dataframe_from_tensor(scalar)
+
+    # from tensor with given columns
+    df = dataframe_from_tensor(tensor, columns=list("abcdefghij"))
+    pd.testing.assert_index_equal(df.dtypes.index, pd.Index(list("abcdefghij")))
+
+    # test series from tensor
+    tensor = mt.random.rand(10, chunk_size=4)
+    series = series_from_tensor(tensor, name="a")
+
+    assert series.dtype == tensor.dtype
+    assert series.name == "a"
+    pd.testing.assert_index_equal(series.index_value.to_pandas(), pd.RangeIndex(10))
+
+    d = OrderedDict(
+        [(0, mt.tensor(np.random.rand(4))), (1, mt.tensor(np.random.rand(4)))]
+    )
+    df = dataframe_from_1d_tileables(d)
+    pd.testing.assert_index_equal(df.columns_value.to_pandas(), pd.RangeIndex(2))
+
+    series = series_from_tensor(mt.random.rand(4))
+    pd.testing.assert_index_equal(series.index_value.to_pandas(), pd.RangeIndex(4))
+
+    series = series_from_tensor(mt.random.rand(4), index=[1, 2, 3])
+    pd.testing.assert_index_equal(series.op.index, pd.Index([1, 2, 3]))
+
+    series = series_from_tensor(
+        mt.random.rand(4), index=pd.Index([1, 2, 3], name="my_index")
+    )
+    pd.testing.assert_index_equal(series.op.index, pd.Index([1, 2, 3], name="my_index"))
+    assert series.index_value.name == "my_index"
+
+    with pytest.raises(TypeError):
+        series_from_tensor(mt.ones((10, 10)))
+
+    # index has wrong shape
+    with pytest.raises(ValueError):
+        dataframe_from_tensor(mt.random.rand(4, 3), index=mt.random.rand(5))
+
+    # columns have wrong shape
+    with pytest.raises(ValueError):
+        dataframe_from_tensor(mt.random.rand(4, 3), columns=["a", "b"])
+
+    # index should be 1-d
+    with pytest.raises(ValueError):
+        dataframe_from_tensor(
+            mt.tensor(np.random.rand(3, 2)), index=mt.tensor(np.random.rand(3, 2))
+        )
+
+    # 1-d tensors should have same shape
+    with pytest.raises(ValueError):
+        dataframe_from_1d_tileables(
+            OrderedDict(
+                [(0, mt.tensor(np.random.rand(3))), (1, mt.tensor(np.random.rand(2)))]
+            )
+        )
+
+    # index has wrong shape
+    with pytest.raises(ValueError):
+        dataframe_from_1d_tileables(
+            {0: mt.tensor(np.random.rand(3))}, index=mt.tensor(np.random.rand(2))
+        )
+
+    # columns have wrong shape
+    with pytest.raises(ValueError):
+        dataframe_from_1d_tileables(
+            {0: mt.tensor(np.random.rand(3))}, columns=["a", "b"]
+        )
+
+    # index should be 1-d
+    with pytest.raises(ValueError):
+        series_from_tensor(mt.random.rand(4), index=mt.random.rand(4, 3))
+
+
+def test_from_odps_table():
+    odps_entry = ODPS.from_environments()
+    table_name = tn("test_from_odps_table_src")
+    odps_entry.delete_table(table_name, if_exists=True)
+    test_table = odps_entry.create_table(
+        table_name, "col1 string, col2 bigint, col3 double", lifecycle=1
+    )
+
+    part_table_name = tn("test_parted_from_odps_table_src")
+    odps_entry.delete_table(part_table_name, if_exists=True)
+    test_parted_table = odps_entry.create_table(
+        part_table_name,
+        ("col1 string, col2 bigint, col3 double", "pt string"),
+        lifecycle=1,
+    )
+
+    df = read_odps_table(table_name)
+    assert df.op.table_name == test_table.full_table_name
+    assert df.index_value.name is None
+    assert isinstance(df.index_value.value, IndexValue.RangeIndex)
+    assert df.op.get_columns() == ["col1", "col2", "col3"]
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series(
+            [np.dtype("O"), np.dtype("int64"), np.dtype("float64")],
+            index=["col1", "col2", "col3"],
+        ),
+    )
+
+    with pytest.raises(ValueError):
+        read_odps_table(test_table, columns=["col3", "col4"])
+    with pytest.raises(ValueError):
+        read_odps_table(test_table, index_col="col4")
+    with pytest.raises(ValueError):
+        read_odps_table(test_table, columns=["col1", "col2"], index_col="col2")
+
+    df = read_odps_table(test_table, columns=["Col1", "col2"])
+    assert df.op.table_name == test_table.full_table_name
+    assert df.index_value.name is None
+    assert isinstance(df.index_value.value, IndexValue.RangeIndex)
+    assert df.op.get_columns() == ["col1", "col2"]
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series([np.dtype("O"), np.dtype("int64")], index=["col1", "col2"]),
+    )
+
+    df = read_odps_table(test_table, index_col="col1")
