Perfect, hardly. In fact it is since I'm so im-perfect that I like to have the build failing when a unit test fails. You can still commit code and continue to develop with your failing test in Maven today. My fear with the switch is developers will turn it on and unknowningly commit code that fails.
================================================================= Jeffrey D. Brekke Quad/Graphics [EMAIL PROTECTED] http://www.qg.com > -----Original Message----- > From: Martin van den Bemt [mailto:[EMAIL PROTECTED]] > Sent: Wednesday, May 29, 2002 5:31 PM > To: Turbine Maven Developers List > Subject: RE: Test behavior > > > Extra info ;) > I assume that a dist target also tests the code (never used > dist btw..) > eg the Eclipse project also makes a dist with failed tests... > > Cut & past from a mail of them : > > Integration build 200205291235 is now available for download from > eclipse.org. > <start> > NEW! HPUX-MOTIF drops now available > > Failed Unit Tests: > org.eclipse.jdt.ui.tests > org.eclipse.jdt.ui.tests.refactoring > org.eclipse.update.tests.core > > <end> > > Mvgr, > Martin > > On Wed, 2002-05-29 at 17:51, Martin van den Bemt wrote: > > You are living in a perfect world, my friend ;) > > > > However the code I am working on is already there and I am writing > > testcases for it. Since there isn't any documentation of intended > > behaviour, I just make a test of the classes, with the > behaviour I guess > > is intended. Which leaves almost all tests to fail.. Since I cannot > > change the code in this system, without actually knowing if > everything > > still works, I have to wait for someone else to pick the > failures up and > > make sure it get's ok.. > > Since I don't use maven:test indiviudally (but called > together with some > > other targets from another build file, I don't get a complete run. > > > > > > Mvgr, > > Martin > > > > On Wed, 2002-05-29 at 17:35, Brekke, Jeff wrote: > > > I don't understand what your reasoning here. Can you > explain again why > > > if the unit test is failing you want a successful build? > I try to write > > > tests first, which would always be failing, then make them pass. > > > > > > ================================================================= > > > Jeffrey D. Brekke Quad/Graphics > > > [EMAIL PROTECTED] http://www.qg.com > > > > > > > > > > -----Original Message----- > > > > From: Martin van den Bemt [mailto:[EMAIL PROTECTED]] > > > > Sent: Wednesday, May 29, 2002 10:00 AM > > > > To: Turbine Maven Developers List > > > > Subject: Test behaviour > > > > > > > > > > > > Hi everyone, > > > > > > > > The default test behaviour of maven is to fail the build, > > > > Since I don't want that in some cases, since I am failing tests > > > > intentionally (why?? : I am writing testcases for > something I don't > > > > know the intended behaviour of and the current > behaviour strikes me as > > > > odd or buggy, I will write the test as I see fit, and > ask someone who > > > > knows to get it the it's supposed to be -; hope you > still with me..) > > > > > > > > For this to work I have attached a patch, which allows a > > > > property to be > > > > set in your favorit properties file. Default behaviour is > > > > still to fail > > > > the test though.. > > > > > > > > Have fun.. > > > > > > > > Mvgr, > > > > Martin > > > > > > > > > > > > > > > > > > -- > > > To unsubscribe, e-mail: > <mailto:[EMAIL PROTECTED]> > > > For additional commands, e-mail: > <mailto:[EMAIL PROTECTED]> > > > > > > > > > > > > > > -- > > To unsubscribe, e-mail: > <mailto:[EMAIL PROTECTED]> > > For additional commands, e-mail: > <mailto:[EMAIL PROTECTED]> > > > > > > > > -- > To unsubscribe, e-mail: > <mailto:[EMAIL PROTECTED]> > For additional commands, e-mail: > <mailto:[EMAIL PROTECTED]> > -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
