Merge preparser Scanner with main JavaScript scanner.

src/scanner-base.cc

 // Copyright 2010 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
 //       copyright notice, this list of conditions and the following
 //       disclaimer in the documentation and/or other materials provided
@@ skipping 462 matching lines @@
       case '?':
         token = Select(Token::CONDITIONAL);
         break;
 
       case '~':
         token = Select(Token::BIT_NOT);
         break;
 
       default:
         if (ScannerConstants::kIsIdentifierStart.get(c0_)) {
-          token = ScanIdentifier();
+          token = ScanIdentifierOrKeyword();
         } else if (IsDecimalDigit(c0_)) {
           token = ScanNumber(false);
         } else if (SkipWhiteSpace()) {
           token = Token::WHITESPACE;
         } else if (c0_ < 0) {
           token = Token::EOS;
         } else {
           token = Select(Token::ILLEGAL);
         }
         break;
@@ skipping 58 matching lines @@
   // should be illegal, but they are commonly handled
   // as non-escaped characters by JS VMs.
   AddLiteralChar(c);
 }
 
 
 Token::Value JavaScriptScanner::ScanString() {
   uc32 quote = c0_;
   Advance();  // consume quote
 
-  LiteralScope literal(this);
+  LiteralScope literal(this, kLiteralString);
   while (c0_ != quote && c0_ >= 0
          && !ScannerConstants::kIsLineTerminator.get(c0_)) {
     uc32 c = c0_;
     Advance();
     if (c == '\\') {
       if (c0_ < 0) return Token::ILLEGAL;
       ScanEscape();
     } else {
       AddLiteralChar(c);
     }
@@ skipping 10 matching lines @@
   while (IsDecimalDigit(c0_))
     AddLiteralCharAdvance();
 }
 
 
 Token::Value JavaScriptScanner::ScanNumber(bool seen_period) {
   ASSERT(IsDecimalDigit(c0_));  // the first digit of the number or the fraction
 
   enum { DECIMAL, HEX, OCTAL } kind = DECIMAL;
 
-  LiteralScope literal(this);
+  LiteralScope literal(this, kLiteralNumber);
   if (seen_period) {
     // we have already seen a decimal point of the float
     AddLiteralChar('.');
     ScanDecimalDigits();  // we know we have at least one digit
 
   } else {
     // if the first character is '0' we must check for octals and hex
     if (c0_ == '0') {
       AddLiteralCharAdvance();
 
@@ skipping 66 matching lines @@
   if (c0_ != 'u') return unibrow::Utf8::kBadChar;
   Advance();
   uc32 c = ScanHexEscape('u', 4);
   // We do not allow a unicode escape sequence to start another
   // unicode escape sequence.
   if (c == '\\') return unibrow::Utf8::kBadChar;
   return c;
 }
 
 
-Token::Value JavaScriptScanner::ScanIdentifier() {
+Token::Value JavaScriptScanner::ScanIdentifierOrKeyword() {
   ASSERT(ScannerConstants::kIsIdentifierStart.get(c0_));
-
-  LiteralScope literal(this);
+  LiteralScope literal(this, kLiteralIdentifier);
   KeywordMatcher keyword_match;
-
   // Scan identifier start character.
   if (c0_ == '\\') {
     uc32 c = ScanIdentifierUnicodeEscape();
     // Only allow legal identifier start characters.
     if (!ScannerConstants::kIsIdentifierStart.get(c)) return Token::ILLEGAL;
     AddLiteralChar(c);
-    keyword_match.Fail();
-  } else {
-    AddLiteralChar(c0_);
-    keyword_match.AddChar(c0_);
-    Advance();
+    return ScanIdentifierSuffix(&literal);
+  }
+
+  uc32 first_char = c0_;
+  Advance();
+  AddLiteralChar(first_char);
+  if (!keyword_match.AddChar(first_char)) {
+    return ScanIdentifierSuffix(&literal);
   }
 
