anon/jeger
1
0
Форкнуть 0
Код Задачи Запросы на слияние Релизы Вики Активность

.

Просмотр исходного кода
Этот коммит содержится в:
anon 2023-09-02 17:39:52 +02:00
родитель df5bbb2e7d
Коммит 7c88282816
13 изменённых файлов: 896 добавлений и 776 удалений
Показать статистику Скачать Patch файл Скачать Diff файл Показать все файлы Свернуть все файлы

2
.gdbinit Обычный файл
Просмотреть файл

@ -0,0 +1,2 @@
source debug/regex.pretty_print.py
source debug/stage1.gdb

3
.gitignore поставляемый Обычный файл
Просмотреть файл

@ -0,0 +1,3 @@
regtest
*.out
.gdb_history

10
Makefile Обычный файл
Просмотреть файл

@ -0,0 +1,10 @@
CXXFLAGS := -fuse-ld=mold -ggdb
OUT := regtest
main:
g++ ${CXXFLAGS} source/main.cpp source/vector.c source/regex.c -o ${OUT}
run:
${OUT}
test: run

1
debug/stage1.gdb Обычный файл
Просмотреть файл

@ -0,0 +1 @@
b regex_match

3
debug/stage2.gdb Обычный файл
Просмотреть файл

@ -0,0 +1,3 @@
break regex_assert
continue
print * regex

18
documentation/README.md Обычный файл
Просмотреть файл

@ -0,0 +1,18 @@
# Abstraction
+---------------------+
| |
| |
| State register |
| |
| |
+---------------------+
+---------------------------------+
| State transition table |
+---------------------------------+
+---------------------------------+
| Fallback transition table |
+---------------------------------+

136
source/main.cpp Обычный файл
Просмотреть файл

@ -0,0 +1,136 @@
#include <stdio.h>
#include "regex.h"
static int test_counter = 0;
static int passed_tests = 0;
static int positives = 0;
static int positive_successes = 0;
static int negatives = 0;
static int negative_successes = 0;
static void
TEST(const char * const what,
const char * const on,
const bool expect){
regex_t * r = regex_compile(what);
bool result = regex_search(r, on);
bool passed = (result == expect);
expect ? ++positives : ++negatives;
if (passed) {
printf("\033[32;1mSuccess\033[0;1m. - \033[0m");
expect ? ++positive_successes : ++negative_successes;
} else {
printf("\033[31;1mFailiour\033[0;1m. - \033[0m");
}
printf("'%12s'\033[1m vs \033[0m'%12s'\033[1m:\033[0m Result = %d, Expected = %d\n", what, on, result, expect);
if (passed) {
++passed_tests;
}
++test_counter;
}
signed main() {
TEST(R"del(abc)del","abc",true);
TEST(R"del(efg1)del","efg1",true);
TEST(R"del(nig)del","ger",false);
TEST(R"del(ss)del","sss",true);
TEST(R"del(sss)del","ss",false);
puts("");
TEST(R"del(ab+c)del","abc",true);
TEST(R"del(ef+g1)del","effffg1",true);
TEST(R"del(efg1+)del","efg",false);
TEST(R"del(efg1+)del","efg1",true);
TEST(R"del(efg1+)del","efg11",true);
puts("");
TEST(R"del(a+a)del","aaa",true);
TEST(R"del(a+a)del","aa",true);
TEST(R"del(a+a)del","a",false);
TEST(R"del(a+a)del","aaa",true);
TEST(R"del(a+\+)del","aaa",false);
puts("");
TEST(R"del(ab*c)del","abc",true);
TEST(R"del(ef*g1)del","effffg1",true);
TEST(R"del(efg1*)del","efg",true);
TEST(R"del(efg1*)del","efg1",true);
TEST(R"del(efg1*)del","efg11",true);
puts("");
TEST(R"del(ne.)del","net",true);
TEST(R"del(ne.)del","ne",false);
TEST(R"del(ne.+)del","neoo",true);
TEST(R"del(ne.*)del","neoo",true);
TEST(R"del(ne.*)del","ne",true);
puts("");
TEST(R"del(ne.)del","ne\t",true);
TEST(R"del(ne\t)del","ne",false);
TEST(R"del(ne\t)del","ne\t",true);
TEST(R"del(ne\t)del","net",false);
TEST(R"del(ne)del","ne\t",true);
puts("");
TEST(R"del(\sa)del"," a",true);
TEST(R"del(\sa)del"," a ",true);
TEST(R"del(\wi)del","hi",true);
TEST(R"del(\w+)del","asd",true);
TEST(R"del(\w*)del","",true);
puts("");
TEST(R"del([A-Za-z]+)del","HelloWorld",true);
TEST(R"del([A-Za-z]+g)del","HelloWorldg",true);
TEST(R"del([A-Za-z]+g)del","g",false);
TEST(R"del([A-Za-z]*g)del","g",true);
TEST(R"del([A-Za-z]+1)del","1",false);
puts("");
TEST(R"del(^\^)del","^^",true);
TEST(R"del(^\^)del"," ^",false);
TEST(R"del(^ \^)del"," ^",true);
TEST(R"del(^a*)del","asd",true);
TEST(R"del(^)del","",true);
puts("");
TEST(R"del(\<test)del","test",true);
TEST(R"del(test\>)del","test",true);
TEST(R"del(\<test)del","atest",false);
TEST(R"del(test\>)del","testa",false);
TEST(R"del(\<test\>)del","test",true);
//TEST(R"del(\d{3})del","123",true);
//TEST(R"del(^\w+@\w+\.\w+$)del","example@email.com",true);
//TEST(R"del(\b\w+\b)del","This is a test",true);
//TEST(R"del(^[A-Za-z]+\s\d+)del","OpenAI 123",true);
//TEST(R"del([0-9]{4}-[0-9]{2}-[0-9]{2})del","2023-08-22",true);
//TEST(R"del(^[^abc]+$)del","def123",true);
//TEST(R"del(\b\d{5}\b)del","12345 67890",true);
//TEST(R"del(^[A-Z][a-z]+$)del","OpenAI",true);
//TEST(R"del(\d{3}-\d{2}-\d{4})del","123-45-6789",true);
//TEST(R"del(^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})del","192.168.1.1",true);
//TEST(R"del(^\w{8,12})del","Password123", false);
if(test_counter == passed_tests) {
fputs("\033[32m", stdout);
}
printf("\nPassed %d out of %d tests.\033[0m\n", passed_tests, test_counter);
printf("\tPositives: %d/%d\n", positive_successes, positives);
printf("\tNegatives: %d/%d\n", negative_successes, negatives);
}

