def _quick_concat_or_none(dfs, dimension_columns, partition_columns):
dfs = list(dfs)
if dfs:
return quick_concat(
dfs=dfs,
dimension_columns=dimension_columns,
partition_columns=partition_columns,
)
else:
return None
def query_cube(
cube,
store,
conditions=None,
datasets=None,
dimension_columns=None,
partition_by=None,
payload_columns=None,
):
"""
Query cube.
.. note::
In case of ``partition_by=None`` (default case), only a single partition is generated. If this one will be
empty (e.g. due to the provided conditions), an empty list will be returned, and a single-element list
otherwise.
Parameters
----------
cube: Cube
Cube specification.
store: simplekv.KeyValueStore
KV store that preserves the cube.
conditions: Union[None, Condition, Iterable[Condition], Conjunction]
Conditions that should be applied, optional.
datasets: Union[None, Iterable[str], Dict[str, kartothek.core.dataset.DatasetMetadata]]
Datasets to query, must all be part of the cube. May be either the result of :meth:`discover_datasets`, a list
of Ktk_cube dataset ID or ``None`` (in which case auto-discovery will be used).
dimension_columns: Union[None, str, Iterable[str]]
Dimension columns of the query, may result in projection. If not provided, dimension columns from cube
specification will be used.
partition_by: Union[None, str, Iterable[str]]
By which column logical partitions should be formed. If not provided, a single partition will be generated.
payload_columns: Union[None, str, Iterable[str]]
Which columns apart from ``dimension_columns`` and ``partition_by`` should be returned.
Returns
-------
dfs: List[pandas.DataFrame]
List of non-empty DataFrames, order by ``partition_by``. Column of DataFrames is alphabetically ordered. Data
types are provided on best effort (they are restored based on the preserved data, but may be different due to
Pandas NULL-handling, e.g. integer columns may be floats).
"""
intention, _empty, groups = plan_query(
cube=cube,
store=store,
conditions=conditions,
datasets=datasets,
dimension_columns=dimension_columns,
partition_by=partition_by,
payload_columns=payload_columns,
)
dfs = [load_group(group=g, store=store, cube=cube) for g in groups]
dfs = [df for df in dfs if not df.empty]
if not intention.partition_by and (len(dfs) > 0):
dfs = [
quick_concat(
dfs=dfs,
dimension_columns=intention.dimension_columns,
partition_columns=cube.partition_columns,
)
]
return dfs