X-Git-Url: https://git.camperquake.de/gitweb.cgi?a=blobdiff_plain;f=videosite%2FCollegeHumorGrabber.pm;h=c7851e37ff87ed7eb735a452dcb05249c67e7aa8;hb=8d918b07ad100a4d7b96cf5df640cef0a8a0411d;hp=bbf54a688aac683fab225aca387df9641d98a041;hpb=6bfa02043de8e679102e4d3babd2cc1e6e49824b;p=videosite.git

diff --git a/videosite/CollegeHumorGrabber.pm b/videosite/CollegeHumorGrabber.pm
index bbf54a6..c7851e3 100644
--- a/videosite/CollegeHumorGrabber.pm
+++ b/videosite/CollegeHumorGrabber.pm
@@ -9,7 +9,6 @@ use videosite::GrabberBase;
 @ISA = qw(videosite::GrabberBase);
 
 use videosite::HTMLHelper;
-use LWP::Simple qw(!get);
 use XML::Simple;
 use Data::Dumper;
 
@@ -17,15 +16,16 @@ use strict;
 
 sub new {
     my $class = shift;
-    my $self = $class->SUPER::new();
-
-    $self->{'NAME'} = 'collegehumor';
-    $self->{'PATTERNS'} = ['(http://www.collegehumor.com/video:(\d+))'];
-
-    bless($self, $class);
-    $self->_prepare_parameters();
-
-    return $self;
+    my $self = $class->SUPER::new(
+        NAME => 'collegehumor',
+        _SELFTESTURL => 'http://www.collegehumor.com/video/5635400/pixar-intro-parody',
+        _SELFTESTTITLE => 'Pixar Intro Parody',
+        PATTERNS => ['(http://www.collegehumor.com/video:(\d+))',
+                     '(http://www.collegehumor.com/video/(\d+))'],
+        @_,
+    );
+
+    return bless($self, $class);
 }
 
 sub _parse {
@@ -49,7 +49,7 @@ 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;
     }
@@ -60,20 +60,7 @@ 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.
-    $p = videosite::HTMLHelper->new();
-    unless(defined($content = $p->load(sprintf('http://www.collegehumor.com/video:%s', $2)))) {
-        $self->error('Could not download HTML');
-        return undef;
-    }
-
-    $t = $p->findnodes('h1[@id="item_title"]');
-    if (defined($t)) {
-        $metadata->{'TITLE'} = $t->{'_content'}->[0];
-    }
+    $metadata->{'TITLE'} = $t->{'video'}->{'caption'};
 
     unless(defined($metadata->{'DLURL'}) && defined($metadata->{'TITLE'})) {
         $self->error('Could not extract download URL and title');