"""``GenericDataset`` loads/saves data from/to a data file using an underlying
filesystem (e.g.: local, S3, GCS). It uses polars to handle the
type of read/write target.
"""
import warnings
from copy import deepcopy
from io import BytesIO
from pathlib import PurePosixPath
from typing import Any, Dict
import fsspec
import polars as pl
from kedro.io.core import Version, get_filepath_str, get_protocol_and_path
from kedro_datasets import KedroDeprecationWarning
from kedro_datasets._io import AbstractVersionedDataset, DatasetError
[docs]class GenericDataset(AbstractVersionedDataset[pl.DataFrame, pl.DataFrame]):
"""``polars.GenericDataset`` loads/saves data from/to a data file using an underlying
filesystem (e.g.: local, S3, GCS). It uses polars to handle the dynamically select the
appropriate type of read/write on a best effort basis.
Example usage for the `YAML API <https://kedro.readthedocs.io/en/stable/data/\
data_catalog_yaml_examples.html>`_:
.. code-block:: yaml
cars:
type: polars.GenericDataset
file_format: parquet
filepath: s3://data/01_raw/company/cars.parquet
load_args:
low_memory: True
save_args:
compression: "snappy"
Example using Python API:
::
>>> from kedro_datasets.polars import GenericDataset
>>> import polars as pl
>>>
>>> data = pl.DataFrame({'col1': [1, 2], 'col2': [4, 5],
... 'col3': [5, 6]})
>>>
>>> dataset = GenericDataset(filepath='test.parquet', file_format='parquet')
>>> dataset.save(data)
>>> reloaded = dataset.load()
>>> assert data.frame_equal(reloaded)
"""
DEFAULT_LOAD_ARGS = {} # type: Dict[str, Any]
DEFAULT_SAVE_ARGS = {} # type: Dict[str, Any]
[docs] def __init__( # noqa: PLR0913
self,
filepath: str,
file_format: str,
load_args: Dict[str, Any] = None,
save_args: Dict[str, Any] = None,
version: Version = None,
credentials: Dict[str, Any] = None,
fs_args: Dict[str, Any] = None,
):
"""Creates a new instance of ``GenericDataset`` pointing to a concrete data file
on a specific filesystem. The appropriate polars load/save methods are dynamically
identified by string matching on a best effort basis.
Args:
filepath: Filepath in POSIX format to a file prefixed with a protocol like
`s3://`.
If prefix is not provided, `file` protocol (local filesystem)
will be used.
The prefix should be any protocol supported by ``fsspec``.
Key assumption: The first argument of either load/save method points to
a filepath/buffer/io type location. There are some read/write targets such
as 'clipboard' or 'records' that will fail since they do not take a filepath
like argument.
file_format: String which is used to match the appropriate load/save method on a
best effort basis. For example if 'csv' is passed, the `polars.read_csv` and
`polars.DataFrame.write_csv` methods will be identified. An error will
be raised unless there is at least one matching `read_<file_format>`
or `write_<file_format>`.
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/io.html
All defaults are preserved.
save_args: Polars options for saving files.
Here you can find all available arguments:
https://pola-rs.github.io/polars/py-polars/html/reference/io.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``).
metadata: Any arbitrary metadata.
This is ignored by Kedro, but may be consumed by users or external plugins.
Raises:
DatasetError: Will be raised if at least less than one appropriate read or write
methods are identified.
"""
self._file_format = file_format.lower()
_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)
if protocol == "file":
_fs_args.setdefault("auto_mkdir", True)
self._protocol = protocol
self._fs = fsspec.filesystem(self._protocol, **_credentials, **_fs_args)
super().__init__(
filepath=PurePosixPath(path),
version=version,
exists_function=self._fs.exists,
glob_function=self._fs.glob,
)
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)
_fs_open_args_save.setdefault("mode", "wb")
self._fs_open_args_load = _fs_open_args_load
self._fs_open_args_save = _fs_open_args_save
def _load(self) -> pl.DataFrame:
load_path = get_filepath_str(self._get_load_path(), self._protocol)
load_method = getattr(pl, f"read_{self._file_format}", None)
if not load_method:
raise DatasetError(
f"Unable to retrieve 'polars.read_{self._file_format}' method, please"
" ensure that your "
"'file_format' parameter has been defined correctly as per the Polars"
" API"
" https://pola-rs.github.io/polars/py-polars/html/reference/io.html"
)
with self._fs.open(load_path, **self._fs_open_args_load) as fs_file:
return load_method(fs_file, **self._load_args)
def _save(self, data: pl.DataFrame) -> None:
save_path = get_filepath_str(self._get_save_path(), self._protocol)
save_method = getattr(data, f"write_{self._file_format}", None)
if not save_method:
raise DatasetError(
f"Unable to retrieve 'polars.DataFrame.write_{self._file_format}' "
"method, please "
"ensure that your 'file_format' parameter has been defined correctly as"
" per the Polars API "
"https://pola-rs.github.io/polars/py-polars/html/reference/io.html"
)
buf = BytesIO()
save_method(buf, **self._save_args)
with self._fs.open(save_path, **self._fs_open_args_save) as fs_file:
fs_file.write(buf.getvalue())
self._invalidate_cache()
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)
def _describe(self) -> Dict[str, Any]:
return {
"file_format": self._file_format,
"filepath": self._filepath,
"protocol": self._protocol,
"load_args": self._load_args,
"save_args": self._save_args,
"version": self._version,
}
def _release(self) -> None:
super()._release()
self._invalidate_cache()
def _invalidate_cache(self) -> None:
"""Invalidate underlying filesystem caches."""
filepath = get_filepath_str(self._filepath, self._protocol)
self._fs.invalidate_cache(filepath)
_DEPRECATED_CLASSES = {
"GenericDataSet": GenericDataset,
}
def __getattr__(name):
if name in _DEPRECATED_CLASSES:
alias = _DEPRECATED_CLASSES[name]
warnings.warn(
f"{repr(name)} has been renamed to {repr(alias.__name__)}, "
f"and the alias will be removed in Kedro-Datasets 2.0.0",
KedroDeprecationWarning,
stacklevel=2,
)
return alias
raise AttributeError(f"module {repr(__name__)} has no attribute {repr(name)}")