[Isolates] Static mutable data of Scanner class moved to ScannerCharacterClasses / Isolate.

src/scanner.cc

 // Copyright 2006-2008 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 16 matching lines @@
 
 #include "v8.h"
 
 #include "ast.h"
 #include "handles.h"
 #include "scanner.h"
 
 namespace v8 {
 namespace internal {
 
-// ----------------------------------------------------------------------------
-// Character predicates
-
-
-unibrow::Predicate<IdentifierStart, 128> Scanner::kIsIdentifierStart;
-unibrow::Predicate<IdentifierPart, 128> Scanner::kIsIdentifierPart;
-unibrow::Predicate<unibrow::LineTerminator, 128> Scanner::kIsLineTerminator;
-unibrow::Predicate<unibrow::WhiteSpace, 128> Scanner::kIsWhiteSpace;
-
-
-StaticResource<Scanner::Utf8Decoder> Scanner::utf8_decoder_;
-
 
 // ----------------------------------------------------------------------------
 // UTF8Buffer
 
 UTF8Buffer::UTF8Buffer() : data_(NULL), limit_(NULL) { }
 
 
 UTF8Buffer::~UTF8Buffer() {
   if (data_ != NULL) DeleteArray(data_);
 }
@@ skipping 269 matching lines @@
   }
   // On fallthrough, it's a failure.
   state_ = UNMATCHABLE;
 }
 
 
 // ----------------------------------------------------------------------------
 // Scanner
 
 Scanner::Scanner(ParserMode pre)
-    : stack_overflow_(false), is_pre_parsing_(pre == PREPARSE) { }
+    : stack_overflow_(false), is_pre_parsing_(pre == PREPARSE),
+      character_classes_(Isolate::Current()->
+        scanner_character_classes()) { }
 
 
 void Scanner::Initialize(Handle<String> source,
                          ParserLanguage language) {
   safe_string_input_buffer_.Reset(source.location());
   Init(source, &safe_string_input_buffer_, 0, source->length(), language);
 }
 
 
 void Scanner::Initialize(Handle<String> source,
@@ skipping 121 matching lines @@
   return source_pos() != start_position;
 }
 
 
 bool Scanner::SkipJavaScriptWhiteSpace() {
   int start_position = source_pos();
 
   while (true) {
     // We treat byte-order marks (BOMs) as whitespace for better
     // compatibility with Spidermonkey and other JavaScript engines.
-    while (kIsWhiteSpace.get(c0_) || IsByteOrderMark(c0_)) {
+    while (character_classes_->is_white_space_.get(c0_) ||
+        IsByteOrderMark(c0_)) {
       // IsWhiteSpace() includes line terminators!
-      if (kIsLineTerminator.get(c0_)) {
+      if (character_classes_->is_line_terminator_.get(c0_)) {
         // Ignore line terminators, but remember them. This is necessary
         // for automatic semicolon insertion.
         has_line_terminator_before_next_ = true;
       }
       Advance();
     }
 
     // If there is an HTML comment end '-->' at the beginning of a
     // line (with only whitespace in front of it), we treat the rest
     // of the line as a comment. This is in line with the way
@@ skipping 19 matching lines @@
 
 
 Token::Value Scanner::SkipSingleLineComment() {
   Advance();
 
   // The line terminator at the end of the line is not considered
   // to be part of the single-line comment; it is recognized
   // separately by the lexical grammar and becomes part of the
   // stream of input elements for the syntactic grammar (see
   // ECMA-262, section 7.4, page 12).
-  while (c0_ >= 0 && !kIsLineTerminator.get(c0_)) {
+  while (c0_ >= 0 &&
+      !character_classes_->is_line_terminator_.get(c0_)) {
     Advance();
   }
 
   return Token::WHITESPACE;
 }
 
 
 Token::Value Scanner::SkipMultiLineComment() {
   ASSERT(c0_ == '*');
   Advance();
@@ skipping 208 matching lines @@
 
 
 Token::Value Scanner::ScanJsonIdentifier(const char* text,
                                          Token::Value token) {
   StartLiteral();
   while (*text != '\0') {
     if (c0_ != *text) return Token::ILLEGAL;
     Advance();
     text++;
   }
-  if (kIsIdentifierPart.get(c0_)) return Token::ILLEGAL;
+  if (character_classes_->is_identifier_part_.get(c0_))
+      return Token::ILLEGAL;
   TerminateLiteral();
   return token;
 }
 
 
 void Scanner::ScanJavaScript() {
   next_.literal_buffer = NULL;
   Token::Value token;
   has_line_terminator_before_next_ = false;
   do {
@@ skipping 203 matching lines @@
 
       case '?':
         token = Select(Token::CONDITIONAL);
         break;
 
       case '~':
         token = Select(Token::BIT_NOT);
         break;
 
       default:
-        if (kIsIdentifierStart.get(c0_)) {
+        if (character_classes_->is_identifier_start_.get(c0_)) {
           token = ScanIdentifier();
         } 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);
         }
@@ skipping 58 matching lines @@
   }
   return x;
 }
 
 
 void Scanner::ScanEscape() {
   uc32 c = c0_;
   Advance();
 
