Fix broken metadata

diff --git a/src/metadata.js b/src/metadata.js
index f3805616866b8070abc9745c907d2ea2fa7c28c9..f2585214ccab4f6272a87d55644e1b88a6c8dd51 100644 (file)
--- a/src/metadata.js
+++ b/src/metadata.js
@@ -4,8 +4,28 @@
 'use strict';
 
 var Metadata = PDFJS.Metadata = (function MetadataClosure() {
+  function fixMetadata(meta) {
+    return meta.replace(/>\\376\\377([^<]+)/g, function(all, codes) {
+      var bytes = codes.replace(/\\([0-3])([0-7])([0-7])/g,
+                                function(code, d1, d2, d3) {
+        return String.fromCharCode(d1 * 64 + d2 * 8 + d3 * 1);
+      });
+      var chars = '';
+      for (var i = 0; i < bytes.length; i += 2) {
+        var code = bytes.charCodeAt(i) * 256 + bytes.charCodeAt(i + 1);
+        chars += code >= 32 && code < 127 && code != 60 && code != 62 &&
+          code != 38 && false ? String.fromCharCode(code) :
+          '&#x' + (0x10000 + code).toString(16).substring(1) + ';';
+      }
+      return '>' + chars;
+    });
+  }
+
   function Metadata(meta) {
     if (typeof meta === 'string') {
+      // Ghostscript produces invalid metadata
+      meta = fixMetadata(meta);
+
       var parser = new DOMParser();
       meta = parser.parseFromString(meta, 'application/xml');
     } else if (!(meta instanceof Document)) {

diff --git a/test/unit/metadata_spec.js b/test/unit/metadata_spec.js
new file mode 100644 (file)
index 0000000..2d4b839
--- /dev/null
+++ b/test/unit/metadata_spec.js
@@ -0,0 +1,18 @@
+/* -*- Mode: Java; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
+/* vim: set shiftwidth=2 tabstop=2 autoindent cindent expandtab: */
+
+'use strict';
+
+describe('metadata', function() {
+  describe('incorrect_xmp', function() {
+    it('should fix the incorrect XMP data', function() {
+      var invalidXMP = '<x:xmpmeta xmlns:x=\'adobe:ns:meta/\'>' +
+        '<rdf:RDF xmlns:rdf=\'http://www.w3.org/1999/02/22-rdf-syntax-ns#\'>' +
+        '<rdf:Description xmlns:dc=\'http://purl.org/dc/elements/1.1/\'>' +
+        '<dc:title>\\376\\377\\000P\\000D\\000F\\000&</dc:title>' +
+        '</rdf:Description></rdf:RDF></x:xmpmeta>';
+      var meta = new Metadata(invalidXMP);
+      expect(meta.get('dc:title')).toEqual('PDF&');
+    });
+  });
+});

diff --git a/test/unit/unit_test.html b/test/unit/unit_test.html
index 49de9dbde9093a0bd23eade499d137b2816a01b5..cdd0c297f9963a417ad4aef20adeaa211b262108 100644 (file)
--- a/test/unit/unit_test.html
+++ b/test/unit/unit_test.html
@@ -41,6 +41,7 @@
   <script type="text/javascript" src="crypto_spec.js"></script>
   <script type="text/javascript" src="stream_spec.js"></script>
   <script type="text/javascript" src="api_spec.js"></script>
+  <script type="text/javascript" src="metadata_spec.js"></script>
   <script type="text/javascript">
     'use strict';