FileDataset

FileDataset is used to load and save data to files using the Ibis framework.

kedro_datasets.ibis.FileDataset

FileDataset(
    filepath,
    file_format="parquet",
    *,
    table_name=None,
    connection=None,
    load_args=None,
    save_args=None,
    version=None,
    metadata=None
)

Bases: ConnectionMixin, AbstractVersionedDataset[Table, Table]

FileDataset loads/saves data from/to a specified file format.

Examples:

Using the YAML API:

cars:
  type: ibis.FileDataset
  filepath: data/01_raw/company/cars.csv
  file_format: csv
  table_name: cars
  connection:
    backend: duckdb
    database: company.db
  load_args:
    sep: ","
    nullstr: "#NA"
  save_args:
    sep: ","
    nullstr: "#NA"

motorbikes:
  type: ibis.FileDataset
  filepath: s3://your_bucket/data/02_intermediate/company/motorbikes/
  file_format: delta
  table_name: motorbikes
  connection:
    backend: polars

Using the Python API:

>>> import ibis
>>> from kedro_datasets.ibis import FileDataset
>>>
>>> data = ibis.memtable({"col1": [1, 2], "col2": [4, 5], "col3": [5, 6]})
>>>
>>> dataset = FileDataset(
...     filepath=tmp_path / "test.csv",
...     file_format="csv",
...     table_name="test",
...     connection={"backend": "duckdb", "database": tmp_path / "file.db"},
... )
>>> dataset.save(data)
>>> reloaded = dataset.load()
>>> assert data.execute().equals(reloaded.execute())

FileDataset connects to the Ibis backend object constructed from the connection configuration. The backend key provided in the config can be any of the supported backends <https://ibis- project.org/install>. The remaining dictionary entries will be passed as arguments to the underlying connect() method (e.g. ibis.duckdb.connect() <https://ibis-project.org/backends/duckdb #ibis.duckdb.connect>).

The read method corresponding to the given file_format (e.g. read_csv() <https://ibis-project.org/backends/ duckdb#ibis.backends.duckdb.Backend.read_csv>_) is used to load the file with the backend. Note that only the data is loaded; no link to the underlying file exists past FileDataset.load().

Parameters:

• filepath (str) –

  Path to a file to register as a table. Most useful for loading data into your data warehouse (for testing). On save, the backend exports data to the specified path.

• file_format (str, default: 'parquet' ) –

  String specifying the file format for the file. Defaults to writing execution results to a Parquet file.

• table_name (str | None, default: None ) –

  The name to use for the created table (on load).

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

  Configuration for connecting to an Ibis backend. If not provided, connect to DuckDB in in-memory mode.

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

  Additional arguments passed to the Ibis backend's read_{file_format} method.

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

  Additional arguments passed to the Ibis backend's to_{file_format} method.

• 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.

• 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/ibis/file_dataset.py

def __init__(  # noqa: PLR0913
    self,
    filepath: str,
    file_format: str = "parquet",
    *,
    table_name: str | None = None,
    connection: dict[str, Any] | None = None,
    load_args: dict[str, Any] | None = None,
    save_args: dict[str, Any] | None = None,
    version: Version | None = None,
    metadata: dict[str, Any] | None = None,
) -> None:
    """Creates a new ``FileDataset`` pointing to the given filepath.

    ``FileDataset`` connects to the Ibis backend object constructed
    from the connection configuration. The `backend` key provided in
    the config can be any of the `supported backends <https://ibis-\
    project.org/install>`_. The remaining dictionary entries will be
    passed as arguments to the underlying ``connect()`` method (e.g.
    `ibis.duckdb.connect() <https://ibis-project.org/backends/duckdb\
    #ibis.duckdb.connect>`_).

    The read method corresponding to the given ``file_format`` (e.g.
    `read_csv() <https://ibis-project.org/backends/\
    duckdb#ibis.backends.duckdb.Backend.read_csv>`_) is used to load
    the file with the backend. Note that only the data is loaded; no
    link to the underlying file exists past ``FileDataset.load()``.

