cpp_slex_lexer.hpp

C++的一个好库。。。现在很流行

/*=============================================================================
    Boost.Wave: A Standard compliant C++ preprocessor library

    SLex (Spirit Lex) based C++ lexer
    
    http://www.boost.org/

    Copyright (c) 2001-2005 Hartmut Kaiser. Distributed under the Boost 
    Software License, Version 1.0. (See accompanying file 
    LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
=============================================================================*/

#if !defined(SLEX_LEXER_HPP_5E8E1DF0_BB41_4938_B7E5_A4BB68222FF6_INCLUDED)
#define SLEX_LEXER_HPP_5E8E1DF0_BB41_4938_B7E5_A4BB68222FF6_INCLUDED

#include <string>
#if defined(BOOST_SPIRIT_DEBUG)
#include <iostream>
#endif // defined(BOOST_SPIRIT_DEBUG)

#include <boost/assert.hpp>
#include <boost/spirit/core.hpp>

#include <boost/wave/wave_config.hpp>
#include <boost/wave/language_support.hpp>
#include <boost/wave/token_ids.hpp>
#include <boost/wave/util/file_position.hpp>
#include <boost/wave/util/time_conversion_helper.hpp>
#include <boost/wave/cpplexer/validate_universal_char.hpp>
#include <boost/wave/cpplexer/convert_trigraphs.hpp>
#include <boost/wave/cpplexer/cpplexer_exceptions.hpp>

#include "../slex_interface.hpp"
#include "../slex_token.hpp"
#include "../slex_iterator.hpp"

#include "lexer.hpp"   // "spirit/lexer.hpp"

///////////////////////////////////////////////////////////////////////////////
namespace boost {
namespace wave {
namespace cpplexer {
namespace slex {
namespace lexer {

///////////////////////////////////////////////////////////////////////////////
//  The following numbers are the arraysizes of the token regex's which we
//  need to specify to make the CW compiler happy (at least up to V9.5).
#if BOOST_WAVE_SUPPORT_MS_EXTENSIONS != 0
#define INIT_DATA_SIZE        176
#define INIT_DATA_CPP_SIZE    15
#else
#define INIT_DATA_SIZE        159
#define INIT_DATA_CPP_SIZE    15
#endif

///////////////////////////////////////////////////////////////////////////////
// 
//  encapsulation of the boost::spirit::slex based cpp lexer
//
///////////////////////////////////////////////////////////////////////////////

///////////////////////////////////////////////////////////////////////////////
//  The following lexer_base class was necessary to workaround a CodeWarrior 
//  bug (at least up to CW V9.5).
template <typename IteratorT, typename PositionT>
class lexer_base 
:   public boost::spirit::lexer<
        boost::wave::util::position_iterator<IteratorT, PositionT> >
{
protected:
    typedef boost::wave::util::position_iterator<IteratorT, PositionT> 
        iterator_type;
    typedef typename std::iterator_traits<IteratorT>::value_type  char_type;
    typedef boost::spirit::lexer<iterator_type>                   base_type;

    lexer_base();
    
// initialization data (regular expressions for the token definitions)
    struct lexer_data {
        token_id tokenid;                       // token data
        char_type const *tokenregex;            // associated token to match
        typename base_type::callback_t tokencb; // associated callback function
        unsigned int lexerstate;                // valid for lexer state
    };
};

///////////////////////////////////////////////////////////////////////////////
template <typename IteratorT, typename PositionT>
class lexer 
:   public lexer_base<IteratorT, PositionT>
{
public:
    typedef boost::wave::cpplexer::slex_token<PositionT>  token_type;
    
    void init_dfa(boost::wave::language_support language);

// get time of last compilation
    static std::time_t get_compilation_time() 
        { return compilation_time.get_time(); }

// helper for calculation of the time of last compilation
    static boost::wave::util::time_conversion_helper compilation_time;

private:
    typedef lexer_base<IteratorT, PositionT> base_type;

    static typename base_type::lexer_data const init_data[INIT_DATA_SIZE];          // common patterns
    static typename base_type::lexer_data const init_data_cpp[INIT_DATA_CPP_SIZE];  // C++ only patterns
};

