Changeset 7235

Timestamp:

2004-04-28T13:18:45+12:00 (20 years ago)

Author:

kjdon

Message:

fixed a couple of bugs and added a bit of output to do with extracting titles

File:

• trunk/gsdl/perllib/plugins/HTMLPlug.pm (modified) (4 diffs)

trunk/gsdl/perllib/plugins/HTMLPlug.pm

@@ -604 +604 @@
     # one of the tags we want to match.
 
+    # special case for title - we want to remember if its been found
+    my $found_title = 0;
     # this assumes that ">" won't appear. (I don't think it's allowed to...)
     $html_header =~ /^/; # match the start of the string, for \G assertion
+   
     while ($html_header =~ m/\G.*?<meta(.*?)>/sig) {
     my $metatag=$1;
@@ -651 +654 @@
         $tag = $find_fields{lc($tag)};
     }
+    if (lc($tag) eq "title") {
+        $found_title = 1;
+    }
     print $outhandle " extracted \"$tag\" metadata \"$value\"\n" 
         if ($self->{'verbosity'} > 2);
@@ -658 +664 @@
     
     # TITLE: extract the document title
-    if (exists $find_fields{'title'} && $find_fields{'title'} == 0) {
+    if (exists $find_fields{'title'} && !$found_title) {
     # we want a title, and didn't find one in the meta tags
     # see if there's a <title> tag
     my $title;
-    if ($html_header =~ /<title[^>]*>([^<]*)<\/title[^>]*>/is) {
+    my $from = "";
+    if ($html_header =~ /<title[^>]*>([^<]+)<\/title[^>]*>/is) {
         $title = $1;
+        $from = "<title> tags";
     }
     if (!defined $title) {
+        $from = "first 100 chars";
         # if no title use first 100 or so characters
         $title = $$textref;
@@ -685 +694 @@
     $title =~ s/^\s+//s; # in case title_sub introduced any...
     $doc_obj->add_utf8_metadata ($section, 'Title', $title);
-    print $outhandle " extracted Title metadata \"$title\"\n" 
+    print $outhandle " extracted Title metadata \"$title\" from $from\n" 
         if ($self->{'verbosity'} > 2);
-    }
+    } 
 
     # Special, for metadata names such as tagH1 - extracts