On Wednesday 26 September 2007 4:59 pm, Eric Wilhelm wrote: > # from Michael G Schwern > # on Wednesday 26 September 2007 16:40: > > Write a Build.PL with a compatibility Makefile.PL and make sure that > > Apache::Test is in the MakeMaker PREREQ_PM. You appear to be doing > > that so it should work. Is it not? > > Note that the build class is trying to be 'Apache::TestMB', I think this > implies that 'Apache::TestMB' should be in configure_requires in the > META.yml. > > Does CPANPLUS have configure_requires now?
configure_requires would be nice, but what supports it at this point? > Not sure about doing a hand-coded Makefile.PL with Apache::TestMM (still > sounds like a configure_requires thing.) The eval looks like it would > work, but why bother? > > echo \ > 'die "Please install Module::Build and run Build.PL instead.\n"' \ > > Makefile.PL I tell ya, that's -mighty- tempting at this point. What I'd like, though, is to at least stem/slow the number of CPAN testers failures or at least get them turned into something more like "no idea; build environment not suitable for this module". Its all perception... right now I've got people asking me how I can claim that the module works when all they're seeing is CPAN Tester failures. They're not going to go to the trouble of actually reading the failures to find out why they failed, they're just seeing high failure rates and asking questions. -- Graham TerMarsch
