ceph/src/boost/libs/wave/samples/cpp_tokens/slex/cpp_slex_lexer.hpp

   1 /*=============================================================================
   2     Boost.Wave: A Standard compliant C++ preprocessor library
   3
   4     SLex (Spirit Lex) based C++ lexer
   5
   6     http://www.boost.org/
   7
   8     Copyright (c) 2001-2012 Hartmut Kaiser. Distributed under the Boost
   9     Software License, Version 1.0. (See accompanying file
  10     LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
  11 =============================================================================*/
  12
  13 #if !defined(SLEX_LEXER_HPP_5E8E1DF0_BB41_4938_B7E5_A4BB68222FF6_INCLUDED)
  14 #define SLEX_LEXER_HPP_5E8E1DF0_BB41_4938_B7E5_A4BB68222FF6_INCLUDED
  15
  16 #include <string>
  17 #if defined(BOOST_SPIRIT_DEBUG)
  18 #include <iostream>
  19 #endif // defined(BOOST_SPIRIT_DEBUG)
  20
  21 #include <boost/assert.hpp>
  22 #include <boost/spirit/include/classic_core.hpp>
  23
  24 #include <boost/wave/wave_config.hpp>
  25 #include <boost/wave/language_support.hpp>
  26 #include <boost/wave/token_ids.hpp>
  27 #include <boost/wave/util/file_position.hpp>
  28 #include <boost/wave/util/time_conversion_helper.hpp>
  29 #include <boost/wave/cpplexer/validate_universal_char.hpp>
  30 #include <boost/wave/cpplexer/convert_trigraphs.hpp>
  31 #include <boost/wave/cpplexer/cpplexer_exceptions.hpp>
  32 #if BOOST_WAVE_SUPPORT_PRAGMA_ONCE != 0
  33 #include <boost/wave/cpplexer/detect_include_guards.hpp>
  34 #endif
  35 #include <boost/wave/cpplexer/cpp_lex_interface.hpp>
  36
  37 #include "../slex_interface.hpp"
  38 #include "../slex_token.hpp"
  39 #include "../slex_iterator.hpp"
  40
  41 #include "lexer.hpp"   // "spirit/lexer.hpp"
  42
  43 ///////////////////////////////////////////////////////////////////////////////
  44 namespace boost {
  45 namespace wave {
  46 namespace cpplexer {
  47 namespace slex {
  48 namespace lexer {
  49
  50 ///////////////////////////////////////////////////////////////////////////////
  51 //  The following numbers are the array sizes of the token regex's which we
  52 //  need to specify to make the CW compiler happy (at least up to V9.5).
  53 #if BOOST_WAVE_SUPPORT_MS_EXTENSIONS != 0
  54 #define INIT_DATA_SIZE              175
  55 #else
  56 #define INIT_DATA_SIZE              158
  57 #endif
  58 #define INIT_DATA_CPP_SIZE          15
  59 #define INIT_DATA_PP_NUMBER_SIZE    2
  60 #define INIT_DATA_CPP0X_SIZE        15
  61
  62 ///////////////////////////////////////////////////////////////////////////////
  63 //
  64 //  encapsulation of the boost::spirit::classic::slex based cpp lexer
  65 //
  66 ///////////////////////////////////////////////////////////////////////////////
  67
  68 ///////////////////////////////////////////////////////////////////////////////
  69 //  The following lexer_base class was necessary to workaround a CodeWarrior
  70 //  bug (at least up to CW V9.5).
  71 template <typename IteratorT, typename PositionT>
  72 class lexer_base
  73 :   public boost::spirit::classic::lexer<
  74         boost::wave::util::position_iterator<IteratorT, PositionT> >
  75 {
  76 protected:
  77     typedef boost::wave::util::position_iterator<IteratorT, PositionT>
  78         iterator_type;
  79     typedef typename std::iterator_traits<IteratorT>::value_type  char_type;
  80     typedef boost::spirit::classic::lexer<iterator_type> base_type;
  81
  82     lexer_base();
  83
  84 // initialization data (regular expressions for the token definitions)
  85     struct lexer_data {
  86         token_id tokenid;                       // token data
  87         char_type const *tokenregex;            // associated token to match
  88         typename base_type::callback_t tokencb; // associated callback function
  89         unsigned int lexerstate;                // valid for lexer state
  90     };
  91 };
  92
  93 ///////////////////////////////////////////////////////////////////////////////
  94 template <typename IteratorT, typename PositionT>
  95 class lexer
  96 :   public lexer_base<IteratorT, PositionT>
  97 {
  98 public:
  99     typedef boost::wave::cpplexer::slex_token<PositionT>  token_type;
 100
 101     void init_dfa(boost::wave::language_support language);
 102
 103 // get time of last compilation
 104     static std::time_t get_compilation_time()
 105         { return compilation_time.get_time(); }
 106
 107 // helper for calculation of the time of last compilation
 108     static boost::wave::util::time_conversion_helper compilation_time;
 109
 110 private:
 111     typedef lexer_base<IteratorT, PositionT> base_type;
 112
 113     static typename base_type::lexer_data const init_data[INIT_DATA_SIZE];          // common patterns
 114     static typename base_type::lexer_data const init_data_cpp[INIT_DATA_CPP_SIZE];  // C++ only patterns
 115     static typename base_type::lexer_data const init_data_pp_number[INIT_DATA_PP_NUMBER_SIZE];  // pp-number only patterns
 116     static typename base_type::lexer_data const init_data_cpp0x[INIT_DATA_CPP0X_SIZE];  // C++0X only patterns
 117 };
 118
 119 ///////////////////////////////////////////////////////////////////////////////
 120 //  data required for initialization of the lexer (token definitions)
 121 #define OR                  "|"
 122 #define Q(c)                "\\" c
 123 #define TRI(c)              Q("?") Q("?") c
 124
 125 // definition of some sub-token regexps to simplify the regex definitions
 126 #define BLANK               "[ \\t]"
 127 #define CCOMMENT            \
 128     Q("/") Q("*") "[^*]*" Q("*") "+" "(" "[^/*][^*]*" Q("*") "+" ")*" Q("/")
 129
 130 #define PPSPACE             "(" BLANK OR CCOMMENT ")*"
 131
 132 #define OCTALDIGIT          "[0-7]"
 133 #define DIGIT               "[0-9]"
 134 #define HEXDIGIT            "[0-9a-fA-F]"
 135 #define OPTSIGN             "[-+]?"
