[es6] Correct parsing of regular expression literal flags.

src/scanner.h

 // Copyright 2011 the V8 project authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 // Features shared by parsing and pre-parsing scanners.
 
 #ifndef V8_SCANNER_H_
 #define V8_SCANNER_H_
 
 #include "src/allocation.h"
@@ skipping 427 matching lines @@
   // Returns true if there was a line terminator before the peek'ed token,
   // possibly inside a multi-line comment.
   bool HasAnyLineTerminatorBeforeNext() const {
     return has_line_terminator_before_next_ ||
            has_multiline_comment_before_next_;
   }
 
   // Scans the input as a regular expression pattern, previous
   // character(s) must be /(=). Returns true if a pattern is scanned.
   bool ScanRegExpPattern(bool seen_equal);
-  // Returns true if regexp flags are scanned (always since flags can
-  // be empty).
-  bool ScanRegExpFlags();
+  // Scans the input as regular expression flags. Returns the flags on success.
+  Maybe<RegExp::Flags> ScanRegExpFlags();
 
   // Scans the input as a template literal
   Token::Value ScanTemplateStart();
   Token::Value ScanTemplateContinuation();
 
   const LiteralBuffer* source_url() const { return &source_url_; }
   const LiteralBuffer* source_mapping_url() const {
     return &source_mapping_url_;
   }
 
@@ skipping 302 matching lines @@
   bool has_line_terminator_before_next_;
   // Whether there is a multi-line comment that contains a
   // line-terminator after the current token, and before the next.
   bool has_multiline_comment_before_next_;
 };
 
 }  // namespace internal
 }  // namespace v8
 
 #endif  // V8_SCANNER_H_