On Tue, Nov 27, 2001 at 02:21:45PM -0500, Rocco Caputo wrote: > On Tue, Nov 27, 2001 at 09:48:43AM -0800, Andrew A. Chen wrote: > > > > Just a few comments on this... there appear to be several warnings on the > > tests below on Kernel.pm, line 1267. After googling for a bit, I found > > this reference to waitpid() on win32 not particularly liking -1 if you want > > a non-blocking wait... I have yet to fully understand Kernel.pm, so this > > may or may not apply, but the original message is at > > http://aspn.activestate.com/ASPN/Mail/Message/perl-win32-users/555975 and > > below are the test results. Cheers. :) > > Dear gods! Must I track child PIDs and poll for them explicitly? > [moaning and wailing and gnashing of teeth]
I just posted a testcase and patch to perl5-porters that makes waitpid(-1,WNOHANG) do the right thing on MSWin32. This should fix some of Wheel::Run's problems on Windows. -- Rocco Caputo / [EMAIL PROTECTED] / poe.perl.org / poe.sourceforge.net
