kjs/lexer.h

   1 // -*- c-basic-offset: 2 -*-
   2 /*
   3  *  This file is part of the KDE libraries
   4  *  Copyright (C) 1999-2000 Harri Porten (porten@kde.org)
   5  *  Copyright (C) 2007 Apple Inc.
   6  *
   7  *  This library is free software; you can redistribute it and/or
   8  *  modify it under the terms of the GNU Library General Public
   9  *  License as published by the Free Software Foundation; either
  10  *  version 2 of the License, or (at your option) any later version.
  11  *
  12  *  This library is distributed in the hope that it will be useful,
  13  *  but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  15  *  Library General Public License for more details.
  16  *
  17  *  You should have received a copy of the GNU Library General Public License
  18  *  along with this library; see the file COPYING.LIB.  If not, write to
  19  *  the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
  20  *  Boston, MA 02110-1301, USA.
  21  *
  22  */
  23
  24 #ifndef Lexer_h
  25 #define Lexer_h
  26
  27 #include "ustring.h"
  28 #include <wtf/Vector.h>
  29
  30 namespace KJS {
  31
  32   class Identifier;
  33   class RegExp;
  34
  35   class Lexer : Noncopyable {
  36   public:
  37     void setCode(int startingLineNumber, const UChar *c, unsigned int len);
  38     int lex();
  39
  40     int lineNo() const { return yylineno; }
  41
  42     bool prevTerminator() const { return terminator; }
  43
  44     enum State { Start,
  45                  IdentifierOrKeyword,
  46                  Identifier,
  47                  InIdentifierOrKeyword,
  48                  InIdentifier,
  49                  InIdentifierStartUnicodeEscapeStart,
  50                  InIdentifierStartUnicodeEscape,
  51                  InIdentifierPartUnicodeEscapeStart,
  52                  InIdentifierPartUnicodeEscape,
  53                  InSingleLineComment,
  54                  InMultiLineComment,
  55                  InNum,
  56                  InNum0,
  57                  InHex,
  58                  InOctal,
  59                  InDecimal,
  60                  InExponentIndicator,
  61                  InExponent,
  62                  Hex,
  63                  Octal,
  64                  Number,
  65                  String,
  66                  Eof,
  67                  InString,
  68                  InEscapeSequence,
  69                  InHexEscape,
  70                  InUnicodeEscape,
  71                  Other,
  72                  Bad };
  73
  74     bool scanRegExp();
  75     const UString& pattern() const { return m_pattern; }
  76     const UString& flags() const { return m_flags; }
  77
  78     static unsigned char convertHex(int);
  79     static unsigned char convertHex(int c1, int c2);
  80     static UChar convertUnicode(int c1, int c2, int c3, int c4);
  81     static bool isIdentStart(int);
  82     static bool isIdentPart(int);
  83     static bool isHexDigit(int);
  84
  85     bool sawError() const { return error; }
  86
  87     void clear();
  88
  89   private:
  90     friend Lexer& lexer();
  91     Lexer();
  92
  93     int yylineno;
  94     bool done;
  95     Vector<char> m_buffer8;
  96     Vector<UChar> m_buffer16;
  97     bool terminator;
  98     bool restrKeyword;
  99     // encountered delimiter like "'" and "}" on last run
 100     bool delimited;
 101     bool skipLF;
 102     bool skipCR;
 103     bool eatNextIdentifier;
 104     int stackToken;
 105     int lastToken;
 106
 107     State state;
 108     void setDone(State);
 109     unsigned int pos;
 110     void shift(unsigned int p);
 111     void nextLine();
 112     int lookupKeyword(const char *);
 113
 114     bool isWhiteSpace() const;
 115     bool isLineTerminator();
 116     static bool isOctalDigit(int);
 117
 118     int matchPunctuator(int c1, int c2, int c3, int c4);
 119     static unsigned short singleEscape(unsigned short);
 120     static unsigned short convertOctal(int c1, int c2, int c3);
 121
 122     void record8(int);
 123     void record16(int);
 124     void record16(UChar);
 125
 126     KJS::Identifier* makeIdentifier(const Vector<UChar>& buffer);
 127     UString* makeUString(const Vector<UChar>& buffer);
 128
 129     const UChar* code;
 130     unsigned int length;
 131     int yycolumn;
 132     int atLineStart;
 133     bool error;
 134
 135     // current and following unicode characters (int to allow for -1 for end-of-file marker)
 136     int current, next1, next2, next3;
 137
 138     Vector<UString*> m_strings;
 139     Vector<KJS::Identifier*> m_identifiers;
 140
 141     UString m_pattern;
 142     UString m_flags;
 143   };
 144
 145   Lexer& lexer(); // Returns the singletone JavaScript lexer.
 146
 147 } // namespace KJS
 148
 149 #endif // Lexer_h