libhl/source/hl.c

255 lines
5.5 KiB
C
Raw Normal View History

2023-08-29 13:09:55 -04:00
#include "hl.h"
#include <assert.h>
vector_t token_table = {
.data = NULL,
.element_size = sizeof(token_t *),
.element_count = 0UL
};
display_t * display_table = NULL;
// -------------------------
// ### Library Mangement ###
// -------------------------
hl_group_t * special_hl = NULL;
hl_group_t * control_hl = NULL;
hl_group_t * keyword_hl = NULL;
hl_group_t * block_hl = NULL;
hl_group_t * separator_hl = NULL;
hl_group_t * operator_hl = NULL;
hl_group_t * comment_hl = NULL;
hl_group_t * string_literal_hl = NULL;
// --------------------------------
// ### Constructors/Destructors ###
// --------------------------------
void new_display_mode(display_t * mode) {
HASH_ADD_STR(display_table,
key,
mode);
}
int free_token(token_t * token) {
free(token->hl);
regex_free(token->syntax);
return 0;
}
int append_token(token_t * token) {
vector_push(&token_table, &token);
return 0;
}
token_t * new_symbol_token(const char * const c,
hl_group_t * const g) {
token_t * mt = (token_t*)malloc(sizeof(token_t));
mt->hl = g;
mt->t = KEYSYMBOL;
mt->syntax = regex_compile(c);
append_token(mt);
return mt;
}
int new_symbol_tokens(const char * const * symbols,
hl_group_t * const g) {
int i = 0;
while (*symbols) {
if(new_symbol_token(*symbols, g)) {
++i;
} else {
assert(!(bool)"Kinda failed to new symbol token thing.");
}
++symbols;
}
return i;
}
int new_char_tokens(const char * str,
hl_group_t * const g) {
int i = 0;
char buffer[3];
buffer[0] = '\\';
buffer[2] = '\0';
for(const char * s = str; *s != '\0'; s++) {
buffer[1] = *s;
if(new_symbol_token(is_magic(*s) ? buffer : buffer + 1, g)) {
++i;
} else {
assert(!(bool)"Kinda failed to new char token thing.");
}
}
return i;
}
token_t * new_keyword_token(const char * const word,
hl_group_t * const g) {
2023-09-21 03:06:40 -04:00
size_t word_length = strlen(word);
char * new_word = (char*)malloc(word_length + 4 + 1);
2023-08-29 13:09:55 -04:00
2023-09-21 03:06:40 -04:00
memcpy(new_word, "\\<", 2);
memcpy(new_word + 2, word, word_length);
strcpy(new_word + 2 + word_length, "\\>");
2023-08-29 13:09:55 -04:00
token_t * mt = (token_t*)malloc(sizeof(token_t));
mt->hl = g;
mt->t = KEYWORD;
2023-09-21 03:06:40 -04:00
mt->syntax = regex_compile(new_word);
2023-08-29 13:09:55 -04:00
append_token(mt);
return mt;
}
int new_keyword_tokens(const char * const * words,
hl_group_t * const g) {
int i = 0;
while (*words) {
if(new_keyword_token(*words, g)) {
++i;
}
++words;
}
return i;
}
token_t * new_region_token(const char * start,
const char * end,
2023-09-18 17:28:24 -04:00
hl_group_t * g) {
2023-08-29 13:09:55 -04:00
char buffer[100];
buffer[0] = '\0';
strcat(buffer, start);
strcat(buffer, "[\\d\\D]*");
strcat(buffer, end);
token_t * mt = (token_t*)malloc(sizeof(token_t));
mt->hl = g;
mt->t = KEYSYMBOL;
mt->syntax = regex_compile(buffer);
append_token(mt);
return mt;
}
token_t * new_token(const char * const word,
const token_type_t t,
hl_group_t * const g) {
switch (t) {
case KEYSYMBOL: {
return new_symbol_token(word, g);
}
case KEYWORD: {
return new_keyword_token(word, g);
}
case MATCH: {
token_t * mt = (token_t*)malloc(sizeof(token_t));
mt->hl = g;
mt->t = MATCH;
mt->syntax = regex_compile(word);
append_token(mt);
} break;
case REGION: {
} break;
}
return NULL;
}
// --------------------
// ### Highlighting ###
// --------------------
void render_string(const char * const string,
2023-09-18 17:28:24 -04:00
const char * const mode) {
2023-09-20 16:43:47 -04:00
display_t * display;
HASH_FIND_STR(display_table,
mode,
display);
typedef struct {
const token_t * t;
const match_t * m;
int i;
} result_t;
result_t * const r = (result_t *)malloc(sizeof(result_t) * 1024); // XXX: dont
int rrs = 0;
2023-09-20 18:47:04 -04:00
for (size_t i = 0; i < token_table.element_count; i++) {
2023-09-20 16:43:47 -04:00
token_t * t = *(token_t**)vector_get(&token_table,
i);
match_t * match = regex_match(t->syntax, string, true);
2023-09-23 11:06:44 -04:00
if (is_sentinel(match)) {
2023-09-20 16:43:47 -04:00
free(match);
continue;
}
r[rrs++] = (result_t){
.t = t,
.m = match,
.i = 0,
};
}
2023-08-29 13:09:55 -04:00
for (const char * s = string; *s != '\00';) {
2023-09-20 16:43:47 -04:00
const result_t sentinel = (result_t){NULL, &(match_t){ -1, -1}, -1};
const result_t * max;
max = &sentinel;
for (int h = 0; h < rrs; h++) {
result_t * const current_result = r + h;
2023-09-23 11:06:44 -04:00
for (int j = 0; !is_sentinel(&(current_result->m[j])); j++) {
2023-09-20 16:43:47 -04:00
if (current_result->m[j].position == (s - string)) {
if (current_result->m[j].width > max->m->width) {
current_result->i = j;
max = current_result;
}
break;
}
2023-08-29 13:09:55 -04:00
}
}
2023-09-20 16:43:47 -04:00
if (max != &sentinel) {
2023-09-20 17:37:07 -04:00
const match_t * mymatch = &(max->m[max->i]);
const int padding = mymatch->position - (s - string);
2023-09-20 16:43:47 -04:00
if (padding) {
display->callback(s,
padding,
NULL);
2023-08-29 13:09:55 -04:00
}
2023-09-20 16:43:47 -04:00
display->callback(s + padding,
2023-09-20 17:37:07 -04:00
mymatch->width,
2023-09-20 16:43:47 -04:00
max->t->hl->attributes);
2023-09-20 17:37:07 -04:00
s += padding + mymatch->width;
2023-08-29 13:09:55 -04:00
} else {
2023-09-20 16:43:47 -04:00
display->callback(s, 1, NULL);
2023-08-29 13:09:55 -04:00
++s;
}
}
}
int hl_init(void) {
return 0;
}
int hl_deinit(void) {
2023-09-20 16:43:47 -04:00
//for (size_t i = 0; i < token_table.element_count; i++) {
// free_token(*(token_t**)vector_get(&token_table, i));
//}
2023-08-29 13:09:55 -04:00
return 0;
}