I’m afraid I’m not sure what you mean, but you don’t want the whole chain of tests to stop in the middle just because of a single test case failed, do you? I agree that the output can be improved to be more intuitive though.
Apparently PyTest seems to be getting popular. It would be nice for it to be integrated into core library so that we can use it without defining special dependency etc (AFAIUC, your current approach requires a special test runner script). Would you interested in going further down this road (if so I’d definitely like to help)?