On Wed, May 1, 2013 at 12:57 PM, Manfred Lotz <manfred.l...@arcor.de> wrote:
> On Wed, 1 May 2013 12:00:18 -0700
> Charles DeRykus <dery...@gmail.com> wrote:
>
> > > ...
> > > Thanks for your detailed explanations. I think that close should
> > > work as I cannot see any reason why a failure of a command closes
> > > the pipe prematurely.
> >
> > Actually, everything is ok until the close which checks both the
> > status from reaping the child and the close itself:
> >
>
> >> OK, the issue is that close is more than just close. Looking from this
> >> side it works like designed.
>
> 'perldoc -f close' hints at this although it could be clearer that
>
$? could potentially cause close to
fail:
...
If the filehandle came from a piped open, "close" returns false
if one of the other syscalls involved fails or if its program
exits with non-zero status. If the only problem was that the
program exited non-zero, $! will be set to 0. Closing a pipe
also waits for the process executing on the pipe to exit--in
case you wish to look at the output of the pipe afterwards--and
implicitly puts the exit status value of that command into $?
and "${^CHILD_ERROR_NATIVE}".
--
Charles DeRykus