///////////////////////////////////////////////////////////////////////////////
//  data required for initialization of the lexer (token definitions)
#define OR      "|"
#define Q(c)    "\\" c
#define TRI(c)  Q("?") Q("?") c

// definition of some subtoken regexps to simplify the regex definitions
#define BLANK               "[ \\t]"
#define CCOMMENT            \
    Q("/") Q("*") "[^*]*" Q("*") "+" "(" "[^/*][^*]*" Q("*") "+" ")*" Q("/")
        
#define PPSPACE             "(" BLANK OR CCOMMENT ")*"

#define OCTALDIGIT          "[0-7]"
#define DIGIT               "[0-9]"
#define HEXDIGIT            "[0-9a-fA-F]"
#define SIGN                "[-+]?"
#define EXPONENT            "(" "[eE]" SIGN "[0-9]+" ")"

#define INTEGER             \
    "(" "(0x|0X)" HEXDIGIT "+" OR "0" OCTALDIGIT "*" OR "[1-9]" DIGIT "*" ")"
            
#define INTEGER_SUFFIX      "(" "[uU][lL]?|[lL][uU]?" ")"
#define LONGINTEGER_SUFFIX  "(" "[uU]" "(" "ll" OR "LL" ")" OR \
                            "(" "ll" OR "LL" ")" "[uU]" "?" ")"
#define FLOAT_SUFFIX        "(" "[fF][lL]?|[lL][fF]?" ")"
#define CHAR_SPEC           "L?"

#define BACKSLASH           "(" Q("\\") OR TRI(Q("/")) ")"
#define ESCAPESEQ           BACKSLASH "(" \
                                "[abfnrtv?'\"]" OR \
                                BACKSLASH OR \
                                "x" HEXDIGIT "+" OR \
                                OCTALDIGIT OCTALDIGIT "?" OCTALDIGIT "?" \
                            ")"
#define HEXQUAD             HEXDIGIT HEXDIGIT HEXDIGIT HEXDIGIT 
#define UNIVERSALCHAR       BACKSLASH "(" \
                                "u" HEXQUAD OR \
                                "U" HEXQUAD HEXQUAD \
                            ")" 

#define POUNDDEF            "(" "#" OR TRI("=") OR Q("%:") ")"
#define NEWLINEDEF          "(" "\\n" OR "\\r" OR "\\r\\n" ")"

#if BOOST_WAVE_SUPPORT_INCLUDE_NEXT != 0
#define INCLUDEDEF          "(include|include_next)"
#else
#define INCLUDEDEF          "include"
#endif

///////////////////////////////////////////////////////////////////////////////
//  sexer state constants
#define LEXER_STATE_NORMAL  0
#define LEXER_STATE_PP      1

#define NUM_LEXER_STATES    1

//  helper for initializing token data
#define TOKEN_DATA(id, regex)                                                 \
        { T_##id, regex, 0, LEXER_STATE_NORMAL }                              \
    /**/

#define TOKEN_DATA_EX(id, regex, callback)                                    \
        { T_##id, regex, callback, LEXER_STATE_NORMAL }                       \
    /**/

