Experimental parser: parsing \u stuff correctly.

src/lexer/lexer.re

 // Portions of this code based on re2c:
 //   (re2c/examples/push.re)
 // Copyright 2013 the V8 project authors. All rights reserved.
 // Redistribution and use in source and binary forms, with or without
 // modification, are permitted provided that the following conditions are
 // met:
 //
 //     * Redistributions of source code must retain the above copyright
 //       notice, this list of conditions and the following disclaimer.
 //     * Redistributions in binary form must reproduce the above
@@ skipping 26 matching lines @@
 // - Run-time lexing modifications: harmony number literals, keywords depending
 //   on harmony_modules, harmony_scoping
 // - Escaping the string literals (like the baseline does)
 // - Error recovery after illegal tokens.
 
 enum Condition {
   kConditionNormal,
   kConditionDoubleQuoteString,
   kConditionSingleQuoteString,
   kConditionIdentifier,
+  kConditionIdentifierIllegal,
   kConditionSingleLineComment,
   kConditionMultiLineComment,
   kConditionHtmlComment
 };
 
 #if defined(WIN32)
 
   typedef signed char   int8_t;
   typedef signed short  int16_t;
   typedef signed int   int32_t;
@@ skipping 133 matching lines @@
     re2c:define:YYCTYPE = uint8_t;
     re2c:define:YYLIMIT = limit_;
     re2c:define:YYMARKER = marker_;
 
     eof = "\000";
     any = [\000-\377];
     whitespace_char = [ \t\v\f\r];
     whitespace = whitespace_char+;
     identifier_start_ = [$_a-zA-Z];
     identifier_char = [$_a-zA-Z0-9];
-    not_identifier_char = any\identifier_char;
-    illegal_after_identifier = [\\];
+    not_identifier_char = any\identifier_char\[\\];
     line_terminator = [\n\r]+;
     digit = [0-9];
     hex_digit = [0-9a-fA-F];
     maybe_exponent = ('e' [-+]? digit+)?;
 
     <Normal> "break" not_identifier_char      { PUSH_TOKEN_LOOKAHEAD(Token::BREAK); }
     <Normal> "case" not_identifier_char       { PUSH_TOKEN_LOOKAHEAD(Token::CASE); }
     <Normal> "catch" not_identifier_char      { PUSH_TOKEN_LOOKAHEAD(Token::CATCH); }
     <Normal> "class" not_identifier_char      { PUSH_TOKEN_LOOKAHEAD(Token::FUTURE_RESERVED_WORD); }
     <Normal> "const" not_identifier_char      { PUSH_TOKEN_LOOKAHEAD(Token::CONST); }
@@ skipping 99 matching lines @@
     <Normal> "~"           { PUSH_TOKEN(Token::BIT_NOT); }
     <Normal> ","           { PUSH_TOKEN(Token::COMMA); }
 
     <Normal> line_terminator  { PUSH_LINE_TERMINATOR(); }
     <Normal> whitespace       { SKIP(); }
 
     <Normal> ["]           :=> DoubleQuoteString
     <Normal> [']           :=> SingleQuoteString
 
     <Normal> identifier_start_    :=> Identifier
+    <Normal> "\\u0000"            :=> IdentifierIllegal
+    <Normal> "\\u" [0-9a-fA-F]{4}    :=> Identifier
+    <Normal> "\\"                 { PUSH_TOKEN(Token::ILLEGAL); }
 
     <Normal> eof           { PUSH_EOF_AND_RETURN();}
     <Normal> any           { PUSH_TOKEN(Token::ILLEGAL); }
 
     <DoubleQuoteString> "\\\""  { goto yy0; }
     <DoubleQuoteString> '"'     { PUSH_TOKEN(Token::STRING);}
     <DoubleQuoteString> "\\" "\n" "\r"? { goto yy0; }
     <DoubleQuoteString> "\\" "\r" "\n"? { goto yy0; }
     <DoubleQuoteString> "\n"    => Normal { PUSH_TOKEN_LOOKAHEAD(Token::ILLEGAL); }
     <DoubleQuoteString> "\r"    => Normal { PUSH_TOKEN_LOOKAHEAD(Token::ILLEGAL); }
     <DoubleQuoteString> eof     { TERMINATE_ILLEGAL(); }
     <DoubleQuoteString> any     { goto yy0; }
 
     <SingleQuoteString> "\\'"   { goto yy0; }
     <SingleQuoteString> "'"     { PUSH_TOKEN(Token::STRING);}
     <SingleQuoteString> "\\" "\n" "\r"? { goto yy0; }
     <SingleQuoteString> "\\" "\r" "\n"? { goto yy0; }
     <SingleQuoteString> "\n"    => Normal { PUSH_TOKEN_LOOKAHEAD(Token::ILLEGAL); }
     <SingleQuoteString> "\r"    => Normal { PUSH_TOKEN_LOOKAHEAD(Token::ILLEGAL); }
     <SingleQuoteString> eof     { TERMINATE_ILLEGAL(); }
     <SingleQuoteString> any     { goto yy0; }
 
     <Identifier> identifier_char+  { goto yy0; }
-    <Identifier> illegal_after_identifier { PUSH_TOKEN(Token::ILLEGAL); }
+    <Identifier> "\\u0000"         :=> IdentifierIllegal
+    <Identifier> "\\u" [0-9a-fA-F]{4} { goto yy0; }
+    <Identifier> "\\"              { PUSH_TOKEN(Token::ILLEGAL); }
     <Identifier> any               { PUSH_TOKEN_LOOKAHEAD(Token::IDENTIFIER); }
 
+    <IdentifierIllegal> identifier_char+  { goto yy0; }
+    <IdentifierIllegal> "\\"+              { goto yy0; }
+    <IdentifierIllegal> any               { PUSH_TOKEN_LOOKAHEAD(Token::ILLEGAL); }
+
     <SingleLineComment> line_terminator { PUSH_LINE_TERMINATOR();}
     <SingleLineComment> eof             { PUSH_LINE_TERMINATOR();}
     <SingleLineComment> any             { goto yy0; }
 
     <MultiLineComment> [*][//]  { PUSH_LINE_TERMINATOR();}
     <MultiLineComment> eof      { TERMINATE_ILLEGAL(); }
     <MultiLineComment> any      { goto yy0; }
 
     <HtmlComment> eof        { TERMINATE_ILLEGAL(); }
     <HtmlComment> "-->"      { PUSH_LINE_TERMINATOR();}
@@ skipping 26 matching lines @@
     size_t start_offset = start_ - buffer_;
     memmove(buffer_, start_, limit_ - start_);
     marker_ -= start_offset;
     cursor_ -= start_offset;
     limit_ -= start_offset;
     start_ -= start_offset;
     real_start_ += start_offset;
   }
   return 0;
 }