use IPC::Run qw(run);

my ($out, $err) = ('', '');
run(['/path/to/prog', @args],
    '>',  sub { $out .= $_[0]; print $_[0] },
    '2>', sub { $err .= $_[0]; print $_[0] });
[download]

--
Ilya Martynov, ilya@iponweb.net
CTO IPonWEB (UK) Ltd
Quality Perl Programming and Unix Support UK managed @ offshore prices - http://www.iponweb.net
Personal website - http://martynov.org

This seems to have worked the best of all, after I turned off buffering for STDOUT/STDERR, with code such as this:

my ($ofh) = select(STDOUT); $| = 1; select($ofh);
$ofh = select(STDERR); $| = 1; select($ofh);

my ($out, $err);

run (
    [ '/usr/sbin/pkgrm', 'VRTSvxvm' ],
    '>', sub { $out .= $_[0]; print $_[0] },
    '2>', sub { $err .= $_[0]; print $_[0] },
    );
[download]

Now, if the program I was running demands that it connects to a pty before it will unbuffer STDOUT/STDERR properly, what could I do? I've tried the following, but it doesn't send any output to the screen:

use IPC::Run qw(run);

my ($out, $err);

run (
    [ '/usr/sbin/pkgrm', 'VRTSvxvm' ],
    '>', sub { $out .= $_[0]; print $_[0] },
    '>', '>pty>',
    '2>', sub { $err .= $_[0]; print $_[0] },
    '2>', '>pty>'
    );
[download]

I'm not sure it is correct syntax. From IPC::Run docs:

       '>pty>' means '&>pty>', not '1>pty>'
           The pseudo terminal redirects both stdout and stderr
           unless you specify a file descriptor.  If you want to
           grab stderr separately, do this:

              start \@cmd, '<pty<', \$in, '>pty>', \$out, '2>', \$err 
+;
[download]

So probably your code should look like (untested):

use IPC::Run qw(run);

my ($out, $err);

run (
    [ '/usr/sbin/pkgrm', 'VRTSvxvm' ],
    '>pty>', sub { $out .= $_[0]; print STDOUT $_[0] },
    '2>', sub { $err .= $_[0]; print STDERR $_[0] },
    );
[download]

--
Ilya Martynov, ilya@iponweb.net
CTO IPonWEB (UK) Ltd
Quality Perl Programming and Unix Support UK managed @ offshore prices - http://www.iponweb.net
Personal website - http://martynov.org