This has a problem with the multi-digit requirement t-rex revealed to us here, and unfortunately there's no simple way to fix it by the use of  \d+ or any other trick I can think of:

c:\@Work\Perl\monks>perl -wMstrict -le
"printf qq{$_ } for grep {/\d/} split '', q(<1,22>:<5,7>:<333,0>);
"
1 2 2 5 7 3 3 3 0
[download]

Update: Actually, if one were married to split, it could be made to work with a rather more complex regex:

c:\@Work\Perl\monks>perl -wMstrict -MData::Dump -le
"my @d = grep length, split qr{ >:< | [<,>] }xms, q(<1,22>:<5,7>:<333,
+0>);
 dd @d;
"
(1, 22, 5, 7, 333, 0)
[download]

Still fairly simple with split:

#!/usr/bin/perl -l

# http://perlmonks.org/?node_id=1176478

use strict;
use warnings;

my $string = '<1,22>:<5,7>:<333,0>';
my @x = grep /\d/, split /\D+/, $string;
print for @x;
[download]

Even simpler: no need for grep.

my @x = split /\D+/, $string;

Update: tybalt89 and AnomalousMonk are both right, of course. My bad :(

Aha! A short, sweet split pattern++. I never thought of that one. But I still prefer  m{ \d+ }xmsg extraction.

---

Give a man a fish:  <%-{-{-{-<

"...a problem with the multi-digit requirement"

Yes. In the moment i clicked on "create" i noticed it. Believe me ;-)

Best regards, Karl

«The Crux of the Biscuit is the Apostrophe»