 136 #define EXPSTART            "[eE]" "[-+]"
 137 #define EXPONENT            "(" "[eE]" OPTSIGN "[0-9]+" ")"
 138 #define NONDIGIT            "[a-zA-Z_]"
 139
 140 #define INTEGER             \
 141     "(" "(0x|0X)" HEXDIGIT "+" OR "0" OCTALDIGIT "*" OR "[1-9]" DIGIT "*" ")"
 142
 143 #define INTEGER_SUFFIX      "(" "[uU][lL]?|[lL][uU]?" ")"
 144 #if BOOST_WAVE_SUPPORT_MS_EXTENSIONS != 0
 145 #define LONGINTEGER_SUFFIX  "(" "[uU]" "(" "[lL][lL]" ")" OR \
 146                                 "(" "[lL][lL]" ")" "[uU]" "?" OR \
 147                                 "i64" \
 148                             ")"
 149 #else
 150 #define LONGINTEGER_SUFFIX  "(" "[uU]" "(" "[lL][lL]" ")" OR \
 151                             "(" "[lL][lL]" ")" "[uU]" "?" ")"
 152 #endif
 153 #define FLOAT_SUFFIX        "(" "[fF][lL]?" OR "[lL][fF]?" ")"
 154 #define CHAR_SPEC           "L?"
 155 #define EXTCHAR_SPEC        "(" "[uU]" OR "u8" ")"
 156
 157 #define BACKSLASH           "(" Q("\\") OR TRI(Q("/")) ")"
 158 #define ESCAPESEQ           "(" BACKSLASH "(" \
 159                                 "[abfnrtv?'\"]" OR \
 160                                 BACKSLASH OR \
 161                                 "x" HEXDIGIT "+" OR \
 162                                 OCTALDIGIT OCTALDIGIT "?" OCTALDIGIT "?" \
 163                             "))"
 164 #define HEXQUAD             "(" HEXDIGIT HEXDIGIT HEXDIGIT HEXDIGIT ")"
 165 #define UNIVERSALCHAR       "(" BACKSLASH "(" \
 166                                 "u" HEXQUAD OR \
 167                                 "U" HEXQUAD HEXQUAD \
 168                             "))"
 169
 170 #define POUNDDEF            "(" "#" OR TRI("=") OR Q("%:") ")"
 171 #define NEWLINEDEF          "(" "\n" OR "\r" OR "\r\n" ")"
 172
 173 #if BOOST_WAVE_SUPPORT_INCLUDE_NEXT != 0
 174 #define INCLUDEDEF          "(include|include_next)"
 175 #else
 176 #define INCLUDEDEF          "include"
 177 #endif
 178
 179 #define PP_NUMBERDEF        Q(".") "?" DIGIT "(" DIGIT OR NONDIGIT OR EXPSTART OR Q(".") ")*"
 180
 181 ///////////////////////////////////////////////////////////////////////////////
 182 //  lexer state constants
 183 #define LEXER_STATE_NORMAL  0
 184 #define LEXER_STATE_PP      1
 185
 186 #define NUM_LEXER_STATES    1
 187
 188 //  helper for initializing token data
 189 #define TOKEN_DATA(id, regex)                                                 \
 190         { T_##id, regex, 0, LEXER_STATE_NORMAL }                              \
 191     /**/
 192
 193 #define TOKEN_DATA_EX(id, regex, callback)                                    \
 194         { T_##id, regex, callback, LEXER_STATE_NORMAL }                       \
 195     /**/
 196
 197 ///////////////////////////////////////////////////////////////////////////////
 198 // common C++/C99 token definitions
 199 template <typename IteratorT, typename PositionT>
 200 typename lexer_base<IteratorT, PositionT>::lexer_data const
 201 lexer<IteratorT, PositionT>::init_data[INIT_DATA_SIZE] =
 202 {
 203     TOKEN_DATA(AND, "&"),
 204     TOKEN_DATA(ANDAND, "&&"),
 205     TOKEN_DATA(ASSIGN, "="),
 206     TOKEN_DATA(ANDASSIGN, "&="),
 207     TOKEN_DATA(OR, Q("|")),
 208     TOKEN_DATA(OR_TRIGRAPH, TRI("!")),
 209     TOKEN_DATA(ORASSIGN, Q("|=")),
 210     TOKEN_DATA(ORASSIGN_TRIGRAPH, TRI("!=")),
 211     TOKEN_DATA(XOR, Q("^")),
 212     TOKEN_DATA(XOR_TRIGRAPH, TRI("'")),
 213     TOKEN_DATA(XORASSIGN, Q("^=")),
 214     TOKEN_DATA(XORASSIGN_TRIGRAPH, TRI("'=")),
 215     TOKEN_DATA(COMMA, ","),
 216     TOKEN_DATA(COLON, ":"),
 217     TOKEN_DATA(DIVIDEASSIGN, Q("/=")),
 218     TOKEN_DATA(DIVIDE, Q("/")),
 219     TOKEN_DATA(DOT, Q(".")),
 220     TOKEN_DATA(ELLIPSIS, Q(".") Q(".") Q(".")),
 221     TOKEN_DATA(EQUAL, "=="),
 222     TOKEN_DATA(GREATER, ">"),
 223     TOKEN_DATA(GREATEREQUAL, ">="),
 224     TOKEN_DATA(LEFTBRACE, Q("{")),
 225     TOKEN_DATA(LEFTBRACE_ALT, "<" Q("%")),
 226     TOKEN_DATA(LEFTBRACE_TRIGRAPH, TRI("<")),
 227     TOKEN_DATA(LESS, "<"),
 228     TOKEN_DATA(LESSEQUAL, "<="),
 229     TOKEN_DATA(LEFTPAREN, Q("(")),
