On 5/3/2011 1:01 AM, Jonathan M Davis wrote:
They do not, however, catch everything, and they do not stop Phobos developers from making mistakes.
The mistakes, as I see them, is in not running the test suite before checking in Phobos code.
The master branch of Phobos should always be able to run the entire test suite successfully (even if some tests have to be disabled for that to happen). Code that breaks that should be on side branches.
The problem with the test suite not running successfully is it completely stalls development for everyone else, because they cannot test their new code.
_______________________________________________ phobos mailing list [email protected] http://lists.puremagic.com/mailman/listinfo/phobos
