Changeset 10600

Timestamp:

2005-09-07T09:18:27+12:00 (18 years ago)

Author:

chi

Message:

modifications for deal with document title (as the first H1 heading) and also store the
metadata title retrieved from document at the top_section.

File:

• trunk/gsdl/perllib/plugins/StructuredHTMLPlug.pm (modified) (7 diffs)

trunk/gsdl/perllib/plugins/StructuredHTMLPlug.pm

@@ -92 +92 @@
     print $outhandle "StructuredHTMLPlug: processing $file\n"
         if $self->{'verbosity'} > 1;
+    
     my @head_and_body = split(/<body/i,$$textref);
     my $head = shift(@head_and_body);
     my $body_text = join("<body", @head_and_body);
-    
+    $head =~ m/<title>(.+)<\/title>/i;
+    my $doctitle = $1 if defined $1;
     if (defined $self->{'extracted_word_metadata_fields'} && $self->{'extracted_word_metadata_fields'}=~ /\S/) {
     my @doc_properties = split(/<xml>/i,$head);
@@ -108 +110 @@
     # get rid of TOC and TOF sections and their title
     if ($self->{'delete_toc'} == 1){
-    #line-height:150%;mso-ansi-language:FR'>Contents<o:p></o:p></span></b></p>
-    # get rid of Table of Contents title and Table of Figures 
-    # these two lines don't work - how can we do this properlly??
     if (defined $self->{'toc_header'}&& $self->{'toc_header'} =~ /\S/){
         $body_text =~ s/<p class=(($self->{'toc_header'})[^>]*)>(.+?)<\/p>//isg;
@@ -118 +117 @@
     }
     }
-
+    
     if (defined $self->{'title_header'} && $self->{'title_header'}=~ /\S/){
     $self->{'title_header'} =~ s/^(\()(.*)(\))/$2/is;
-    $body_text =~ s/<p class=(($self->{'title_header'})[^>]*)>(.+?)<\/p>/<p class=$1><title>$3<\/title><\/p>/isg;
+    #$body_text =~ s/<p class=(($self->{'title_header'})[^>]*)>(.+?)<\/p>/<p class=$1><title>$3<\/title><\/p>/isg;
+    #$doctitle = $3;
+    $body_text =~ s/<p class=(($self->{'title_header'})[^>]*)>(.+?)<\/p>/<p class=$1><h1>$3<\/h1><\/p>/isg;
+    #$body_text =~ m/<p class=(($self->{'title_header'})[^>]*)>(.+?)<\/p>/isg;
+    #$doctitle = "<h1>".$3."<\/h1>" if defined $3;
     }
 
@@ -142 +145 @@
     $body_text =~ s/(<p[^>]*><o:p>&nbsp;<\/o:p><\/p>)//isg;
     
+    $section_text .= "<!--\n<Section>\n-->\n";
+    #my $top_section_tag = "<!--\n<Section>\n-->\n";
+    #$body_text =~ s/(<div.*)/$top_section_text$doctitle$1/i;
+    #$body_text =~ s/(<div.*)/$top_section_tag$1/i;
     my $body = "<body".$body_text;
     
     my $section_text = $head;
-    $section_text .= "<!--\n<Section>\n-->\n";
     
     # split HTML text on <h1>, <h2> etc tags
@@ -203 +209 @@
             }
 
-            my $spacing = "  " x $hnum;         
+            my $spacing = "  " x $hnum;
             $section_text .= "<!--\n";
             $section_text .= $spacing."<Section>\n";
@@ -219 +225 @@
         else {
 ###     print STDERR "***** hc = <h$hc\n\n";
-        
         }
         $section_text .= "<h$hc";
@@ -275 +280 @@
     $doc_obj->add_utf8_metadata ($cursection, "srcicon",  "_icondoc_");
     $doc_obj->add_utf8_metadata ($cursection, "/srclink", "</a>");
-
-    my $file_size = -s $filename;
+    $doc_obj->add_utf8_metadata ($cursection, "Title", $doctitle);
+    my $file_size = -s $filename;
     if ($file_size>1024)
     {