 230     TOKEN_DATA(LEFTBRACKET, Q("[")),
 231     TOKEN_DATA(LEFTBRACKET_ALT, "<:"),
 232     TOKEN_DATA(LEFTBRACKET_TRIGRAPH, TRI(Q("("))),
 233     TOKEN_DATA(MINUS, Q("-")),
 234     TOKEN_DATA(MINUSASSIGN, Q("-=")),
 235     TOKEN_DATA(MINUSMINUS, Q("-") Q("-")),
 236     TOKEN_DATA(PERCENT, Q("%")),
 237     TOKEN_DATA(PERCENTASSIGN, Q("%=")),
 238     TOKEN_DATA(NOT, "!"),
 239     TOKEN_DATA(NOTEQUAL, "!="),
 240     TOKEN_DATA(OROR, Q("|") Q("|")),
 241     TOKEN_DATA(OROR_TRIGRAPH, TRI("!") Q("|") OR Q("|") TRI("!") OR TRI("!") TRI("!")),
 242     TOKEN_DATA(PLUS, Q("+")),
 243     TOKEN_DATA(PLUSASSIGN, Q("+=")),
 244     TOKEN_DATA(PLUSPLUS, Q("+") Q("+")),
 245     TOKEN_DATA(ARROW, Q("->")),
 246     TOKEN_DATA(QUESTION_MARK, Q("?")),
 247     TOKEN_DATA(RIGHTBRACE, Q("}")),
 248     TOKEN_DATA(RIGHTBRACE_ALT, Q("%>")),
 249     TOKEN_DATA(RIGHTBRACE_TRIGRAPH, TRI(">")),
 250     TOKEN_DATA(RIGHTPAREN, Q(")")),
 251     TOKEN_DATA(RIGHTBRACKET, Q("]")),
 252     TOKEN_DATA(RIGHTBRACKET_ALT, ":>"),
 253     TOKEN_DATA(RIGHTBRACKET_TRIGRAPH, TRI(Q(")"))),
 254     TOKEN_DATA(SEMICOLON, ";"),
 255     TOKEN_DATA(SHIFTLEFT, "<<"),
 256     TOKEN_DATA(SHIFTLEFTASSIGN, "<<="),
 257     TOKEN_DATA(SHIFTRIGHT, ">>"),
 258     TOKEN_DATA(SHIFTRIGHTASSIGN, ">>="),
 259     TOKEN_DATA(STAR, Q("*")),
 260     TOKEN_DATA(COMPL, Q("~")),
 261     TOKEN_DATA(COMPL_TRIGRAPH, TRI("-")),
 262     TOKEN_DATA(STARASSIGN, Q("*=")),
 263     TOKEN_DATA(ASM, "asm"),
 264     TOKEN_DATA(AUTO, "auto"),
 265     TOKEN_DATA(BOOL, "bool"),
 266     TOKEN_DATA(FALSE, "false"),
 267     TOKEN_DATA(TRUE, "true"),
 268     TOKEN_DATA(BREAK, "break"),
 269     TOKEN_DATA(CASE, "case"),
 270     TOKEN_DATA(CATCH, "catch"),
 271     TOKEN_DATA(CHAR, "char"),
 272     TOKEN_DATA(CLASS, "class"),
 273     TOKEN_DATA(CONST, "const"),
 274     TOKEN_DATA(CONSTCAST, "const_cast"),
 275     TOKEN_DATA(CONTINUE, "continue"),
 276     TOKEN_DATA(DEFAULT, "default"),
 277     TOKEN_DATA(DELETE, "delete"),
 278     TOKEN_DATA(DO, "do"),
 279     TOKEN_DATA(DOUBLE, "double"),
 280     TOKEN_DATA(DYNAMICCAST, "dynamic_cast"),
 281     TOKEN_DATA(ELSE, "else"),
 282     TOKEN_DATA(ENUM, "enum"),
 283     TOKEN_DATA(EXPLICIT, "explicit"),
 284     TOKEN_DATA(EXPORT, "export"),
 285     TOKEN_DATA(EXTERN, "extern"),
 286     TOKEN_DATA(FLOAT, "float"),
 287     TOKEN_DATA(FOR, "for"),
 288     TOKEN_DATA(FRIEND, "friend"),
 289     TOKEN_DATA(GOTO, "goto"),
 290     TOKEN_DATA(IF, "if"),
 291     TOKEN_DATA(INLINE, "inline"),
 292     TOKEN_DATA(INT, "int"),
 293     TOKEN_DATA(LONG, "long"),
 294     TOKEN_DATA(MUTABLE, "mutable"),
 295     TOKEN_DATA(NAMESPACE, "namespace"),
 296     TOKEN_DATA(NEW, "new"),
 297     TOKEN_DATA(OPERATOR, "operator"),
 298     TOKEN_DATA(PRIVATE, "private"),
 299     TOKEN_DATA(PROTECTED, "protected"),
 300     TOKEN_DATA(PUBLIC, "public"),
 301     TOKEN_DATA(REGISTER, "register"),
 302     TOKEN_DATA(REINTERPRETCAST, "reinterpret_cast"),
 303     TOKEN_DATA(RETURN, "return"),
 304     TOKEN_DATA(SHORT, "short"),
 305     TOKEN_DATA(SIGNED, "signed"),
 306     TOKEN_DATA(SIZEOF, "sizeof"),
 307     TOKEN_DATA(STATIC, "static"),
 308     TOKEN_DATA(STATICCAST, "static_cast"),
 309     TOKEN_DATA(STRUCT, "struct"),
 310     TOKEN_DATA(SWITCH, "switch"),
 311     TOKEN_DATA(TEMPLATE, "template"),
 312     TOKEN_DATA(THIS, "this"),
 313     TOKEN_DATA(THROW, "throw"),
 314     TOKEN_DATA(TRY, "try"),
 315     TOKEN_DATA(TYPEDEF, "typedef"),
 316     TOKEN_DATA(TYPEID, "typeid"),
 317     TOKEN_DATA(TYPENAME, "typename"),
 318     TOKEN_DATA(UNION, "union"),
 319     TOKEN_DATA(UNSIGNED, "unsigned"),
 320     TOKEN_DATA(USING, "using"),
 321     TOKEN_DATA(VIRTUAL, "virtual"),
 322     TOKEN_DATA(VOID, "void"),
 323     TOKEN_DATA(VOLATILE, "volatile"),
 324     TOKEN_DATA(WCHART, "wchar_t"),
 325     TOKEN_DATA(WHILE, "while"),
 326     TOKEN_DATA(PP_DEFINE, POUNDDEF PPSPACE "define"),
 327     TOKEN_DATA(PP_IF, POUNDDEF PPSPACE "if"),
 328     TOKEN_DATA(PP_IFDEF, POUNDDEF PPSPACE "ifdef"),
