It should insert into test_cases ... instead.
The 1/0 idiom shouldn't be used for failures that could come from unexpected input but only to catch logic bugs. I wonder how many of these could be check constraints instead.
Also, what it reports as failure is something we routinely accept as success, so it should probably be redesigned.
ref https://github.com/kumc-bmi/i2p-transform/blob/master/Oracle/cdm_transform_tests.sql#L129
It should
insert into test_cases ...instead.The
1/0idiom shouldn't be used for failures that could come from unexpected input but only to catch logic bugs. I wonder how many of these could be check constraints instead.Also, what it reports as failure is something we routinely accept as success, so it should probably be redesigned.
ref https://github.com/kumc-bmi/i2p-transform/blob/master/Oracle/cdm_transform_tests.sql#L129