#!/usr/bin/perl
use strict;
use warnings;
use CGI::Carp qw(fatalsToBrowser);
use LWP;
use vars '@ISA'; # for get_basic_credentials
@ISA = 'LWP::UserAgent';
use HTML::Parser;
use DateTime; # for the special url
use CGI qw/:cgi/; # because it's a link, so I think cgi mode is better
my $cgi = CGI->new;
print $cgi->header(-charset=>'utf-8');
my $dt = DateTime->now;
my $url = 'http://koala.ilog.fr/twikiirc/bin/irclogger_log/perlchina?date=' . $dt->ymd . ',' . $dt->day_abbr; # like 2005-05-18,Wed
my $agent = __PACKAGE__->new;
my $request = HTTP::Request->new(GET => $url);
my $response = $agent->request($request);
$response->is_success or die $response->message;
print 'Parse links ....
'; #$response->content;
my $parser = HTML::Parser->new(api_version => 3);
$parser->handler(start => \&got_links, 'tagname, attr');
$parser->parse($response->content);
$parser->eof;
sub got_links {
my ($tagname, $attr) = @_;
if ($tagname eq 'a' && $attr->{href} && $attr->{href} !~ /^[\/\?]/) {
print '{href}, '\'>', $attr->{href}, '
';
}
}
sub get_basic_credentials {
#my ($self, $realm, $url) = @_;
return ('perlchina', 'perlchina'); # the perlchina irc log site's username&password
}