+    assert df.op.table_name == test_table.full_table_name
+    assert df.index_value.name == "col1"
+    assert isinstance(df.index_value.value, IndexValue.Index)
+    assert df.index.dtype == np.dtype("O")
+    assert df.op.get_columns() == ["col2", "col3"]
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series([np.dtype("int64"), np.dtype("float64")], index=["col2", "col3"]),
+    )
+
+    with pytest.raises(ValueError):
+        read_odps_table(test_table, append_partitions=True)
+
+    df = read_odps_table(test_parted_table, append_partitions=True)
+    assert df.op.append_partitions is True
+    assert df.op.get_columns() == ["col1", "col2", "col3", "pt"]
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series(
+            [np.dtype("O"), np.dtype("int64"), np.dtype("float64"), np.dtype("O")],
+            index=["col1", "col2", "col3", "pt"],
+        ),
+    )
+
+    df = read_odps_table(
+        test_parted_table, columns=["col1", "col2", "pt"], partitions="pt=20240103"
+    )
+    assert df.op.append_partitions is True
+    assert df.op.partitions == ["pt=20240103"]
+    assert df.op.get_columns() == ["col1", "col2", "pt"]
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series(
+            [np.dtype("O"), np.dtype("int64"), np.dtype("O")],
+            index=["col1", "col2", "pt"],
+        ),
+    )
+
+    test_table.drop()
+    test_parted_table.drop()
+
+
+def test_from_odps_query():
+    odps_entry = ODPS.from_environments()
+    table1_name = tn("test_from_odps_query_src1")
+    table2_name = tn("test_from_odps_query_src2")
+    odps_entry.delete_table(table1_name, if_exists=True)
+    odps_entry.delete_table(table2_name, if_exists=True)
+    test_table = odps_entry.create_table(
+        table1_name, "col1 string, col2 bigint, col3 double", lifecycle=1
+    )
+    # need some data to produce complicated plans
+    odps_entry.write_table(test_table, [["A", 10, 3.5]])
+    test_table2 = odps_entry.create_table(
+        table2_name, "col1 string, col2 bigint, col3 double", lifecycle=1
+    )
+    odps_entry.write_table(test_table2, [["A", 10, 4.5]])
+
+    with pytest.raises(ValueError) as err_info:
+        read_odps_query(f"CREATE TABLE dummy_table AS SELECT * FROM {table1_name}")
+    assert "instant query" in err_info.value.args[0]
+
+    query1 = f"SELECT * FROM {table1_name} WHERE col1 > 10"
+    df = read_odps_query(query1)
+    assert df.op.query == query1
+    assert df.index_value.name is None
+    assert isinstance(df.index_value.value, IndexValue.RangeIndex)
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series(
+            [np.dtype("O"), np.dtype("int64"), np.dtype("float64")],
+            index=["col1", "col2", "col3"],
+        ),
+    )
+
+    df = read_odps_query(query1, index_col="col1")
+    assert df.op.query == query1
+    assert df.index_value.name == "col1"
+    assert isinstance(df.index_value.value, IndexValue.Index)
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series([np.dtype("int64"), np.dtype("float64")], index=["col2", "col3"]),
+    )
+
+    query2 = (
+        f"SELECT t1.col1, t1.col2, t1.col3 as c31, t2.col3 as c32 "
+        f"FROM {table1_name} t1 "
+        f"INNER JOIN {table2_name} t2 "
+        f"ON t1.col1 = t2.col1 AND t1.col2 = t2.col2"
+    )
+    df = read_odps_query(query2, index_col=["col1", "col2"])
+    assert df.op.query == query2
+    assert df.index_value.names == ["col1", "col2"]
+    assert isinstance(df.index_value.value, IndexValue.MultiIndex)
+    pd.testing.assert_series_equal(
+        df.dtypes,
+        pd.Series([np.dtype("float64"), np.dtype("float64")], index=["c31", "c32"]),
+    )
+
+    test_table.drop()
+    test_table2.drop()
+
+
+def test_date_range():
+    with pytest.raises(TypeError):
+        _ = date_range("2020-1-1", periods="2")
+
+    with pytest.raises(ValueError):
+        _ = date_range("2020-1-1", "2020-1-10", periods=10, freq="D")
+
+    with pytest.raises(ValueError):
+        _ = date_range(pd.NaT, periods=10)
+
+    expected = pd.date_range("2020-1-1", periods=9.0, name="date")
+
+    dr = date_range("2020-1-1", periods=9.0, name="date", chunk_size=3)
+    assert isinstance(dr, DatetimeIndex)
+    assert dr.shape == (9,)
+    assert dr.dtype == expected.dtype
+    assert isinstance(dr.index_value.value, IndexValue.DatetimeIndex)
+    assert dr.index_value.min_val == expected.min()
+    assert dr.index_value.min_val_close is True