718
source/regex.c Обычный файл
Просмотреть файл

@ -0,0 +1,718 @@
/* XXX:
* as it turns out returning a range of match objects is a
* high profile performance issue regarding regex, especially when highlighting.
* now as it stands we search an array of tokens for every position on a string.
* which sounds ok, until one realizes that searching from any position revails a range,
* where (future) matches can or cannot be found. meaning we are computing the same thing
* repeatedly, practically resulting in a bruteforcing situation where instead of eliminating
* certain non-matches, we blindly hammer character by character.
*/
#include "regex.h"
#include <assert.h>
#include <string.h>
#include <limits.h>
#include <stdlib.h>
#define JEGER_INIT_STATE 2
// ------------------
// ### Char tests ###
// ------------------
static bool is_quantifier(const char c) {
for (const char * s = "+*?="; *s != '\00'; s++) {
if (*s == c) {
return true;
}
}
return false;
}
bool is_magic(const char c) {
if (is_quantifier(c)) {
return true;
}
for (const char * s = "\\[].^"; *s != '\00'; s++) {
if (*s == c) {
return true;
}
}
return false;
}
// -----------------
// ### Char sets ###
// -----------------
#define JEGER_CHAR_SET_at "@"
#define JEGER_CHAR_SET_underscore "_"
#define JEGER_CHAR_SET_lower "abcdefghijklmnopqrstuwxyz"
#define JEGER_CHAR_SET_upper "ABCDEFGHIJKLMNOPQRSTUWXYZ"
#define JEGER_CHAR_SET_digits "0123456789"
#define JEGER_CHAR_SET_octal_digits "01234567"
#define JEGER_CHAR_SET_lower_hex "abcdef"
#define JEGER_CHAR_SET_upper_hex "ABCDEF"
#define JEGER_CHAR_SET_oct_241_to_277 \
"\241\242\243\244\245" \
"\246\247\250\251\252" \
"\253\254\255\256\257" \
"\260\261\262\263\264" \
"\265\266\267\270\271" \
"\272\273\274\275\276" \
"\277"
#define JEGER_CHAR_SET_oct_300_to_337 \
"\300\301\302\303\304" \
"\305\306\307\310\311" \
"\312\313\314\315\316" \
"\317\320\321\322\323" \
"\324\325\326\327\330" \
"\331\332\333\334\335" \
"\336\337"
#define JEGER_CHAR_SET_file_extra "/.-_+,#$%~="
#define JEGER_CHAR_SET_whitespace " \t\v\n"
// ----------------------
// ### Internal Types ###
// ----------------------
typedef struct {
int in;
char input;
int to;
int width;
int match_width;
} delta_t;
typedef struct {
int in;
int to;
int width;
int match_width;
} offshoot_t;
enum {
DO_CATCH = 0x00000001 << 0,
IS_NEGATIVE = 0x00000001 << 1,
IS_AT_THE_BEGINNING = 0x00000001 << 2,
DO_SKIP = 0x00000001 << 3,
FORCE_START_OF_STRING = 0x00000001 << 4,
};
typedef struct {
// XXX:
int flags;
// these might be obsolite but im leaving them for now
bool do_loop_hook;
bool do_follow_hook;
bool do_loop_shoot;
bool do_follow_shoot;
// ---
int state;
int width;
char * whitelist;
char * blacklist;
regex_t * regex;
} compiler_state;
// ----------------------------------
// ### Regex creation/destruction ###
// ----------------------------------
static const int HALT_AND_CATCH_FIRE = INT_MIN;
#define ASSERT_HALT(a) ((a == HALT_AND_CATCH_FIRE) ? HALT_AND_CATCH_FIRE : (cs->state + a))
static
void HOOK_ALL( int from,
const char * const str,
int to,
compiler_state * cs) {
for (const char * s = str; *s != '\0'; s++) {
delta_t * delta = (delta_t *)malloc(sizeof(delta_t));
delta->in = cs->state + from;
delta->input = *s;
delta->to = ASSERT_HALT(to);
delta->width = cs->width;
vector_push(&cs->regex->delta_table,
&delta);
}
}
static
void ABSOLUTE_OFFSHOOT(int from,
int to,
int width,
int match_width,
compiler_state * cs) {
offshoot_t * offshoot = (offshoot_t *)malloc(sizeof(offshoot_t));
offshoot->in = from;
offshoot->to = to;
offshoot->width = width;
offshoot->match_width = match_width;
vector_push(&cs->regex->catch_table,
&offshoot);
}
static
void OFFSHOOT(int from,
int to,
int width,
int match_width,
compiler_state * cs) {
ABSOLUTE_OFFSHOOT(cs->state + from, ASSERT_HALT(to), width, match_width, cs);
}
static
int escape_1_to_1(const char c, compiler_state * cs) {
char * target_list = (cs->flags & IS_NEGATIVE) ? cs->blacklist : cs->whitelist;
switch (c) {
case 't': {
strcat(target_list, "\t");
} return 1;
case 'n': {
strcat(target_list, "\n");
} return 1;
case 'r': {
strcat(target_list, "\r");
} return 1;
case 'b': {
strcat(target_list, "\b");
} return 1;
case '[': {
strcat(target_list, "[");
} return 1;
case ']': {
strcat(target_list, "]");
} return 1;
case '.': {
strcat(target_list, ".");
} return 1;
case '^': {
strcat(target_list, "^");
} return 1;
case '=': {
strcat(target_list, "=");
} return 1;
case '?': {
strcat(target_list, "?");
} return 1;
case '+': {
strcat(target_list, "+");
} return 1;
case '*': {
strcat(target_list, "*");
} return 1;
case '\\': {
strcat(target_list, "\\");
} return 1;
}
return 0;
}
static
int escape_1_to_N(const char c, compiler_state * cs) {
char * target_list = (cs->flags & IS_NEGATIVE) ? cs->blacklist : cs->whitelist;
switch(c) {
case 'i': {
const char identifier_chars[] = JEGER_CHAR_SET_at
JEGER_CHAR_SET_underscore
JEGER_CHAR_SET_digits
JEGER_CHAR_SET_oct_300_to_337
;
strcpy(target_list, identifier_chars);
return sizeof(identifier_chars)-1;
};
case 'I': {
const char identifier_chars[] = JEGER_CHAR_SET_at
JEGER_CHAR_SET_underscore
JEGER_CHAR_SET_oct_300_to_337
