def store_dataset_from_ddf(
ddf: dd.DataFrame,
store: StoreInput,
dataset_uuid: str,
table: str = SINGLE_TABLE,
secondary_indices: Optional[List[str]] = None,
shuffle: bool = False,
repartition_ratio: Optional[SupportsFloat] = None,
num_buckets: int = 1,
sort_partitions_by: Optional[Union[List[str], str]] = None,
delete_scope: Optional[Iterable[Mapping[str, str]]] = None,
metadata: Optional[Mapping] = None,
df_serializer: Optional[DataFrameSerializer] = None,
metadata_merger: Optional[Callable] = None,
metadata_version: int = DEFAULT_METADATA_VERSION,
partition_on: Optional[List[str]] = None,
bucket_by: Optional[Union[List[str], str]] = None,
overwrite: bool = False,
):
"""
Store a dataset from a dask.dataframe.
"""
partition_on = normalize_arg("partition_on", partition_on)
secondary_indices = normalize_arg("secondary_indices", secondary_indices)
sort_partitions_by = normalize_arg("sort_partitions_by", sort_partitions_by)
bucket_by = normalize_arg("bucket_by", bucket_by)
store = normalize_arg("store", store)
delete_scope = dask.delayed(normalize_arg)("delete_scope", delete_scope)
if table is None:
raise TypeError("The parameter `table` is not optional.")
ds_factory = _ensure_factory(
dataset_uuid=dataset_uuid, store=store, factory=None, load_dataset_metadata=True
)
if not overwrite:
raise_if_dataset_exists(dataset_uuid=dataset_uuid, store=store)
mps = _write_dataframe_partitions(
ddf=ddf,
store=store,
dataset_uuid=dataset_uuid,
table=table,
secondary_indices=secondary_indices,
shuffle=shuffle,
repartition_ratio=repartition_ratio,
num_buckets=num_buckets,
sort_partitions_by=sort_partitions_by,
df_serializer=df_serializer,
metadata_version=metadata_version,
partition_on=partition_on,
bucket_by=bucket_by,
)
return dask.delayed(store_dataset_from_partitions)(
mps,
store=ds_factory.store_factory if ds_factory else store,
dataset_uuid=ds_factory.dataset_uuid if ds_factory else dataset_uuid,
dataset_metadata=metadata,
metadata_merger=metadata_merger,
)
def store_dataframes_as_dataset__iter(
df_generator,
store,
dataset_uuid=None,
metadata=None,
partition_on=None,
df_serializer=None,
overwrite=False,
metadata_storage_format=DEFAULT_METADATA_STORAGE_FORMAT,
metadata_version=DEFAULT_METADATA_VERSION,
secondary_indices=None,
):
"""
Store `pd.DataFrame` s iteratively as a partitioned dataset with multiple tables (files).
Useful for datasets which do not fit into memory.
Parameters
----------
Returns
-------
dataset: kartothek.core.dataset.DatasetMetadata
The stored dataset.
"""
if dataset_uuid is None:
dataset_uuid = gen_uuid()
if not overwrite:
raise_if_dataset_exists(dataset_uuid=dataset_uuid, store=store)
new_partitions = []
for df in df_generator:
mp = parse_input_to_metapartition(df,
metadata_version=metadata_version)
if partition_on:
mp = mp.partition_on(partition_on)
if secondary_indices:
mp = mp.build_indices(secondary_indices)
# Store dataframe, thereby clearing up the dataframe from the `mp` metapartition
mp = mp.store_dataframes(store=store,
dataset_uuid=dataset_uuid,
df_serializer=df_serializer)
# Add `kartothek.io_components.metapartition.MetaPartition` object to list to track partitions
new_partitions.append(mp)
# Store metadata and return `kartothek.DatasetMetadata` object
return store_dataset_from_partitions(
partition_list=new_partitions,
dataset_uuid=dataset_uuid,
store=store,
dataset_metadata=metadata,
metadata_storage_format=metadata_storage_format,
)
def store_delayed_as_dataset(
delayed_tasks: List[Delayed],
store,
dataset_uuid=None,
metadata=None,
df_serializer=None,
overwrite=False,
metadata_merger=None,
metadata_version=naming.DEFAULT_METADATA_VERSION,
partition_on=None,
metadata_storage_format=naming.DEFAULT_METADATA_STORAGE_FORMAT,
table_name: str = SINGLE_TABLE,
secondary_indices=None,
) -> Delayed:
"""
Transform and store a list of dictionaries containing
dataframes to a kartothek dataset in store.