 329     TOKEN_DATA(PP_IFNDEF, POUNDDEF PPSPACE "ifndef"),
 330     TOKEN_DATA(PP_ELSE, POUNDDEF PPSPACE "else"),
 331     TOKEN_DATA(PP_ELIF, POUNDDEF PPSPACE "elif"),
 332     TOKEN_DATA(PP_ENDIF, POUNDDEF PPSPACE "endif"),
 333     TOKEN_DATA(PP_ERROR, POUNDDEF PPSPACE "error"),
 334     TOKEN_DATA(PP_QHEADER, POUNDDEF PPSPACE \
 335         INCLUDEDEF PPSPACE Q("\"") "[^\\n\\r\"]+" Q("\"")),
 336     TOKEN_DATA(PP_HHEADER, POUNDDEF PPSPACE \
 337         INCLUDEDEF PPSPACE "<" "[^\\n\\r>]+" ">"),
 338     TOKEN_DATA(PP_INCLUDE, POUNDDEF PPSPACE \
 339         INCLUDEDEF PPSPACE),
 340     TOKEN_DATA(PP_LINE, POUNDDEF PPSPACE "line"),
 341     TOKEN_DATA(PP_PRAGMA, POUNDDEF PPSPACE "pragma"),
 342     TOKEN_DATA(PP_UNDEF, POUNDDEF PPSPACE "undef"),
 343     TOKEN_DATA(PP_WARNING, POUNDDEF PPSPACE "warning"),
 344 #if BOOST_WAVE_SUPPORT_MS_EXTENSIONS != 0
 345     TOKEN_DATA(MSEXT_INT8, "__int8"),
 346     TOKEN_DATA(MSEXT_INT16, "__int16"),
 347     TOKEN_DATA(MSEXT_INT32, "__int32"),
 348     TOKEN_DATA(MSEXT_INT64, "__int64"),
 349     TOKEN_DATA(MSEXT_BASED, "_?" "_based"),
 350     TOKEN_DATA(MSEXT_DECLSPEC, "_?" "_declspec"),
 351     TOKEN_DATA(MSEXT_CDECL, "_?" "_cdecl"),
 352     TOKEN_DATA(MSEXT_FASTCALL, "_?" "_fastcall"),
 353     TOKEN_DATA(MSEXT_STDCALL, "_?" "_stdcall"),
 354     TOKEN_DATA(MSEXT_TRY , "__try"),
 355     TOKEN_DATA(MSEXT_EXCEPT, "__except"),
 356     TOKEN_DATA(MSEXT_FINALLY, "__finally"),
 357     TOKEN_DATA(MSEXT_LEAVE, "__leave"),
 358     TOKEN_DATA(MSEXT_INLINE, "_?" "_inline"),
 359     TOKEN_DATA(MSEXT_ASM, "_?" "_asm"),
 360     TOKEN_DATA(MSEXT_PP_REGION, POUNDDEF PPSPACE "region"),
 361     TOKEN_DATA(MSEXT_PP_ENDREGION, POUNDDEF PPSPACE "endregion"),
 362 #endif // BOOST_WAVE_SUPPORT_MS_EXTENSIONS != 0
 363 //  TOKEN_DATA(OCTALINT, "0" OCTALDIGIT "*" INTEGER_SUFFIX "?"),
 364 //  TOKEN_DATA(DECIMALINT, "[1-9]" DIGIT "*" INTEGER_SUFFIX "?"),
 365 //  TOKEN_DATA(HEXAINT, "(0x|0X)" HEXDIGIT "+" INTEGER_SUFFIX "?"),
 366     TOKEN_DATA(LONGINTLIT, INTEGER LONGINTEGER_SUFFIX),
 367     TOKEN_DATA(INTLIT, INTEGER INTEGER_SUFFIX "?"),
 368     TOKEN_DATA(FLOATLIT,
 369         "(" DIGIT "*" Q(".") DIGIT "+" OR DIGIT "+" Q(".") ")"
 370         EXPONENT "?" FLOAT_SUFFIX "?" OR
 371         DIGIT "+" EXPONENT FLOAT_SUFFIX "?"),
 372     TOKEN_DATA(CCOMMENT, CCOMMENT),
 373     TOKEN_DATA(CPPCOMMENT, Q("/") Q("/[^\\n\\r]*") NEWLINEDEF ),
 374     TOKEN_DATA(CHARLIT, CHAR_SPEC "'"
 375                 "(" ESCAPESEQ OR UNIVERSALCHAR OR "[^\\n\\r\\\\']" ")+" "'"),
 376     TOKEN_DATA(STRINGLIT, CHAR_SPEC Q("\"")
 377                 "(" ESCAPESEQ OR UNIVERSALCHAR OR "[^\\n\\r\\\\\"]" ")*" Q("\"")),
 378 #if BOOST_WAVE_USE_STRICT_LEXER != 0
 379     TOKEN_DATA(IDENTIFIER, "([a-zA-Z_]" OR UNIVERSALCHAR ")([a-zA-Z0-9_]" OR UNIVERSALCHAR ")*"),
 380 #else
 381     TOKEN_DATA(IDENTIFIER, "([a-zA-Z_$]" OR UNIVERSALCHAR ")([a-zA-Z0-9_$]" OR UNIVERSALCHAR ")*"),
 382 #endif
 383     TOKEN_DATA(SPACE, "[ \t\v\f]+"),
 384 //    TOKEN_DATA(SPACE2, "[\\v\\f]+"),
 385     TOKEN_DATA(CONTLINE, Q("\\") "\n"),
 386     TOKEN_DATA(NEWLINE, NEWLINEDEF),
 387     TOKEN_DATA(POUND_POUND, "##"),
 388     TOKEN_DATA(POUND_POUND_ALT, Q("%:") Q("%:")),
 389     TOKEN_DATA(POUND_POUND_TRIGRAPH, TRI("=") TRI("=")),
 390     TOKEN_DATA(POUND, "#"),
 391     TOKEN_DATA(POUND_ALT, Q("%:")),
 392     TOKEN_DATA(POUND_TRIGRAPH, TRI("=")),
 393     TOKEN_DATA(ANY_TRIGRAPH, TRI(Q("/"))),
 394     TOKEN_DATA(ANY, "."),     // this should be the last recognized token
 395     { token_id(0) }           // this should be the last entry
 396 };
 397
 398 ///////////////////////////////////////////////////////////////////////////////
 399 // C++ only token definitions
 400 template <typename IteratorT, typename PositionT>
 401 typename lexer_base<IteratorT, PositionT>::lexer_data const
 402 lexer<IteratorT, PositionT>::init_data_cpp[INIT_DATA_CPP_SIZE] =
 403 {
 404     TOKEN_DATA(AND_ALT, "bitand"),
 405     TOKEN_DATA(ANDASSIGN_ALT, "and_eq"),
 406     TOKEN_DATA(ANDAND_ALT, "and"),
 407     TOKEN_DATA(OR_ALT, "bitor"),