;
strcpy(target_list, identifier_chars);
return sizeof(identifier_chars)-1;
};
case 'k': {
const char keyword_chars[] = JEGER_CHAR_SET_at
JEGER_CHAR_SET_underscore
JEGER_CHAR_SET_digits
JEGER_CHAR_SET_oct_300_to_337
;
strcpy(target_list, keyword_chars);
return sizeof(keyword_chars)-1;
};
case 'K': {
const char keyword_chars[] = JEGER_CHAR_SET_at
JEGER_CHAR_SET_underscore
JEGER_CHAR_SET_oct_300_to_337
;
strcpy(target_list, keyword_chars);
return sizeof(keyword_chars)-1;
};
case 'f': {
const char filename_chars[] = JEGER_CHAR_SET_at
JEGER_CHAR_SET_digits
JEGER_CHAR_SET_file_extra
;
strcpy(target_list, filename_chars);
return sizeof(filename_chars)-1;
};
case 'F': {
const char filename_chars[] = JEGER_CHAR_SET_at
JEGER_CHAR_SET_file_extra
;
strcpy(target_list, filename_chars);
return sizeof(filename_chars)-1;
};
case 'p': {
const char printable_chars[] = JEGER_CHAR_SET_at
JEGER_CHAR_SET_oct_241_to_277
JEGER_CHAR_SET_oct_300_to_337
;
strcpy(target_list, printable_chars);
return sizeof(printable_chars)-1;
};
case 'P': {
const char printable_chars[] = JEGER_CHAR_SET_at
JEGER_CHAR_SET_oct_241_to_277
JEGER_CHAR_SET_oct_300_to_337
;
strcpy(target_list, printable_chars);
return sizeof(printable_chars)-1;
};
case 's': {
const char whitespace_chars[] = JEGER_CHAR_SET_whitespace;
strcpy(target_list, whitespace_chars);
return sizeof(whitespace_chars)-1;
};
case 'd': {
const char digit_chars[] = JEGER_CHAR_SET_digits;
strcpy(target_list, digit_chars);
return sizeof(digit_chars)-1;
};
case 'x': {
const char hex_chars[] = JEGER_CHAR_SET_digits
JEGER_CHAR_SET_lower_hex
JEGER_CHAR_SET_upper_hex
;
strcpy(target_list, hex_chars);
return sizeof(hex_chars)-1;
};
case 'o': {
const char oct_chars[] = JEGER_CHAR_SET_octal_digits;
strcpy(target_list, oct_chars);
return sizeof(oct_chars)-1;
};
case 'w': {
const char word_chars[] = JEGER_CHAR_SET_underscore
JEGER_CHAR_SET_digits
JEGER_CHAR_SET_lower
JEGER_CHAR_SET_upper
;
strcpy(target_list, word_chars);
return sizeof(word_chars)-1;
};
case 'h': {
const char very_word_chars[] = JEGER_CHAR_SET_underscore
JEGER_CHAR_SET_lower
JEGER_CHAR_SET_upper
;
strcpy(target_list, very_word_chars);
return sizeof(very_word_chars)-1;
};
case 'a': {
const char alpha_chars[] = JEGER_CHAR_SET_lower
JEGER_CHAR_SET_upper
;
strcpy(target_list, alpha_chars);
return sizeof(alpha_chars)-1;
};
case 'l': {
const char lower_alpha_chars[] = JEGER_CHAR_SET_lower;
strcpy(target_list, lower_alpha_chars);
return sizeof(lower_alpha_chars)-1;
};
case 'u': {
const char upper_alpha_chars[] = JEGER_CHAR_SET_upper;
strcpy(target_list, upper_alpha_chars);
return sizeof(upper_alpha_chars)-1;
};
}
return 0;
}
static
int escape_to_negative(const char c,
compiler_state * const cs) {
switch (c) {
case 'D': {
const char digit_chars[] = JEGER_CHAR_SET_digits;
strcpy(cs->blacklist, digit_chars);
cs->flags |= IS_NEGATIVE;
return sizeof(digit_chars)-1;
};
}
return 0;
}
static
int compile_hologram(const char c,
compiler_state * const cs) {
static
const char very_word_chars[] = JEGER_CHAR_SET_underscore
JEGER_CHAR_SET_lower
JEGER_CHAR_SET_upper
;
switch (c) {
case '^': {
cs->whitelist[0] = '\n';
cs->whitelist[1] = '\0';
HOOK_ALL(0, cs->whitelist, 0, cs);
cs->flags |= DO_SKIP;
if (cs->flags & IS_AT_THE_BEGINNING) {
cs->flags |= FORCE_START_OF_STRING;
} else {
cs->state += 1;
}
return 1;
};
case '<': {
if (cs->flags & IS_AT_THE_BEGINNING) {
ABSOLUTE_OFFSHOOT(0, 3, 0, 0, cs); //XXX: figure out how to move this
}
cs->flags |= DO_SKIP;
cs->flags |= IS_NEGATIVE;
strcat(cs->blacklist, very_word_chars);
OFFSHOOT(0, 0, 1, 0, cs); //XXX: figure out how to move this
++cs->state;
return sizeof(very_word_chars)-1;
};
case '>': {
cs->flags |= DO_SKIP;
cs->flags |= IS_NEGATIVE;
strcat(cs->blacklist, very_word_chars);
OFFSHOOT(0, 1, 0, 0, cs); //XXX: figure out how to move this
++cs->state; // XXX: the current bug arises from the state being increased before the blacklist is hooked
return sizeof(very_word_chars)-1;
}
}
return 0;
}
static
int compile_dot(compiler_state * cs) {
cs->flags |= DO_CATCH;
return true;
}
static
int compile_escape(const char c,
compiler_state * const cs) {
return escape_1_to_1(c, cs)
|| escape_1_to_N(c, cs)
|| escape_to_negative(c, cs)
;
}
static
int compile_range(const char * const range,
compiler_state * const cs) {
assert((range[0] == '[') && "Not a range.");
const char * s;
if (range[1] == '^') {
cs->flags |= IS_NEGATIVE;
s = range + 2;
} else {
s = range + 1;
}
char * target_list = (cs->flags & IS_NEGATIVE) ? cs->blacklist : cs->whitelist;
for (; *s != ']'; s++) {
assert((*s != '\0') && "Unclosed range.");
char c = *s;
if (c == '\\') {
s += 1;
assert(compile_escape(*s, cs) && "Unknown escape.");
} else if (*(s+1) == '-') {
char end = *(s+2);
assert((c < end) && "Endless range.");
for (char cc = c; cc < end+1; cc++) {
strncat(target_list, &cc, 1);
strncat(target_list, "\0", 1);
}
s += 2;
} else {
strncat(target_list, &c, 1);
}
}
return ((s - range) + 1);
}
void filter_blacklist(const char * whitelist,
const char * blacklist,
char * filtered) {
for (; *blacklist != '\0'; blacklist++) {
for(; *whitelist != '\0'; whitelist++) {
if (*blacklist == *whitelist) {
