static int c;
token_t lexer_token;
+symbol_t *symbol_L;
static FILE *input;
static char buf[1024 + MAX_PUTBACK];
static const char *bufend;
static const char *bufpos;
static strset_t stringset;
-//static FILE **input_stack;
-//static char **buf_stack;
static void error_prefix_at(const char *input_name, unsigned linenr)
{
static inline void put_back(int pc)
{
- char *p = (char*) bufpos - 1;
- bufpos--;
- assert(p >= buf);
+ assert(bufpos >= buf);
+ assert(bufpos < buf+MAX_PUTBACK || *bufpos == pc);
+
+ char *p = buf + (bufpos - buf);
*p = pc;
+ /* going backwards in the buffer is legal as long as it's not more often
+ * than MAX_PUTBACK */
+ bufpos--;
+
#ifdef DEBUG_CHARS
printf("putback '%c'\n", pc);
#endif
lexer_token.source_position.linenr++; \
code;
+#define eat(c_type) do { assert(c == c_type); next_char(); } while(0)
+
static void maybe_concat_lines(void)
{
- next_char();
+ eat('\\');
+
switch(c) {
MATCH_NEWLINE(return;)
{
next_real_char();
+#if 0
/* filter trigraphs */
if(UNLIKELY(c == '\\')) {
maybe_concat_lines();
}
end_of_next_char:
+#endif
+ (void) maybe_concat_lines;
#ifdef DEBUG_CHARS
printf("nchar '%c'\n", c);
#else
}
}
+static void parse_integer_suffix(void)
+{
+ if(c == 'U' || c == 'U') {
+ /* TODO do something with the suffixes... */
+ next_char();
+ if(c == 'L' || c == 'l') {
+ next_char();
+ if(c == 'L' || c == 'l') {
+ next_char();
+ }
+ }
+ } else if(c == 'l' || c == 'L') {
+ next_char();
+ if(c == 'l' || c == 'L') {
+ next_char();
+ if(c == 'u' || c == 'U') {
+ next_char();
+ }
+ } else if(c == 'u' || c == 'U') {
+ next_char();
+ }
+ }
+}
+
+static void parse_floating_suffix(void)
+{
+ switch(c) {
+ /* TODO: do something usefull with the suffixes... */
+ case 'f':
+ case 'F':
+ case 'l':
+ case 'L':
+ next_char();
+ break;
+ default:
+ break;
+ }
+}
+
static void parse_number_hex(void)
{
assert(c == 'x' || c == 'X');
} else if ('a' <= c && c <= 'f') {
value = 16 * value + c - 'a' + 10;
} else {
- lexer_token.type = T_INTEGER;
+ parse_integer_suffix();
+
+ lexer_token.type = T_INTEGER;
lexer_token.v.intvalue = value;
return;
}
next_char();
}
+
+ if(c == '.' || c == 'p' || c == 'P') {
+ next_char();
+ panic("Hex floating point numbers not implemented yet");
+ }
}
static void parse_number_oct(void)
{
int value = 0;
+ while(c >= '0' && c <= '7') {
+ value = 8 * value + c - '0';
+ next_char();
+ }
+ if (c == '8' || c == '9') {
+ parse_error("invalid octal number");
+ lexer_token.type = T_ERROR;
+ return;
+ }
+
+ lexer_token.type = T_INTEGER;
+ lexer_token.v.intvalue = value;
+
+ parse_integer_suffix();
+}
+
+static void parse_floatingpoint_exponent(long double value)
+{
+ unsigned int expo = 0;
+ long double factor = 10.;
+
+ if(c == '-') {
+ next_char();
+ factor = 0.1;
+ } else if(c == '+') {
+ next_char();
+ }
+
+ while(c >= '0' && c <= '9') {
+ expo = 10 * expo + (c - '0');
+ next_char();
+ }
+
while(1) {
- if ('0' <= c && c <= '7') {
- value = 8 * value + c - '0';
- } else if (c == '8' || c == '9') {
- parse_error("invalid octal number");
- lexer_token.type = T_ERROR;
- return;
- } else {
- lexer_token.type = T_INTEGER;
- lexer_token.v.intvalue = value;
- return;
- }
+ if(expo & 1)
+ value *= factor;
+ expo >>= 1;
+ if(expo == 0)
+ break;
+ factor *= factor;
+ }
+
+ lexer_token.type = T_FLOATINGPOINT;
+ lexer_token.v.floatvalue = value;
+
+ parse_floating_suffix();
+}
+
+static void parse_floatingpoint_fract(int integer_part)
+{
+ long double value = integer_part;
+ long double factor = 1.;
+
+ while(c >= '0' && c <= '9') {
+ factor *= 0.1;
+ value += (c - '0') * factor;
next_char();
}
+
+ if(c == 'e' || c == 'E') {
+ next_char();
+ parse_floatingpoint_exponent(value);
+ return;
+ }
+
+ lexer_token.type = T_FLOATINGPOINT;
+ lexer_token.v.floatvalue = value;
+
+ parse_floating_suffix();
}
static void parse_number_dec(void)
{
int value = 0;
- for(;;) {
- if (isdigit(c)) {
- value = 10 * value + c - '0';
- } else {
- lexer_token.type = T_INTEGER;