 408     TOKEN_DATA(ORASSIGN_ALT, "or_eq"),
 409     TOKEN_DATA(OROR_ALT, "or"),
 410     TOKEN_DATA(XORASSIGN_ALT, "xor_eq"),
 411     TOKEN_DATA(XOR_ALT, "xor"),
 412     TOKEN_DATA(NOTEQUAL_ALT, "not_eq"),
 413     TOKEN_DATA(NOT_ALT, "not"),
 414     TOKEN_DATA(COMPL_ALT, "compl"),
 415 #if BOOST_WAVE_SUPPORT_IMPORT_KEYWORD != 0
 416     TOKEN_DATA(IMPORT, "import"),
 417 #endif
 418     TOKEN_DATA(ARROWSTAR, Q("->") Q("*")),
 419     TOKEN_DATA(DOTSTAR, Q(".") Q("*")),
 420     TOKEN_DATA(COLON_COLON, "::"),
 421     { token_id(0) }       // this should be the last entry
 422 };
 423
 424 ///////////////////////////////////////////////////////////////////////////////
 425 // C++ only token definitions
 426 template <typename IteratorT, typename PositionT>
 427 typename lexer_base<IteratorT, PositionT>::lexer_data const
 428 lexer<IteratorT, PositionT>::init_data_pp_number[INIT_DATA_PP_NUMBER_SIZE] =
 429 {
 430     TOKEN_DATA(PP_NUMBER, PP_NUMBERDEF),
 431     { token_id(0) }       // this should be the last entry
 432 };
 433
 434 ///////////////////////////////////////////////////////////////////////////////
 435 // C++ only token definitions
 436
 437 #define T_EXTCHARLIT      token_id(T_CHARLIT|AltTokenType)
 438 #define T_EXTSTRINGLIT    token_id(T_STRINGLIT|AltTokenType)
 439 #define T_EXTRAWSTRINGLIT token_id(T_RAWSTRINGLIT|AltTokenType)
 440
 441 template <typename IteratorT, typename PositionT>
 442 typename lexer_base<IteratorT, PositionT>::lexer_data const
 443 lexer<IteratorT, PositionT>::init_data_cpp0x[INIT_DATA_CPP0X_SIZE] =
 444 {
 445     TOKEN_DATA(EXTCHARLIT, EXTCHAR_SPEC "'"
 446                 "(" ESCAPESEQ OR UNIVERSALCHAR OR "[^\\n\\r\\\\']" ")+" "'"),
 447     TOKEN_DATA(EXTSTRINGLIT, EXTCHAR_SPEC Q("\"")
 448                 "(" ESCAPESEQ OR UNIVERSALCHAR OR "[^\\n\\r\\\\\"]" ")*" Q("\"")),
 449     TOKEN_DATA(RAWSTRINGLIT, CHAR_SPEC "R" Q("\"")
 450                 "(" ESCAPESEQ OR UNIVERSALCHAR OR "[^\\\\\"]" ")*" Q("\"")),
 451     TOKEN_DATA(EXTRAWSTRINGLIT, EXTCHAR_SPEC "R" Q("\"")
 452                 "(" ESCAPESEQ OR UNIVERSALCHAR OR "[^\\\\\"]" ")*" Q("\"")),
 453     TOKEN_DATA(ALIGNAS, "alignas"),
 454     TOKEN_DATA(ALIGNOF, "alignof"),
 455     TOKEN_DATA(CHAR16_T, "char16_t"),
 456     TOKEN_DATA(CHAR32_T, "char32_t"),
 457     TOKEN_DATA(CONSTEXPR, "constexpr"),
 458     TOKEN_DATA(DECLTYPE, "decltype"),
 459     TOKEN_DATA(NOEXCEPT, "noexcept"),
 460     TOKEN_DATA(NULLPTR, "nullptr"),
 461     TOKEN_DATA(STATICASSERT, "static_assert"),
 462     TOKEN_DATA(THREADLOCAL, "threadlocal"),
 463     { token_id(0) }       // this should be the last entry
 464 };
 465
 466 ///////////////////////////////////////////////////////////////////////////////
 467 //  undefine macros, required for regular expression definitions
 468 #undef INCLUDEDEF
 469 #undef POUNDDEF
 470 #undef CCOMMENT
 471 #undef PPSPACE
 472 #undef DIGIT
 473 #undef OCTALDIGIT
 474 #undef HEXDIGIT
 475 #undef NONDIGIT
 476 #undef OPTSIGN
 477 #undef EXPSTART
 478 #undef EXPONENT
 479 #undef LONGINTEGER_SUFFIX
 480 #undef INTEGER_SUFFIX
 481 #undef INTEGER
 482 #undef FLOAT_SUFFIX
 483 #undef CHAR_SPEC
 484 #undef BACKSLASH
 485 #undef ESCAPESEQ
 486 #undef HEXQUAD
 487 #undef UNIVERSALCHAR
 488 #undef PP_NUMBERDEF
 489
 490 #undef Q
 491 #undef TRI
 492 #undef OR
 493
 494 #undef TOKEN_DATA
 495 #undef TOKEN_DATA_EX
 496
 497 ///////////////////////////////////////////////////////////////////////////////
 498 // initialize cpp lexer with token data
 499 template <typename IteratorT, typename PositionT>
 500 inline
 501 lexer_base<IteratorT, PositionT>::lexer_base()
 502 :   base_type(NUM_LEXER_STATES)
 503 {
 504 }
 505
 506 template <typename IteratorT, typename PositionT>
 507 inline void
 508 lexer<IteratorT, PositionT>::init_dfa(boost::wave::language_support lang)
 509 {
 510     if (this->has_compiled_dfa())
 511         return;
 512
 513 // if pp-numbers should be preferred, insert the corresponding rule first
 514     if (boost::wave::need_prefer_pp_numbers(lang)) {
