X-Git-Url: https://git.xonotic.org/?p=xonotic%2Fgmqcc.git;a=blobdiff_plain;f=lexer.h;h=24ac39a63f7f574d1e9a8a7dfc1f8236cc51612a;hp=ae8812b261c5be6e68252a78704e2e32438b59c7;hb=69b55ccc03b56af1f6c05eb45866ab198307487f;hpb=b76ffe8304dd59cf53c35f53b4b42f5a2fbac2bf diff --git a/lexer.h b/lexer.h index ae8812b..24ac39a 100644 --- a/lexer.h +++ b/lexer.h @@ -1,33 +1,57 @@ -#ifndef GMQCC_LEXER_HDR_ -#define GMQCC_LEXER_HDR_ +/* + * Copyright (C) 2012, 2013 + * Wolfgang Bumiller + * + * Permission is hereby granted, free of charge, to any person obtaining a copy of + * this software and associated documentation files (the "Software"), to deal in + * the Software without restriction, including without limitation the rights to + * use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies + * of the Software, and to permit persons to whom the Software is furnished to do + * so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef GMQCC_LEXER_HDR +#define GMQCC_LEXER_HDR typedef struct token_s token; -#include "ast.h" - struct token_s { - int ttype; + int ttype; - MEM_VECTOR_MAKE(char, value); + char *value; - union { - vector v; - int i; - double f; - int t; /* type */ - } constval; + union { + vector v; + int i; + double f; + int t; /* type */ + } constval; - struct token_s *next; - struct token_s *prev; +#if 0 + struct token_s *next; + struct token_s *prev; +#endif - lex_ctx ctx; + lex_ctx ctx; }; +#if 0 token* token_new(); void token_delete(token*); token* token_copy(const token *cp); void token_delete_all(token *t); token* token_copy_all(const token *cp); +#endif /* Lexer * @@ -45,13 +69,29 @@ enum { TOKEN_KEYWORD, /* loop */ + TOKEN_DOTS, /* 3 dots, ... */ + + TOKEN_ATTRIBUTE_OPEN, /* [[ */ + TOKEN_ATTRIBUTE_CLOSE, /* ]] */ + + TOKEN_VA_ARGS, /* for the ftepp only */ + TOKEN_VA_ARGS_ARRAY, /* for the ftepp only */ + TOKEN_VA_COUNT, /* to get the count of vaargs */ + TOKEN_STRINGCONST, /* not the typename but an actual "string" */ TOKEN_CHARCONST, TOKEN_VECTORCONST, TOKEN_INTCONST, TOKEN_FLOATCONST, - TOKEN_EOF, + TOKEN_WHITE, + TOKEN_EOL, + + /* if we add additional tokens before this, the exposed API + * should not be broken anyway, but EOF/ERROR/... should + * still be at the bottom + */ + TOKEN_EOF = 1024, /* We use '< TOKEN_ERROR', so TOKEN_FATAL must come after it and any * other error related tokens as well @@ -60,48 +100,47 @@ enum { TOKEN_FATAL /* internal error, eg out of memory */ }; -static const char *_tokennames[] = { - "TOKEN_START", - "TOKEN_IDENT", - "TOKEN_TYPENAME", - "TOKEN_OPERATOR", - "TOKEN_KEYWORD", - "TOKEN_STRINGCONST", - "TOKEN_CHARCONST", - "TOKEN_VECTORCONST", - "TOKEN_INTCONST", - "TOKEN_FLOATCONST", - "TOKEN_EOF", - "TOKEN_ERROR", - "TOKEN_FATAL", -}; -typedef int -_all_tokennames_added_[ - ((TOKEN_FATAL - TOKEN_START + 1) == - (sizeof(_tokennames)/sizeof(_tokennames[0]))) - ? 