#! /usr/bin/perl
use strict;
use warnings;
use URI::Find::Schemeless;
use HTML::Entities qw(encode_entities);

my $text = q~
   hallo dies ist keine.url
   dies ist aber eine: www.intertivity.com
   ftp.irgendwas.de/test/thisfile mailto:perl@intertivityNOSP4M.com od
+er so
   yeah perl@intertivityNOSP4M.com http://www.intertivity.com/
   ~;

my $finder = URI::Find::Schemeless->new
   ( 
      sub {
         my ($uri, $originalUri) = @_;
         return q/<a href="/ . encode_entities("$uri") . q/">/ . encod
+e_entities($originalUri) . q~</a>~;
      }
   );

my $howManyFound = $finder->find(\$text);

print "$howManyFound URIs found\n";
print "$text\n";
[download]

Others have mentioned how to find URLs, but you may want to double check whether or not the URLs are actually valid, using a module like LWP::Simple or LWP::UserAgent For instance, for all urls @urls you find, you might do something like:

# assuming you've already populated @urls
# and done:
use LWP::UserAgent;
use strict;
use warnings;

# try this:
my @old_urls = @urls;
@urls = ();
my $user_agent = LWP::UserAgent->new;
while (@old_urls) {
  my $url = shift (@old_urls);
  my $response = $user_agent->get($url);
  if ($response->is_success) {
     push @urls, $url;
     # or, if you want to get more detailed:
     # push @urls, {
     #              url => $url,
     #              type => $response->content_type,
     #             };
   }
}
[download]