X-Git-Url: https://git.camperquake.de/gitweb.cgi?a=blobdiff_plain;f=videosite%2FCollegeHumorGrabber.pm;h=4388a312f38903887f07dfa29439d24bab874bfe;hb=400621043a9ee2217a07e2afe5cedf23a2491d69;hp=9fdb87e456e3c9f5fcc9b11a363920cee28a5e9a;hpb=d738e03f4f2d70a41eba8b77177826d1ff62f42b;p=videosite.git

diff --git a/videosite/CollegeHumorGrabber.pm b/videosite/CollegeHumorGrabber.pm
index 9fdb87e..4388a31 100644
--- a/videosite/CollegeHumorGrabber.pm
+++ b/videosite/CollegeHumorGrabber.pm
@@ -1,8 +1,14 @@
+# Grabber for collegehumor.com
+#
+# (c) 2007 by Ralf Ertzinger <ralf@camperquake.de>
+# licensed under GNU GPL v2
+
 package CollegeHumorGrabber;
 
 use GrabberBase;
 @ISA = qw(GrabberBase);
 
+use 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|;
@@ -36,7 +43,8 @@ sub _parse {
 
     $metadata->{'URL'} = $url;
     $metadata->{'ID'} = $2;
-    $metadata->{'TYPE'} = 'collegehumor';
+    $metadata->{'TYPE'} = 'video';
+    $metadata->{'SOURCE'} = $self->{'NAME'};
     $metadata->{'TITLE'} = undef;
     $metadata->{'DLURL'} = undef;
 
@@ -52,7 +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.
+    $p = HTMLHelper->new();
+    unless(defined($content = $p->load(sprintf('http://www.collegehumor.com/video:%s', $2)))) {
+        $self->error('Could not download HTML');
+        return undef;
+    }
+
+    $t = $p->findnodes('meta[@name="title"]');
+    if (defined($t)) {
+        $metadata->{'TITLE'} = $t->{'content'};
+    }
 
     unless(defined($metadata->{'DLURL'}) && defined($metadata->{'TITLE'})) {
         $self->error('Could not extract download URL and title');