Re: compat Makefile.PL bombs when Build.PL does exit(0)

2008-12-12 Thread David Golden
At least on Win32, M::B generates Build and then calls pl2bat on it. So there's still a Build file. CPAN calls "perl Build" anyway, I think. (I can't check at the moment.) -- David On Fri, Dec 12, 2008 at 2:11 PM, Ken Williams wrote: > On Thu, Dec 11, 2008 at 10:58 PM, David Golden wrote: >

Re: compat Makefile.PL bombs when Build.PL does exit(0)

2008-12-12 Thread Eric Wilhelm
# from Ken Williams # on Friday 12 December 2008 11:11: >Testing for -r "Build" might be problematic for platforms that have >filename extensions for executables (e.g. .bat on DOS and whatever the >VMS equivalent is). I was going to say Module::Build->build_script(), but my newfound strictures i

Re: compat Makefile.PL bombs when Build.PL does exit(0)

2008-12-12 Thread Ken Williams
On Thu, Dec 11, 2008 at 10:58 PM, David Golden wrote: > > Anyway, here's a simple patch that should "fix" it (note -- untested). It > does change the return value, but I'm not sure if that's important or not. Testing for -r "Build" might be problematic for platforms that have filename extensions

Re: compat Makefile.PL bombs when Build.PL does exit(0)

2008-12-11 Thread David Golden
My guess is that the "bug" might still there, but I think this is a case of user-error, or rather a use-case that Module::Build::Compat shouldn't need to support. (a) they can now disable reporting via CPAN Testers preferences website (b) they can not use a passthrough Makefile at all and just pu

compat Makefile.PL bombs when Build.PL does exit(0)

2008-12-11 Thread Eric Wilhelm
Hi all, I don't have time to triage all of the M::B bugs and we really need to get some of these cleared out. This one looks like it may have been resolved within the last year and possibly been made a non-issue by recent cpantesters improvements. Could I please have a volunteer to look into