 515         for (int j = 0; 0 != init_data_pp_number[j].tokenid; ++j) {
 516             this->register_regex(init_data_pp_number[j].tokenregex,
 517                 init_data_pp_number[j].tokenid, init_data_pp_number[j].tokencb,
 518                 init_data_pp_number[j].lexerstate);
 519         }
 520     }
 521
 522 // if in C99 mode, some of the keywords are not valid
 523     if (!boost::wave::need_c99(lang)) {
 524         for (int j = 0; 0 != init_data_cpp[j].tokenid; ++j) {
 525             this->register_regex(init_data_cpp[j].tokenregex,
 526                 init_data_cpp[j].tokenid, init_data_cpp[j].tokencb,
 527                 init_data_cpp[j].lexerstate);
 528         }
 529     }
 530
 531 // if in C++0x mode, add all new keywords
 532 #if BOOST_WAVE_SUPPORT_CPP0X != 0
 533     if (boost::wave::need_cpp0x(lang)) {
 534         for (int j = 0; 0 != init_data_cpp0x[j].tokenid; ++j) {
 535             this->register_regex(init_data_cpp0x[j].tokenregex,
 536                 init_data_cpp0x[j].tokenid, init_data_cpp0x[j].tokencb,
 537                 init_data_cpp0x[j].lexerstate);
 538         }
 539     }
 540 #endif
 541
 542     for (int i = 0; 0 != init_data[i].tokenid; ++i) {
 543         this->register_regex(init_data[i].tokenregex, init_data[i].tokenid,
 544             init_data[i].tokencb, init_data[i].lexerstate);
 545     }
 546 }
 547
 548 ///////////////////////////////////////////////////////////////////////////////
 549 // get time of last compilation of this file
 550 template <typename IteratorT, typename PositionT>
 551 boost::wave::util::time_conversion_helper
 552     lexer<IteratorT, PositionT>::compilation_time(__DATE__ " " __TIME__);
 553
 554 ///////////////////////////////////////////////////////////////////////////////
 555 }   // namespace lexer
 556
 557 ///////////////////////////////////////////////////////////////////////////////
 558 //
 559 template <typename IteratorT, typename PositionT>
 560 inline void
 561 init_lexer (lexer::lexer<IteratorT, PositionT> &lexer,
 562     boost::wave::language_support language, bool force_reinit = false)
 563 {
 564     if (lexer.has_compiled_dfa())
 565         return;     // nothing to do
 566
 567     using std::ifstream;
 568     using std::ofstream;
 569     using std::ios;
 570     using std::cerr;
 571     using std::endl;
 572
 573 ifstream dfa_in("wave_slex_lexer.dfa", ios::in|ios::binary);
 574
 575     lexer.init_dfa(language);
 576     if (force_reinit || !dfa_in.is_open() ||
 577         !lexer.load (dfa_in, (long)lexer.get_compilation_time()))
 578     {
 579 #if defined(BOOST_SPIRIT_DEBUG)
 580         cerr << "Compiling regular expressions for slex ...";
 581 #endif // defined(BOOST_SPIRIT_DEBUG)
 582
 583         dfa_in.close();
 584         lexer.create_dfa();
 585
 586     ofstream dfa_out ("wave_slex_lexer.dfa", ios::out|ios::binary|ios::trunc);
 587
 588         if (dfa_out.is_open())
 589             lexer.save (dfa_out, (long)lexer.get_compilation_time());
 590
 591 #if defined(BOOST_SPIRIT_DEBUG)
 592         cerr << " Done." << endl;
 593 #endif // defined(BOOST_SPIRIT_DEBUG)
 594     }
 595 }
 596
 597 ///////////////////////////////////////////////////////////////////////////////
 598 //
 599 //  lex_functor
 600 //
 601 ///////////////////////////////////////////////////////////////////////////////
 602
 603 template <typename IteratorT, typename PositionT = wave::util::file_position_type>
 604 class slex_functor
 605 :   public slex_input_interface<
 606         typename lexer::lexer<IteratorT, PositionT>::token_type
 607     >
 608 {
 609 public:
 610
 611     typedef boost::wave::util::position_iterator<IteratorT, PositionT>
 612           iterator_type;
 613     typedef typename std::iterator_traits<IteratorT>::value_type    char_type;
 614     typedef BOOST_WAVE_STRINGTYPE                                   string_type;
 615     typedef typename lexer::lexer<IteratorT, PositionT>::token_type token_type;
 616
 617     slex_functor(IteratorT const &first_, IteratorT const &last_,
