G'day Vincent / p5p, Vincent Pit wrote: > Your t/fork.t test fails on some Solaris systems (see > http://bbbike.radzeit.de/~slaven/cpantestersmatrix.cgi?dist=autodie+1.99). Thanks for the bug report; I've got it on my TODO list for the 2.00 autodie release. This may end up being an author's test, but the Solaris bug (where some resources aren't defined) I should be able to detect and skip on reliably. > Actually, I'm not sure about adding to the core a test that uses a > non-core module. Even if it is skipped when the required module is absent. I agree here. Any test that will *never* get run because it depends upon non-core modules shouldn't be in core; it's just needless bloat. Luckily, it's very easy for me to remove them from my build script that merges my changes. If the current patches are applied they'll just get skipped, so this shouldn't be a show-stopper. Rather than send through patches-on-patches, I'll wait until the current patches are committed, and then generate patches on that. Alternatively, when those patches do get merged, the following files can be removed without harm or loss of functionality: t/lib/autodie/fork.t # Depends on BSD::Resource t/lib/autodie/lex58.t # Depends on IPC::System::Simple t/lib/autodie/system.t # Depends on IPC::System::Simple Cheerio, Paul -- Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/ Director of Training | Ph: +61 3 9354 6001 Perl Training Australia | Fax: +61 3 9354 2681Thread Previous