1 : -1]; - typedef struct { - FILE *file; - char *name; - size_t line; - size_t sline; /* line at the start of a token */ + char *name; + int value; +} frame_macro; - char peek[256]; - size_t peekpos; +typedef struct lex_file_s { + FILE *file; + const char *open_string; + size_t open_string_length; + size_t open_string_pos; - token *tok; + char *name; + size_t line; + size_t sline; /* line at the start of a token */ - struct { - bool noops; - } flags; -} lex_file; + int peek[256]; + size_t peekpos; + + bool eof; + + token tok; /* not a pointer anymore */ -MEM_VECTOR_PROTO(lex_file, char, token); + struct { + bool noops; + bool nodigraphs; /* used when lexing string constants */ + bool preprocessing; /* whitespace and EOLs become actual tokens */ + bool mergelines; /* backslash at the end of a line escapes the newline */ + } flags; + + int framevalue; + frame_macro *frames; + char *modelname; + + size_t push_line; +} lex_file; lex_file* lex_open (const char *file); +lex_file* lex_open_string(const char *str, size_t len, const char *name); void lex_close(lex_file *lex); int lex_do (lex_file *lex); +void lex_cleanup(void); /* Parser * @@ -120,7 +159,7 @@ typedef struct { unsigned int operands; unsigned int id; unsigned int assoc; - unsigned int prec; + signed int prec; unsigned int flags; } oper_info; @@ -128,18 +167,24 @@ typedef struct { #define opid2(a,b) ((a<<8)|b) #define opid3(a,b,c) ((a<<16)|(b<<8)|c) -static const oper_info operators[] = { - { "++", 1, opid3('S','+','+'), ASSOC_LEFT, 16, OP_SUFFIX}, - { "--", 1, opid3('S','-','-'), ASSOC_LEFT, 16, OP_SUFFIX}, +static const oper_info c_operators[] = { + { "(", 0, opid1('('), ASSOC_LEFT, 99, OP_PREFIX}, /* paren expression - non function call */ - { ".", 2, opid1('.'), ASSOC_LEFT, 15, 0 }, + { "++", 1, opid3('S','+','+'), ASSOC_LEFT, 17, OP_SUFFIX}, + { "--", 1, opid3('S','-','-'), ASSOC_LEFT, 17, OP_SUFFIX}, + { ".", 2, opid1('.'), ASSOC_LEFT, 17, 0 }, + { "(", 0, opid1('('), ASSOC_LEFT, 17, 0 }, /* function call */ + { "[", 2, opid1('['), ASSOC_LEFT, 17, 0 }, /* array subscript */ + + { "++", 1, opid3('+','+','P'), ASSOC_RIGHT, 16, OP_PREFIX }, + { "--", 1, opid3('-','-','P'), ASSOC_RIGHT, 16, OP_PREFIX }, - { "!", 1, opid2('!', 'P'), ASSOC_RIGHT, 14, 0 }, - { "~", 1, opid2('~', 'P'), ASSOC_RIGHT, 14, 0 }, + { "**", 2, opid2('*', '*'), ASSOC_RIGHT, 15, 0 }, + + { "!", 1, opid2('!', 'P'), ASSOC_RIGHT, 14, OP_PREFIX }, + { "~", 1, opid2('~', 'P'), ASSOC_RIGHT, 14, OP_PREFIX }, { "+", 1, opid2('+','P'), ASSOC_RIGHT, 14, OP_PREFIX }, { "-", 1, opid2('-','P'), ASSOC_RIGHT, 14, OP_PREFIX }, - { "++", 1, opid3('+','+','P'), ASSOC_RIGHT, 14, OP_PREFIX }, - { "--", 1, opid3('-','-','P'), ASSOC_RIGHT, 14, OP_PREFIX }, /* { "&", 1, opid2('&','P'), ASSOC_RIGHT, 14, OP_PREFIX }, */ { "*", 2, opid1('*'), ASSOC_LEFT, 13, 0 }, @@ -154,6 +199,7 @@ static const oper_info operators[] = { { "<", 2, opid1('<'), ASSOC_LEFT, 10, 0 }, { ">", 2, opid1('>'), ASSOC_LEFT, 10, 0 }, + { "<=>", 2, opid3('<','=','>'), ASSOC_LEFT, 10, 0 }, { "<=", 2, opid2('<','='), ASSOC_LEFT, 10, 0 }, { ">=", 2, opid2('>','='), ASSOC_LEFT, 10, 0 }, @@ -183,34 +229,112 @@ static const oper_info operators[] = { { "&=", 2, opid2('&','='), ASSOC_RIGHT, 2, 0 }, { "^=", 2, opid2('^','='), ASSOC_RIGHT, 2, 0 }, { "|=", 2, opid2('|','='), ASSOC_RIGHT, 2, 0 }, + { "&~=", 2, opid3('&','~','='), ASSOC_RIGHT, 2, 0 }, + + { ":", 0, opid2(':','?'), ASSOC_RIGHT, 1, 0 }, - { ",", 2, opid1(','), ASSOC_LEFT, 1, 0 } + { ",", 2, opid1(','), ASSOC_LEFT, 0, 0 } }; -static const size_t operator_count = (sizeof(operators) / sizeof(operators[0])); +static const size_t c_operator_count = (sizeof(c_operators) / sizeof(c_operators[0])); -typedef struct -{ - lex_file *lex; - int error; - lex_ctx ctx; +static const oper_info fte_operators[] = { + { "(", 0, opid1('('), ASSOC_LEFT, 99, OP_PREFIX}, /* paren expression - non function call */ - token *tokens; - token *lastok; + { "++", 1, opid3('S','+','+'), ASSOC_LEFT, 15, OP_SUFFIX}, + { "--", 1, opid3('S','-','-'), ASSOC_LEFT, 15, OP_SUFFIX}, + { ".", 2, opid1('.'), ASSOC_LEFT, 15, 0 }, + { "(", 0, opid1('('), ASSOC_LEFT, 15, 0 }, /* function call */ + { "[", 2, opid1('['), ASSOC_LEFT, 15, 0 }, /* array subscript */ - token *tok; /* current token */ + { "!", 1, opid2('!', 'P'), ASSOC_RIGHT, 14, OP_PREFIX }, + { "+", 1, opid2('+','P'), ASSOC_RIGHT, 14, OP_PREFIX }, + { "-", 1, opid2('-','P'), ASSOC_RIGHT, 14, OP_PREFIX }, + { "++", 1, opid3('+','+','P'), ASSOC_RIGHT, 14, OP_PREFIX }, + { "--", 1, opid3('-','-','P'), ASSOC_RIGHT, 14, OP_PREFIX }, - MEM_VECTOR_MAKE(ast_value*, globals); -} parse_file; + { "*", 2, opid1('*'), ASSOC_LEFT, 13, 0 }, + { "/", 2, opid1('/'), ASSOC_LEFT, 13, 0 }, + { "&", 2, opid1('&'), ASSOC_LEFT, 13, 0 }, + { "|", 2, opid1('|'), ASSOC_LEFT, 13, 0 }, -MEM_VECTOR_PROTO(parse_file, ast_value*, globals); + { "+", 2, opid1('+'), ASSOC_LEFT, 12, 0 }, + { "-", 2, opid1('-'), ASSOC_LEFT, 12, 0 }, + + { "<<", 2, opid2('<','<'), ASSOC_LEFT, 11, 0 }, + { ">>", 2, opid2('>','>'), ASSOC_LEFT, 11, 0 }, + + { "<", 2, opid1('<'), ASSOC_LEFT, 10, 0 }, + { ">", 2, opid1('>'), ASSOC_LEFT, 10, 0 }, + { "<=", 2, opid2('<','='), ASSOC_LEFT, 10, 0 }, + { ">=", 2, opid2('>','='), ASSOC_LEFT, 10, 0 }, + { "==", 2, opid2('=','='), ASSOC_LEFT, 10, 0 }, + { "!