Parameters
----------
"""
store = lazy_store(store)
if dataset_uuid is None:
dataset_uuid = gen_uuid()
if not overwrite:
raise_if_dataset_exists(dataset_uuid=dataset_uuid, store=store)
raise_if_indices_overlap(partition_on, secondary_indices)
input_to_mps = partial(
parse_input_to_metapartition,
metadata_version=metadata_version,
table_name=table_name,
)
mps = map_delayed(input_to_mps, delayed_tasks)
if partition_on:
mps = map_delayed(MetaPartition.partition_on, mps, partition_on=partition_on)
if secondary_indices:
mps = map_delayed(MetaPartition.build_indices, mps, columns=secondary_indices)
mps = map_delayed(
MetaPartition.store_dataframes,
mps,
store=store,
df_serializer=df_serializer,
dataset_uuid=dataset_uuid,
)
return delayed(store_dataset_from_partitions)(
mps,
dataset_uuid=dataset_uuid,
store=store,
dataset_metadata=metadata,
metadata_merger=metadata_merger,
metadata_storage_format=metadata_storage_format,
)
def create_empty_dataset_header(
store,
dataset_uuid,
table_meta,
partition_on=None,
metadata=None,
overwrite=False,
metadata_storage_format=DEFAULT_METADATA_STORAGE_FORMAT,
metadata_version=DEFAULT_METADATA_VERSION,
):
"""
Create an dataset header without any partitions. This may be used in combination
with :func:`~kartothek.io.eager.write_single_partition` to create implicitly partitioned datasets.
.. note::
The created dataset will **always** have explicit_partition==False
.. warning::
This function should only be used in very rare occasions. Usually you're better off using
full end-to-end pipelines.
Parameters
----------
"""
store = lazy_store(store)()
if not overwrite:
raise_if_dataset_exists(dataset_uuid=dataset_uuid, store=store)
for table, schema in table_meta.items():
table_meta[table] = make_meta(schema,
origin=table,
partition_keys=partition_on)
store_schema_metadata(
schema=table_meta[table],
dataset_uuid=dataset_uuid,
store=store,
table=table,
)
dataset_builder = DatasetMetadataBuilder(
uuid=dataset_uuid,
metadata_version=metadata_version,
partition_keys=partition_on,
explicit_partitions=False,
table_meta=table_meta,
)
if metadata:
for key, value in metadata.items():
dataset_builder.add_metadata(key, value)
if metadata_storage_format.lower() == "json":
store.put(*dataset_builder.to_json())
elif metadata_storage_format.lower() == "msgpack":
store.put(*dataset_builder.to_msgpack())
else:
raise ValueError(
"Unknown metadata storage format encountered: {}".format(
metadata_storage_format))
return dataset_builder.to_dataset()
def store_dataframes_as_dataset(
store,
dataset_uuid,
dfs,
metadata=None,
partition_on=None,
df_serializer=None,
overwrite=False,
metadata_storage_format=DEFAULT_METADATA_STORAGE_FORMAT,
metadata_version=DEFAULT_METADATA_VERSION,
):
"""
Utility function to store a list of dataframes as a partitioned dataset with multiple tables (files).
Useful for very small datasets where all data fits into memory.
Parameters
----------
dfs : dict of pd.DataFrame or pd.DataFrame
The dataframe(s) to be stored. If only a single dataframe is passed, it will be stored as the `core` table.
Returns
-------
The stored dataset
"""
if dataset_uuid is None:
dataset_uuid = gen_uuid()
if isinstance(dfs, dict):
dfs = {"data": [(table, df) for table, df in dfs.items()]}
if not overwrite:
raise_if_dataset_exists(dataset_uuid=dataset_uuid, store=store)
mp = parse_input_to_metapartition(dfs, metadata_version)
if partition_on:
mp = MetaPartition.partition_on(mp, partition_on)
mps = mp.store_dataframes(store=store,
dataset_uuid=dataset_uuid,
df_serializer=df_serializer)
return store_dataset_from_partitions(
partition_list=mps,
dataset_uuid=dataset_uuid,
store=store,
dataset_metadata=metadata,
metadata_storage_format=metadata_storage_format,
)
def _store_bag_as_dataset_parallel(
bag: db.Bag,
store: KeyValueStore,
cube: Cube,
ktk_cube_dataset_ids: Iterable[str],
metadata: Optional[Dict[str, Dict[str, Any]]],
existing_datasets,
overwrite: bool = False,
update: bool = False,
delete_scopes=None,
df_serializer: Optional[ParquetSerializer] = None,
) -> db.Bag:
"""
Vendored, simplified and modified version of kartotheks ``store_bag_as_dataset`` which cannot be easily used to
store datasets in parallel (e.g. from a dict).
