I'm coming in a little late to this conversation, seems like you've
got the situation well in hand.
I should have anticipated that perlbug might get flooded by this
little project. :(
Anyhow, just a few points-of-order...
On Tue, Feb 20, 2001 at 12:21:44AM -0800, Richard Soderberg wrote:
> On Tue, 20 Feb 2001, H.Merijn Brand wrote:
> > On Mon, 19 Feb 2001 18:52:46 -0500 (EST), [EMAIL PROTECTED] wrote:
> > Is it wrong to presume that people/builders/testers/smokers building
> > bleadperl *ALSO* /read/ p5p?
Most smokers will *not* be active readers of p5p (I don't). Most
people who volunteered expressed that they have plenty of spare
computing power, but not alot of tuits to devote. It cannot be a
requirement that a smoker must skim p5p before sending in their
report. It must be near 100% automated, otherwise it will not work.
This is why failure reports kept streaming in, but it looks like
you're already on the way towards solving it. Just keep in mind that
whatever you come up with must require no regular human intervention.
> Unfortunately, I'm not aware of code that allows me to easily seperate
> known test failures from unexpected test failures barring some interesting
> modifications to the test suite.
The interesting modifications were done a few days ago. :) The test
suite now honors todo tests, expressing a test which is expected to
fail and will be fixed shortly. See t/op/chop.t test #34 for an
example and the latest Test::Harness docs for details.
--
Michael G Schwern <[EMAIL PROTECTED]> http://www.pobox.com/~schwern/
Perl6 Quality Assurance <[EMAIL PROTECTED]> Kwalitee Is Job One
