On Fri, Jun 26, 2009 at 10:49:38PM -0500, Craig A. Berry wrote: > > 2.) All of the calls to IPC::Cmd::run pass a buffer to capture the > output of the subprocess. But IPC::Cmd::run silently ignores that > buffer and never puts anything in it when it's using system() to do > the subprocess (as it is on VMS). In most cases that just means an > error message is less informative than it might be, but in the case of > _find_prereqs, the output is what's of interest and you don't get any > :-(. So I grabbed the coattails of the ON_WIN32 code that's already > there and just read the prereq data that's in the file. > > I have one remaining test failure: > > not ok 35 - $ENV PERL5_CPANPLUS_IS_EXECUTING set correctly during execution > # Failed test ' $ENV PERL5_CPANPLUS_IS_EXECUTING set correctly > during execution' > # at [-.lib.cpanplus.dist.build.t]02_cpanplus-dist-build.t line 232. > # '[ERROR] [Fri Jun 26 22:21:21 2009] Build.PL failed: > # [ERROR] [Fri Jun 26 22:21:21 2009] Unable to create a new > distribution object for 'Foo::Bar' -- cannot continue' > # doesn't match > '(?-xism:ENV=D0\:\[CRAIG\.BLEAD\.LIB\.CPANPLUS\.DIST\.BUILD\.T\.dummy\-cpanplus\.5_11_0\.build\.Foo\-Bar\-0_01\] > Build\.PL)' > ok 36 - ENV var now unset > 1..36 > # Looks like you failed 1 test of 36. > > I think this is also because output from the executed command is not > captured. Maybe we should just skip that test on VMS, or maybe we > should just use plain old backticks, which I experimented with a bit > and they seem to work fine, but I expect ya'll had some reason for not > doing things that way. In any case, if you have a better suggestion > than just skipping test #35, let me know and I'll take it for a spin. Capturing in a consistent manner is a PITA. Anyways, your patch has been applied as is. I've skipped the env var test on VMS as well. 0.35_02 released to CPAN and a patch to follow. Many thanks, -- Chris Williams aka BinGOs PGP ID 0x4658671F http://www.gumbynet.org.uk ==========================Thread Previous | Thread Next