Slash Boxes
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

use Perl Log In

Log In

[ Create a new account ]

ajt (2546)

  (email not shown publicly)

UK based. Perl, XML/HTTP, SAP, Debian hacker.

  • PerlMonks: ajt []
  • Local LUG: AdamTrickett []
  • Debian Administration: ajt []
  • LinkedIn: drajt []

Journal of ajt (2546)

Wednesday March 31, 2004
04:35 AM

CPAN Testers

[ #18136 ]

This week I uploaded version 0.13 of XML-RSS-Tools to CPAN. Some nice soul ran a test of it for me on their SPARC box, and it failed some tests. The problem is, I don't know what's actually wrong, the test report I got doesn't have enough useful information in it. I know it passes all tests on Linux/5.6.1, Linux/5.8.0 and Windows/5.8.0, with and without XML cataog support. The only problem may be with XML::LibXML & XML::LibXSLT, which periodically change their behaviour - and have tripped me up before.

What I should look into is making the test failures more diagnostic, so I have a better idea of what went wrong, and hence how to fix it.

The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
More | Login | Reply
Loading... please wait.
  • I just started messing around with it on my Sunblade. I'm already hitting make test failures with XML::LibXML, so maybe there's an issue there. I'll try to keep you posted.
    • Ta, as good as XML::LibXML [] & libxml2 [] are as a combination, they are a bit of a moving target - which where I fear my problem lies. Do let me know if you are able to dig deep enough to find out why it's not working, I will then do my best to fix it!

      -- "It's not magic, it's work..."
  • In a case like this, I write back to the tester and politely ask for a rerun of the specific test, either through prove, or with HARNESS_VERBOSE=1.


    • The same tester found a problem with 0.13_6, but never replied to my email. Thankfully in that case I was able to figure out what the problem was without their input. Ironically their system tested and passed version 0.13_7, which as far as I can tell is identical to 0.13 with respect to the test suite and the code in question...

      I think that libxml can return different XML depending upon which version you have, small white space changes are (mostly) valid in XML, but make checking that it's correct a real

      -- "It's not magic, it's work..."