Parenthesis grouping into regexes.

[aramisf]

Hi everyone!

I was writing a script to filter ip addresses. And used two (simple and incomplete) regexes:

my $ip4 = "([0-9]{1,3}\.){3}[0-9]{1,3}";

my $ip6 = "([a-f0-9]{4}\:)+?";

my $prefix = "/\d\d";

They aren't (and don't need to be) fully correct. I understand that these regexes match an invalid ips like 999.888.777.666, but the ips I have on the list are ok.

My question resides here:

my $pattern = "^($ip4|$ip6)($prefix)?";

It is intented to catch and ip address, followed or not by a prefix. But the resulting match does not catch the prefix:

open (FILE, "<$some_file") or die "Error: $!\n";

while (<FILE>) {

    next if !~ m[$pattern];

    print "Hey, I found \$1: $1 \$2: $2 \$3: $3\n";
}
close FILE;

I don't understand how could one use the groups inside two or three levels into parenthesis.

You see, the $pattern variable looks this ugly:

^(([0-9]{1,3}\.){3}[0-9]{1,3}|([a-f0-9]{4}\:)+?)(/\d\d)?

Considering this situation, how the matches are set into $1, $2... variables?

Is there a way to set the ip address into $1 and the $prefix into $2 ?

If possible, I'd like to use $ip4 instead of:

"[0-9]{1,3}.[0-9]{1,3}.[0-9]{1,3}.[0-9]{1,3}"

### Update ###

There was a little problem about my english, sorry about that. Where I typed 'mask' I wanted to mean 'prefix' (the '/\d\d' part of an ip address). Maybe this is what Kenosis asked (if I understand the question).