Re: Bugs & regression tests

As we discussed in the PDM, this should be pretty much automated. It means a serious upgrade to the already not-very-cutting-edge bugs database that we have. There should be a dedicated field for a reproducing code snippet (if available); When the bug is closed, the person closing the bug would have the option of turning this code snippet into a tester script. If we find a person that has time to rewrite the bugs database, we should also apply some common standard practices, such as normalizing the database. Of course, until all that happens, the QA team can try to do all that manually, but it'd be much better if it was semi-automated. Zeev On Fri, 30 Jun 2000, Andrei Zmievski wrote: > What would also be good I think, is if the QA team wrote a regression > test for each valid bug fix to make sure that bugs don't reappear. > > -Andrei > > "Science is like sex: sometimes something useful comes out, > but that is not the reason we are doing it" -- Richard Feynman > > -- Zeev Suraski <zeev@zend.com> http://www.zend.com/