///////////////////////////////////////////////////////////////////////////////
// common C++/C99 token definitions
template <typename IteratorT, typename PositionT>
typename lexer_base<IteratorT, PositionT>::lexer_data const 
lexer<IteratorT, PositionT>::init_data[INIT_DATA_SIZE] = 
{
    TOKEN_DATA(AND, "&"),
    TOKEN_DATA(ANDAND, "&&"),
    TOKEN_DATA(ASSIGN, "="),
    TOKEN_DATA(ANDASSIGN, "&="),
    TOKEN_DATA(OR, Q("|")),
    TOKEN_DATA(OR_TRIGRAPH, TRI("!")),
    TOKEN_DATA(ORASSIGN, Q("|=")),
    TOKEN_DATA(ORASSIGN_TRIGRAPH, TRI("!=")),
    TOKEN_DATA(XOR, Q("^")),
    TOKEN_DATA(XOR_TRIGRAPH, TRI("'")),
    TOKEN_DATA(XORASSIGN, Q("^=")),
    TOKEN_DATA(XORASSIGN_TRIGRAPH, TRI("'=")),
    TOKEN_DATA(COMMA, ","),
    TOKEN_DATA(COLON, ":"),
    TOKEN_DATA(DIVIDEASSIGN, Q("/=")),
    TOKEN_DATA(DIVIDE, Q("/")),
    TOKEN_DATA(DOT, Q(".")),
    TOKEN_DATA(ELLIPSIS, Q(".") Q(".") Q(".")),
    TOKEN_DATA(EQUAL, "=="),
    TOKEN_DATA(GREATER, ">"),
    TOKEN_DATA(GREATEREQUAL, ">="),
    TOKEN_DATA(LEFTBRACE, Q("{")),
    TOKEN_DATA(LEFTBRACE_ALT, "<" Q("%")),
    TOKEN_DATA(LEFTBRACE_TRIGRAPH, TRI("<")),
    TOKEN_DATA(LESS, "<"),
    TOKEN_DATA(LESSEQUAL, "<="),
    TOKEN_DATA(LEFTPAREN, Q("(")),
    TOKEN_DATA(LEFTBRACKET, Q("[")),
    TOKEN_DATA(LEFTBRACKET_ALT, "<:"),
    TOKEN_DATA(LEFTBRACKET_TRIGRAPH, TRI(Q("("))),
    TOKEN_DATA(MINUS, Q("-")),
    TOKEN_DATA(MINUSASSIGN, Q("-=")),
    TOKEN_DATA(MINUSMINUS, Q("-") Q("-")),
    TOKEN_DATA(PERCENT, Q("%")),
    TOKEN_DATA(PERCENTASSIGN, Q("%=")),
    TOKEN_DATA(NOT, "!"),
    TOKEN_DATA(NOTEQUAL, "!="),
    TOKEN_DATA(OROR, Q("|") Q("|")),
    TOKEN_DATA(OROR_TRIGRAPH, TRI("!") Q("|") OR Q("|") TRI("!") OR TRI("!") TRI("!")),
    TOKEN_DATA(PLUS, Q("+")),
    TOKEN_DATA(PLUSASSIGN, Q("+=")),
    TOKEN_DATA(PLUSPLUS, Q("+") Q("+")),
    TOKEN_DATA(ARROW, Q("->")),
    TOKEN_DATA(QUESTION_MARK, Q("?")),
    TOKEN_DATA(RIGHTBRACE, Q("}")),
    TOKEN_DATA(RIGHTBRACE_ALT, Q("%>")),
    TOKEN_DATA(RIGHTBRACE_TRIGRAPH, TRI(">")),
    TOKEN_DATA(RIGHTPAREN, Q(")")),
    TOKEN_DATA(RIGHTBRACKET, Q("]")),
    TOKEN_DATA(RIGHTBRACKET_ALT, ":>"),
    TOKEN_DATA(RIGHTBRACKET_TRIGRAPH, TRI(Q(")"))),
    TOKEN_DATA(SEMICOLON, ";"),
    TOKEN_DATA(SHIFTLEFT, "<<"),
    TOKEN_DATA(SHIFTLEFTASSIGN, "<<="),
    TOKEN_DATA(SHIFTRIGHT, ">>"),
    TOKEN_DATA(SHIFTRIGHTASSIGN, ">>="),
    TOKEN_DATA(STAR, Q("*")),
