A lot of the 'test_api' level tests are there to test the actual HTTP response codes wrt different errors. (So have 1 success level and 1 failure level test for each parameter based on different error codes.)
So I think the test coverage is reasonable as it stands, but we did switch things to using a module.
A lot of the 'test_api' level tests are there to test the actual HTTP response codes wrt different errors. (So have 1 success level and 1 failure level test for each parameter based on different error codes.)
So I think the test coverage is reasonable as it stands, but we did switch things to using a module.