Use `stringToPDFString` to sanitizing bad "Prefix" entries in Page Label dictionaries

It seems that certain bad PDF generators can create badly encoded "Prefix" entries for Page Labels, one example being http://ukjewishfilm.org/wp-content/uploads/2015/09/Jewish-Film-Festival-Programme-ONLINE.pdf.

Unfortunately I didn't come across such a PDF file while adding the API support for Page Labels, but with them now being used in the viewer I just found this issue. With this patch, we now display the Page Labels in the same way as Adobe Reader.

src/core/obj.js

@@ -302,8 +302,9 @@ var Catalog = (function CatalogClosure() {
           assert(!s || isName(s), 'Invalid style in PageLabel dictionary.');
           style = (s ? s.name : null);
 
-          prefix = labelDict.get('P') || '';
-          assert(isString(prefix), 'Invalid prefix in PageLabel dictionary.');
+          var p = labelDict.get('P') || '';
+          assert(isString(p), 'Invalid prefix in PageLabel dictionary.');
+          prefix = stringToPDFString(p);
 
           start = labelDict.get('St') || 1;
           assert(isInt(start), 'Invalid start in PageLabel dictionary.');

test/pdfs/.gitignore

@@ -40,6 +40,7 @@
 !issue7544.pdf
 !issue7598.pdf
 !issue7665.pdf
+!bad-PageLabels.pdf
 !filled-background.pdf
 !ArabicCIDTrueType.pdf
 !ThuluthFeatures.pdf

test/unit/api_spec.js

@@ -503,14 +503,24 @@ describe('api', function() {
         return pdfDoc.getPageLabels();
       });
 
-      Promise.all([promise0, promise1, promise2]).then(function (pageLabels) {
+      // PageLabels with bad "Prefix" entries.
+      var url3 = new URL('../pdfs/bad-PageLabels.pdf', window.location).href;
+      var loadingTask3 = new PDFJS.getDocument(url3);
+      var promise3 = loadingTask3.promise.then(function (pdfDoc) {
+        return pdfDoc.getPageLabels();
+      });
+
+      Promise.all([promise0, promise1, promise2, promise3]).then(
+          function (pageLabels) {
         expect(pageLabels[0]).toEqual(['i', 'ii', 'iii', '1']);
         expect(pageLabels[1]).toEqual(['Front Page1']);
         expect(pageLabels[2]).toEqual(['1', '2']);
+        expect(pageLabels[3]).toEqual(['X1']);
 
         loadingTask0.destroy();
         loadingTask1.destroy();
         loadingTask2.destroy();
+        loadingTask3.destroy();
         done();
       }).catch(function (reason) {
         done.fail(reason);