[regexp] step back if starting unicode regexp within surrogate pair.

src/regexp/regexp-ast.h

 // Copyright 2016 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.
 
 #ifndef V8_REGEXP_REGEXP_AST_H_
 #define V8_REGEXP_REGEXP_AST_H_
 
 #include "src/objects.h"
 #include "src/utils.h"
 #include "src/zone.h"
@@ skipping 293 matching lines @@
   // Returns a value representing the standard character set if is_standard()
   // returns true.
   // Currently used values are:
   // s : unicode whitespace
   // S : unicode non-whitespace
   // w : ASCII word character (digit, letter, underscore)
   // W : non-ASCII word character
   // d : ASCII digit
   // D : non-ASCII digit
   // . : non-unicode non-newline
-  // * : All characters
+  // * : All characters, for advancing unanchored regexp
   uc16 standard_type() { return set_.standard_set_type(); }
   ZoneList<CharacterRange>* ranges(Zone* zone) { return set_.ranges(zone); }
   bool is_negated() { return is_negated_; }
 
  private:
   CharacterSet set_;
   bool is_negated_;
 };
 
 
@@ skipping 185 matching lines @@
   RegExpEmpty* AsEmpty() override;
   bool IsEmpty() override;
   int min_match() override { return 0; }
   int max_match() override { return 0; }
 };
 
 }  // namespace internal
 }  // namespace v8
 
 #endif  // V8_REGEXP_REGEXP_AST_H_