Change getPDFFileNameFromURL to ignore data: URLs for performance reasons (issue 8263)

The patch also changes the `defaultFilename` to use the ES6 default parameter notation, and fixes the formatting of the JSDoc comment.

Finally, since `getPDFFileNameFromURL` currently has no unit-tests, a few basic ones are added to avoid regressions.

test/unit/ui_utils_spec.js

@@ -17,16 +17,20 @@
 (function (root, factory) {
   if (typeof define === 'function' && define.amd) {
     define('pdfjs-test/unit/ui_utils_spec', ['exports',
-           'pdfjs-web/ui_utils'], factory);
+      'pdfjs-web/ui_utils', 'pdfjs/shared/util'], factory);
   } else if (typeof exports !== 'undefined') {
-    factory(exports, require('../../web/ui_utils.js'));
+    factory(exports, require('../../web/ui_utils.js'),
+      require('../../src/shared/util.js'));
   } else {
-    factory((root.pdfjsTestUnitUiUtilsSpec = {}), root.pdfjsWebUiUtils);
+    factory((root.pdfjsTestUnitUiUtilsSpec = {}), root.pdfjsWebUiUtils,
+      root.pdfjsSharedUtil);
   }
-}(this, function (exports, webUiUtils) {
+}(this, function (exports, webUiUtils, sharedUtil) {
 
 var binarySearchFirstItem = webUiUtils.binarySearchFirstItem;
+var getPDFFileNameFromURL = webUiUtils.getPDFFileNameFromURL;
 var EventBus = webUiUtils.EventBus;
+var createObjectURL = sharedUtil.createObjectURL;
 
 describe('ui_utils', function() {
   describe('binary search', function() {
@@ -57,6 +61,130 @@ describe('ui_utils', function() {
     });
   });
 
+  describe('getPDFFileNameFromURL', function() {
+    it('gets PDF filename', function() {
+      // Relative URL
+      expect(getPDFFileNameFromURL('/pdfs/file1.pdf')).toEqual('file1.pdf');
+      // Absolute URL
+      expect(getPDFFileNameFromURL(
+        'http://www.example.com/pdfs/file2.pdf')).toEqual('file2.pdf');
+    });
+
+    it('gets fallback filename', function() {
+      // Relative URL
+      expect(getPDFFileNameFromURL('/pdfs/file1.txt')).toEqual('document.pdf');
+      // Absolute URL
+      expect(getPDFFileNameFromURL(
+        'http://www.example.com/pdfs/file2.txt')).toEqual('document.pdf');
+    });
+
+    it('gets custom fallback filename', function() {
+      // Relative URL
+      expect(getPDFFileNameFromURL('/pdfs/file1.txt', 'qwerty1.pdf')).
+        toEqual('qwerty1.pdf');
+      // Absolute URL
+      expect(getPDFFileNameFromURL('http://www.example.com/pdfs/file2.txt',
+        'qwerty2.pdf')).toEqual('qwerty2.pdf');
+
+      // An empty string should be a valid custom fallback filename.
+      expect(getPDFFileNameFromURL('/pdfs/file3.txt', '')).toEqual('');
+    });
+
+    it('gets PDF filename from URL containing leading/trailing whitespace',
+        function() {
+      // Relative URL
+      expect(getPDFFileNameFromURL(
+        '   /pdfs/file1.pdf   ')).toEqual('file1.pdf');
+      // Absolute URL
+      expect(getPDFFileNameFromURL(
+        '   http://www.example.com/pdfs/file2.pdf   ')).toEqual('file2.pdf');
+    });
+
+    it('gets PDF filename from query string', function() {
+      // Relative URL
+      expect(getPDFFileNameFromURL(
+        '/pdfs/pdfs.html?name=file1.pdf')).toEqual('file1.pdf');
+      // Absolute URL
+      expect(getPDFFileNameFromURL(
+        'http://www.example.com/pdfs/pdf.html?file2.pdf')).toEqual('file2.pdf');
+    });
+
+    it('gets PDF filename from hash string', function() {
+      // Relative URL
+      expect(getPDFFileNameFromURL(
+        '/pdfs/pdfs.html#name=file1.pdf')).toEqual('file1.pdf');
+      // Absolute URL
+      expect(getPDFFileNameFromURL(
+        'http://www.example.com/pdfs/pdf.html#file2.pdf')).toEqual('file2.pdf');
+    });
+
+    it('gets correct PDF filename when multiple ones are present', function() {
+      // Relative URL
+      expect(getPDFFileNameFromURL(
+        '/pdfs/file1.pdf?name=file.pdf')).toEqual('file1.pdf');
+      // Absolute URL
+      expect(getPDFFileNameFromURL(
+        'http://www.example.com/pdfs/file2.pdf#file.pdf')).toEqual('file2.pdf');
+    });
+
+    it('gets PDF filename from URI-encoded data', function() {
+      var encodedUrl = encodeURIComponent(
+        'http://www.example.com/pdfs/file1.pdf');
+      expect(getPDFFileNameFromURL(encodedUrl)).toEqual('file1.pdf');
+
+      var encodedUrlWithQuery = encodeURIComponent(
+        'http://www.example.com/pdfs/file.txt?file2.pdf');
+      expect(getPDFFileNameFromURL(encodedUrlWithQuery)).toEqual('file2.pdf');
+    });
+
+    it('gets PDF filename from data mistaken for URI-encoded', function() {
+      expect(getPDFFileNameFromURL('/pdfs/%AA.pdf')).toEqual('%AA.pdf');
+
+      expect(getPDFFileNameFromURL('/pdfs/%2F.pdf')).toEqual('%2F.pdf');
+    });
+
+    it('gets PDF filename from (some) standard protocols', function() {
+      // HTTP
+      expect(getPDFFileNameFromURL('http://www.example.com/file1.pdf')).
+        toEqual('file1.pdf');
+      // HTTPS
+      expect(getPDFFileNameFromURL('https://www.example.com/file2.pdf')).
+        toEqual('file2.pdf');
+      // File
+      expect(getPDFFileNameFromURL('file:///path/to/files/file3.pdf')).
+        toEqual('file3.pdf');
+      // FTP
+      expect(getPDFFileNameFromURL('ftp://www.example.com/file4.pdf')).
+        toEqual('file4.pdf');
+    });
+
+    it('gets PDF filename from query string appended to "blob:" URL',
+        function() {
+      var typedArray = new Uint8Array([1, 2, 3, 4, 5]);
+      var blobUrl = createObjectURL(typedArray, 'application/pdf');
+      // Sanity check to ensure that a "blob:" URL was returned.
+      expect(blobUrl.indexOf('blob:') === 0).toEqual(true);
+
+      expect(getPDFFileNameFromURL(blobUrl + '?file.pdf')).toEqual('file.pdf');
+    });
+
+    it('gets fallback filename from query string appended to "data:" URL',
+        function() {
+      var typedArray = new Uint8Array([1, 2, 3, 4, 5]);
+      var dataUrl = createObjectURL(typedArray, 'application/pdf',
+                                    /* forceDataSchema = */ true);
+      // Sanity check to ensure that a "data:" URL was returned.
+      expect(dataUrl.indexOf('data:') === 0).toEqual(true);
+
+      expect(getPDFFileNameFromURL(dataUrl + '?file1.pdf')).
+        toEqual('document.pdf');
+
+      // Should correctly detect a "data:" URL with leading whitespace.
+      expect(getPDFFileNameFromURL('     ' + dataUrl + '?file2.pdf')).
+        toEqual('document.pdf');
+    });
+  });
+
   describe('EventBus', function () {
     it('dispatch event', function () {
       var eventBus = new EventBus();