    TOKEN_DATA(COMPL, Q("~")),
    TOKEN_DATA(COMPL_TRIGRAPH, TRI("-")),
    TOKEN_DATA(STARASSIGN, Q("*=")),
    TOKEN_DATA(ASM, "asm"),
    TOKEN_DATA(AUTO, "auto"),
    TOKEN_DATA(BOOL, "bool"),
    TOKEN_DATA(FALSE, "false"),
    TOKEN_DATA(TRUE, "true"),
    TOKEN_DATA(BREAK, "break"),
    TOKEN_DATA(CASE, "case"),
    TOKEN_DATA(CATCH, "catch"),
    TOKEN_DATA(CHAR, "char"),
    TOKEN_DATA(CLASS, "class"),
    TOKEN_DATA(CONST, "const"),
    TOKEN_DATA(CONSTCAST, "const_cast"),
    TOKEN_DATA(CONTINUE, "continue"),
    TOKEN_DATA(DEFAULT, "default"),
//    TOKEN_DATA(DEFINED, "defined"),
    TOKEN_DATA(DELETE, "delete"),
    TOKEN_DATA(DO, "do"),
    TOKEN_DATA(DOUBLE, "double"),
    TOKEN_DATA(DYNAMICCAST, "dynamic_cast"),
    TOKEN_DATA(ELSE, "else"),
    TOKEN_DATA(ENUM, "enum"),
    TOKEN_DATA(EXPLICIT, "explicit"),
    TOKEN_DATA(EXPORT, "export"),
    TOKEN_DATA(EXTERN, "extern"),
    TOKEN_DATA(FLOAT, "float"),
    TOKEN_DATA(FOR, "for"),
    TOKEN_DATA(FRIEND, "friend"),
    TOKEN_DATA(GOTO, "goto"),
    TOKEN_DATA(IF, "if"),
    TOKEN_DATA(INLINE, "inline"),
    TOKEN_DATA(INT, "int"),
    TOKEN_DATA(LONG, "long"),
    TOKEN_DATA(MUTABLE, "mutable"),
    TOKEN_DATA(NAMESPACE, "namespace"),
    TOKEN_DATA(NEW, "new"),
    TOKEN_DATA(OPERATOR, "operator"),
    TOKEN_DATA(PRIVATE, "private"),
    TOKEN_DATA(PROTECTED, "protected"),
    TOKEN_DATA(PUBLIC, "public"),
    TOKEN_DATA(REGISTER, "register"),
    TOKEN_DATA(REINTERPRETCAST, "reinterpret_cast"),
    TOKEN_DATA(RETURN, "return"),
    TOKEN_DATA(SHORT, "short"),
    TOKEN_DATA(SIGNED, "signed"),
    TOKEN_DATA(SIZEOF, "sizeof"),
    TOKEN_DATA(STATIC, "static"),
    TOKEN_DATA(STATICCAST, "static_cast"),
    TOKEN_DATA(STRUCT, "struct"),
    TOKEN_DATA(SWITCH, "switch"),
    TOKEN_DATA(TEMPLATE, "template"),
    TOKEN_DATA(THIS, "this"),
    TOKEN_DATA(THROW, "throw"),
    TOKEN_DATA(TRY, "try"),
    TOKEN_DATA(TYPEDEF, "typedef"),
    TOKEN_DATA(TYPEID, "typeid"),
    TOKEN_DATA(TYPENAME, "typename"),
    TOKEN_DATA(UNION, "union"),
    TOKEN_DATA(UNSIGNED, "unsigned"),
    TOKEN_DATA(USING, "using"),
    TOKEN_DATA(VIRTUAL, "virtual"),
    TOKEN_DATA(VOID, "void"),
    TOKEN_DATA(VOLATILE, "volatile"),
    TOKEN_DATA(WCHART, "wchar_t"),
    TOKEN_DATA(WHILE, "while"),
    TOKEN_DATA(PP_DEFINE, POUNDDEF PPSPACE "define"),
    TOKEN_DATA(PP_IF, POUNDDEF PPSPACE "if"),
