well, i don't see mention of why you couldn't pre-process the data (I also don't see code for that matter : ), but what you could do is strip off the starting space and trailing space first and then feed it to split. You may also want to note that you don't need the /g in the regex for split.

use strict;

my $string = '       this          that           other        ';

$string =~ s/^\s+//;
$string =~ s/\s+$//;

my ($one, $two, $three) = split /\s+/, $string;

print "$one\n$two\n$three\n\n";
[download]

which then does:

/home/jptxs > perl -w stripNsplit
this
that
other

/home/jptxs >
[download]

"sometimes when you make a request for the head you don't
want the big, fat body...don't you go snickering."
                                         -- Nathan Torkington UoP2K a.k.a gnat