X-Git-Url: https://git.camperquake.de/gitweb.cgi?a=blobdiff_plain;f=videosite%2FCollegeHumorGrabber.pm;h=bbf54a688aac683fab225aca387df9641d98a041;hb=4f73c0b48954f5f4f30ee96779ecaa3fca847305;hp=c325ed359de56b04e8485b50280d04cd460ad667;hpb=35306939804820c033c8e25ab5aa5134f5dc5735;p=videosite.git

diff --git a/videosite/CollegeHumorGrabber.pm b/videosite/CollegeHumorGrabber.pm
index c325ed3..bbf54a6 100644
--- a/videosite/CollegeHumorGrabber.pm
+++ b/videosite/CollegeHumorGrabber.pm
@@ -1,8 +1,14 @@
-package CollegeHumorGrabber;
+# Grabber for collegehumor.com
+#
+# (c) 2007 by Ralf Ertzinger <ralf@camperquake.de>
+# licensed under GNU GPL v2
 
-use GrabberBase;
-@ISA = qw(GrabberBase);
+package videosite::CollegeHumorGrabber;
 
+use videosite::GrabberBase;
+@ISA = qw(videosite::GrabberBase);
+
+use videosite::HTMLHelper;
 use LWP::Simple qw(!get);
 use XML::Simple;
 use Data::Dumper;
@@ -29,6 +35,7 @@ sub _parse {
     my $content;
     my $metadata = {};
     my $p = XML::Simple->new();
+    my @accum;
     my $t;
 
     $url =~ m|$pattern|;
@@ -37,7 +44,7 @@ sub _parse {
     $metadata->{'URL'} = $url;
     $metadata->{'ID'} = $2;
     $metadata->{'TYPE'} = 'video';
-    $metadata->{'SOURCE'} = 'collegehumor';
+    $metadata->{'SOURCE'} = $self->{'NAME'};
     $metadata->{'TITLE'} = undef;
     $metadata->{'DLURL'} = undef;
 
@@ -53,7 +60,20 @@ sub _parse {
     }
 
     $metadata->{'DLURL'} = $t->{'video'}->{'file'};
-    $metadata->{'TITLE'} = $t->{'video'}->{'caption'};
+
+    # 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];
+    }
 
     unless(defined($metadata->{'DLURL'}) && defined($metadata->{'TITLE'})) {
         $self->error('Could not extract download URL and title');