Yeah, your comments on the unit test ordering make sense. Thanks everybody for your advice! Regarding Simon's comment "btw: it seems you are using 'test/unit'. :-)" I do have one more question. I was originally using this in my test file: class SomeTest < RUNIT::TestCase but noticed that I could use Test::Unit instead and get output that I preferred sometimes (if I don't need to see that stack trace for example): class SomeTest < Test::Unit::TestCase but Simon's comment would seem to indicate that I shouldn't be doing that. If you subclass Test::Unit::TestCase in your test file, then the self.suite methods in RUNIT::TestCase and Test::Unit::TestCase are both called. However, if you subclass RUNIT::TestCase then the self.suite method in Test::Unit::TestCase is never called. I found it strange that the method was never called, so I assumed that subclassing Test::Unit::TestCase was the way to go. I saw several examples online that subclassed RUNIT::TestCase but I thought they might have been old examples or something. So is subclassing RUNIT::TestCase definitely the right thing to do for some reason? Thanks, -Jordan At 04:48 05/06/24 +0900, Jim Freeze wrote: >* Jordan Gilliland <jordan / ce-lab.net> [2005-06-23 21:56:40 +0900]: > > > Is this a bug or is it simply not part of the unit testing methodology to > > have a sequence of tests, assuming some order-independent set of tests > > instead? > >No, not a bug. The Unit Test police make sure that >tests are run in random order. > >If you need ordering, then you must handle it. The easy >way to do this is to put order dependent tests in the >same test_X method, or have that method call the tests, >where you have defined your order dependent tests >in methods that do not start with 'test'. > >-- >Jim Freeze > > > >---- X-Spam-Report ---- > >No, hits=-2.5 required=5.0 > > 0.1 RCVD_BY_IP Received by mail server with no name > -2.6 BAYES_00 BODY: Bayesian spam probability is 0 to 1% > [score: 0.0000]