CSVDataset

CSVDataset loads and saves data to/from CSV files using Polars.

kedro_datasets.polars.CSVDataset

CSVDataset(
    *,
    filepath,
    load_args=None,
    save_args=None,
    version=None,
    credentials=None,
    fs_args=None,
    metadata=None
)

Bases: AbstractVersionedDataset[DataFrame, DataFrame]

CSVDataset loads/saves data from/to a CSV file using an underlying filesystem (e.g.: local, S3, GCS). It uses polars to handle the CSV file.

Examples:

Using the YAML API:

cars:
  type: polars.CSVDataset
  filepath: data/01_raw/company/cars.csv
  load_args:
    sep: ","
    parse_dates: False
  save_args:
    has_header: False
    null_value: "somenullstring"

motorbikes:
  type: polars.CSVDataset
  filepath: s3://your_bucket/data/02_intermediate/company/motorbikes.csv
  credentials: dev_s3

Using the Python API:

>>> import sys
>>>
>>> import polars as pl
>>> import pytest
>>> from kedro_datasets.polars import CSVDataset
>>>
>>> if sys.platform.startswith("win"):
...     pytest.skip("this doctest fails on Windows CI runner")
...
>>> data = pl.DataFrame({"col1": [1, 2], "col2": [4, 5], "col3": [5, 6]})
>>>
>>> dataset = CSVDataset(filepath=tmp_path / "test.csv")
>>> dataset.save(data)
>>> reloaded = dataset.load()
>>> assert data.equals(reloaded)

Parameters:

• filepath (str | PathLike) –

  Filepath as a string or path-like object in POSIX format to a CSV file prefixed with a protocol s3://. If prefix is not provided, file protocol (local filesystem) will be used. The prefix should be any protocol supported by fsspec. Note: http(s) doesn't support versioning.

• load_args (dict[str, Any] | None, default: None ) –

  Polars options for loading CSV files. Here you can find all available arguments: https://pola-rs.github.io/polars/py-polars/html/reference/api/polars.read_csv.html#polars.read_csv All defaults are preserved, but we explicitly use rechunk=True for seaborn compatibility.

• save_args (dict[str, Any] | None, default: None ) –

  Polars options for saving CSV files. Here you can find all available arguments: https://pola-rs.github.io/polars/py-polars/html/reference/api/polars.DataFrame.write_csv.html All defaults are preserved.

• version (Version | None, default: None ) –

  If specified, should be an instance of kedro.io.core.Version. If its load attribute is None, the latest version will be loaded. If its save attribute is None, save version will be autogenerated.

• credentials (dict[str, Any] | None, default: None ) –

  Credentials required to get access to the underlying filesystem. E.g. for GCSFileSystem it should look like {"token": None}.

• fs_args (dict[str, Any] | None, default: None ) –

  Extra arguments to pass into underlying filesystem class constructor (e.g. {"project": "my-project"} for GCSFileSystem). Defaults are preserved, apart from the open_args_save mode which is set to w.

• metadata (dict[str, Any] | None, default: None ) –

  Any arbitrary metadata. This is ignored by Kedro, but may be consumed by users or external plugins.

Source code in kedro_datasets/polars/csv_dataset.py

def __init__(  # noqa: PLR0913
    self,
    *,
    filepath: str | os.PathLike,
    load_args: dict[str, Any] | None = None,
    save_args: dict[str, Any] | None = None,
    version: Version | None = None,
    credentials: dict[str, Any] | None = None,
    fs_args: dict[str, Any] | None = None,
    metadata: dict[str, Any] | None = None,
) -> None:
    """Creates a new instance of ``CSVDataset`` pointing to a concrete CSV file
    on a specific filesystem.

    Args:
        filepath: Filepath as a string or path-like object in POSIX format to a CSV file prefixed with a protocol
            `s3://`.
            If prefix is not provided, `file` protocol (local filesystem)
            will be used.
            The prefix should be any protocol supported by ``fsspec``.
            Note: `http(s)` doesn't support versioning.
        load_args: Polars options for loading CSV files.
            Here you can find all available arguments:
            https://pola-rs.github.io/polars/py-polars/html/reference/api/polars.read_csv.html#polars.read_csv
            All defaults are preserved, but we explicitly use `rechunk=True` for `seaborn`
            compatibility.
        save_args: Polars options for saving CSV files.
            Here you can find all available arguments:
            https://pola-rs.github.io/polars/py-polars/html/reference/api/polars.DataFrame.write_csv.html
            All defaults are preserved.
        version: If specified, should be an instance of
            ``kedro.io.core.Version``. If its ``load`` attribute is
            None, the latest version will be loaded. If its ``save``
            attribute is None, save version will be autogenerated.
        credentials: Credentials required to get access to the underlying filesystem.
            E.g. for ``GCSFileSystem`` it should look like `{"token": None}`.
        fs_args: Extra arguments to pass into underlying filesystem class constructor
            (e.g. `{"project": "my-project"}` for ``GCSFileSystem``).
            Defaults are preserved, apart from the `open_args_save` `mode` which is set to `w`.

