def expect_column_value_lengths_to_be_between(self,
column,
min_value=None,
max_value=None,
mostly=None,
result_format=None,
include_config=False,
catch_exceptions=None,
meta=None):
if min_value is None and max_value is None:
return column.withColumn('__success', lit(True))
elif min_value is None:
return column.withColumn(
'__success',
when(length_(column[0]) <= max_value,
lit(True)).otherwise(lit(False)))
elif max_value is None:
return column.withColumn(
'__success',
when(length_(column[0]) >= min_value,
lit(True)).otherwise(lit(False)))
# FIXME: whether the below condition is enforced seems to be somewhat inconsistent
# else:
# if min_value > max_value:
# raise ValueError("minvalue cannot be greater than max_value")
return column.withColumn(
'__success',
when((min_value <= length_(column[0])) &
(length_(column[0]) <= max_value),
lit(True)).otherwise(lit(False)))
def expect_column_value_lengths_to_equal(
self,
column,
value, # int
mostly=None,
result_format=None,
include_config=True,
catch_exceptions=None,
meta=None,
):
return column.withColumn('__success', when(length_(column[0]) == value, lit(True)).otherwise(lit(False)))