 618             PositionT const &pos_, boost::wave::language_support language_)
 619     :   first(first_, last_, pos_), language(language_), at_eof(false)
 620     {
 621         // initialize lexer dfa tables
 622         init_lexer(lexer, language_);
 623     }
 624     virtual ~slex_functor() {}
 625
 626 // get the next token from the input stream
 627     token_type& get(token_type& result)
 628     {
 629         if (!at_eof) {
 630             do {
 631             // generate and return the next token
 632             std::string value;
 633             PositionT pos = first.get_position();   // begin of token position
 634             token_id id = token_id(lexer.next_token(first, last, &value));
 635
 636                 if ((token_id)(-1) == id)
 637                     id = T_EOF;     // end of input reached
 638
 639             string_type token_val(value.c_str());
 640
 641                 if (boost::wave::need_emit_contnewlines(language) ||
 642                     T_CONTLINE != id)
 643                 {
 644                 //  The cast should avoid spurious warnings about missing case labels
 645                 //  for the other token ids's.
 646                     switch (static_cast<unsigned int>(id)) {
 647                     case T_IDENTIFIER:
 648                     // test identifier characters for validity (throws if
 649                     // invalid chars found)
 650                         if (!boost::wave::need_no_character_validation(language)) {
 651                             using boost::wave::cpplexer::impl::validate_identifier_name;
 652                             validate_identifier_name(token_val,
 653                                 pos.get_line(), pos.get_column(), pos.get_file());
 654                         }
 655                         break;
 656
 657                     case T_EXTCHARLIT:
 658                     case T_EXTSTRINGLIT:
 659                     case T_EXTRAWSTRINGLIT:
 660                         id = token_id(id & ~AltTokenType);
 661                         BOOST_FALLTHROUGH;
 662
 663                     case T_CHARLIT:
 664                     case T_STRINGLIT:
 665                     case T_RAWSTRINGLIT:
 666                     // test literal characters for validity (throws if invalid
 667                     // chars found)
 668                         if (boost::wave::need_convert_trigraphs(language)) {
 669                             using boost::wave::cpplexer::impl::convert_trigraphs;
 670                             token_val = convert_trigraphs(token_val);
 671                         }
 672                         if (!boost::wave::need_no_character_validation(language)) {
 673                             using boost::wave::cpplexer::impl::validate_literal;
 674                             validate_literal(token_val,
 675                                 pos.get_line(), pos.get_column(), pos.get_file());
 676                         }
 677                         break;
 678
 679                     case T_LONGINTLIT:  // supported in C99 and long_long mode
 680                         if (!boost::wave::need_long_long(language)) {
 681                         // syntax error: not allowed in C++ mode
 682                             BOOST_WAVE_LEXER_THROW(
 683                                 boost::wave::cpplexer::lexing_exception,
 684                                 invalid_long_long_literal, value.c_str(),
 685                                 pos.get_line(), pos.get_column(),
 686                                 pos.get_file().c_str());
 687                         }
 688                         break;
 689
 690 #if BOOST_WAVE_SUPPORT_INCLUDE_NEXT != 0
 691                     case T_PP_HHEADER:
 692                     case T_PP_QHEADER:
 693                     case T_PP_INCLUDE:
 694                     // convert to the corresponding ..._next token, if appropriate
 695                         {
 696                         // Skip '#' and whitespace and see whether we find an
 697                         // 'include_next' here.