        metadata: Any arbitrary metadata.
            This is ignored by Kedro, but may be consumed by users or external plugins.
    """
    _fs_args = deepcopy(fs_args) or {}
    _fs_open_args_load = _fs_args.pop("open_args_load", {})
    _fs_open_args_save = _fs_args.pop("open_args_save", {})
    _credentials = deepcopy(credentials) or {}

    protocol, path = get_protocol_and_path(filepath, version)
    if protocol == "file":
        _fs_args.setdefault("auto_mkdir", True)

    self._protocol = protocol
    self._storage_options = {**_credentials, **_fs_args}
    self._fs = fsspec.filesystem(self._protocol, **self._storage_options)

    self.metadata = metadata

    super().__init__(
        filepath=PurePosixPath(path),
        version=version,
        exists_function=self._fs.exists,
        glob_function=self._fs.glob,
    )

    # Handle default load and save and fs arguments
    self._load_args = {**self.DEFAULT_LOAD_ARGS, **(load_args or {})}
    self._save_args = {**self.DEFAULT_SAVE_ARGS, **(save_args or {})}
    self._fs_open_args_load = {
        **self.DEFAULT_FS_ARGS.get("open_args_load", {}),
        **(_fs_open_args_load or {}),
    }
    self._fs_open_args_save = {
        **self.DEFAULT_FS_ARGS.get("open_args_save", {}),
        **(_fs_open_args_save or {}),
    }

    if "storage_options" in self._save_args or "storage_options" in self._load_args:
        logger.warning(
            "Dropping 'storage_options' for %s, "
            "please specify them under 'fs_args' or 'credentials'.",
            self._filepath,
        )
        self._save_args.pop("storage_options", None)
        self._load_args.pop("storage_options", None)

DEFAULT_FS_ARGS class-attribute instance-attribute

DEFAULT_FS_ARGS = {
    "open_args_save": {"mode": "w", "encoding": "utf-8"}
}

DEFAULT_LOAD_ARGS class-attribute instance-attribute

DEFAULT_LOAD_ARGS = {'rechunk': True}

DEFAULT_SAVE_ARGS class-attribute instance-attribute

DEFAULT_SAVE_ARGS = {}

_fs instance-attribute

_fs = filesystem(_protocol, **(_storage_options))

_fs_open_args_load instance-attribute

_fs_open_args_load = {
    None: get("open_args_load", {}),
    None: _fs_open_args_load or {},
}

_fs_open_args_save instance-attribute

_fs_open_args_save = {
    None: get("open_args_save", {}),
    None: _fs_open_args_save or {},
}

_load_args instance-attribute

_load_args = {
    None: DEFAULT_LOAD_ARGS,
    None: load_args or {},
}

_protocol instance-attribute

_protocol = protocol

_save_args instance-attribute

_save_args = {
    None: DEFAULT_SAVE_ARGS,
    None: save_args or {},
}

_storage_options instance-attribute

_storage_options = {None: _credentials, None: _fs_args}

metadata instance-attribute

metadata = metadata

_describe

_describe()

Source code in kedro_datasets/polars/csv_dataset.py

def _describe(self) -> dict[str, Any]:
    return {
        "filepath": self._filepath,
        "protocol": self._protocol,
        "load_args": self._load_args,
        "save_args": self._save_args,
        "version": self._version,
    }

_exists

_exists()

Source code in kedro_datasets/polars/csv_dataset.py

def _exists(self) -> bool:
    try:
        load_path = get_filepath_str(self._get_load_path(), self._protocol)
    except DatasetError:
        return False

    return self._fs.exists(load_path)

_invalidate_cache

_invalidate_cache()

Invalidate underlying filesystem caches.

Source code in kedro_datasets/polars/csv_dataset.py

def _invalidate_cache(self) -> None:
    """Invalidate underlying filesystem caches."""
    filepath = get_filepath_str(self._filepath, self._protocol)
    self._fs.invalidate_cache(filepath)

_release

_release()

Source code in kedro_datasets/polars/csv_dataset.py

def _release(self) -> None:
    super()._release()
    self._invalidate_cache()

load

load()

Source code in kedro_datasets/polars/csv_dataset.py

def load(self) -> pl.DataFrame:
    load_path = str(self._get_load_path())
    if self._protocol == "file":
        # file:// protocol seems to misbehave on Windows
        # (<urlopen error file not on local host>),
        # so we don't join that back to the filepath;
        # storage_options also don't work with local paths
        return pl.read_csv(load_path, **self._load_args)

    load_path = f"{self._protocol}{PROTOCOL_DELIMITER}{load_path}"
    return pl.read_csv(
        load_path, storage_options=self._storage_options, **self._load_args
    )

preview

preview(nrows=5)

Generate a preview of the dataset with a specified number of rows.

Parameters:

• nrows (int, default: 5 ) –

  The number of rows to include in the preview. Defaults to 5.

Returns:

• dict –

  A dictionary containing the data in a split format.

Source code in kedro_datasets/polars/csv_dataset.py

def preview(self, nrows: int = 5) -> TablePreview:
    """
    Generate a preview of the dataset with a specified number of rows.

    Args:
        nrows: The number of rows to include in the preview. Defaults to 5.

    Returns:
        dict: A dictionary containing the data in a split format.
    """
    # Create a copy so it doesn't contaminate the original dataset
    dataset_copy = self._copy()
    data = dataset_copy.load().limit(nrows if type(nrows) is int else 5)
    data_dict = data.to_pandas().to_dict(orient="split")
    return TablePreview(data_dict)

save

save(data)

Source code in kedro_datasets/polars/csv_dataset.py

def save(self, data: pl.DataFrame) -> None:
    save_path = get_filepath_str(self._get_save_path(), self._protocol)

    with self._fs.open(save_path, **self._fs_open_args_save) as fs_file:
        data.write_csv(file=fs_file, **self._save_args)

    self._invalidate_cache()