[[FIX]] Correct interpretation of whitespace

src/lex.js

@@ -1514,7 +1514,7 @@ Lexer.prototype = {
     this.from = this.char;
 
     // Move to the next non-space character.
-    while (/\s/.test(this.peek())) {
+    while (reg.whitespace.test(this.peek())) {
       this.from += 1;
       this.skip();
     }

src/reg.js

@@ -36,3 +36,11 @@ exports.fallsThrough = /^\s*falls?\sthrough\s*$/;
 // very conservative rule (eg: only one space between the start of the comment and the first character)
 // to relax the maxlen option
 exports.maxlenException = /^(?:(?:\/\/|\/\*|\*) ?)?[^ ]+$/;
+
+// Node.js releases prior to version 8 include a version of the V8 engine which
+// incorrectly interprets the character class escape `\s`. The following
+// regular expression may be replaced with `/\s/` when JSHint removes support
+// for Node.js versions prior to 8.
+// Source:
+// https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp
+exports.whitespace = /[ \f\n\r\t\v\u00a0\u1680\u2000-\u200a\u2028\u2029\u202f\u205f\u3000\ufeff]/;

tests/test262/expectations.txt

@@ -288,7 +288,6 @@ language/identifiers/vals-rus-alpha-upper-via-escape-hex.js
 language/rest-parameters/array-pattern.js
 language/rest-parameters/no-alias-arguments.js
 language/rest-parameters/object-pattern.js
-language/white-space/mongolian-vowel-separator.js
 language/arguments-object/10.5-7-b-2-s.js
 language/arguments-object/10.5-7-b-3-s.js
 language/arguments-object/10.6-10-c-ii-2.js