Hello Gisle
HTML::TokeParser can not be used in this situation. It needs to have the whole document available at contruction time. You will have to use HTML::Parser directly instead. For this case I would think that HTML::LinkExtor should also be perfect. It can give you a callback each time it finds a link.
Yes I know TokeParser needs the whole document, but It would be great If there was any trick to use it with callbacks. The script I sent only it's an example. I can't use HTML::LinkExtor cause I need to extract links and embedded objects, so I will read deeply about HTML processing with trees.
Make you callback something like:
my $p;
$req_cb = sub {
my($data, $resp) = @_;
$p ||= HTML::LinkExtor->new(sub { print "Found link\n" }, $resp->base);
$p->parse($data);
}
Best Regards,
Andres
_________________________________________________________________
MSN. M�s �til Cada D�a http://www.msn.es/intmap/
