def test_str_tupleset_fail_unstable(param):
with pytest.raises(
TypeError,
match="which has type {} has an unstable iteration order".format(
type(param).__name__),
):
converter_str_tupleset(param)
def _process_dimension_columns(dimension_columns, cube):
"""
Process and check given dimension columns.
Parameters
----------
dimension_columns: Optional[Iterable[str]]
Dimension columns of the query, may result in projection.
cube: Cube
Cube specification.
Returns
-------
dimension_columns: Tuple[str, ...]
Real dimension columns.
"""
if dimension_columns is None:
return cube.dimension_columns
else:
dimension_columns = converter_str_tupleset(dimension_columns)
missing = set(dimension_columns) - set(cube.dimension_columns)
if missing:
raise ValueError(
"Following dimension columns were requested but are missing from the cube: {missing}".format(
missing=", ".join(sorted(missing))
)
)
if len(dimension_columns) == 0:
raise ValueError("Dimension columns cannot be empty.")
return dimension_columns
def _process_partition_by(partition_by, cube, all_available_columns, indexed_columns):
"""
Process and check given partition-by columns.
Parameters
----------
partition_by: Optional[Iterable[str]]
By which column logical partitions should be formed.
cube: Cube
Cube specification.
all_available_columns: Set[str]
All columns that are available for query.
indexed_columns: Dict[str, Set[str]]
Indexed columns per ktk_cube dataset ID.
Returns
-------
partition_by: Tuple[str, ...]
Real partition-by columns, may be empty.
"""
if partition_by is None:
return []
else:
partition_by = converter_str_tupleset(partition_by)
partition_by_set = set(partition_by)
missing_available = partition_by_set - all_available_columns
if missing_available:
raise ValueError(
"Following partition-by columns were requested but are missing from the cube: {missing}".format(
missing=", ".join(sorted(missing_available))
)
)
missing_indexed = partition_by_set - reduce(
set.union, indexed_columns.values(), set()
)
if missing_indexed:
raise ValueError(
"Following partition-by columns are not indexed and cannot be used: {missing}".format(
missing=", ".join(sorted(missing_indexed))
)
)
return partition_by
def test_str_tupleset_fail_duplicates():
with pytest.raises(ValueError,
match="Tuple-set contains duplicates: foo, foo"):
converter_str_tupleset([b"foo", "foo"])
def test_str_tupleset_ok(param, expected):
actual = converter_str_tupleset(param)
assert isinstance(actual, tuple)
assert actual == expected
assert all(isinstance(x, str) for x in actual)