pastebin - collaborative debugging

# gdns.pl
# Index all subdomains from a domain without using zone transfer or dns resolving.
# Intended use for this script is to enumerate all subdomains for a given domain.
# Author:  DiabloHorn
# Date: 16-02-2008
# Version: 0.3

#changes
# fixed the - bug
# fixed the hang bug when it reached the end of the google results
#

###TODO###
# implement rotating proxy support

###POSSIBLE BUGS###
# the need to ctrl+c if results are less then 100

use LWP::UserAgent;
use HTML::LinkExtor;
use URI::URL;

$numArgs = $#ARGV + 1;

if($numArgs != 1){
        print "gdns.pl <base domain>\n";
        print "Example: gdns.pl kd-team.com\n";
        print "Should find all subdomains indexed by google.";
        exit(1);
}
#intentionally not using quotemeta, cause it fucks up in the search
$searchitem = $ARGV[0];

#if you change this....change the regexes.
my $baseSEngine = "http://www.google.com";

#start url for searching
$url = URI->new("$baseSEngine/search?hl=en&q=site%3A$searchitem");
$ua = LWP::UserAgent->new;
$ua->agent('Opera/9.20 (Windows NT 6.0; U; en)'); #this should help us a little to fool google.

print "[*] starting subdomain search on $searchitem\n";

#hash containing all found sub domains
my %allurls = ();
#hash containing all the "next" urls from google
my %nexturls = ();

#callback for each request to parse the page
sub callback {
 my($tag, %attr) = @_;
 #for this poc we are only interested in the <a href> tags
 return if $tag ne 'a';
 my @links = values %attr;
 foreach $link(@links){
        #extract all urls that contain the base domain
        if($link =~ m!(^(http://|https://|ftp://|irc://)(([a-zA-Z0-9\-\.]*)(\.+))*$searchitem)!io){
                if (!exists $allurls{$1}){
                        $allurls{$1} = $1;
                        print "$1\n";
                }
        }

        #extract the google next urls
        if($link =~ m!/search\?q=site:$searchitem&hl=\w+&start=\d+&sa=\w!io){
                if (!exists $nexturls{$link}){
                        $nexturls{$link} = $link;
                }
        }
 }

}

#setup the callback
$p = HTML::LinkExtor->new(\&callback);

# Request document and parse it as it arrives
$res = $ua->request(HTTP::Request->new(GET => $url),sub {$p->parse($_[0])});

$visitedGURLS = 0;
#for the moment beeing assume 10000 results.
while(1){
        if($visitedGURLS == scalar keys(%nexturls)){
                last;
        }
        foreach $nurl(sort keys(%nexturls)){
                my $value = $nexturls{$nurl};
                #prevent parsing pages twice
                if($value ne "visited"){
                        my $temp = URI->new($baseSEngine.$value);
                        #you can comment this out if you only want clean finds.
                        #print "[*] searching next page $temp\n";
                        $res = $ua->request(HTTP::Request->new(GET => $temp),sub {$p->parse($_[0])});
                        $nexturls{$nurl} = "visited";
                        $visitedGURLS++;
                        sleep 3; #try and prevent getting blocked by google
                }
        }
}