On Sep 2, 2008, at 2:04 PM, David Golden wrote:
On Tue, Sep 2, 2008 at 2:24 PM, chromatic <[EMAIL PROTECTED]> wrote:
... but every time I see yet another arcane cantrip to add to my
projects to
work around brokenness in CPAN Testers clients, a little bit more
of my
motivation to care slips away.
You only need one and you'll never need to learn another:
exit if $ENV{AUTOMATED_TESTING};
Which removed the usefulness of those that do testing correctly and
submit useful reports
... not just for CPAN Testers, but for actual users.
Maybe CPAN Testers is too easy a target. Maybe the real blame lies
elsewhere.
CPAN Testers is the lightning rod because it sits at the end of a
horrible antiquated toolchain that is not user friendly and where
every author has had the luxury to tell "actual users" about
dependency failures via a free-form console interface.
It's compounded by the original design that has a many-to-many
(testers-to-authors) approach to communication, which makes applying
any sort of quality standards or control next to impossible.
True.
You seem to want to have CPAN Testers only send failure reports when
it's a *real* failure. That requires one of the following:
* Artificial Intelligence (or a sufficiently good regex replacement)
sufficient to read console output and and interpret it to your
criteria
True. However I would contest that the Makefile.PL or Build.PL cannot
be "known" to be a failure of the distribution, so the "Artificial
Intelligence" that you have programmed into CPAN testers is flawed.
The UNKNOWN response was specifically added to CPAN testers to
indicate that it is not known if the distribution passes the test suite.
* A structured way of specifying dependencies and a consistent way to
evaluate whether they are met
Well when distributions that DO specify in META.yml what prerequisites
are needs and the distribution still has fail reports due to "Cannot
locate Foo/Bar.pm" when it was in the META.yml then I consider that a
bug in the testing and/or reporting and providing invalid results to
the very users CPAN testers was intented to help
All the cantrips you hate are just different variations on the second
choice because there's no consensus for a single way to do it.
Maybe it's nostalgia, but does anyone else miss the days when you
could upload
a new distribution to the CPAN and maybe someday get a couple of
"Hey,
thanks!" messages, rather than a whole pile of "Your stupid crappy
software
for jerks is broken, you toad!" messages...
<grin>You'd rather get spammed with all the PASS emails too?</grin>
No, I would rather CPAN testers be useful.
Graham.