web/ui_utils.js

@@ -353,15 +353,26 @@ function noContextMenuHandler(e) {
   e.preventDefault();
 }
 
+function isDataSchema(url) {
+  var i = 0, ii = url.length;
+  while (i < ii && url[i].trim() === '') {
+    i++;
+  }
+  return url.substr(i, 5).toLowerCase() === 'data:';
+}
+
 /**
  * Returns the filename or guessed filename from the url (see issue 3455).
- * url {String} The original PDF location.
+ * @param {string} url - The original PDF location.
- * defaultFilename {string} The value to return if the file name is unknown.
+ * @param {string} defaultFilename - The value returned if the filename is
- * @return {String} Guessed PDF file name.
+ *   unknown, or the protocol is unsupported.
+ * @returns {string} Guessed PDF filename.
  */
-function getPDFFileNameFromURL(url, defaultFilename) {
+function getPDFFileNameFromURL(url, defaultFilename = 'document.pdf') {
-  if (typeof defaultFilename === 'undefined') {
+  if (isDataSchema(url)) {
-    defaultFilename = 'document.pdf';
+    console.warn('getPDFFileNameFromURL: ' +
+                 'ignoring "data:" URL for performance reasons.');
+    return defaultFilename;
   }
   var reURI = /^(?:(?:[^:]+:)?\/\/[^\/]+)?([^?#]*)(\?[^#]*)?(#.*)?$/;
   //            SCHEME        HOST         1.PATH  2.QUERY   3.REF
@@ -369,8 +380,8 @@ function getPDFFileNameFromURL(url, defaultFilename) {
   var reFilename = /[^\/?#=]+\.pdf\b(?!.*\.pdf\b)/i;
   var splitURI = reURI.exec(url);
   var suggestedFilename = reFilename.exec(splitURI[1]) ||
-                           reFilename.exec(splitURI[2]) ||
+                          reFilename.exec(splitURI[2]) ||
-                           reFilename.exec(splitURI[3]);
+                          reFilename.exec(splitURI[3]);
   if (suggestedFilename) {
     suggestedFilename = suggestedFilename[0];
     if (suggestedFilename.indexOf('%') !== -1) {