goto long_continue;
}
}
strncat(filtered, blacklist, 1);
long_continue:
;
}
}
regex_t * regex_compile(const char * const pattern) {
regex_t * regex = (regex_t *)malloc(sizeof(regex_t));
regex->str = strdup(pattern);
vector_init(&regex->delta_table, sizeof(delta_t*), 0UL);
vector_init(&regex->catch_table, sizeof(offshoot_t*), 0UL);
// this is plain retarded
char whitelist[64];
char blacklist[64];
compiler_state cs = {
.flags = IS_AT_THE_BEGINNING,
.state = JEGER_INIT_STATE,
.width = 0,
.whitelist = whitelist,
.blacklist = blacklist,
.regex = regex,
};
for (const char * s = pattern; *s != '\00';) {
// Reset the compiler
assert(!is_quantifier(*s) && "Pattern starts with quantifier.");
whitelist[0] = '\0';
blacklist[0] = '\0';
cs.flags &= IS_AT_THE_BEGINNING;
/**/
cs.do_loop_hook = false;
cs.do_follow_hook = false;
cs.do_loop_shoot = false;
cs.do_follow_shoot = false;
/**/
cs.width = 1;
// Translate char
switch (*s) {
case '^': {
compile_hologram(*s, &cs);
} break;
case '.': {
compile_dot(&cs);
} break;
case '\\': {
s += 1;
assert((compile_escape(*s, &cs) || compile_hologram(*s, &cs)) && "Unknown escape.");
} break;
case '[': {
s += compile_range(s, &cs) - 1;
} break;
default: { // Literal
whitelist[0] = *s;
whitelist[1] = '\0';
} break;
}
s += 1;
// Compile blacklist
if (*blacklist) {
char filtered_blacklist[64];
filtered_blacklist[0] = '\0';
filter_blacklist(whitelist, blacklist, filtered_blacklist);
HOOK_ALL(0, filtered_blacklist, HALT_AND_CATCH_FIRE, &cs);
}
if (cs.flags & DO_SKIP) {
goto long_continue;
}
// Compile with quantifier
switch (*s) {
case '=':
case '?': {
HOOK_ALL(0, whitelist, +1, &cs);
if ((cs.flags & DO_CATCH) || (cs.flags & IS_NEGATIVE)) {
OFFSHOOT(0, +1, 1, 1, &cs);
}
s += 1;
} break;
case '*': {
HOOK_ALL(0, whitelist, 0, &cs);
if ((cs.flags & DO_CATCH)
|| (cs.flags & IS_NEGATIVE)) {
OFFSHOOT(0, 0, 1, 1, &cs);
}
s += 1;
} break;
case '+': {
HOOK_ALL(0, whitelist, +1, &cs);
if ((cs.flags & DO_CATCH)
|| (cs.flags & IS_NEGATIVE)) {
OFFSHOOT(0, +1, 1, 1, &cs);
}
++cs.state;
HOOK_ALL(0, whitelist, 0, &cs);
if ((cs.flags & DO_CATCH)
|| (cs.flags & IS_NEGATIVE)) {
OFFSHOOT(0, 0, 1, 1, &cs);
}
s += 1;
} break;
default: { // Literal
HOOK_ALL(0, whitelist, +1, &cs);
if ((cs.flags & DO_CATCH)
|| (cs.flags & IS_NEGATIVE)) {
OFFSHOOT(0, +1, 1, 1, &cs);
}
++cs.state;
} break;
}
long_continue:
cs.flags &= !IS_AT_THE_BEGINNING;
}
// Init state hookups
ABSOLUTE_OFFSHOOT(0, 2, 0, 0, &cs);
if (cs.flags & FORCE_START_OF_STRING) {
ABSOLUTE_OFFSHOOT(1, HALT_AND_CATCH_FIRE, 0, 0, &cs);
} else {
ABSOLUTE_OFFSHOOT(1, 2, 0, 0, &cs);
}
regex->accepting_state = cs.state;
return regex;
}
int regex_free(regex_t * const regex) {
free(regex->str);
vector_free(&regex->delta_table);
vector_free(&regex->catch_table);
free(regex);
return 0;
}
// -----------------
// ### Searching ###
// -----------------
static
const offshoot_t * catch_table_lookup(const regex_t * const regex,
const int * const state) {
for (size_t i = 0; i < regex->catch_table.element_count; i++){
const offshoot_t * const offshoot = *(offshoot_t**)vector_get(&regex->catch_table, i);
if (offshoot->in == *state) {
return offshoot;
}
}
return NULL;
}
static
bool regex_assert(const regex_t * const regex,
const char * const string,
int state,
match_t * const match) {
if (state == HALT_AND_CATCH_FIRE) {
return false;
}
bool last_stand = false;
const char * s = string;
LOOP: {
if (*s == '\0') {
last_stand = true;
goto PERFORM_CATCH_LOOKUP;
}
// Jump search for the correct state
const int jump = 10;
size_t i = jump;
while (i < regex->delta_table.element_count) {
const delta_t * const delta = *(delta_t**)vector_get(&regex->delta_table, i);
if (delta->in >= state) {
break;
}
i += jump;
}
i -= jump;
// Linear search finish up
for (; i < regex->delta_table.element_count; i++) {
const delta_t * const delta = *(delta_t**)vector_get(&regex->delta_table, i);
if (delta->in > state) {
break;
}
if ((delta->in == state)
&& (delta->input == *s)) {
const int r = regex_assert(regex, s + delta->width, delta->to, match);
if(r){
if ((match->position != -1)
&& (delta->match_width)) {
match->position = (s - string);
}
match->width += delta->match_width;
return r;
}
}
}
}
PERFORM_CATCH_LOOKUP: {
const offshoot_t * const my_catch = catch_table_lookup(regex, &state);
if (my_catch && (!my_catch->width || !last_stand)) {
state = my_catch->to;
s += my_catch->width;
match->width += my_catch->match_width;
goto LOOP;
}
}
return (state == regex->accepting_state);
}
match_t * regex_match(const regex_t * const regex,
const char * const string,
const bool is_start_of_string) {
if (regex == NULL) {
return NULL;
}
match_t * match = (match_t *)malloc(sizeof(match_t));
if (string == NULL) {
match->position = -1;
match->width = 0;
return match;
}
const int initial_state = (int)(!is_start_of_string);
// XXX: this should be called in a loop, always restarting from the last char of the last match
if (regex_assert(regex, string, initial_state, match)) {
return match;
} else {
return NULL;
}
}
bool regex_search(const regex_t * const regex,
const char * const string) {
return (bool)regex_match(regex, string, true);
}

