[Breaking this out into a separate subject] On Jan 8, 2008 3:01 PM, Barbie <[EMAIL PROTECTED]> wrote: > On Tue, Jan 08, 2008 at 01:56:48PM -0500, David Golden wrote: > > On Jan 8, 2008 1:11 PM, Barbie <[EMAIL PROTECTED]> wrote: > > > It's mostly to handle exclusions, avoid duplication, prempt NA > > > submissions (a NA prereq will submit a NA in place of a FAIL for the > > > current dist), > > > > I thought we agreed in discussing semantics about NA that > > distributions failing prereqs should not have reports submitted at > > all. (ca. CPAN::Reporter 0.46)
I may be throwing you off when I use the term "failing prerequisites" -- I mean that prerequisites aren't satisfied, not that they get a FAIL grade. > If the prereq has been reported as NA, then if the dist being tested > fails, that means the dist being tested is also NA. The example I had at > the time was trying to test on 5.6.1 where a dist required a module that > required 5.8, this therefore meant the dist being tested also had a NA > filed against it. > > The only time it won't get an NA report is if the prereq is actually > optional and the dist passes. > > Note that a FAIL prereq and a NA prereq are different things. Perhaps this is how the logic has been done in the past, but I don't think it's a good idea. Example: * Foo-1.23 depends on Bar * Bar-3.14 tests NA on Win32, Perl 5.8.8 * Foo-1.23 gets marked NA on Win32, Perl 5.8.8 * Bar-3.15 fixes Win32 problems and tests PASS on Perl 5.8.8 After Bar is fixed, the NA on Foo is wrong. I think test reports should only be based on a distribution's own tests. If prerequisites can't be satisfied -- whether due to failure of tests (FAIL) or due to lack of platform support (NA) -- no report should be sent. The same hold true for prerequisites that can't be installed because of missing libraries. With Devel::CheckLib, missing library dependencies fail silently -- no Makefile, no Build file, no tests and no reports. They just *aren't there*. So while you note that a FAIL prereq and an NA prereq aren't the same thing, that's almost beside the point -- what matters is (a) whether prerequisites are *present* (in @INC) and (b) whether they meet required version specification. FAIL/NA on distributions may or may not have anything to do with the success or failure of subsequent distributions with dependencies. David