   // Skip escaped newlines.
-  if (kIsLineTerminator.get(c)) {
+  if (character_classes_->is_line_terminator_.get(c)) {
     // Allow CR+LF newlines in multiline string literals.
     if (IsCarriageReturn(c) && IsLineFeed(c0_)) Advance();
     // Allow LF+CR newlines in multiline string literals.
     if (IsLineFeed(c) && IsCarriageReturn(c0_)) Advance();
     return;
   }
 
   switch (c) {
     case '\'':  // fall through
     case '"' :  // fall through
@@ skipping 21 matching lines @@
   // as non-escaped characters by JS VMs.
   AddChar(c);
 }
 
 
 Token::Value Scanner::ScanString() {
   uc32 quote = c0_;
   Advance();  // consume quote
 
   StartLiteral();
-  while (c0_ != quote && c0_ >= 0 && !kIsLineTerminator.get(c0_)) {
+  while (c0_ != quote && c0_ >= 0 &&
+      !character_classes_->is_line_terminator_.get(c0_)) {
     uc32 c = c0_;
     Advance();
     if (c == '\\') {
       if (c0_ < 0) return Token::ILLEGAL;
       ScanEscape();
     } else {
       AddChar(c);
     }
   }
   if (c0_ != quote) {
@@ skipping 93 matching lines @@
       // we must have at least one decimal digit after 'e'/'E'
       return Token::ILLEGAL;
     ScanDecimalDigits();
   }
   TerminateLiteral();
 
   // The source character immediately following a numeric literal must
   // not be an identifier start or a decimal digit; see ECMA-262
   // section 7.8.3, page 17 (note that we read only one decimal digit
   // if the value is 0).
-  if (IsDecimalDigit(c0_) || kIsIdentifierStart.get(c0_))
+  if (IsDecimalDigit(c0_) ||
+      character_classes_->is_identifier_start_.get(c0_))
     return Token::ILLEGAL;
 
   return Token::NUMBER;
 }
 
 
 uc32 Scanner::ScanIdentifierUnicodeEscape() {
   Advance();
   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 Scanner::ScanIdentifier() {
-  ASSERT(kIsIdentifierStart.get(c0_));
+  ASSERT(character_classes_->is_identifier_start_.get(c0_));
 
   StartLiteral();
   KeywordMatcher keyword_match;
 
   // Scan identifier start character.
   if (c0_ == '\\') {
     uc32 c = ScanIdentifierUnicodeEscape();
     // Only allow legal identifier start characters.
-    if (!kIsIdentifierStart.get(c)) return Token::ILLEGAL;
+    if (!character_classes_->is_identifier_start_.get(c))
+        return Token::ILLEGAL;
     AddChar(c);
     keyword_match.Fail();
   } else {
     AddChar(c0_);
     keyword_match.AddChar(c0_);
     Advance();
   }
 
   // Scan the rest of the identifier characters.
-  while (kIsIdentifierPart.get(c0_)) {
+  while (character_classes_->is_identifier_part_.get(c0_)) {
     if (c0_ == '\\') {
       uc32 c = ScanIdentifierUnicodeEscape();
       // Only allow legal identifier part characters.
-      if (!kIsIdentifierPart.get(c)) return Token::ILLEGAL;
+      if (!character_classes_->is_identifier_part_.get(c))
+          return Token::ILLEGAL;
       AddChar(c);
       keyword_match.Fail();
     } else {
       AddChar(c0_);
       keyword_match.AddChar(c0_);
       Advance();
     }
   }
   TerminateLiteral();
 
   return keyword_match.token();
 }
 
 
 
-bool Scanner::IsIdentifier(unibrow::CharacterStream* buffer) {
+bool ScannerCharacterClasses::IsIdentifier(unibrow::CharacterStream* buffer) {
   // Checks whether the buffer contains an identifier (no escape).
   if (!buffer->has_more()) return false;
-  if (!kIsIdentifierStart.get(buffer->GetNext())) return false;
+  if (!is_identifier_start_.get(buffer->GetNext())) return false;
   while (buffer->has_more()) {
-    if (!kIsIdentifierPart.get(buffer->GetNext())) return false;
+    if (!is_identifier_part_.get(buffer->GetNext())) return false;
   }
   return true;
 }
 
 
 bool Scanner::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.
   StartLiteral();
   if (seen_equal)
     AddChar('=');
 
   while (c0_ != '/' || in_character_class) {
-    if (kIsLineTerminator.get(c0_) || c0_ < 0)
+    if (character_classes_->is_line_terminator_.get(c0_) || c0_ < 0)
       return false;
     if (c0_ == '\\') {  // escaped character
       AddCharAdvance();
-      if (kIsLineTerminator.get(c0_) || c0_ < 0)
+      if (character_classes_->is_line_terminator_.get(c0_) || c0_ < 0)
         return false;
       AddCharAdvance();
     } else {  // unescaped character
       if (c0_ == '[')
         in_character_class = true;
       if (c0_ == ']')
         in_character_class = false;
       AddCharAdvance();
     }
   }
   Advance();  // consume '/'
 
   TerminateLiteral();
 
   return true;
 }
 
 bool Scanner::ScanRegExpFlags() {
   // Scan regular expression flags.
   StartLiteral();
-  while (kIsIdentifierPart.get(c0_)) {
+  while (character_classes_->is_identifier_part_.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.
         AddChar(c);
         continue;
       }
     }
     AddCharAdvance();
   }
   TerminateLiteral();
 
   next_.location.end_pos = source_pos() - 1;
   return true;
 }
 
 } }  // namespace v8::internal