X-Git-Url: https://git.camperquake.de/gitweb.cgi?a=blobdiff_plain;f=videosite%2FCollegeHumorGrabber.pm;h=bbf54a688aac683fab225aca387df9641d98a041;hb=4f73c0b48954f5f4f30ee96779ecaa3fca847305;hp=b25043cd61841f2987601aae09b478b86f74ca5e;hpb=7f8669fc619a87e496066fb6f92f3df127e985c5;p=videosite.git

diff --git a/videosite/CollegeHumorGrabber.pm b/videosite/CollegeHumorGrabber.pm
index b25043c..bbf54a6 100644
--- a/videosite/CollegeHumorGrabber.pm
+++ b/videosite/CollegeHumorGrabber.pm
@@ -3,14 +3,14 @@
 # (c) 2007 by Ralf Ertzinger <ralf@camperquake.de>
 # licensed under GNU GPL v2
 
-package CollegeHumorGrabber;
+package videosite::CollegeHumorGrabber;
 
-use GrabberBase;
-@ISA = qw(GrabberBase);
+use videosite::GrabberBase;
+@ISA = qw(videosite::GrabberBase);
 
+use videosite::HTMLHelper;
 use LWP::Simple qw(!get);
 use XML::Simple;
-use HTML::Parser;
 use Data::Dumper;
 
 use strict;
@@ -64,24 +64,15 @@ sub _parse {
     # The XML does not contain the full title of the video, for
     # reasons possibly known to some jerk at CollegeHumor.
     # So we'll have to parse the actual HTML, too.
-    unless(defined($content = LWP::Simple::get(sprintf('http://www.collegehumor.com/video:%s', $2)))) {
+    $p = videosite::HTMLHelper->new();
+    unless(defined($content = $p->load(sprintf('http://www.collegehumor.com/video:%s', $2)))) {
         $self->error('Could not download HTML');
         return undef;
     }
-    $p = HTML::Parser->new(api_version => 3);
-
-    $p->handler(start => \@accum, "tagname, attr");
-    $p->report_tags(qw(meta));
-    $p->utf8_mode(1);
-    $p->parse($content);
-
-    # Look for the title in the meta tags
-    foreach $t (@accum) {
-        if ('meta' eq $t->[0]) {
-            if (exists($t->[1]->{'name'}) and ('title' eq $t->[1]->{'name'})) {
-                $metadata->{'TITLE'} = $t->[1]->{'content'};
-            }
-        }
+
+    $t = $p->findnodes('h1[@id="item_title"]');
+    if (defined($t)) {
+        $metadata->{'TITLE'} = $t->{'_content'}->[0];
     }
 
     unless(defined($metadata->{'DLURL'}) && defined($metadata->{'TITLE'})) {