Quoting Jon Loeliger (2017-11-10 23:11:36) > First, this is draconian for no really good reason. Second, it should be > fixed. Third, I would do that except I am stupid and need a clue where > or how to fix this situation so the tests are less draconian. (Can we > get a "less than 0" test instead of "equal to -1"?)
When writing the logic, it was sufficient (so far) for everybody to just know what is the correct value to expect (0, -1, ..). If there really is a situation when an API call can have a random (from the client's view) return value, I can think of two approaches to deal with this... 1.) introduce a special case "expected_retval=None" which just skips verifying the return value (in framework) leaving this up to the caller 2.) allow passing a functor instead of the expected value, which fulfills the role of validator and have the call it, supplying the actual "retval" from API response message as a parameter to the functor to figure out whether it's one of the expected ones or nor ... Thanks, Klement _______________________________________________ vpp-dev mailing list vpp-dev@lists.fd.io https://lists.fd.io/mailman/listinfo/vpp-dev
