> I haven't tested it, but the Right solution feels like we should make > the tests not call each other. Either they should be combined into > one test, or the common code should be factored out into a helper > method. What benefit does having the called test give, if the *exact* > same code is being tested elsewhere?
I suppose they could be combined now. When originally written the first test, a copy, passed but second test, the same copy followed by a revert, failed. If I used just used one test then it would have been an XFAIL and the working operation could have broken without triggering an additional FAIL. -- Philip
