admiralfreezbee has asked for the wisdom of the Perl Monks concerning the following question:

I'm trying to match two kind of lines with one regular expression: 
00:04:23:a5:ce:6d > ff:ff:ff:ff:ff:ff, ethertype IPv6 (0x86dd), length
+ 66: (hlim 255, next-header: UDP (17), length: 12) ::.61701 > ff02::1
+:2:.547: [udp sum ok] dhcp solicit (xid=493907)

[download]
and 
00:04:23:a5:ce:6d > ff:ff:ff:ff:ff:ff, ethertype IPv6 (0x86dd), length
+ 66: ::.61701 > ff02::1:2:.547: [udp sum ok] dhcp solicit (xid=493907
+) (len12, hlim 255)

[download]
The only difference between the two is that the len and hlim differ in place.
I'm using this regexpression:
$line=~/^(\S+) > (\S+), \S+ IPv6 \S+ \S+ \S+ .* \S+ > \S+\.547: .* dhc
+p6 solicit \(xid=(\d+)/

[download]
Can anyone tell me what's wrong with it? Thanks!

Replies are listed 'Best First'.
Re: Help on regular expressions
by ikegami (Patriarch) on Sep 21, 2005 at 14:23 UTC

    Seems to me the following would do the trick:

    $line=~/^(\S+) [ ] > [ ] (\S+), .* IPv6 .* \.547 .* xid=(\d+)/x;

    [download]
[reply]
[d/l]
Re: Help on regular expressions
by goldclaw (Scribe) on Sep 21, 2005 at 14:54 UTC
    I assume the 6 in dhcp6 in your regexp is a typo you introduced while posting. Other than that, did you notice how many fields are between IPv6 and > in your second line? How many fields do you have in your regexp? Otherwise I would also recommend using ".*?" instead of ".*". Unless you really know what you do, that is usually what you intend. (.*? means match a sequence of any charachters, but as few as possible...) gc
[reply]
Re: Help on regular expressions
by izut (Chaplain) on Sep 21, 2005 at 14:29 UTC
    I see you don't need anything between ff:ff:ff:ff:ff:ff and xid, so this works: 
    my $line = "00:04:23:a5:ce:6d > ff:ff:ff:ff:ff:ff, ethertype IPv6 (0x8
+6dd), length 66: ::.61701 > ff02::1:2:.547: [udp sum ok] dhcp solicit
+ (xid=493907) (len12, hlim 255)";
my @match = $line =~ /^(\S+) > (\S+),.*IPv6.*\.547:\.*xid=(\d+)/;
print "@match\n";

    [download]
    Update: You want to match IPv6 and .547, haven't read that.

    Igor S. Lopes - izut
    surrender to perl. your code, your rules.
[reply]
[d/l]
[select]

      Hi,

      You might want to use the \\x trick:

      #!/usr/bin/perl

use strict;
use warnings;

my $data;

$data = '00:04:23:a5:ce:6d > ff:ff:ff:ff:ff:ff, ethertype IPv6 (0x86dd
+), length 66: (hlim 255, next-header: UDP (17), length: 12) ::.61701 
+> ff02::1:2:.547: [udp sum ok] dhcp solicit (xid=493907)';

if ($data =~ /^
    (\S+)\s # i need this one
    \>\s
    (\S+),\s # this one as well
    \S+\sIPv6\s
    \(
        \S+
    \),\s\S+\s\d+\:\s
    \(
        hlim\s\d+,\s
        next-header:\s\S+\s\(\d+\),\s
        length:\s\d+
    \)\s
    (\S+)\s # just as this one
    \>\s
    (\S+)\s # this one
    \[udp\ssum\sok\]\sdhcp\ssolicit\s
    \(
        \S+\=
        (\d+) # and the final number
    \)
        /x) {
    print ">$1<\n";
    print ">$2<\n";
    print ">$3<\n";
    print ">$4<\n";
    print ">$5<\n";
}

      [download]

      It makes the regex a little easier to understand.

      --
      if ( 1 ) { $postman->ring() for (1..2); }
[reply]
[d/l]

Back to Seekers of Perl Wisdom