libabacus/src/lexer.c

110 lines
3.3 KiB
C
Raw Normal View History

2018-02-10 13:52:33 -08:00
#include "lexer.h"
2018-02-10 17:09:35 -08:00
#include "ll.h"
2018-02-11 22:32:42 -08:00
#include "util.h"
2018-02-10 17:09:35 -08:00
#include <ctype.h>
2018-04-21 14:09:01 -07:00
#include <stdlib.h>
libab_result libab_lexer_init(libab_lexer* lexer) {
size_t i;
libab_result result = LIBAB_SUCCESS;
2018-04-21 14:09:01 -07:00
const char* words[] = {".",
"[a-zA-Z][a-zA-Z0-9_]*",
"[0-9]+(\\.[0-9]*)?",
"if",
"else",
"while",
"do",
"->",
"fun",
"return"};
libab_lexer_token tokens[] = {
2018-04-21 14:09:01 -07:00
TOKEN_CHAR, TOKEN_ID, TOKEN_NUM, TOKEN_KW_IF,
TOKEN_KW_ELSE, TOKEN_KW_WHILE, TOKEN_KW_DO, TOKEN_KW_ARROW,
TOKEN_KW_FUN, TOKEN_KW_RETURN };
2018-04-21 14:09:01 -07:00
const size_t count = sizeof(tokens) / sizeof(libab_lexer_token);
eval_config_init(&lexer->config);
2018-04-21 14:09:01 -07:00
for (i = 0; i < count && result == LIBAB_SUCCESS; i++) {
result = libab_convert_lex_result(
2018-04-21 14:09:01 -07:00
eval_config_add(&lexer->config, words[i], tokens[i]));
}
2018-04-21 14:09:01 -07:00
if (result != LIBAB_SUCCESS) {
2018-03-16 23:09:11 -07:00
eval_config_free(&lexer->config);
}
return result;
}
2018-02-10 17:09:35 -08:00
struct lexer_state {
size_t line;
size_t line_from;
const char* source;
ll* matches;
};
int _lexer_foreach_convert_match(void* data, va_list args) {
libab_result result = LIBAB_SUCCESS;
libab_lexer_match* new_match;
2018-02-10 17:09:35 -08:00
match* match = data;
struct lexer_state* state = va_arg(args, struct lexer_state*);
char first_char = state->source[match->from];
2018-04-21 14:09:01 -07:00
if (isspace(first_char)) {
/* Skip */
2018-04-21 14:09:01 -07:00
} else if (first_char == '\n') {
2018-02-10 17:09:35 -08:00
state->line++;
state->line_from = match->to;
2018-04-21 14:09:01 -07:00
} else if ((new_match = malloc(sizeof(*new_match)))) {
new_match->type = match->pattern;
2018-02-10 17:09:35 -08:00
new_match->from = match->from;
new_match->to = match->to;
new_match->line_from = state->line_from;
new_match->line = state->line;
result = libab_convert_ds_result(ll_append(state->matches, new_match));
2018-04-21 14:09:01 -07:00
if (result != LIBAB_SUCCESS) {
2018-02-10 17:09:35 -08:00
free(new_match);
}
} else {
result = LIBAB_MALLOC;
}
return result;
}
2018-04-21 14:09:01 -07:00
libab_result libab_lexer_lex(libab_lexer* lexer, const char* string,
ll* lex_into) {
2018-02-10 17:09:35 -08:00
libab_result result;
ll raw_matches;
struct lexer_state state;
ll_init(&raw_matches);
state.line = 0;
state.line_from = 0;
state.matches = lex_into;
state.source = string;
result = libab_convert_lex_result(
2018-04-21 14:09:01 -07:00
eval_all(string, 0, &lexer->config, &raw_matches));
2018-02-10 17:09:35 -08:00
2018-04-21 14:09:01 -07:00
if (result == LIBAB_SUCCESS) {
result = (libab_result)ll_foreach(&raw_matches, NULL, compare_always,
_lexer_foreach_convert_match, &state);
2018-02-10 17:09:35 -08:00
}
2018-04-21 14:09:01 -07:00
if (result != LIBAB_SUCCESS) {
ll_foreach(lex_into, NULL, compare_always,
libab_lexer_foreach_match_free);
2018-02-10 17:09:35 -08:00
}
ll_foreach(&raw_matches, NULL, compare_always, eval_foreach_match_free);
ll_free(&raw_matches);
return result;
}
libab_result libab_lexer_free(libab_lexer* lexer) {
return libab_convert_lex_result(eval_config_free(&lexer->config));
}
int libab_lexer_foreach_match_free(void* data, va_list args) {
2018-04-21 14:09:01 -07:00
free((libab_lexer_match*)data);
2018-02-10 17:09:35 -08:00
return 0;
}