[regexp] back refs must not start/end in the middle of a surrogate pair

src/regexp/regexp-macro-assembler.cc

 // Copyright 2012 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.
 
 #include "src/regexp/regexp-macro-assembler.h"
 
 #include "src/assembler.h"
 #include "src/isolate-inl.h"
 #include "src/regexp/regexp-stack.h"
 #include "src/simulator.h"
@@ skipping 70 matching lines @@
         if (s1[0] != s2[0]) {
           return 0;
         }
       }
     }
   }
   return 1;
 }
 
 
+void RegExpMacroAssembler::CheckNotInSurrogatePair(int cp_offset,
+                                                   Label* on_failure) {
+  Label ok;
+  // Check that current character is not a trail surrogate.
+  LoadCurrentCharacter(cp_offset, &ok);
+  CheckCharacterNotInRange(kTrailSurrogateStart, kTrailSurrogateEnd, &ok);
+  // Check that previous character is not a lead surrogate.
+  LoadCurrentCharacter(cp_offset - 1, &ok);
+  CheckCharacterInRange(kLeadSurrogateStart, kLeadSurrogateEnd, on_failure);
+  Bind(&ok);
+}
+
+
 #ifndef V8_INTERPRETED_REGEXP  // Avoid unused code, e.g., on ARM.
 
 NativeRegExpMacroAssembler::NativeRegExpMacroAssembler(Isolate* isolate,
                                                        Zone* zone)
     : RegExpMacroAssembler(isolate, zone) {}
 
 
 NativeRegExpMacroAssembler::~NativeRegExpMacroAssembler() {
 }
 
@@ skipping 224 matching lines @@
   }
   *stack_base = new_stack_base;
   intptr_t stack_content_size = old_stack_base - stack_pointer;
   return new_stack_base - stack_content_size;
 }
 
 #endif  // V8_INTERPRETED_REGEXP
 
 }  // namespace internal
 }  // namespace v8