+    assert dr.index_value.max_val == expected.max()
+    assert dr.index_value.max_val_close is True
+    assert dr.index_value.is_unique == expected.is_unique
+    assert dr.index_value.is_monotonic_increasing == expected.is_monotonic_increasing
+    assert dr.name == expected.name
+
+
+def test_resolve_task_sector():
+    input_path = os.path.join(os.path.dirname(__file__), "test-data", "task-input.txt")
+    with open(input_path, "r") as f:
+        sector = f.read()
+    actual_sector = _resolve_task_sector("job0", sector)
+
+    assert actual_sector.job_name == "job0"
+    assert actual_sector.task_name == "M1"
+    assert actual_sector.output_target == "Screen"
+    assert len(actual_sector.schema) == 78
+    assert actual_sector.schema[0] == ColumnSchema("unnamed: 0", "bigint", "")
+    assert actual_sector.schema[1] == ColumnSchema("id", "bigint", "id_alias")
+    assert actual_sector.schema[2] == ColumnSchema("listing_url", "string", "")

maxframe/dataframe/datastore/__init__.py

@@ -0,0 +1,26 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from .to_odps import to_odps_table
+
+
+def _install():
+    from ..core import DATAFRAME_TYPE
+
+    for t in DATAFRAME_TYPE:
+        t.to_odps_table = to_odps_table
+
+
+_install()
+del _install

maxframe/dataframe/datastore/core.py

@@ -0,0 +1,19 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from ..operators import DataFrameOperator, DataFrameOperatorMixin
+
+
+class DataFrameDataStore(DataFrameOperator, DataFrameOperatorMixin):
+    pass

maxframe/dataframe/datastore/to_csv.py

@@ -0,0 +1,227 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from ... import opcodes
+from ...serialization.serializables import (
+    AnyField,
+    BoolField,
+    DictField,
+    Int32Field,
+    Int64Field,
+    KeyField,
+    ListField,
+    StringField,
+)
+from ..utils import parse_index
+from .core import DataFrameDataStore
+
+
+class DataFrameToCSV(DataFrameDataStore):
+    _op_type_ = opcodes.TO_CSV
+
+    input = KeyField("input")
+    path = AnyField("path")
+    sep = StringField("sep")
+    na_rep = StringField("na_rep")
+    float_format = StringField("float_format")
+    columns = ListField("columns")
+    header = AnyField("header")
+    index = BoolField("index")
+    index_label = AnyField("index_label")
+    mode = StringField("mode")
+    encoding = StringField("encoding")
+    compression = AnyField("compression")
+    quoting = Int32Field("quoting")
+    quotechar = StringField("quotechar")
+    line_terminator = StringField("line_terminator")
+    chunksize = Int64Field("chunksize")
+    date_format = StringField("date_format")
+    doublequote = BoolField("doublequote")
+    escapechar = StringField("escapechar")
+    decimal = StringField("decimal")
+    storage_options = DictField("storage_options")
+
+    def __init__(self, output_types=None, **kw):
+        super().__init__(_output_types=output_types, **kw)
+
+    @property
+    def one_file(self):
+        # if wildcard in path, write csv into multiple files
+        return "*" not in self.path
+
+    @property
+    def output_stat(self):
+        return self.output_stat
+
+    @property
+    def output_limit(self):
+        return 1 if not self.output_stat else 2
+
+    def _set_inputs(self, inputs):
+        super()._set_inputs(inputs)
+        self._input = self._inputs[0]
+
+    def __call__(self, df):
+        index_value = parse_index(df.index_value.to_pandas()[:0], df)
+        if df.ndim == 2:
+            columns_value = parse_index(
+                df.columns_value.to_pandas()[:0], store_data=True
+            )
+            return self.new_dataframe(
+                [df],
+                shape=(0, 0),
+                dtypes=df.dtypes[:0],
+                index_value=index_value,
+                columns_value=columns_value,
+            )
+        else:
+            return self.new_series(
+                [df], shape=(0,), dtype=df.dtype, index_value=index_value
+            )
+
+
+def to_csv(
+    df,
+    path,
+    sep=",",
+    na_rep="",
+    float_format=None,
+    columns=None,
+    header=True,
+    index=True,
+    index_label=None,
+    mode="w",
+    encoding=None,
+    compression="infer",
+    quoting=None,
+    quotechar='"',
+    line_terminator=None,
+    chunksize=None,
+    date_format=None,
+    doublequote=True,
+    escapechar=None,
+    decimal=".",
+    storage_options=None,
+):
+    r"""
+    Write object to a comma-separated values (csv) file.