   // Scan the rest of the identifier characters.
   while (ScannerConstants::kIsIdentifierPart.get(c0_)) {
+    if (c0_ != '\\') {
+      uc32 next_char = c0_;
+      Advance();
+      AddLiteralChar(next_char);
+      if (keyword_match.AddChar(next_char)) continue;
+    }
+    // Fallthrough if no loner able to complete keyword.
+    return ScanIdentifierSuffix(&literal);
+  }
+  literal.Complete();
+
+  return keyword_match.token();
+}
+
+
+Token::Value JavaScriptScanner::ScanIdentifierSuffix(LiteralScope* literal) {
+  // Scan the rest of the identifier characters.
+  while (ScannerConstants::kIsIdentifierPart.get(c0_)) {
     if (c0_ == '\\') {
       uc32 c = ScanIdentifierUnicodeEscape();
       // Only allow legal identifier part characters.
       if (!ScannerConstants::kIsIdentifierPart.get(c)) return Token::ILLEGAL;
       AddLiteralChar(c);
-      keyword_match.Fail();
     } else {
       AddLiteralChar(c0_);
-      keyword_match.AddChar(c0_);
       Advance();
     }
   }
-  literal.Complete();
+  literal->Complete();
 
-  return keyword_match.token();
+  return Token::IDENTIFIER;
 }
 
 
-
 bool JavaScriptScanner::ScanRegExpPattern(bool seen_equal) {
   // Scan: ('/' | '/=') RegularExpressionBody '/' RegularExpressionFlags
   bool in_character_class = false;
 
   // Previous token is either '/' or '/=', in the second case, the
   // pattern starts at =.
   next_.location.beg_pos = source_pos() - (seen_equal ? 2 : 1);
   next_.location.end_pos = source_pos() - (seen_equal ? 1 : 0);
 
   // Scan regular expression body: According to ECMA-262, 3rd, 7.8.5,
   // the scanner should pass uninterpreted bodies to the RegExp
   // constructor.
-  LiteralScope literal(this);
+  LiteralScope literal(this, kLiteralRegExp);
   if (seen_equal)
     AddLiteralChar('=');
 
   while (c0_ != '/' || in_character_class) {
     if (ScannerConstants::kIsLineTerminator.get(c0_) || c0_ < 0) return false;
     if (c0_ == '\\') {  // escaped character
       AddLiteralCharAdvance();
       if (ScannerConstants::kIsLineTerminator.get(c0_) || c0_ < 0) return false;
       AddLiteralCharAdvance();
     } else {  // unescaped character
       if (c0_ == '[') in_character_class = true;
       if (c0_ == ']') in_character_class = false;
       AddLiteralCharAdvance();
     }
   }
   Advance();  // consume '/'
 
   literal.Complete();
 
   return true;
 }
 
+
 bool JavaScriptScanner::ScanRegExpFlags() {
   // Scan regular expression flags.
-  LiteralScope literal(this);
+  LiteralScope literal(this, kLiteralRegExpFlags);
   while (ScannerConstants::kIsIdentifierPart.get(c0_)) {
     if (c0_ == '\\') {
       uc32 c = ScanIdentifierUnicodeEscape();
       if (c != static_cast<uc32>(unibrow::Utf8::kBadChar)) {
         // We allow any escaped character, unlike the restriction on
         // IdentifierPart when it is used to build an IdentifierName.
         AddLiteralChar(c);
         continue;
       }
     }
@@ skipping 93 matching lines @@
       if (MatchKeywordStart(input, "finally", 1, Token::FINALLY)) return;
       if (MatchKeywordStart(input, "for", 1, Token::FOR)) return;
       if (MatchKeywordStart(input, "function", 1, Token::FUNCTION)) return;
       break;
     case I:
       if (MatchKeyword(input, 'f', KEYWORD_MATCHED, Token::IF)) return;
       if (MatchKeyword(input, 'n', IN, Token::IN)) return;
       break;
     case IN:
       token_ = Token::IDENTIFIER;
-      if (MatchKeywordStart(input, "instanceof", 2, Token::INSTANCEOF)) {
-        return;
-      }
+      if (MatchKeywordStart(input, "instanceof", 2, Token::INSTANCEOF)) return;
       break;
     case N:
       if (MatchKeywordStart(input, "native", 1, Token::NATIVE)) return;
       if (MatchKeywordStart(input, "new", 1, Token::NEW)) return;
       if (MatchKeywordStart(input, "null", 1, Token::NULL_LITERAL)) return;
       break;
     case T:
       if (MatchState(input, 'h', TH)) return;
       if (MatchState(input, 'r', TR)) return;
       if (MatchKeywordStart(input, "typeof", 1, Token::TYPEOF)) return;
@@ skipping 15 matching lines @@
       if (MatchKeywordStart(input, "with", 1, Token::WITH)) return;
       break;
     case UNMATCHABLE:
       break;
   }
   // On fallthrough, it's a failure.
   state_ = UNMATCHABLE;
 }
 
 } }  // namespace v8::internal