4
src/regex.h → source/regex.h
Просмотреть файл

@ -21,8 +21,8 @@ typedef struct {
extern regex_t * regex_compile(const char * const pattern);
extern int regex_free(regex_t * const regex);
extern bool regex_search(regex_t * regex, const char * const string);
extern match_t * regex_match(regex_t * regex, const char * const string, const bool start_of_string, const int string_offset);
extern bool regex_search(const regex_t * const regex, const char * const string);
extern match_t * regex_match(const regex_t * const regex, const char * const string, const bool start_of_string);
extern bool is_magic(const char c);

10
src/vector.c → source/vector.c
Просмотреть файл

@ -1,7 +1,3 @@
/* vector.c
* Copyright 2023 Anon Anonson, Ognjen 'xolatile' Milan Robovic, Emil Williams
* SPDX Identifier: GPL-3.0-only / NO WARRANTY / NO GUARANTEE */
#include "vector.h"
#include <stdio.h>
@ -18,7 +14,7 @@ void vector_init(vector_t * vector,
vector->element_size = element_size;
vector->element_count = element_count;
vector->data = calloc(vector->element_count, vector->element_size);
vector->data = (char *)calloc(vector->element_count, vector->element_size);
assert(vector->data);
}
@ -29,8 +25,8 @@ void vector_push(vector_t * vector,
vector->element_count += 1;
vector->data = realloc(vector->data,
vector->element_size * vector->element_count);
vector->data = (char *)realloc(vector->data,
vector->element_size * vector->element_count);
assert(vector->data);

0
src/vector.h → source/vector.h
Просмотреть файл

76
src/main.c
Просмотреть файл

@ -1,76 +0,0 @@
// @COMPILECMD g++ $@ -o regtest -O0 -ggdb -pg -fno-inline
#include <stdio.h>
#include "regex.hpp"
#define TEST(a, b, expected) do { \
r = regex_compile(a); \
bool result = regex_search(r, b); \
bool passed = (result == expected); \
if (passed) { printf("Success. - "); } else { printf("Failiour. - "); } \
printf("%s vs %s: Result = %d, Expected = %d\n", #a, #b, result, expected); \
++num_tests; \
if (passed) { ++passed_tests; } \
} while(0)
signed main() {
int num_tests = 0;
int passed_tests = 0;
regex_t * r;
TEST(R"del(abc)del","abc",true);
TEST(R"del(efg1)del","efg1",true);
TEST(R"del(nig)del","ger",false);
puts("");
TEST(R"del(ab+c)del","abc",true);
TEST(R"del(ef+g1)del","effffg1",true);
TEST(R"del(ni*g?)del","ngg",false);
puts("");
TEST(R"del(ne.)del","net",true);
TEST(R"del(ne.)del","ne",false);
TEST(R"del(ne.+)del","neoo",true);
puts("");
TEST(R"del(ne.)del","ne\t",true);
TEST(R"del(ne\t)del","ne",false);
TEST(R"del(ne\t)del","ne\t",true);
puts("");
TEST(R"del(\sa)del"," a",true);
TEST(R"del(\wi)del","hi",true);
TEST(R"del(\w+)del","asd",true);
puts("");
TEST(R"del([A-Za-z]+)del","HelloWorld",true);
TEST(R"del([A-Za-z]+g)del","HelloWorldg",true);
TEST(R"del([A-Za-z]+g)del","g",false);
puts("");
TEST(R"del(a+a)del","aaa",true);
TEST(R"del(a+a)del","aa",true);
TEST(R"del(a+a)del","a",false);
//++num_tests; TEST(R"del(\d{3})del","123",true);
//++num_tests; TEST(R"del(^\w+@\w+\.\w+$)del","example@email.com",true);
//++num_tests; TEST(R"del(\b\w+\b)del","This is a test",true);
//++num_tests; TEST(R"del(^[A-Za-z]+\s\d+)del","OpenAI 123",true);
//++num_tests; TEST(R"del([0-9]{4}-[0-9]{2}-[0-9]{2})del","2023-08-22",true);
//++num_tests; TEST(R"del(^[^abc]+$)del","def123",true);
//++num_tests; TEST(R"del(\b\d{5}\b)del","12345 67890",true);
//++num_tests; TEST(R"del(^[A-Z][a-z]+$)del","OpenAI",true);
//++num_tests; TEST(R"del(\d{3}-\d{2}-\d{4})del","123-45-6789",true);
//++num_tests; TEST(R"del(^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})del","192.168.1.1",true);
//++num_tests; TEST(R"del(^\w{8,12})del","Password123", false);
printf("\nPassed %d out of %d tests.\n", passed_tests, num_tests);
}