 698                             typename string_type::size_type start = value.find("include");
 699                             if (0 == value.compare(start, 12, "include_next", 12))
 700                                 id = token_id(id | AltTokenType);
 701                             break;
 702                         }
 703 #endif // BOOST_WAVE_SUPPORT_INCLUDE_NEXT != 0
 704
 705                     case T_EOF:
 706                     // T_EOF is returned as a valid token, the next call will
 707                     // return T_EOI, i.e. the actual end of input
 708                         at_eof = true;
 709                         token_val.clear();
 710                         break;
 711
 712                     case T_OR_TRIGRAPH:
 713                     case T_XOR_TRIGRAPH:
 714                     case T_LEFTBRACE_TRIGRAPH:
 715                     case T_RIGHTBRACE_TRIGRAPH:
 716                     case T_LEFTBRACKET_TRIGRAPH:
 717                     case T_RIGHTBRACKET_TRIGRAPH:
 718                     case T_COMPL_TRIGRAPH:
 719                     case T_POUND_TRIGRAPH:
 720                     case T_ANY_TRIGRAPH:
 721                         if (boost::wave::need_convert_trigraphs(language))
 722                         {
 723                             using boost::wave::cpplexer::impl::convert_trigraph;
 724                             token_val = convert_trigraph(token_val);
 725                         }
 726                         break;
 727                     }
 728
 729                     result = token_type(id, token_val, pos);
 730 #if BOOST_WAVE_SUPPORT_PRAGMA_ONCE != 0
 731                     return guards.detect_guard(result);
 732 #else
 733                     return result;
 734 #endif
 735                 }
 736
 737             // skip the T_CONTLINE token
 738             } while (true);
 739         }
 740         return result = token_type();   // return T_EOI
 741     }
 742
 743     void set_position(PositionT const &pos)
 744     {
 745         // set position has to change the file name and line number only
 746         first.get_position().set_file(pos.get_file());
 747         first.get_position().set_line(pos.get_line());
 748     }
 749
 750 #if BOOST_WAVE_SUPPORT_PRAGMA_ONCE != 0
 751     bool has_include_guards(std::string& guard_name) const
 752         { return guards.detected(guard_name); }
 753 #endif
 754
 755 private:
 756     iterator_type first;
 757     iterator_type last;
 758     boost::wave::language_support language;
 759     static lexer::lexer<IteratorT, PositionT> lexer;   // needed only once
 760
 761     bool at_eof;
 762
 763 #if BOOST_WAVE_SUPPORT_PRAGMA_ONCE != 0
 764     include_guards<token_type> guards;
 765 #endif
 766 };
 767
 768 template <typename IteratorT, typename PositionT>
 769 lexer::lexer<IteratorT, PositionT> slex_functor<IteratorT, PositionT>::lexer;
 770
 771 #undef T_EXTCHARLIT
 772 #undef T_EXTSTRINGLIT
 773 #undef T_EXTRAWSTRINGLIT
 774
 775 ///////////////////////////////////////////////////////////////////////////////
 776 //
 777 //  The 'new_lexer' function allows the opaque generation of a new lexer object.
 778 //  It is coupled to the iterator type to allow to decouple the lexer/iterator
 779 //  configurations at compile time.
 780 //
 781 //  This function is declared inside the cpp_slex_token.hpp file, which is
 782 //  referenced by the source file calling the lexer and the source file, which
 783 //  instantiates the lex_functor. But it is defined here, so it will be
 784 //  instantiated only while compiling the source file, which instantiates the
 785 //  lex_functor. While the cpp_slex_token.hpp file may be included everywhere,
 786 //  this file (cpp_slex_lexer.hpp) should be included only once. This allows
 787 //  to decouple the lexer interface from the lexer implementation and reduces
 788 //  compilation time.
 789 //
 790 ///////////////////////////////////////////////////////////////////////////////
 791
 792 ///////////////////////////////////////////////////////////////////////////////
 793 //
 794 //  The new_lexer_gen<>::new_lexer function (declared in cpp_slex_token.hpp)
 795 //  should be defined inline, if the lex_functor shouldn't be instantiated
 796 //  separately from the lex_iterator.
 797 //
 798 //  Separate (explicit) instantiation helps to reduce compilation time.
 799 //
 800 ///////////////////////////////////////////////////////////////////////////////
 801
 802 #if BOOST_WAVE_SEPARATE_LEXER_INSTANTIATION != 0
 803 #define BOOST_WAVE_SLEX_NEW_LEXER_INLINE
 804 #else
 805 #define BOOST_WAVE_SLEX_NEW_LEXER_INLINE inline
 806 #endif
 807
 808 template <typename IteratorT, typename PositionT>
 809 BOOST_WAVE_SLEX_NEW_LEXER_INLINE
 810 lex_input_interface<slex_token<PositionT> > *
 811 new_lexer_gen<IteratorT, PositionT>::new_lexer(IteratorT const &first,
 812     IteratorT const &last, PositionT const &pos,
 813     boost::wave::language_support language)
 814 {
 815     return new slex_functor<IteratorT, PositionT>(first, last, pos,
 816         language);
 817 }
 818
 819 #undef BOOST_WAVE_SLEX_NEW_LEXER_INLINE
 820
 821 ///////////////////////////////////////////////////////////////////////////////
 822 }   // namespace slex
 823 }   // namespace cpplexer
 824 }   // namespace wave
 825 }   // namespace boost
 826
 827 #endif // !defined(SLEX_LEXER_HPP_5E8E1DF0_BB41_4938_B7E5_A4BB68222FF6_INCLUDED)