Changeset 24028 for main/trunk/greenstone2

Timestamp:

2011-05-18T20:42:54+12:00 (13 years ago)

Author:

ak19

Message:

Bugfix for PagedImagePlugin dealing incorrectly with accented characters when working with item files in XML.

File:

• main/trunk/greenstone2/perllib/plugins/PagedImagePlugin.pm (modified) (4 diffs)

main/trunk/greenstone2/perllib/plugins/PagedImagePlugin.pm

@@ -134 +134 @@
 package PagedImagePlugin;
 
+use Encode;
 use ReadXMLFile;
 use ReadTextFile;
@@ -485 +486 @@
     $self->{'current_section'} = $doc_obj->get_parent_section($self->{'current_section'});
     } elsif ($element eq "Metadata") {
-    my $meta_name = $self->{'metadata_name'};
+    
+    # text read in by XML::Parser is in Perl's binary byte value
+    # form ... need to explicitly make it UTF-8     
+    my $meta_name = decode("utf-8",$self->{'metadata_name'});
+    my $metadata_value = decode("utf-8",$self->{'metadata_value'});
+    
     if ($meta_name =~ /\./) {
         $meta_name = "ex.$meta_name";
     }
-    $doc_obj->add_utf8_metadata ($self->{'current_section'}, $meta_name, $self->{'metadata_value'});
+    
+    $doc_obj->add_utf8_metadata ($self->{'current_section'}, $meta_name, $metadata_value);
     $self->{'metadata_name'} = "";
     $self->{'metadata_value'} = "";
@@ -627 +634 @@
     my $num = 0;
     while (defined ($line = <ITEMFILE>)) {
+    
+    # text read in by XML::Parser is in Perl's binary byte value
+    # form ... need to explicitly make it UTF-8
+    $line = decode("utf-8",$line);
+    
     next unless $line =~ /\w/;
     chomp $line;
@@ -706 +718 @@
 
     my $text="";
-    &ReadTextFile::read_file($self, $filename_full_path, $encoding, $language, \$text);
+    &ReadTextFile::read_file($self, $filename_full_path, $encoding, $language, \$text); # already decoded as utf8
     if (!length ($text)) {
     # It's a bit unusual but not out of the question to have no text, so just give a warning