X-Git-Url: http://nsz.repo.hu/git/?a=blobdiff_plain;f=lexer.c;h=5ca734a9256a171d0ac75c1898de04d0b8dc96c2;hb=23e241346f8f62eb0bdd52fe5a0c62b38fcb4fd8;hp=4956dfcb9ec9b3c78cfd9d03a615fbb6ab1fec0c;hpb=9a4380489c081acfc4400860b39b7722f2d31763;p=cparser diff --git a/lexer.c b/lexer.c index 4956dfc..5ca734a 100644 --- a/lexer.c +++ b/lexer.c @@ -1,7 +1,27 @@ +/* + * This file is part of cparser. + * Copyright (C) 2007-2008 Matthias Braun + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + */ #include #include "diagnostic.h" #include "lexer.h" +#include "symbol_t.h" #include "token_t.h" #include "symbol_table_t.h" #include "adt/error.h" @@ -12,6 +32,7 @@ #include "target_architecture.h" #include "parser.h" #include "warning.h" +#include "lang_features.h" #include #include @@ -27,14 +48,6 @@ #define strtold(s, e) strtod(s, e) #endif -#if defined HAS_SIGNED_CHAR -typedef signed char char_type; -#elif defined HAS_UNSIGNED_CHAR -typedef unsigned char char_type; -#else -# error signedness of char not determined -#endif - static int c; token_t lexer_token; symbol_t *symbol_L; @@ -43,6 +56,7 @@ static char buf[1024 + MAX_PUTBACK]; static const char *bufend; static const char *bufpos; static strset_t stringset; +bool allow_dollar_in_symbol = true; /** * Prints a parse error message at the current token. @@ -51,13 +65,28 @@ static strset_t stringset; */ static void parse_error(const char *msg) { - errorf(lexer_token.source_position, "%s", msg); + errorf(&lexer_token.source_position, "%s", msg); +} + +/** + * Prints an internal error message at the current token. + * + * @param msg the error message + */ +static NORETURN internal_error(const char *msg) +{ + internal_errorf(&lexer_token.source_position, "%s", msg); } static inline void next_real_char(void) { assert(bufpos <= bufend); if (bufpos >= bufend) { + if (input == NULL) { + c = EOF; + return; + } + size_t s = fread(buf + MAX_PUTBACK, 1, sizeof(buf) - MAX_PUTBACK, input); if(s == 0) { @@ -70,6 +99,11 @@ static inline void next_real_char(void) c = *bufpos++; } +/** + * Put a character back into the buffer. + * + * @param pc the character to put back + */ static inline void put_back(int pc) { assert(bufpos > buf); @@ -112,6 +146,10 @@ static void maybe_concat_lines(void) c = '\\'; } +/** + * Set c to the next input character, ie. + * after expanding trigraphs. + */ static inline void next_char(void) { next_real_char(); @@ -157,6 +195,7 @@ end_of_next_char:; } #define SYMBOL_CHARS \ + case '$': if (!allow_dollar_in_symbol) goto dollar_sign; \ case 'a': \ case 'b': \ case 'c': \ @@ -223,6 +262,10 @@ end_of_next_char:; case '8': \ case '9': +/** + * Read a symbol from the input and build + * the lexer_token. + */ static void parse_symbol(void) { symbol_t *symbol; @@ -240,6 +283,7 @@ static void parse_symbol(void) break; default: +dollar_sign: goto end_symbol; } } @@ -307,7 +351,7 @@ static void parse_integer_suffix(bool is_oct_hex) if(v >= TARGET_LONG_MIN && v <= TARGET_LONG_MAX) { lexer_token.datatype = type_long; return; - } else if(is_oct_hex && v >= 0 && v <= TARGET_ULONG_MAX) { + } else if(is_oct_hex && v >= 0 && (unsigned long long)v <= (unsigned long long)TARGET_ULONG_MAX) { lexer_token.datatype = type_unsigned_long; return; } @@ -336,7 +380,7 @@ static void parse_integer_suffix(bool is_oct_hex) static void parse_floating_suffix(void) { switch(c) { - /* TODO: do something usefull with the suffixes... */ + /* TODO: do something useful with the suffixes... */ case 'f': case 'F': next_char(); @@ -437,43 +481,107 @@ end: return v; } +/** + * Parses a hex number including hex floats and set the + * lexer_token. + */ static void parse_number_hex(void) { + bool is_float = false; assert(c == 'x' || c == 'X'); next_char(); + obstack_1grow(&symbol_obstack, '0'); + obstack_1grow(&symbol_obstack, 