I have six distributions on the CPAN, yet I've never had a single test fail. I could pat myself on the back, but I do wonder if this is because I didn't write enough tests, skipped corner cases, or in the case of AI::NeuralNet::Simple, I didn't know how to test it (and hence have only 18 tests). A couple of days ago, I finally received a report of a test failure. It was on AI::NeuralNet::Simple. I wasn't surprised because the bulk of the code is written in C, so I naturally assumed that I was going to have some obvious error that a C neophyte like myself had overlooked.
Nope. The module depends Inline::C and a couple of test modules. Somewhere in the modules I use is a dependency on Sub::Uplevel. I haven't tracked down who uses it yet, but I'm kind of bummed out that my perfect (cough) track record was ruined by someone else's module. I don't want to put Sub::Uplevel in my Makefile.PL, so I'll have to get off my rear and find out who left it out of their dependencies.
Ovid, that test was reported from my system. Sorry =D
A similar failure was reported from testing Test::Class (source).
Adrian, the author of Test::Class, suggested that my installation of Test::Exception was hosed, thus making Sub::Uplevel barf. After re-installing Test::Exception his distribution passed (source).
I will try yours again tonight and see if it passes.
Re:Test::Exception...
Ovid on 2004-02-09T18:18:22
What curious timing. I was just about to post my research into this. I just decided to look into this and saw that Sub::Uplevel was coming from Test::Exception, but it's listed in the Makefile.PL. Test::Exception might be hosed, but it may just be a simple case of reinstalling Sub::Uplevel. Clearly you do have Test::Exception installed, otherwise the Sub::Uplevel call wouldn't have been made (though a clean install wouldn't hurt).
And no worries. It was bound to happen sooner or later
:) Re:Test::Exception...
LTjake on 2004-02-10T01:22:53
Writing Makefile for AI::NeuralNet::Simple
PERL_DL_NONLAZY=1/usr/bin/perl5.8.0 "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t
t/10nn_simple....ok
All tests successful.
Files=1, Tests=18, 10 wallclock secs ( 9.27 cusr + 0.51 csys = 9.78 CPU)