691
src/regex.c
Просмотреть файл

@ -1,691 +0,0 @@
/* regex.c
* Copyright 2023 Anon Anonson, Ognjen 'xolatile' Milan Robovic, Emil Williams
* SPDX Identifier: GPL-3.0-only / NO WARRANTY / NO GUARANTEE */
/* XXX:
* as it turns out returning a range of match objects is a
* high profile performance issue regarding regex, especially when highlighting.
* now as it stands we search an array of tokens for every position on a string.
* which sounds ok, until one realizes that searching from any position revails a range,
* where (future) matches can or cannot be found. meaning we are computing the same thing
* repeatedly, practically resulting in a bruteforcing situation where instead of eliminating
* certain non-matches, we blindly hammer character by character.
*/
/* XXX:
* the bigass char sets should be global and broken into parts
*/
#include "regex.h"
#include <assert.h>
#include <string.h>
#include <limits.h>
#include <stdlib.h>
// ------------------
// ### Char tests ###
// ------------------
static bool is_quantifier(const char c) {
for (const char * s = "+*?="; *s != '\00'; s++) {
if (*s == c) {
return true;
}
}
return false;
}
bool is_magic(const char c) {
if (is_quantifier(c)) {
return true;
}
for (const char * s = "\\[].^"; *s != '\00'; s++) {
if (*s == c) {
return true;
}
}
return false;
}
// ----------------------
// ### Internal Types ###
// ----------------------
typedef struct {
int in;
char input;
int to;
int width;
int match_width;
} delta_t;
typedef struct {
int in;
int to;
int width;
int match_width;
} offshoot_t;
typedef struct {
// XXX:
// These should share a mask
// Not even sure why they are pointers to begin with
bool * do_catch;
bool * is_negative;
bool is_at_the_beginning;
bool do_skip;
// these might be obsolite but im leaving them for now
bool * do_loop_hook;
bool * do_follow_hook;
bool * do_loop_shoot;
bool * do_follow_shoot;
// ---
int * state;
int * width;
char * whitelist;
char * blacklist;
regex_t * regex;
} compiler_state;
// ----------------------------------
// ### Regex creation/destruction ###
// ----------------------------------
#define HALT_AND_CATCH_FIRE INT_MIN
static void HOOK_ALL( int from,
const char * const str,
int to,
compiler_state * cs) {
int hook_to = (to == HALT_AND_CATCH_FIRE) ? HALT_AND_CATCH_FIRE : ((*cs->state) + to);
for (const char * s = str; *s != '\0'; s++) {
delta_t * delta = malloc(sizeof(delta_t));
delta->in = *cs->state + from;
delta->input = *s;
delta->to = hook_to;
delta->width = *cs->width;
vector_push(&cs->regex->delta_table,
&delta);
}
}
// XXX: align
static void ABSOLUTE_OFFSHOOT(int from,
int to,
int width,
int match_width,
compiler_state * cs) {
offshoot_t * offshoot = malloc(sizeof(offshoot_t));
offshoot->in = from;
offshoot->to = to;
offshoot->width = width;
offshoot->match_width = match_width;
vector_push(&cs->regex->catch_table,
&offshoot);
}
// XXX: align
static void OFFSHOOT(int from,
int to,
int width,
int match_width,
compiler_state * cs) {
ABSOLUTE_OFFSHOOT(*cs->state + from, *cs->state + to, width, match_width, cs);
}
static int escape_1_to_1(const char c, compiler_state * cs) {
char * target_list = (*cs->is_negative) ? cs->blacklist : cs->whitelist;
switch (c) {
case 't': {
strcat(target_list, "\t");
} return 1;
case 'n': {
strcat(target_list, "\n");
} return 1;
case 'r': {
strcat(target_list, "\r");
} return 1;
case 'b': {
strcat(target_list, "\b");
} return 1;
case '[': {
strcat(target_list, "[");
} return 1;
case ']': {
strcat(target_list, "]");
} return 1;
case '.': {
strcat(target_list, ".");
} return 1;
case '^': {
strcat(target_list, "^");
} return 1;
case '=': {
strcat(target_list, "=");
} return 1;
case '?': {
strcat(target_list, "?");
} return 1;
case '+': {
strcat(target_list, "+");
} return 1;
case '*': {
strcat(target_list, "*");
} return 1;
case '\\': {
strcat(target_list, "\\");
} return 1;
}
return 0;
}
static int escape_1_to_N(const char c, compiler_state * cs) {
char * target_list = (*cs->is_negative) ? cs->blacklist : cs->whitelist;
switch(c) {
case 'i': {
const char identifier_chars[] = "@0123456789_"
"\300\301\302\303\304"
"\305\306\307\310\311"
"\312\313\314\315\316"
"\317\320\321\322\323"
"\324\325\326\327\330"
"\331\332\333\334\335"
"\336\337";
strcpy(target_list, identifier_chars);
return sizeof(identifier_chars)-1;
};
case 'I': {
const char identifier_chars[] = "@_"
"\300\301\302\303\304"
"\305\306\307\310\311"
"\312\313\314\315\316"
"\317\320\321\322\323"
"\324\325\326\327\330"
"\331\332\333\334\335"
"\336\337";
strcpy(target_list, identifier_chars);
return sizeof(identifier_chars)-1;
};
case 'k': {
const char keyword_chars[] = "@0123456789_"
"\300\301\302\303\304"
"\305\306\307\310\311"
"\312\313\314\315\316"
"\317\320\321\322\323"
"\324\325\326\327\330"
"\331\332\333\334\335"
"\336\337";
strcpy(target_list, keyword_chars);
return sizeof(keyword_chars)-1;
};
case 'K': {
const char keyword_chars[] = "@_"
"\300\301\302\303\304"
"\305\306\307\310\311"