'x'); + while(isxdigit(c)) { obstack_1grow(&symbol_obstack, (char) c); next_char(); } - obstack_1grow(&symbol_obstack, '\0'); - char *string = obstack_finish(&symbol_obstack); - if(c == '.' || c == 'p' || c == 'P') { + if (c == '.') { + obstack_1grow(&symbol_obstack, (char) c); next_char(); - panic("Hex floating point numbers not implemented yet"); + + while (isxdigit(c)) { + obstack_1grow(&symbol_obstack, (char) c); + next_char(); + } + is_float = true; + } + if (c == 'p' || c == 'P') { + obstack_1grow(&symbol_obstack, (char) c); + next_char(); + + if (c == '-' || c == '+') { + obstack_1grow(&symbol_obstack, (char) c); + next_char(); + } + + while (isxdigit(c)) { + obstack_1grow(&symbol_obstack, (char) c); + next_char(); + } + is_float = true; } + + obstack_1grow(&symbol_obstack, '\0'); + char *string = obstack_finish(&symbol_obstack); if(*string == '\0') { parse_error("invalid hex number"); lexer_token.type = T_ERROR; + obstack_free(&symbol_obstack, string); + return; } - const char *endptr; - lexer_token.type = T_INTEGER; - lexer_token.v.intvalue = parse_int_string(string, &endptr, 16); - if(*endptr != '\0') { - parse_error("hex number literal too long"); + if (is_float) { + char *endptr; + lexer_token.type = T_FLOATINGPOINT; + lexer_token.v.floatvalue = strtold(string, &endptr); + + if(*endptr != '\0') { + parse_error("invalid hex float literal"); + } + + parse_floating_suffix(); + } else { + const char *endptr; + lexer_token.type = T_INTEGER; + lexer_token.v.intvalue = parse_int_string(string + 2, &endptr, 16); + if(*endptr != '\0') { + parse_error("hex number literal too long"); + } + parse_integer_suffix(true); } obstack_free(&symbol_obstack, string); - parse_integer_suffix(true); } +/** + * Returns true if the given char is a octal digit. + * + * @param char the character to check + */ static inline bool is_octal_digit(int chr) { - return '0' <= chr && chr <= '7'; + switch(chr) { + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + return true; + default: + return false; + } } +/** + * Parses a octal number and set the lexer_token. + */ static void parse_number_oct(void) { while(is_octal_digit(c)) { @@ -494,26 +602,30 @@ static void parse_number_oct(void) parse_integer_suffix(true); } +/** + * Parses a decimal including float number and set the + * lexer_token. + */ static void parse_number_dec(void) { bool is_float = false; - while(isdigit(c)) { + while (isdigit(c)) { obstack_1grow(&symbol_obstack, (char) c); next_char(); } - if(c == '.') { + if (c == '.') { obstack_1grow(&symbol_obstack, '.'); next_char(); - while(isdigit(c)) { + while (isdigit(c)) { obstack_1grow(&symbol_obstack, (char) c); next_char(); } is_float = true; } if(c == 'e' || c == 'E') { - obstack_1grow(&symbol_obstack, 'e'); + obstack_1grow(&symbol_obstack, (char) c); next_char(); if(c == '-' || c == '+') { @@ -555,6 +667,9 @@ static void parse_number_dec(void) obstack_free(&symbol_obstack, string); } +/** + * Parses a number and sets the lexer_token. + */ static void parse_number(void) { if (c == '0') { @@ -593,38 +708,83 @@ static void parse_number(void) } } +/** + * Returns the value of a digit. + * The only portable way to do it ... + */ +static int digit_value(int digit) { + switch (digit) { + case '0': return 0; + case '1': return 1; + case '2': return 2; + case '3': return 3; + case '4': return 4; + case '5': return 5; + case '6': return 6; + case '7': return 7; + case '8': return 8; + case '9': return 9; + case 'a': + case 'A': return 10; + case 'b': + case 'B': return 11; + case 'c': + case 'C': return 12; + case 'd': + case 'D': return 13; + case 'e': + case 'E': return 14; + case 'f': + case 'F': return 15; + default: + internal_error("wrong character given"); + } +} + +/** + * Parses an octal character sequence. + * + * @param first_digit the already read first digit + */ static int parse_octal_sequence(const int