On Sun, Oct 11, 2009 at 10:45:04AM +0100, Steve Hay wrote: > I just wanted to point out that yesterday's upgrades to CGI.pm and > Parse-CPAN-Meta both cause test failures, and both failures are > already known about. > > There are existing CPAN RT tickets regarding each. I replied to them > and CCed p5p in the hope of obviating the need for this email, but > neither reply seems to have turned up on p5p. The two tickets are:
It might be in a moderation queue. > http://rt.cpan.org/Public/Bug/Display.html?id=49599 > http://rt.cpan.org/Public/Bug/Display.html?id=47844 I'd already decided to solve the CGI bug by removing t/fast.t from core: http://perl5.git.perl.org/perl.git/commit/440f0407a5bf0dfc Exclude CGI/t/fast.t from the core, as it relies on FCGI, which is not in core. Whilst the test is designed to skip if FCGI is not present, that logic is now broken, and even if it were fixed, the as-implemented skip count is now wrong. We're never going to run this test, as core tests run with just the core's libraries in @INC, so I see no reason to ship it. I think, in general, it is better not to upgrade modules if they're going to cause tests to fail. (And mail the list, to explain why it wasn't possible) than do things that add "known failures" to core. I'm really, really, keen for blead to remain "all tests pass" at all times.* Historically, we've had a big problem with people sending patches against a stable release, even *development* type patches, because they perceive that blead is "unstable" and "not a good thing to hack against". I've not been the only one trying to re-educate people into grabbing a (near) current blead version and basing their work on *that*, but all this relies on that blead version passing tests. If blead doesn't pass tests, it nullifies these efforts. Nicholas Clark * Yes, I know that it's not right now. Kittens are suffering as I type. And I know that the smokers fail on not-that-obscure platforms. There is *nothing* stopping *anyone* on this list from trying to help diagnose what those problems are cause by, and working towards fixing them. It doesn't take a degree in rocket science** ** Please note, I do not have, and did not attempt to convey any impression of having, and form of degree in rocket science. Or any experience of it.
