Changeset 13170

Timestamp:

2006-10-27T13:37:58+13:00 (17 years ago)

Author:

kjdon

Message:

docprint is no longer a docproc object. Its now a perl module that outputs a (GA) XMl representation of a section of a doc obj. Also has an escape_text function

File:

• trunk/gsdl/perllib/docprint.pm (modified) (2 diffs)

trunk/gsdl/perllib/docprint.pm

@@ -1 +1 @@
 ###########################################################################
 #
-# docprint.pm
+# docprint.pm --
 # A component of the Greenstone digital library software
 # from the New Zealand Digital Library Project at the 
 # University of Waikato, New Zealand.
 #
-# Copyright (C) 1999 New Zealand Digital Library Project
+# Copyright (C) 2006 New Zealand Digital Library Project
 #
 # This program is free software; you can redistribute it and/or modify
@@ -24 +24 @@
 ###########################################################################
 
-# This document processor prints a document out to STDOUT
-
+# This is used to output an XML representation of a doc_obj - this will be 
+# GA format.
+# This is used by GAPlugout, doc.pm and incremental_build
 
 package docprint;
 
-use arcinfo;
-use docproc;
-use util;
+use strict;
 
+sub get_section_xml {
+    
+    my ($doc_obj, $section) = @_;
 
-sub BEGIN {
-    @docprint::ISA = ('docproc');
+    my $section_ptr = $doc_obj->_lookup_section ($section);
+    return "" unless defined $section_ptr;
+
+    my $all_text = "<Section>\n";
+    $all_text .= "  <Description>\n";
+    
+    # output metadata
+    foreach my $data (@{$section_ptr->{'metadata'}}) {
+    my $escaped_value = &escape_text($data->[1]);
+    $all_text .= '    <Metadata name="' . $data->[0] . '">' . $escaped_value . "</Metadata>\n";
+    }
+
+    $all_text .= "  </Description>\n";
+
+    # output the text
+    $all_text .= "  <Content>";
+    $all_text .= &escape_text($section_ptr->{'text'});
+    $all_text .= "</Content>\n";
+    
+    # output all the subsections
+    foreach my $subsection (@{$section_ptr->{'subsection_order'}}) {
+    $all_text .= &get_section_xml($doc_obj, "$section.$subsection");
+    }
+    
+    $all_text .=  "</Section>\n";
+
+    # make sure no nasty control characters have snuck through
+    # (XML::Parser will barf on anything it doesn't consider to be
+    # valid UTF-8 text, including things like \c@, \cC etc.)
+    $all_text =~ s/[\x00-\x09\x0B\x0C\x0E-\x1F]//g;
+
+    return $all_text;
 }
 
-sub new {
-    my ($class) = @_;
-    my $self = new docproc ();
+sub escape_text {
+    my ($text) = @_;
+    # special characters in the xml encoding
+    $text =~ s/&/&amp;/g; # this has to be first...
+    $text =~ s/</&lt;/g;
+    $text =~ s/>/&gt;/g;
+    $text =~ s/\"/&quot;/g;
 
-    return bless $self, $class;
+    return $text;
 }
 
-sub process {
-    my $self = shift (@_);
-    my ($doc_obj) = @_;
-
-    # add associated files as metadata to the document
-    my @assoc_files = ();
-    foreach my $assoc_file (@{$doc_obj->get_assoc_files()}) {
-    if (-e $assoc_file->[0]) {
-        $doc_obj->add_metadata ($doc_obj->get_top_section(), 
-                    "gsdlassocfile", 
-                    "$assoc_file->[1]:$assoc_file->[2]");
-    }
-    }
-    
-    # output document
-    $doc_obj->output_section(STDOUT, $doc_obj->get_top_section(),
-                 undef, undef, 0);
-}
-
-
 1;