X-Git-Url: https://git.camperquake.de/gitweb.cgi?a=blobdiff_plain;f=videosite%2FCollegeHumorGrabber.pm;h=7fb3cae4174198e4fb8220dc65d2820a8ba12a60;hb=d4fae22c083e110d1c788a78196f85d1e39b116e;hp=b25043cd61841f2987601aae09b478b86f74ca5e;hpb=7f8669fc619a87e496066fb6f92f3df127e985c5;p=videosite.git

diff --git a/videosite/CollegeHumorGrabber.pm b/videosite/CollegeHumorGrabber.pm
index b25043c..7fb3cae 100644
--- a/videosite/CollegeHumorGrabber.pm
+++ b/videosite/CollegeHumorGrabber.pm
@@ -3,14 +3,13 @@
 # (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 LWP::Simple qw(!get);
+use videosite::HTMLHelper;
 use XML::Simple;
-use HTML::Parser;
 use Data::Dumper;
 
 use strict;
@@ -20,7 +19,10 @@ sub new {
     my $self = $class->SUPER::new();
 
     $self->{'NAME'} = 'collegehumor';
-    $self->{'PATTERNS'} = ['(http://www.collegehumor.com/video:(\d+))'];
+    $self->{_SELFTESTURL} = 'http://www.collegehumor.com/video/5635400/pixar-intro-parody';
+    $self->{_SELFTESTTITLE} = 'Pixar Intro Parody';
+    $self->{'PATTERNS'} = ['(http://www.collegehumor.com/video:(\d+))',
+                           '(http://www.collegehumor.com/video/(\d+))'];
 
     bless($self, $class);
     $self->_prepare_parameters();
@@ -49,7 +51,7 @@ sub _parse {
     $metadata->{'DLURL'} = undef;
 
     # Get the XML file containing the video metadata
-    unless(defined($content = LWP::Simple::get(sprintf('http://www.collegehumor.com/moogaloop/video:%s', $2)))) {
+    unless(defined($content = $self->simple_get(sprintf('http://www.collegehumor.com/moogaloop/video/%s', $2)))) {
         $self->error('Could not download XML metadata');
         return undef;
     }
@@ -60,29 +62,7 @@ sub _parse {
     }
 
     $metadata->{'DLURL'} = $t->{'video'}->{'file'};
-
-    # The XML does not contain the full title of the video, for
-    # reasons possibly known to some jerk at CollegeHumor.
-    # So we'll have to parse the actual HTML, too.
-    unless(defined($content = LWP::Simple::get(sprintf('http://www.collegehumor.com/video:%s', $2)))) {
-        $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'};
-            }
-        }
-    }
+    $metadata->{'TITLE'} = $t->{'video'}->{'caption'};
 
     unless(defined($metadata->{'DLURL'}) && defined($metadata->{'TITLE'})) {
         $self->error('Could not extract download URL and title');