def source(path_or_paths, filesystem=None, partitioning=None, format=None):
"""
Open a (multi-file) data source.
Parameters
----------
path_or_paths : str, pathlib.Path, or list of those
Path to a file or to a directory containing the data files, or
a list of paths.
filesystem : FileSystem, default None
By default will be inferred from the path.
partitioning : Partitioning(Factory), str or list of str
The partitioning scheme specified with the ``partitioning()``
function. A flavor string can be used as shortcut, and with a list of
field names a DirectionaryPartitioning will be inferred.
format : str, default None
Currently only "parquet" is supported.
Returns
-------
DataSource of DataSourceDiscovery
"""
fs, paths_or_selector = _ensure_fs_and_paths(path_or_paths, filesystem)
partitioning = _ensure_partitioning(partitioning)
format = _ensure_format(format or "parquet")
# TODO pass through options
options = FileSystemFactoryOptions()
if isinstance(partitioning, PartitioningFactory):
options.partitioning_factory = partitioning
elif isinstance(partitioning, Partitioning):
options.partitioning = partitioning
return FileSystemSourceFactory(fs, paths_or_selector, format, options)
def source(path_or_paths, filesystem=None, partitioning=None, format=None):
"""
Open a (multi-file) data source.
Parameters
----------
path_or_paths : str, pathlib.Path, or list of those
Path to a file or to a directory containing the data files, or
a list of paths.
filesystem : FileSystem, default None
By default will be inferred from the path.
partitioning : Partitioning(Factory), str or list of str
The partitioning scheme specified with the ``partitioning()``
function. A flavor string can be used as shortcut, and with a list of
field names a DirectionaryPartitioning will be inferred.
format : str
Currently only "parquet" is supported.
Returns
-------
DataSource of DataSourceDiscovery
"""
filesystem, paths_or_selector = _ensure_fs_and_paths(
path_or_paths, filesystem)
partitioning = _ensure_partitioning(partitioning)
format = format or "parquet"
if format == "parquet":
format = ParquetFileFormat()
elif not isinstance(format, FileFormat):
raise ValueError("format '{0}' is not supported".format(format))
# TODO pass through options
options = FileSystemFactoryOptions()
if isinstance(partitioning, PartitioningFactory):
options.partitioning_factory = partitioning
elif isinstance(partitioning, Partitioning):
options.partitioning = partitioning
discovery = FileSystemSourceFactory(filesystem, paths_or_selector, format,
options)
# TODO return Source if a specific schema was passed?
# need to return SourceFactory since `dataset` might need to
# finish the factory with a unified schema
return discovery
def factory(path_or_paths, filesystem=None, partitioning=None,
format=None):
"""
Create a factory which can be used to build a Dataset.
Parameters
----------
path_or_paths : str, pathlib.Path, or list of those
Path to a file or to a directory containing the data files, or
a list of paths.
filesystem : FileSystem, default None
By default will be inferred from the path.
partitioning : Partitioning or PartitioningFactory or str or list of str
The partitioning scheme specified with the ``partitioning()``
function. A flavor string can be used as shortcut, and with a list of
field names a DirectionaryPartitioning will be inferred.
format : str, default None
Currently only "parquet" is supported.
Returns
-------
FileSystemDatasetFactory
"""
if not isinstance(path_or_paths, (list, tuple)):
path_or_paths = [path_or_paths]
partitioning = _ensure_partitioning(partitioning)
format = _ensure_format(format or "parquet")
# TODO pass through options
options = FileSystemFactoryOptions()
if isinstance(partitioning, PartitioningFactory):
options.partitioning_factory = partitioning
elif isinstance(partitioning, Partitioning):
options.partitioning = partitioning
factories = []
for path in path_or_paths:
fs, paths_or_selector = _ensure_fs_and_paths(path, filesystem)
factories.append(FileSystemDatasetFactory(fs, paths_or_selector,
format, options))
if len(factories) == 0:
raise ValueError("Need at least one path")
elif len(factories) == 1:
return factories[0]
else:
return UnionDatasetFactory(factories)