    TOKEN_DATA(PP_IFDEF, POUNDDEF PPSPACE "ifdef"),
    TOKEN_DATA(PP_IFNDEF, POUNDDEF PPSPACE "ifndef"),
    TOKEN_DATA(PP_ELSE, POUNDDEF PPSPACE "else"),
    TOKEN_DATA(PP_ELIF, POUNDDEF PPSPACE "elif"),
    TOKEN_DATA(PP_ENDIF, POUNDDEF PPSPACE "endif"),
    TOKEN_DATA(PP_ERROR, POUNDDEF PPSPACE "error"),
    TOKEN_DATA(PP_QHEADER, POUNDDEF PPSPACE \
        INCLUDEDEF PPSPACE Q("\"") "[^\\n\\r\"]+" Q("\"")),
    TOKEN_DATA(PP_HHEADER, POUNDDEF PPSPACE \
        INCLUDEDEF PPSPACE "<" "[^\\n\\r>]+" ">"),
    TOKEN_DATA(PP_INCLUDE, POUNDDEF PPSPACE \
        INCLUDEDEF PPSPACE),
    TOKEN_DATA(PP_LINE, POUNDDEF PPSPACE "line"),
    TOKEN_DATA(PP_PRAGMA, POUNDDEF PPSPACE "pragma"),
    TOKEN_DATA(PP_UNDEF, POUNDDEF PPSPACE "undef"),
    TOKEN_DATA(PP_WARNING, POUNDDEF PPSPACE "warning"),
#if BOOST_WAVE_SUPPORT_MS_EXTENSIONS != 0
    TOKEN_DATA(MSEXT_INT8, "__int8"),
    TOKEN_DATA(MSEXT_INT16, "__int16"),
    TOKEN_DATA(MSEXT_INT32, "__int32"),
    TOKEN_DATA(MSEXT_INT64, "__int64"),
    TOKEN_DATA(MSEXT_BASED, "_?" "_based"),
    TOKEN_DATA(MSEXT_DECLSPEC, "_?" "_declspec"),
    TOKEN_DATA(MSEXT_CDECL, "_?" "_cdecl"),
    TOKEN_DATA(MSEXT_FASTCALL, "_?" "_fastcall"),
    TOKEN_DATA(MSEXT_STDCALL, "_?" "_stdcall"),
    TOKEN_DATA(MSEXT_TRY , "__try"),
    TOKEN_DATA(MSEXT_EXCEPT, "__except"),
    TOKEN_DATA(MSEXT_FINALLY, "__finally"),
    TOKEN_DATA(MSEXT_LEAVE, "__leave"),
    TOKEN_DATA(MSEXT_INLINE, "_?" "_inline"),
    TOKEN_DATA(MSEXT_ASM, "_?" "_asm"),
    TOKEN_DATA(MSEXT_PP_REGION, POUNDDEF PPSPACE "region"),
    TOKEN_DATA(MSEXT_PP_ENDREGION, POUNDDEF PPSPACE "endregion"),
#endif // BOOST_WAVE_SUPPORT_MS_EXTENSIONS != 0
    TOKEN_DATA(IDENTIFIER, "([a-zA-Z_]" OR UNIVERSALCHAR ")([a-zA-Z0-9_]" OR UNIVERSALCHAR ")*"),
//  TOKEN_DATA(OCTALINT, "0" OCTALDIGIT "*" INTEGER_SUFFIX "?"),
//  TOKEN_DATA(DECIMALINT, "[1-9]" DIGIT "*" INTEGER_SUFFIX "?"),
//  TOKEN_DATA(HEXAINT, "(0x|0X)" HEXDIGIT "+" INTEGER_SUFFIX "?"),
    TOKEN_DATA(INTLIT, INTEGER INTEGER_SUFFIX "?"),
    TOKEN_DATA(LONGINTLIT, INTEGER LONGINTEGER_SUFFIX),
    TOKEN_DATA(FLOATLIT, 
        "(" DIGIT "*" Q(".") DIGIT "+" OR DIGIT "+" Q(".") ")" 
        EXPONENT "?" FLOAT_SUFFIX "?" OR
        DIGIT "+" EXPONENT FLOAT_SUFFIX "?"),
    TOKEN_DATA(CCOMMENT, CCOMMENT),
    TOKEN_DATA(CPPCOMMENT, Q("/") Q("/[^\\n\\r]*") NEWLINEDEF ),
    TOKEN_DATA(CHARLIT, CHAR_SPEC "'"