    Args:
        filepath: Path to a file to register as a table. Most useful
            for loading data into your data warehouse (for testing).
            On save, the backend exports data to the specified path.
        file_format: String specifying the file format for the file.
            Defaults to writing execution results to a Parquet file.
        table_name: The name to use for the created table (on load).
        connection: Configuration for connecting to an Ibis backend.
            If not provided, connect to DuckDB in in-memory mode.
        load_args: Additional arguments passed to the Ibis backend's
            `read_{file_format}` method.
        save_args: Additional arguments passed to the Ibis backend's
            `to_{file_format}` method.
        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.
        metadata: Any arbitrary metadata. This is ignored by Kedro,
            but may be consumed by users or external plugins.
    """
    self._file_format = file_format
    self._table_name = table_name
    self._connection_config = connection or self.DEFAULT_CONNECTION_CONFIG
    self.metadata = metadata

    super().__init__(
        filepath=PurePosixPath(filepath),
        version=version,
        exists_function=lambda filepath: Path(filepath).exists(),
    )

    # Set load and save arguments, overwriting defaults if provided.
    self._load_args = deepcopy(self.DEFAULT_LOAD_ARGS)
    if load_args is not None:
        self._load_args.update(load_args)

    self._save_args = deepcopy(self.DEFAULT_SAVE_ARGS)
    if save_args is not None:
        self._save_args.update(save_args)

DEFAULT_CONNECTION_CONFIG class-attribute

DEFAULT_CONNECTION_CONFIG = {
    "backend": "duckdb",
    "database": ":memory:",
}

DEFAULT_LOAD_ARGS class-attribute

DEFAULT_LOAD_ARGS = {}

DEFAULT_SAVE_ARGS class-attribute

DEFAULT_SAVE_ARGS = {}

_CONNECTION_GROUP class-attribute

_CONNECTION_GROUP = 'ibis'

_connection_config instance-attribute

_connection_config = connection or DEFAULT_CONNECTION_CONFIG

_file_format instance-attribute

_file_format = file_format

_load_args instance-attribute

_load_args = deepcopy(DEFAULT_LOAD_ARGS)

_save_args instance-attribute

_save_args = deepcopy(DEFAULT_SAVE_ARGS)

_table_name instance-attribute

_table_name = table_name

connection property

connection

The Backend instance for the connection configuration.

metadata instance-attribute

metadata = metadata

_connect

_connect()

Source code in kedro_datasets/ibis/file_dataset.py

def _connect(self) -> BaseBackend:
    import ibis  # noqa: PLC0415

    config = deepcopy(self._connection_config)
    backend = getattr(ibis, config.pop("backend"))
    return backend.connect(**config)

_describe

_describe()

Source code in kedro_datasets/ibis/file_dataset.py

def _describe(self) -> dict[str, Any]:
    return {
        "filepath": self._filepath,
        "file_format": self._file_format,
        "table_name": self._table_name,
        "backend": self._connection_config["backend"],
        "load_args": self._load_args,
        "save_args": self._save_args,
        "version": self._version,
    }

_exists

_exists()

Source code in kedro_datasets/ibis/file_dataset.py

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

    return Path(load_path).exists()

load

load()

Source code in kedro_datasets/ibis/file_dataset.py

def load(self) -> ir.Table:
    load_path = self._get_load_path()
    reader = getattr(self.connection, f"read_{self._file_format}")
    return reader(load_path, table_name=self._table_name, **self._load_args)

save

save(data)

Source code in kedro_datasets/ibis/file_dataset.py

def save(self, data: ir.Table) -> None:
    save_path = self._get_save_path()
    Path(save_path).parent.mkdir(parents=True, exist_ok=True)
    writer = getattr(self.connection, f"to_{self._file_format}")
    writer(data, save_path, **self._save_args)