- Convert SnotrGrabber and CollegeHumorGrabber to HTMLHelper

diff --git a/videosite/CollegeHumorGrabber.pm b/videosite/CollegeHumorGrabber.pm
index b25043c..4388a31 100644 (file)
--- a/videosite/CollegeHumorGrabber.pm
+++ b/videosite/CollegeHumorGrabber.pm
@@ -8,9 +8,9 @@ package CollegeHumorGrabber;
 use GrabberBase;
 @ISA = qw(GrabberBase);
 
+use 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)))) {
+    $p = HTMLHelper->new();
+    unless(defined($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 = $p->findnodes('meta[@name="title"]');
+    if (defined($t)) {
+        $metadata->{'TITLE'} = $t->{'content'};
     }
 
     unless(defined($metadata->{'DLURL'}) && defined($metadata->{'TITLE'})) {

diff --git a/videosite/SnotrGrabber.pm b/videosite/SnotrGrabber.pm
index f5acf3d..3e40ff6 100644 (file)
--- a/videosite/SnotrGrabber.pm
+++ b/videosite/SnotrGrabber.pm
@@ -8,8 +8,7 @@ package SnotrGrabber;
 use GrabberBase;
 @ISA = qw(GrabberBase);
 
-use LWP::Simple qw(!get);
-use HTML::Parser;
+use HTMLHelper;
 use Data::Dumper;
 
 use strict;
@@ -33,9 +32,8 @@ sub _parse {
     my $pattern = shift;
     my $content;
     my $metadata = {};
-    my $p;
-    my $t;
-    my @accum;
+    my $p = HTMLHelper->new();
+    my $n;
 
     $url =~ m|$pattern|;
     $url = $1;
@@ -48,26 +46,14 @@ sub _parse {
     $metadata->{'DLURL'} = undef;
 
     # Get the HTML file containing the title
-    unless(defined($content = LWP::Simple::get(sprintf('http://www.snotr.com/video/%s', $2)))) {
+    unless(defined($p->load(sprintf('http://www.snotr.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'};
-                last;
-            }
-        }
+    $n = $p->findnodes('meta[@name="title"]');
+    if (defined($n)) {
+        $metadata->{'TITLE'} = $n->{'content'};
     }
 
     # The actual video URL is insultingly simple.