Source code for kedro.extras.datasets.pandas.parquet_dataset

"""``ParquetDataSet`` loads/saves data from/to a Parquet file using an underlying
filesystem (e.g.: local, S3, GCS). It uses pandas to handle the Parquet file.
from copy import deepcopy
from pathlib import Path, PurePosixPath
from typing import Any, Dict

import fsspec
import pandas as pd
import pyarrow as pa
import pyarrow.parquet as pq

from import (

[docs]class ParquetDataSet(AbstractVersionedDataSet): """``ParquetDataSet`` loads/saves data from/to a Parquet file using an underlying filesystem (e.g.: local, S3, GCS). It uses pandas to handle the Parquet file. Example adding a catalog entry with `YAML API <\ 01_data_catalog.html#using-the-data-catalog-with-the-yaml-api>`_: .. code-block:: yaml >>> boats: >>> type: pandas.ParquetDataSet >>> filepath: data/01_raw/boats.parquet >>> load_args: >>> engine: pyarrow >>> use_nullable_dtypes: True >>> save_args: >>> file_scheme: hive >>> has_nulls: False >>> engine: pyarrow >>> >>> trucks: >>> type: pandas.ParquetDataSet >>> filepath: abfs://container/02_intermediate/trucks.parquet >>> credentials: dev_abs >>> load_args: >>> columns: [name, gear, disp, wt] >>> index: name >>> save_args: >>> compression: GZIP >>> partition_on: [name] Example using Python API: :: >>> from kedro.extras.datasets.pandas import ParquetDataSet >>> import pandas as pd >>> >>> data = pd.DataFrame({'col1': [1, 2], 'col2': [4, 5], >>> 'col3': [5, 6]}) >>> >>> # data_set = ParquetDataSet(filepath="gcs://bucket/test.parquet") >>> data_set = ParquetDataSet(filepath="test.parquet") >>> >>> reloaded = data_set.load() >>> assert data.equals(reloaded) """ DEFAULT_LOAD_ARGS = {} # type: Dict[str, Any] DEFAULT_SAVE_ARGS = {} # type: Dict[str, Any] # pylint: disable=too-many-arguments
[docs] def __init__( self, filepath: 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, ) -> None: """Creates a new instance of ``ParquetDataSet`` pointing to a concrete Parquet file on a specific filesystem. Args: filepath: Filepath in POSIX format to a Parquet 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``. It can also be a path to a directory. If the directory is provided then it can be used for reading partitioned parquet files. Note: `http(s)` doesn't support versioning. load_args: Additional options for loading Parquet file(s). Here you can find all available arguments when reading single file: Here you can find all available arguments when reading partitioned datasets: All defaults are preserved. save_args: Additional saving options for `pyarrow.parquet.write_table` and `pyarrow.Table.from_pandas`. Here you can find all available arguments for `write_table()`: The arguments for `from_pandas()` should be passed through a nested key: `from_pandas`. E.g.: `save_args = {"from_pandas": {"preserve_index": False}}` Here you can find all available arguments for `from_pandas()`: version: If specified, should be an instance of ````. 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``), as well as to pass to the filesystem's `open` method through nested keys `open_args_load` and `open_args_save`. Here you can find all available arguments for `open`: All defaults are preserved. """ _fs_args = deepcopy(fs_args) or {} self._fs_open_args_load = _fs_args.pop("open_args_load", {}) _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._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._from_pandas_args = {} # type: Dict[str, Any] # Handle default load and save arguments 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._from_pandas_args.update(save_args.pop("from_pandas", {})) self._save_args.update(save_args)
def _describe(self) -> Dict[str, Any]: return dict( filepath=self._filepath, protocol=self._protocol, load_args=self._load_args, save_args=self._save_args, version=self._version, ) def _load(self) -> pd.DataFrame: load_path = get_filepath_str(self._get_load_path(), self._protocol) if self._fs.isdir(load_path): # It doesn't work at least on S3 if root folder was created manually # data = ( pq.ParquetDataset(load_path, filesystem=self._fs) .read(**self._load_args) .to_pandas() ) else: with, **self._fs_open_args_load) as fs_file: data = pd.read_parquet(fs_file, **self._load_args) return data def _save(self, data: pd.DataFrame) -> None: save_path = get_filepath_str(self._get_save_path(), self._protocol) if Path(save_path).is_dir(): raise DataSetError( f"Saving {self.__class__.__name__} to a directory is not supported." ) if self._protocol not in HTTP_PROTOCOLS: self._fs.makedirs(Path(save_path).parent.as_posix(), exist_ok=True) table = pa.Table.from_pandas(data, **self._from_pandas_args) pq.write_table( table=table, where=save_path, filesystem=self._fs, **self._save_args ) 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 _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)