+
+    Parameters
+    ----------
+    path : str
+        File path.
+        If path is a string with wildcard e.g. '/to/path/out-*.csv',
+        to_csv will try to write multiple files, for instance,
+        chunk (0, 0) will write data into '/to/path/out-0.csv'.
+        If path is a string without wildcard,
+        all data will be written into a single file.
+    sep : str, default ','
+        String of length 1. Field delimiter for the output file.
+    na_rep : str, default ''
+        Missing data representation.
+    float_format : str, default None
+        Format string for floating point numbers.
+    columns : sequence, optional
+        Columns to write.
+    header : bool or list of str, default True
+        Write out the column names. If a list of strings is given it is
+        assumed to be aliases for the column names.
+    index : bool, default True
+        Write row names (index).
+    index_label : str or sequence, or False, default None
+        Column label for index column(s) if desired. If None is given, and
+        `header` and `index` are True, then the index names are used. A
+        sequence should be given if the object uses MultiIndex. If
+        False do not print fields for index names. Use index_label=False
+        for easier importing in R.
+    mode : str
+        Python write mode, default 'w'.
+    encoding : str, optional
+        A string representing the encoding to use in the output file,
+        defaults to 'utf-8'.
+    compression : str or dict, default 'infer'
+        If str, represents compression mode. If dict, value at 'method' is
+        the compression mode. Compression mode may be any of the following
+        possible values: {'infer', 'gzip', 'bz2', 'zip', 'xz', None}. If
+        compression mode is 'infer' and `path_or_buf` is path-like, then
+        detect compression mode from the following extensions: '.gz',
+        '.bz2', '.zip' or '.xz'. (otherwise no compression). If dict given
+        and mode is 'zip' or inferred as 'zip', other entries passed as
+        additional compression options.
+    quoting : optional constant from csv module
+        Defaults to csv.QUOTE_MINIMAL. If you have set a `float_format`
+        then floats are converted to strings and thus csv.QUOTE_NONNUMERIC
+        will treat them as non-numeric.
+    quotechar : str, default '\"'
+        String of length 1. Character used to quote fields.
+    line_terminator : str, optional
+        The newline character or character sequence to use in the output
+        file. Defaults to `os.linesep`, which depends on the OS in which
+        this method is called ('\n' for linux, '\r\n' for Windows, i.e.).
+    chunksize : int or None
+        Rows to write at a time.
+    date_format : str, default None
+        Format string for datetime objects.
+    doublequote : bool, default True
+        Control quoting of `quotechar` inside a field.
+    escapechar : str, default None
+        String of length 1. Character used to escape `sep` and `quotechar`
+        when appropriate.
+    decimal : str, default '.'
+        Character recognized as decimal separator. E.g. use ',' for
+        European data.
+    Returns
+    -------
+    None or str
+        If path_or_buf is None, returns the resulting csv format as a
+        string. Otherwise returns None.
+
+    See Also
+    --------
+    read_csv : Load a CSV file into a DataFrame.
+
+    Examples
+    --------
+    >>> import maxframe.dataframe as md
+    >>> df = md.DataFrame({'name': ['Raphael', 'Donatello'],
+    ...                    'mask': ['red', 'purple'],
+    ...                    'weapon': ['sai', 'bo staff']})
+    >>> df.to_csv('out.csv', index=False).execute()
+    """
+
+    if mode != "w":  # pragma: no cover
+        raise NotImplementedError("only support to_csv with mode 'w' for now")
+    op = DataFrameToCSV(
+        path=path,
+        sep=sep,
+        na_rep=na_rep,
+        float_format=float_format,
+        columns=columns,
+        header=header,
+        index=index,
+        index_label=index_label,
+        mode=mode,
+        encoding=encoding,
+        compression=compression,
+        quoting=quoting,
+        quotechar=quotechar,
+        line_terminator=line_terminator,
+        chunksize=chunksize,
+        date_format=date_format,
+        doublequote=doublequote,
+        escapechar=escapechar,
+        decimal=decimal,
+        storage_options=storage_options,
+    )
+    return op(df)