NOTE: use Perl; is on undef hiatus. You can read content, but you can't post it. More info will be forthcoming forthcomingly.
All the Perl that's Practical to Extract and Report
Stories, comments, journals, and other submissions on use Perl; are Copyright 1998-2006, their respective owners.
FAIL feed? (Score:2)
One thing that'd be immensely helpful to me is an RSS feed containing just FAILs. As it stands I regularly get 20-80 reports in my RSS reader, most of which are not interesting (PASS, NA, etc). I just want the failure reports!
Re: (Score:2)
Re: (Score:2)
Also ... (Score:2)
Yes, only getting failure reports for actual test failures would be great. I can live with failures because a broken toolchain didn't install the deps properly. But getting fail reports because the path was too long because they're using Red Hat, or whatever, is just damn annoying.
And one more thing! (Score:2)
Sheesh, I should think and just post once.
Anyway, as an author who does have mostly high quality distros (I think), there are still cases where CPAN testers is really useful.
Recently I shepherded the new Class::MOP and Moose releases. These included a lot of big changes, so we did dev releases specifically to get CPAN tester reports back. These reports were very helpful and we cleaned up quite a few bugs that way. This specific use case was great, and not something I'd really done before.
Thanks! (Score:1)
I just wanted to take the opportunity to thank all the people who help me find bugs in environments I don't have access to.
You're doing a great job!
Thanks & feedback (Score:1)
I'd like to join jplindstrom in thanking all CPANTesters (both frontend and backend) for the good service.
One thing that surprised me a little was this "message avalanche" effect, which I thought was an already-dealt-with problem. In particular, I notice that I only get one single FAIL message for each module that I upload to PAUSE that has problems, even when there are more of them. Am I totally misunderstanding this point in your post, or have I been gifted by some obscure bug^Wfeature?
Re: (Score:1)
Some CPAN Testers have configured their machines not to CC authors at all or not to CC authors for certain grades or PL/make/test phases. I suspect your experience is serendipity, not fundamental design.
-- dagolden
Useful, but what is the bigger picture? (Score:1)
What is the system of which CPAN testers is a part?
I don't think you are standing far enough back.
(Fred Brooks in The Mythical Man-month on documentation: Start off at 50,000 feet and come in really slowly.)
What is the function of CPAN testers?
One size is not going to fit all. For example, Ben Bullock, the person said to have ruined the CPAN ratings feed, may be the kind of developer who needs to be hit with a heavy stick.
http://use.perl.org/~ology/journal/37363 [perl.org]
Admittedly, that can be worked out in the futu