first_digit) { assert(is_octal_digit(first_digit)); - int value = first_digit - '0'; + int value = digit_value(first_digit); if (!is_octal_digit(c)) return value; - value = 8 * value + c - '0'; + value = 8 * value + digit_value(c); next_char(); if (!is_octal_digit(c)) return value; - value = 8 * value + c - '0'; + value = 8 * value + digit_value(c); next_char(); - return (char_type)value; + + if(char_is_signed) { + return (signed char) value; + } else { + return (unsigned char) value; + } } +/** + * Parses a hex character sequence. + */ static int parse_hex_sequence(void) { int value = 0; - while(1) { - if (c >= '0' && c <= '9') { - value = 16 * value + c - '0'; - } else if ('A' <= c && c <= 'F') { - value = 16 * value + c - 'A' + 10; - } else if ('a' <= c && c <= 'f') { - value = 16 * value + c - 'a' + 10; - } else { - break; - } + while(isxdigit(c)) { + value = 16 * value + digit_value(c); next_char(); } - return (char_type)value; + if(char_is_signed) { + return (signed char) value; + } else { + return (unsigned char) value; + } } +/** + * Parse an escape sequence. + */ static int parse_escape_sequence(void) { eat('\\'); @@ -632,7 +792,7 @@ static int parse_escape_sequence(void) int ec = c; next_char(); - switch(ec) { + switch (ec) { case '"': return '"'; case '\'': return '\''; case '\\': return '\\'; @@ -658,12 +818,19 @@ static int parse_escape_sequence(void) case EOF: parse_error("reached end of file while parsing escape sequence"); return EOF; + case 'e': + if (c_mode & _GNUC) + return 27; /* hopefully 27 is ALWAYS the code for ESACAPE */ + /*fallthrough*/ default: parse_error("unknown escape sequence"); return EOF; } } +/** + * Concatenate two strings. + */ string_t concat_strings(const string_t *const s1, const string_t *const s2) { const size_t len1 = s1->size - 1; @@ -685,6 +852,60 @@ string_t concat_strings(const string_t *const s1, const string_t *const s2) #endif } +/** + * Concatenate a string and a wide string. + */ +wide_string_t concat_string_wide_string(const string_t *const s1, const wide_string_t *const s2) +{ + const size_t len1 = s1->size - 1; + const size_t len2 = s2->size - 1; + + wchar_rep_t *const concat = obstack_alloc(&symbol_obstack, (len1 + len2 + 1) * sizeof(*concat)); + const char *const src = s1->begin; + for (size_t i = 0; i != len1; ++i) { + concat[i] = src[i]; + } + memcpy(concat + len1, s2->begin, (len2 + 1) * sizeof(*concat)); + + return (wide_string_t){ concat, len1 + len2 + 1 }; +} + +/** + * Concatenate two wide strings. + */ +wide_string_t concat_wide_strings(const wide_string_t *const s1, const wide_string_t *const s2) +{ + const size_t len1 = s1->size - 1; + const size_t len2 = s2->size - 1; + + wchar_rep_t *const concat = obstack_alloc(&symbol_obstack, (len1 + len2 + 1) * sizeof(*concat)); + memcpy(concat, s1->begin, len1 * sizeof(*concat)); + memcpy(concat + len1, s2->begin, (len2 + 1) * sizeof(*concat)); + + return (wide_string_t){ concat, len1 + len2 + 1 }; +} + +/** + * Concatenate a wide string and a string. + */ +wide_string_t concat_wide_string_string(const wide_string_t *const s1, const string_t *const s2) +{ + const size_t len1 = s1->size - 1; + const size_t len2 = s2->size - 1; + + wchar_rep_t *const concat = obstack_alloc(&symbol_obstack, (len1 + len2 + 1) * sizeof(*concat)); + memcpy(concat, s1->begin, len1 * sizeof(*concat)); + const char *const src = s2->begin; + for (size_t i = 0; i != len2 + 1; ++i) { + concat[i] = src[i]; + } + + return (wide_string_t){ concat, len1 + len2 + 1 }; +} + +/** + * Parse a string literal and set lexer_token. + */ static void parse_string_literal(void) { const unsigned start_linenr = lexer_token.source_position.linenr; @@ -703,7 +924,7 @@ static void parse_string_literal(void) source_position_t source_position; source_position.input_name = lexer_token.source_position.input_name; source_position.linenr = start_linenr; - errorf(source_position, "string has no end"); + errorf(&source_position, "string has no end"); lexer_token.type = T_ERROR; return; } @@ -743,16 +964,22 @@ end_of_string: lexer_token.v.string.size = size; } +/** + * Parse a wide character constant and set lexer_token. + */ static void parse_wide_character_constant(void) { + const unsigned start_linenr = lexer_token.source_position.linenr; + eat('\''); - int found_char = 0; while(1) { switch(c) { - case '\\': - found_char = parse_escape_sequence(); + case '\\': { + wchar_rep_t tc = parse_escape_sequence(); + obstack_grow(&symbol_obstack, &tc, sizeof(tc)); break; + } MATCH_NEWLINE( parse_error("newline while parsing character constant"); @@ -763,30 +990,39 @@ static void parse_wide_character_constant(void) next_char(); goto end_of_wide_char_constant; - case EOF: - parse_error("EOF while parsing character constant"); + case EOF: { + source_position_t source_position = lexer_token.source_position; + source_position.linenr = start_linenr; + errorf(&source_position, "EOF while parsing character constant"); lexer_token.type = T_ERROR; return; + } - default: - if(found_char != 0) { - parse_error("more than 1 characters in character " - "constant"); - goto end_of_wide_char_constant; - } else { - found_char = c; - next_char(); - } + default: { + wchar_rep_t tc = (wchar_rep_t) c; + obstack_grow(&symbol_obstack, &tc, sizeof(tc)); + next_char(); break; } + } } -end_of_wide_char_constant: - lexer_token.type = T_INTEGER; - lexer_token.v.intvalue = found_char; - lexer_token.datatype = type_wchar_t; +end_of_wide_char_constant:; + size_t size = (size_t) obstack_object_size(&symbol_obstack); + assert(size % sizeof(wchar_rep_t) == 0); + size /= sizeof(wchar_rep_t); + + const wchar_rep_t *string = obstack_finish(&symbol_obstack); + + lexer_token.type = T_WIDE_CHARACTER_CONSTANT; + lexer_token.v.wide_string.begin = string; + lexer_token.v.wide_string.size = size; + lexer_token.datatype = type_wchar_t; } +/** + * Parse a wide string literal and set lexer_token. + */ static void parse_wide_string_literal(void) { const unsigned start_linenr = lexer_token.source_position.linenr; @@ -806,7 +1042,7 @@ static void parse_wide_string_literal(void) source_position_t source_position; source_position.input_name = lexer_token.source_position.input_name; source_position.linenr = start_linenr; - errorf(source_position, "string has no end"); + errorf(&source_position, "string has no end"); lexer_token.type = T_ERROR; return; } @@ -849,38 +1085,41 @@ end_of_string:; lexer_token.v.wide_string.size = size; } +/** + * Parse a character constant and set lexer_token. + */ static void parse_character_constant(void) { const unsigned start_linenr = lexer_token.source_position.linenr; eat('\''); - int tc; while(1) { switch(c) { - case '\\': - tc = parse_escape_sequence(); + case '\\': { + int tc = parse_escape_sequence(); obstack_1grow(&symbol_obstack, (char) tc); break; + } MATCH_NEWLINE( parse_error("newline while parsing character constant"); break; ) + case '\'': + next_char(); + goto end_of_char_constant; + case EOF: { source_position_t source_position; source_position.input_name = lexer_token.source_position.input_name; source_position.linenr = start_linenr; - errorf(source_position, "EOF while parsing character constant"); + errorf(&source_position, "EOF while parsing character constant"); lexer_token.type = T_ERROR; return; } - case '\'': - next_char(); - goto end_of_char_constant; - default: obstack_1grow(&symbol_obstack, (char) c); next_char(); @@ -893,12 +1132,15 @@ end_of_char_constant:; const size_t size = (size_t)obstack_object_size(&symbol_obstack); const char *const string = obstack_finish(&symbol_obstack); - lexer_token.type = T_CHARS; + lexer_token.type = T_CHARACTER_CONSTANT; lexer_token.v.string.begin = string; lexer_token.v.string.size = size; lexer_token.datatype = type_int; } +/** + * Skip a multiline comment. + */ static void skip_multiline_comment(void) { unsigned start_linenr = lexer_token.source_position.linenr; @@ -908,7 +1150,10 @@ static void skip_multiline_comment(void) case '/': next_char(); if (c == '*') { - /* TODO: nested comment, warn here */ + /* nested comment, warn here */ + if (warning.comment) { + warningf(&lexer_token.source_position, "'/*' within comment"); + } } break; case '*': @@ -925,7 +1170,7 @@ static void skip_multiline_comment(void) source_position_t source_position; source_position.input_name = lexer_token.source_position.input_name; source_position.linenr = start_linenr; - errorf(source_position, "at end of file while looking for comment end"); + errorf(&source_position, "at end of file while looking for comment end"); return; } @@ -936,6 +1181,9 @@ static void skip_multiline_comment(void) } } +/** + * Skip a single line comment. + */ static void skip_line_comment(void) { while(1) { @@ -947,6 +1195,15 @@ static void skip_line_comment(void) case '\r': return; + case '\\': + next_char(); + if (c == '\n' || c == '\r') { + if (warning.comment) + warningf(&lexer_token.source_position, "multi-line comment"); + return; + } + break; + default: next_char(); break; @@ -954,14 +1211,21 @@ static void skip_line_comment(void) } } +/** The current preprocessor token. */ static token_t pp_token; +/** + * Read the next preprocessor token. + */ static inline void next_pp_token(void) { lexer_next_preprocessing_token(); pp_token = lexer_token; } +/** + * Eat all preprocessor tokens until newline. + */ static void eat_until_newline(void) { while(pp_token.type != '\n' && pp_token.type != T_EOF) { @@ -969,6 +1233,9 @@ static void eat_until_newline(void) } } +/** + * Handle the define directive. + */ static void define_directive(void) { lexer_next_preprocessing_token(); @@ -978,6 +1245,9 @@ static void define_directive(void) } } +/** + * Handle the ifdef directive. + */ static void ifdef_directive(int is_ifndef) { (void) is_ifndef; @@ -986,11 +1256,17 @@ static void ifdef_directive(int is_ifndef) //extect_newline(); } +/** + * Handle the endif directive. + */ static void endif_directive(void) { //expect_newline(); } +/** + * Parse the line directive. + */ static void parse_line_directive(void) { if(pp_token.type != T_INTEGER) { @@ -1007,6 +1283,87 @@ static void parse_line_directive(void) eat_until_newline(); } +/** + * STDC pragmas. + */ +typedef enum stdc_pragma_kind_t { + STDC_UNKNOWN, + STDC_FP_CONTRACT, + STDC_FENV_ACCESS, + STDC_CX_LIMITED_RANGE +} stdc_pragma_kind_t; + +/** + * STDC pragma values. + */ +typedef enum stdc_pragma_value_kind_t { + STDC_VALUE_UNKNOWN, + STDC_VALUE_ON, + STDC_VALUE_OFF, + STDC_VALUE_DEFAULT +} stdc_pragma_value_kind_t; + +/** + * Parse a pragma directive. + */ +static void parse_pragma(void) { + bool unknown_pragma = true; + + next_pp_token(); + if (pp_token.v.symbol->pp_ID == TP_STDC) { + stdc_pragma_kind_t kind = STDC_UNKNOWN; + /* a STDC pragma */ + if (c_mode & _C99) { + next_pp_token(); + + switch (pp_token.v.symbol->pp_ID) { + case TP_FP_CONTRACT: + kind = STDC_FP_CONTRACT; + break; + case TP_FENV_ACCESS: + kind = STDC_FENV_ACCESS; + break; + case TP_CX_LIMITED_RANGE: + kind = STDC_CX_LIMITED_RANGE; + break; + default: + break; + } + if (kind != STDC_UNKNOWN) { + stdc_pragma_value_kind_t value = STDC_VALUE_UNKNOWN; + next_pp_token(); + switch (pp_token.v.symbol->pp_ID) { + case TP_ON: + value = STDC_VALUE_ON; + break; + case TP_OFF: + value = STDC_VALUE_OFF; + break; + case TP_DEFAULT: + value = STDC_VALUE_DEFAULT; + break; + default: + break; + } + if (value != STDC_VALUE_UNKNOWN) { + unknown_pragma = false; + } else { + errorf(&pp_token.source_position, "bad STDC pragma argument"); + } + } + } + } else { + unknown_pragma = true; + } + eat_until_newline(); + if (unknown_pragma && warning.unknown_pragmas) { + warningf(&pp_token.source_position, "encountered unknown #pragma"); + } +} + +/** + * Parse a preprocessor non-null directive. + */ static void parse_preprocessor_identifier(void) { assert(pp_token.type == T_IDENTIFIER); @@ -1041,14 +1398,14 @@ static void parse_preprocessor_identifier(void) parse_error("#error directive: "); break; case TP_pragma: - if (warning.unknown_pragmas) { - warningf(lexer_token.source_position, "encountered unknown #pragma"); - } - eat_until_newline(); + parse_pragma(); break; } } +/** + * Parse a preprocessor directive. + */ static void parse_preprocessor_directive(void) { next_pp_token(); @@ -1060,6 +1417,9 @@ static void parse_preprocessor_directive(void) case T_INTEGER: parse_line_directive(); break; + case '\n': + /* NULL directive, see § 6.10.7 */ + break; default: parse_error("invalid preprocessor directive"); eat_until_newline(); @@ -1080,7 +1440,7 @@ static void parse_preprocessor_directive(void) #define ELSE_CODE(code) \ default: \ - code; \ + code \ } \ } /* end of while(1) */ \ break; @@ -1132,16 +1492,7 @@ void lexer_next_preprocessing_token(void) case '.': MAYBE_PROLOG - case '0': - case '1': - case '2': - case '3': - case '4': - case '5': - case '6': - case '7': - case '8': - case '9': + DIGITS put_back(c); c = '.'; parse_number_dec(); @@ -1197,25 +1548,25 @@ void lexer_next_preprocessing_token(void) ELSE('/') case '%': MAYBE_PROLOG - MAYBE('>', T_PERCENTGREATER) + MAYBE('>', '}') MAYBE('=', T_PERCENTEQUAL) case ':': MAYBE_PROLOG case '%': MAYBE_PROLOG - MAYBE(':', T_PERCENTCOLONPERCENTCOLON) + MAYBE(':', T_HASHHASH) ELSE_CODE( put_back(c); c = '%'; - lexer_token.type = T_PERCENTCOLON; + lexer_token.type = '#'; return; ) - ELSE(T_PERCENTCOLON) + ELSE('#') ELSE('%') case '<': MAYBE_PROLOG - MAYBE(':', T_LESSCOLON) - MAYBE('%', T_LESSPERCENT) + MAYBE(':', '[') + MAYBE('%', '{') MAYBE('=', T_LESSEQUAL) case '<': MAYBE_PROLOG @@ -1241,7 +1592,7 @@ void lexer_next_preprocessing_token(void) ELSE('|') case ':': MAYBE_PROLOG - MAYBE('>', T_COLONGREATER) + MAYBE('>', ']') ELSE(':') case '=': MAYBE_PROLOG @@ -1272,8 +1623,9 @@ void lexer_next_preprocessing_token(void) return; default: +dollar_sign: + errorf(&lexer_token.source_position, "unknown character '%c' found", c); next_char(); - errorf(lexer_token.source_position, "unknown character '%c' found\n", c); lexer_token.type = T_ERROR; return; } @@ -1283,15 +1635,13 @@ void lexer_next_preprocessing_token(void) void lexer_next_token(void) { lexer_next_preprocessing_token(); - if(lexer_token.type != '\n') - return; + while (lexer_token.type == '\n') { newline_found: - do { lexer_next_preprocessing_token(); - } while(lexer_token.type == '\n'); + } - if(lexer_token.type == '#') { + if (lexer_token.type == '#') { parse_preprocessor_directive(); goto newline_found; } @@ -1300,6 +1650,7 @@ newline_found: void init_lexer(void) { strset_init(&stringset); + symbol_L = symbol_table_insert("L"); } void lexer_open_stream(FILE *stream, const char *input_name) @@ -1308,7 +1659,6 @@ void lexer_open_stream(FILE *stream, const char *input_name) lexer_token.source_position.linenr = 0; lexer_token.source_position.input_name = input_name; - symbol_L = symbol_table_insert("L"); bufpos = NULL; bufend = NULL; @@ -1317,6 +1667,20 @@ void lexer_open_stream(FILE *stream, const char *input_name) c = '\n'; } +void lexer_open_buffer(const char *buffer, size_t len, const char *input_name) +{ + input = NULL; + lexer_token.source_position.linenr = 0; + lexer_token.source_position.input_name = input_name; + + bufpos = buffer; + bufend = buffer + len; + + /* place a virtual \n at the beginning so the lexer knows that we're + * at the beginning of a line */ + c = '\n'; +} + void exit_lexer(void) { strset_destroy(&stringset);