"\312\313\314\315\316"
"\317\320\321\322\323"
"\324\325\326\327\330"
"\331\332\333\334\335"
"\336\337";
strcpy(target_list, keyword_chars);
return sizeof(keyword_chars)-1;
};
case 'f': {
const char filename_chars[] = "@0123456789/.-_+,#$%~=";
strcpy(target_list, filename_chars);
return sizeof(filename_chars)-1;
};
case 'F': {
const char filename_chars[] = "@/.-_+,#$%~=";
strcpy(target_list, filename_chars);
return sizeof(filename_chars)-1;
};
case 'p': {
const char printable_chars[] = "@"
"\241\242\243\244\245"
"\246\247\250\251\252"
"\253\254\255\256\257"
"\260\261\262\263\264"
"\265\266\267\270\271"
"\272\273\274\275\276"
"\277"
"\300\301\302\303\304"
"\305\306\307\310\311"
"\312\313\314\315\316"
"\317\320\321\322\323"
"\324\325\326\327\330"
"\331\332\333\334\335"
"\336\337";
strcpy(target_list, printable_chars);
return sizeof(printable_chars)-1;
};
case 'P': {
const char printable_chars[] = "@"
"\241\242\243\244\245"
"\246\247\250\251\252"
"\253\254\255\256\257"
"\260\261\262\263\264"
"\265\266\267\270\271"
"\272\273\274\275\276"
"\277"
"\300\301\302\303\304"
"\305\306\307\310\311"
"\312\313\314\315\316"
"\317\320\321\322\323"
"\324\325\326\327\330"
"\331\332\333\334\335"
"\336\337";
strcpy(target_list, printable_chars);
return sizeof(printable_chars)-1;
};
case 's': {
const char whitespace_chars[] = " \t\v\n";
strcpy(target_list, whitespace_chars);
return sizeof(whitespace_chars)-1;
};
case 'd': {
const char digit_chars[] = "0123456789";
strcpy(target_list, digit_chars);
return sizeof(digit_chars)-1;
};
case 'x': {
const char hex_chars[] = "0123456789"
"abcdef"
"ABCDEF";
strcpy(target_list, hex_chars);
return sizeof(hex_chars)-1;
};
case 'o': {
const char oct_chars[] = "01234567";
strcpy(target_list, oct_chars);
return sizeof(oct_chars)-1;
};
case 'w': {
const char word_chars[] = "0123456789"
"abcdefghijklmnopqrstuwxyz"
"ABCDEFGHIJKLMNOPQRSTUWXYZ"
"_";
strcpy(target_list, word_chars);
return sizeof(word_chars)-1;
};
case 'h': {
const char very_word_chars[] = "abcdefghijklmnopqrstuwxyz"
"ABCDEFGHIJKLMNOPQRSTUWXYZ"
"_";
strcpy(target_list, very_word_chars);
return sizeof(very_word_chars)-1;
};
case 'a': {
const char alpha_chars[] = "abcdefghijklmnopqrstuwxyz"
"ABCDEFGHIJKLMNOPQRSTUWXYZ";
strcpy(target_list, alpha_chars);
return sizeof(alpha_chars)-1;
};
case 'l': {
const char lower_alpha_chars[] = "abcdefghijklmnopqrstuwxyz";
strcpy(target_list, lower_alpha_chars);
return sizeof(lower_alpha_chars)-1;
};
case 'u': {
const char upper_alpha_chars[] = "ABCDEFGHIJKLMNOPQRSTUWXYZ";
strcpy(target_list, upper_alpha_chars);
return sizeof(upper_alpha_chars)-1;
};
}
return 0;
}
static int escape_to_negative(const char c,
compiler_state * cs) {
switch (c) {
case 'D': {
const char digit_chars[] = "0123456789";
strcpy(cs->blacklist, digit_chars);
*cs->is_negative = true;
return sizeof(digit_chars)-1;
};
}
return 0;
}
static int escape_hologram(const char c, compiler_state * cs) {
switch (c) {
case '<': {
if (cs->is_at_the_beginning) {
ABSOLUTE_OFFSHOOT(0, 2, 0, 0, cs);
cs->do_skip = true;
}
const char very_word_chars[] = "abcw";
//const char very_word_chars[] = "abcdefghijklmnopqrstuwxyz"
// "ABCDEFGHIJKLMNOPQRSTUWXYZ"
// "_";
*cs->is_negative = true; // effectless currently; should be used to trigger the following lines in the main compile loop
strcat(cs->blacklist, very_word_chars);
HOOK_ALL(0, cs->blacklist, HALT_AND_CATCH_FIRE, cs);
OFFSHOOT(0, 0, 1, 0, cs);
return sizeof(very_word_chars)-1;
};
case '>': {
const char very_word_chars[] = "abcw";
//const char very_word_chars[] = "abcdefghijklmnopqrstuwxyz"
// "ABCDEFGHIJKLMNOPQRSTUWXYZ"
// "_";
*cs->is_negative = true;
strcat(cs->blacklist, very_word_chars);
return sizeof(very_word_chars)-1;
}
}
return 0;
}
static int compile_dot(compiler_state * cs) {
*cs->do_catch = true;
return true;
}
static int compile_escape(const char c,
compiler_state * cs) {
return escape_1_to_1(c, cs)
|| escape_1_to_N(c, cs)
|| escape_to_negative(c, cs)
|| escape_hologram(c, cs)
;
}
static int compile_range(const char * const range,
compiler_state * cs) {
assert((range[0] == '[') && "Not a range.");
const char * s;
if (range[1] == '^') {
*cs->is_negative = true;
s = range + 2;
} else {
s = range + 1;
}
char * target_list = (*cs->is_negative) ? cs->blacklist : cs->whitelist;
for (; *s != ']'; s++) {
assert((*s != '\0') && "Unclosed range.");
char c = *s;
if (c == '\\') {
s += 1;
assert(compile_escape(*s, cs) && "Unknown escape.");
} else if (*(s+1) == '-') {
char end = *(s+2);
assert((c < end) && "Endless range.");
for (char cc = c; cc < end+1; cc++) {
strncat(target_list, &cc, 1);
strncat(target_list, "\0", 1);
}
s += 2;
} else {
strncat(target_list, &c, 1);
}
}
return ((s - range) + 1);
}
void filter_blacklist(const char * whitelist,
const char * blacklist,
char * filtered) {
for (; *blacklist != '\0'; blacklist++) {
for(; *whitelist != '\0'; whitelist++) {
if (*blacklist == *whitelist) {
goto long_continue;
}
}
strncat(filtered, blacklist, 1);
long_continue:
;
}
}
regex_t * regex_compile(const char * const pattern) {
regex_t * regex = (regex_t *)malloc(sizeof(regex_t));
regex->str = strdup(pattern);
vector_init(&regex->delta_table, sizeof(delta_t*), 0UL);
vector_init(&regex->catch_table, sizeof(offshoot_t*), 0UL);
int state = 2;
// this is plain retarded
bool do_catch;
bool is_negative;
bool do_loop_hook;
bool do_follow_hook;
bool do_loop_shoot;
bool do_follow_shoot;
int width;
char whitelist[64];
char blacklist[64];
compiler_state cs = {
.do_catch = &do_catch,
.is_negative = &is_negative,
.is_at_the_beginning = true,
.do_skip = false,
.state = &state,
.width = &width,
.whitelist = whitelist,
.blacklist = blacklist,
.regex = regex,
};
for (const char * s = pattern; *s != '\00';) {
// Reset the compiler
assert(!is_quantifier(*pattern) && "Pattern starts with quantifier.");
whitelist[0] = '\0';
blacklist[0] = '\0';
do_catch = false;
is_negative = false;
cs.do_skip = false;
/**/
do_loop_hook = false;
do_follow_hook = false;
do_loop_shoot = false;
do_follow_shoot = false;
/**/
width = 1;
// Translate char
switch (*s) {
case '^': {
if (cs.is_at_the_beginning) {
ABSOLUTE_OFFSHOOT(0, 2, 0, 0, &cs);
ABSOLUTE_OFFSHOOT(1, HALT_AND_CATCH_FIRE, 0, 0, &cs);
}
whitelist[0] = '\n';
whitelist[1] = '\0';
HOOK_ALL(0, whitelist, 0, &cs);
if (s != pattern) {
state += 1;
}
cs.do_skip = true;
} break;
case '.': {
compile_dot(&cs);
} break;
case '\\': {
s += 1;
assert(compile_escape(*s, &cs) && "Unknown escape.");
} break;
case '[': {
s += compile_range(s, &cs) - 1;
} break;
default: {
whitelist[0] = *s;
whitelist[1] = '\0';
} break;
}
s += 1;
if (cs.do_skip) {
goto long_continue;
}
// Compile blacklist
if (*blacklist) {
char filtered_blacklist[64];
filtered_blacklist[0] = '\0';
filter_blacklist(whitelist, blacklist, filtered_blacklist);
HOOK_ALL(0, filtered_blacklist, HALT_AND_CATCH_FIRE, &cs);
}
// Compile with quantifier
switch (*s) {
case '=':
case '?': {
do_loop_hook = true;
HOOK_ALL(0, whitelist, +1, &cs);
if (do_catch || is_negative) {
OFFSHOOT(0, +1, 1, 1, &cs);
}
s += 1;
} break;
case '*': {
HOOK_ALL(0, whitelist, 0, &cs);
if (do_catch) {
OFFSHOOT(0, +1, 1, 1, &cs);
} else if (is_negative) {
OFFSHOOT(0, 0, 1, 1, &cs);
}
s += 1;
} break;
case '+': {
HOOK_ALL(0, whitelist, +1, &cs);
if (do_catch || is_negative) {
OFFSHOOT(0, +1, 1, 1, &cs);
}
state += 1;
HOOK_ALL(0, whitelist, 0, &cs);
if (do_catch || is_negative) {
OFFSHOOT(0, 0, 1, 1, &cs);
}
s += 1;
} break;
default: { // Literal
HOOK_ALL(0, whitelist, +1, &cs);
if (do_catch || is_negative) {
OFFSHOOT(0, +1, 1, 1, &cs);
}
state += 1;
} break;
}
long_continue:
cs.is_at_the_beginning = false;
}
regex->accepting_state = state;
return regex;
}
int regex_free(regex_t * const regex) {
free(regex->str);
vector_free(&regex->delta_table);
vector_free(&regex->catch_table);
free(regex);
return 0;
}
// -----------------
// ### Searching ###
// -----------------
// XXX: rename; align
static offshoot_t * catch_(const regex_t * const regex,
int * const state) {
for (size_t i = 0; i < regex->catch_table.element_count; i++){
const offshoot_t * const offshoot = *(offshoot_t**)vector_get(&regex->catch_table, i);
if (offshoot->in == *state) {
*state = offshoot->to;
return offshoot;
}
}
return NULL;
}
static bool regex_assert(const regex_t * const regex,
const char * const string,
const int string_offset, // XXX: useless
int state,
match_t * const match) {
if (state == HALT_AND_CATCH_FIRE) { return false; }
for (const char * s = (string + string_offset); *s != '\00';) {
// XXX: this should be a jump search for the instate and then a linear
for (size_t i = 0; i < regex->delta_table.element_count; i++) {
const delta_t * const delta = *(delta_t**)vector_get(&regex->delta_table, i);
if ((delta->in == state)
&& (delta->input == *s)) {
const int r = regex_assert(regex, string, (s - string) + delta->width, delta->to, match);
if(r){
if ((match->position != -1)
&& (delta->match_width)) {
match->position = (s - string);
}
match->width += delta->match_width;
return r;
}
}
}
const offshoot_t * const catch = catch_(regex, &state);
if ((catch)
&& (state != HALT_AND_CATCH_FIRE)) {
s += catch->width;
match->width += catch->match_width;
continue;
} else {
break;
}
}
return (state == regex->accepting_state);
}
match_t * regex_match( regex_t * regex,
const char * const string,
const bool is_start_of_string,
const int string_offset) { // XXX: remove this useless piece of shit of a parameter nigger
if (regex == NULL) {
return NULL;
}
match_t * m = (match_t *)malloc(sizeof(match_t));
if (string == NULL) {
m->position = -1;
m->width = 0;
return m;
}
const int initial_state = (int)(!is_start_of_string);
// XXX: this should be called in a loop, always restarting from the last char of the last match
if(regex_assert(regex, string, string_offset, initial_state, m)) {
return m;
} else {
free(m);
return NULL;
}
}
bool regex_search( regex_t * regex,
const char * const string) {
return (bool)regex_match(regex, string, true, 0);
}