`delete_scope` is a dictionary mapping the kartothek dataset id to the `delete_scope` of the dataset
(see `update_dataset_from_partitions` for the definition of the single dataset `delete_scope`).
"""
if (not update) and (not overwrite):
for ktk_cube_dataset_id in ktk_cube_dataset_ids:
raise_if_dataset_exists(
dataset_uuid=cube.ktk_dataset_uuid(ktk_cube_dataset_id),
store=store)
mps = bag.map(_multiplex_parse_input_to_metapartition)
# prepare_data_for_ktk already runs `MetaPartition.partition_on` and `MetaPartition.build_indices`, so this is not
# required here anymore
mps = mps.map(_multiplex_store,
store=store,
cube=cube,
df_serializer=df_serializer)
aggregate = partial(
_multiplex_store_dataset_from_partitions_flat,
cube=cube,
existing_datasets=existing_datasets,
metadata=metadata,
store=store,
update=update,
delete_scopes=delete_scopes or {},
)
return mps.reduction(perpartition=list,
aggregate=aggregate,
split_every=False,
out_type=db.Bag)
def _store_bag_as_dataset_parallel(
bag,
store,
cube,
ktk_cube_dataset_ids,
metadata,
existing_datasets,
overwrite=False,
update=False,
):
"""
Vendored, simplified and modified version of kartotheks ``store_bag_as_dataset`` which cannot be easily used to
store datasets in parallel (e.g. from a dict).
"""
if (not update) and (not overwrite):
for ktk_cube_dataset_id in ktk_cube_dataset_ids:
raise_if_dataset_exists(
dataset_uuid=cube.ktk_dataset_uuid(ktk_cube_dataset_id),
store=store)
mps = bag.map(_multiplex_parse_input_to_metapartition)
# prepare_data_for_ktk already runs `MetaPartition.partition_on` and `MetaPartition.build_indices`, so this is not
# required here anymore
mps = mps.map(_multiplex_store, store=store, cube=cube)
aggregate = partial(
_multiplex_store_dataset_from_partitions_flat,
cube=cube,
existing_datasets=existing_datasets,
metadata=metadata,
store=store,
update=update,
)
return mps.reduction(perpartition=list,
aggregate=aggregate,
split_every=False,
out_type=db.Bag)
def store_delayed_as_dataset(
delayed_tasks,
store,
dataset_uuid=None,
metadata=None,
df_serializer=None,
overwrite=False,
metadata_merger=None,
metadata_version=naming.DEFAULT_METADATA_VERSION,
partition_on=None,
metadata_storage_format=naming.DEFAULT_METADATA_STORAGE_FORMAT,
secondary_indices=None,
):
"""
Transform and store a list of dictionaries containing
dataframes to a kartothek dataset in store.
Parameters
----------
delayed_tasks: list of dask.delayed
Every delayed object represents a partition and should be accepted by
:func:`~kartothek.io_components.metapartition.parse_input_to_metapartition`
Returns
-------
A dask.delayed dataset object.
"""
_check_callable(store)
if dataset_uuid is None:
dataset_uuid = gen_uuid()
if not overwrite:
raise_if_dataset_exists(dataset_uuid=dataset_uuid, store=store)
input_to_mps = partial(parse_input_to_metapartition,
metadata_version=metadata_version)
mps = map_delayed(delayed_tasks, input_to_mps)
if partition_on:
mps = map_delayed(mps,
MetaPartition.partition_on,
partition_on=partition_on)
if secondary_indices:
mps = map_delayed(mps,
MetaPartition.build_indices,
columns=secondary_indices)
mps = map_delayed(
mps,
MetaPartition.store_dataframes,
store=store,
df_serializer=df_serializer,
dataset_uuid=dataset_uuid,
)
return delayed(store_dataset_from_partitions)(
mps,
dataset_uuid=dataset_uuid,
store=store,
dataset_metadata=metadata,
metadata_merger=metadata_merger,
metadata_storage_format=metadata_storage_format,
)
def store_bag_as_dataset(
bag,
store,
dataset_uuid=None,
metadata=None,
df_serializer=None,
overwrite=False,
metadata_merger=None,
metadata_version=naming.DEFAULT_METADATA_VERSION,
partition_on=None,
metadata_storage_format=naming.DEFAULT_METADATA_STORAGE_FORMAT,
secondary_indices=None,
):
"""
Transform and store a dask.bag of dictionaries containing
dataframes to a kartothek dataset in store.