- lexer_token.v.intvalue = value;
- return;
- }
+ while(isdigit(c)) {
+ value = 10 * value + c - '0';
next_char();
}
-}
-static void parse_integer_suffix(void)
-{
- if(c == 'U' || c == 'U') {
- /* TODO do something with the suffixes... */
+ if(c == '.') {
next_char();
- if(c == 'L' || c == 'l') {
- next_char();
- if(c == 'L' || c == 'l') {
- next_char();
- }
- }
- } else if(c == 'l' || c == 'L') {
+ parse_floatingpoint_fract(value);
+ return;
+ }
+ if(c == 'e' || c == 'E') {
next_char();
- if(c == 'l' || c == 'L') {
- next_char();
- if(c == 'u' || c == 'U') {
- next_char();
- }
- } else if(c == 'u' || c == 'U') {
- next_char();
- }
+ parse_floatingpoint_exponent(value);
+ return;
}
+ parse_integer_suffix();
+
+ lexer_token.type = T_INTEGER;
+ lexer_token.v.intvalue = value;
}
static void parse_number(void)
next_char();
switch (c) {
case 'X':
- case 'x': parse_number_hex(); break;
- default: parse_number_oct(); break;
+ case 'x':
+ parse_number_hex();
+ break;
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ parse_number_oct();
+ break;
+ case '.':
+ next_char();
+ parse_floatingpoint_fract(0);
+ break;
+ case 'e':
+ case 'E':
+ parse_floatingpoint_exponent(0);
+ break;
+ case '8':
+ case '9':
+ next_char();
+ parse_error("invalid octal number");
+ lexer_token.type = T_ERROR;
+ return;
+ default:
+ put_back(c);
+ c = '0';
+ parse_number_dec();
+ return;
}
} else {
parse_number_dec();
}
-
- parse_integer_suffix();
}
static int parse_octal_sequence(void)
static int parse_escape_sequence(void)
{
- while(1) {
- int ec = c;
- next_char();
+ eat('\\');
- switch(ec) {
- case '"': return '"';
- case '\'': return'\'';
- case '\\': return '\\';
- case '?': return '\?';
- case 'a': return '\a';
- case 'b': return '\b';
- case 'f': return '\f';
- case 'n': return '\n';
- case 'r': return '\r';
- case 't': return '\t';
- case 'v': return '\v';
- case 'x':
- return parse_hex_sequence();
- case '0':
- case '1':
- case '2':
- case '3':
- case '4':
- case '5':
- case '6':
- case '7':
- return parse_octal_sequence();
- case EOF:
- parse_error("reached end of file while parsing escape sequence");
- return EOF;
- default:
- parse_error("unknown escape sequence");
- return EOF;
- }
+ int ec = c;
+ next_char();
+
+ switch(ec) {
+ case '"': return '"';
+ case '\'': return'\'';
+ case '\\': return '\\';
+ case '?': return '\?';
+ case 'a': return '\a';
+ case 'b': return '\b';
+ case 'f': return '\f';
+ case 'n': return '\n';
+ case 'r': return '\r';
+ case 't': return '\t';
+ case 'v': return '\v';
+ case 'x':
+ return parse_hex_sequence();
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ return parse_octal_sequence();
+ case EOF:
+ parse_error("reached end of file while parsing escape sequence");
+ return EOF;
+ default:
+ parse_error("unknown escape sequence");
+ return EOF;
}
}
assert(c == '"');
next_char();
+ int tc;
while(1) {
switch(c) {
case '\\':
- next_char();
- int ec = parse_escape_sequence();
- obstack_1grow(&symbol_obstack, ec);
+ tc = parse_escape_sequence();
+ obstack_1grow(&symbol_obstack, tc);
break;
case EOF:
static void parse_character_constant(void)
{
- assert(c == '\'');
- next_char();
+ eat('\'');
int found_char = 0;
while(1) {
switch(c) {
case '\\':
- next_char();
found_char = parse_escape_sequence();
break;
}
}
-static void parse_preprocessor_directive()
+static void parse_preprocessor_directive(void)
{
next_pp_token();
SYMBOL_CHARS
parse_symbol();
+ /* might be a wide string ( L"string" ) */
+ if(c == '"' && (lexer_token.type == T_IDENTIFIER &&
+ lexer_token.v.symbol == symbol_L)) {
+ parse_string_literal();
+ return;
+ }
return;
DIGITS
void lexer_open_stream(FILE *stream, const char *input_name)
{
input = stream;
- lexer_token.source_position.linenr = 1;
+ lexer_token.source_position.linenr = 0;
lexer_token.source_position.input_name = input_name;
- next_char();
+ symbol_L = symbol_table_insert("L");
+
+ /* 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)