=", 2, opid2('!','='), ASSOC_LEFT, 10, 0 }, -parse_file* parse_open(const char *file); -void parse_file_close(parse_file*); + { "?", 3, opid2('?',':'), ASSOC_RIGHT, 9, 0 }, -bool parse(parse_file*); + { "=", 2, opid1('='), ASSOC_RIGHT, 8, 0 }, + { "+=", 2, opid2('+','='), ASSOC_RIGHT, 8, 0 }, + { "-=", 2, opid2('-','='), ASSOC_RIGHT, 8, 0 }, + { "*=", 2, opid2('*','='), ASSOC_RIGHT, 8, 0 }, + { "/=", 2, opid2('/','='), ASSOC_RIGHT, 8, 0 }, + { "%=", 2, opid2('%','='), ASSOC_RIGHT, 8, 0 }, + { "&=", 2, opid2('&','='), ASSOC_RIGHT, 8, 0 }, + { "|=", 2, opid2('|','='), ASSOC_RIGHT, 8, 0 }, + { "&~=", 2, opid3('&','~','='), ASSOC_RIGHT, 8, 0 }, -bool parse_iskey(parse_file *self, const char *ident); + { "&&", 2, opid2('&','&'), ASSOC_LEFT, 5, 0 }, + { "||", 2, opid2('|','|'), ASSOC_LEFT, 5, 0 }, + + /* Leave precedence 3 for : with -fcorrect-ternary */ + { ",", 2, opid1(','), ASSOC_LEFT, 2, 0 }, + { ":", 0, opid2(':','?'), ASSOC_RIGHT, 1, 0 } +}; +static const size_t fte_operator_count = (sizeof(fte_operators) / sizeof(fte_operators[0])); + +static const oper_info qcc_operators[] = { + { "(", 0, opid1('('), ASSOC_LEFT, 99, OP_PREFIX}, /* paren expression - non function call */ + + { ".", 2, opid1('.'), ASSOC_LEFT, 15, 0 }, + { "(", 0, opid1('('), ASSOC_LEFT, 15, 0 }, /* function call */ + { "[", 2, opid1('['), ASSOC_LEFT, 15, 0 }, /* array subscript */ + + { "!", 1, opid2('!', 'P'), ASSOC_RIGHT, 14, OP_PREFIX }, + { "+", 1, opid2('+','P'), ASSOC_RIGHT, 14, OP_PREFIX }, + { "-", 1, opid2('-','P'), ASSOC_RIGHT, 14, OP_PREFIX }, + + { "*", 2, opid1('*'), ASSOC_LEFT, 13, 0 }, + { "/", 2, opid1('/'), ASSOC_LEFT, 13, 0 }, + { "&", 2, opid1('&'), ASSOC_LEFT, 13, 0 }, + { "|", 2, opid1('|'), ASSOC_LEFT, 13, 0 }, + + { "+", 2, opid1('+'), ASSOC_LEFT, 12, 0 }, + { "-", 2, opid1('-'), ASSOC_LEFT, 12, 0 }, + + { "<", 2, opid1('<'), ASSOC_LEFT, 10, 0 }, + { ">", 2, opid1('>'), ASSOC_LEFT, 10, 0 }, + { "<=", 2, opid2('<','='), ASSOC_LEFT, 10, 0 }, + { ">=", 2, opid2('>','='), ASSOC_LEFT, 10, 0 }, + { "==", 2, opid2('=','='), ASSOC_LEFT, 10, 0 }, + { "!=", 2, opid2('!','='), ASSOC_LEFT, 10, 0 }, + + { "=", 2, opid1('='), ASSOC_RIGHT, 8, 0 }, + { "+=", 2, opid2('+','='), ASSOC_RIGHT, 8, 0 }, + { "-=", 2, opid2('-','='), ASSOC_RIGHT, 8, 0 }, + { "*=", 2, opid2('*','='), ASSOC_RIGHT, 8, 0 }, + { "/=", 2, opid2('/','='), ASSOC_RIGHT, 8, 0 }, + { "%=", 2, opid2('%','='), ASSOC_RIGHT, 8, 0 }, + { "&=", 2, opid2('&','='), ASSOC_RIGHT, 8, 0 }, + { "|=", 2, opid2('|','='), ASSOC_RIGHT, 8, 0 }, + + { "&&", 2, opid2('&','&'), ASSOC_LEFT, 5, 0 }, + { "||", 2, opid2('|','|'), ASSOC_LEFT, 5, 0 }, + + { ",", 2, opid1(','), ASSOC_LEFT, 2, 0 }, +}; +static const size_t qcc_operator_count = (sizeof(qcc_operators) / sizeof(qcc_operators[0])); +extern const oper_info *operators; +extern size_t operator_count; void lexerror(lex_file*, const char *fmt, ...); #endif