This is the dask.bag-equivalent of
:func:`~kartothek.io.dask.delayed.store_delayed_as_dataset`. See there
for more detailed documentation on the different possible input types.
Parameters
----------
bag: dask.bag.Bag
A dask bag containing dictionaries of dataframes or dataframes.
"""
store = lazy_store(store)
if dataset_uuid is None:
dataset_uuid = gen_uuid()
if not overwrite:
raise_if_dataset_exists(dataset_uuid=dataset_uuid, store=store)
raise_if_indices_overlap(partition_on, secondary_indices)
input_to_mps = partial(parse_input_to_metapartition,
metadata_version=metadata_version)
mps = bag.map(input_to_mps)
if partition_on:
mps = mps.map(MetaPartition.partition_on, partition_on=partition_on)
if secondary_indices:
mps = mps.map(MetaPartition.build_indices, columns=secondary_indices)
mps = mps.map(
MetaPartition.store_dataframes,
store=store,
df_serializer=df_serializer,
dataset_uuid=dataset_uuid,
)
aggregate = partial(
_store_dataset_from_partitions_flat,
dataset_uuid=dataset_uuid,
store=store,
dataset_metadata=metadata,
metadata_merger=metadata_merger,
metadata_storage_format=metadata_storage_format,
)
return mps.reduction(perpartition=list,
aggregate=aggregate,
split_every=False)
def test_raise_if_dataset_exists(store_factory, dataset_function):
raise_if_dataset_exists(dataset_uuid="ThisDoesNotExist",
store=store_factory)
with pytest.raises(RuntimeError):
raise_if_dataset_exists(dataset_uuid=dataset_function.uuid,
store=store_factory)
def store_dataset_from_ddf(
ddf: dd.DataFrame,
store: StoreInput,
dataset_uuid: str,
table: str = SINGLE_TABLE,
secondary_indices: Optional[List[str]] = None,
shuffle: bool = False,
repartition_ratio: Optional[SupportsFloat] = None,
num_buckets: int = 1,
sort_partitions_by: Optional[Union[List[str], str]] = None,
metadata: Optional[Mapping] = None,
df_serializer: Optional[DataFrameSerializer] = None,
metadata_merger: Optional[Callable] = None,
metadata_version: int = DEFAULT_METADATA_VERSION,
partition_on: Optional[List[str]] = None,
bucket_by: Optional[Union[List[str], str]] = None,
overwrite: bool = False,
):
"""
Store a dataset from a dask.dataframe.
"""
# normalization done by normalize_args but mypy doesn't recognize this
sort_partitions_by = cast(List[str], sort_partitions_by)
secondary_indices = cast(List[str], secondary_indices)
bucket_by = cast(List[str], bucket_by)
partition_on = cast(List[str], partition_on)
if table is None:
raise TypeError("The parameter `table` is not optional.")
ds_factory = _ensure_factory(dataset_uuid=dataset_uuid,
store=store,
factory=None)
if not overwrite:
raise_if_dataset_exists(dataset_uuid=dataset_uuid, store=store)
mp_ser = _write_dataframe_partitions(
ddf=ddf,
store=ds_factory.store_factory,
dataset_uuid=dataset_uuid,
table=table,
secondary_indices=secondary_indices,
shuffle=shuffle,
repartition_ratio=repartition_ratio,
num_buckets=num_buckets,
sort_partitions_by=sort_partitions_by,
df_serializer=df_serializer,
metadata_version=metadata_version,
partition_on=partition_on,
bucket_by=bucket_by,
)
return mp_ser.reduction(
chunk=_id,
aggregate=_commit_store_from_reduction,
split_every=False,
token="commit-dataset",
meta=object,
aggregate_kwargs={
"store": ds_factory.store_factory,
"dataset_uuid": ds_factory.dataset_uuid,
"dataset_metadata": metadata,
"metadata_merger": metadata_merger,
},
)
