I've written an Integration program that produces files from a source set of data and converts it to a target data file. These files contain field/record information that must be validated by our clients using their validation methods.

I plan to create numerous test cases and expect to receive information on how well the data validated so I can tweak my program.

I want to create a test case that I know will not pass their validation, however I need to know if their validation methods are picking up basics such as the file structure and layout.

I am trying to prevent the situation if all my test cases pass, but later on the files were not in the correct format the client needed.

Do you feel this is a good idea and would I get any valid information from it?



that's not a bogus test case, it's a valid test for expected exceptions

  • 3
    Negative test cases are just as good as positive ones. Does it break the way we expect.... – Jon Raynor Sep 1 '11 at 0:40
  • 1
    Or perhaps more accurately "does it work they way we intended instead of breaking?"? – Andy Hunt Sep 1 '11 at 8:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.