cell AMX_NATIVE_CALL pawn_regex_set(AMX* amx, cell* params)
{
const char *rexp = NULL;
amx_StrParam(amx, params[1], rexp);
if(rexp)
{
int aidi=(int)params[2];
//logprintf("aidi %d", aidi);
if(aidi>=0 && aidi<TotalExpressions)
{
onig_free(rexpression[aidi].RegExpr);
onig_region_free(rexpression[aidi].zreg, 1);
UChar* pattern = (UChar* )rexp;
OnigErrorInfo einfo;
rexpression[aidi].id = onig_new(&rexpression[aidi].RegExpr, pattern, pattern + strlen(rexp), ONIG_OPTION_NONE, ONIG_ENCODING_UTF8, ONIG_SYNTAX_DEFAULT, &einfo);
if(rexpression[aidi].id != ONIG_NORMAL)
{
UChar s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, rexpression[aidi].id, &einfo);
logprintf("[REGEX ERROR]: %s | at index %d. regex_set aborted.", s, aidi);
onig_free(rexpression[aidi].RegExpr);
return -1;
}
rexpression[aidi].zreg = onig_region_new();
//logprintf("aidi %d", aidi);
return 1;
}
logprintf("[REGEX ERROR]: Call regex_set with undefined parameter at index %d", aidi);
return -1;
}
return -1337;
}
// Called with the individual regexp lock held.
regex_t* Regexp::maybe_recompile(STATE, String* string) {
const UChar *pat;
const UChar *end;
OnigEncoding enc;
OnigErrorInfo err_info;
int err;
if(fixed_encoding_) return onig_source_data(state);
Encoding* string_enc = string->encoding(state);
regex_t* onig_encoded = onig_data_encoded(state, string_enc);
if(onig_encoded) return onig_encoded;
enc = string_enc->get_encoding();
Encoding* source_enc = source()->encoding(state);
String* converted = source()->convert_escaped(state, source_enc, fixed_encoding_);
pat = (UChar*)converted->byte_address();
end = pat + converted->byte_size();
int options = onig_source_data(state)->options;
OnigEncoding orig_enc = onig_source_data(state)->enc;
regex_t* reg;
err = onig_new(®, pat, end, options,
enc, ONIG_SYNTAX_RUBY, &err_info);
// If it doesn't work out, then abort and reset the encoding back
// and say that it's forced.
if(err != ONIG_NORMAL) {
err = onig_new(®, pat, end, options,
orig_enc, ONIG_SYNTAX_RUBY, &err_info);
// Ok, wtf. Well, no way to proceed now.
if(err != ONIG_NORMAL) {
UChar onig_err_buf[ONIG_MAX_ERROR_MESSAGE_LEN];
char err_buf[REGEXP_ONIG_ERROR_MESSAGE_LEN];
onig_error_code_to_str(onig_err_buf, err, &err_info);
snprintf(err_buf, REGEXP_ONIG_ERROR_MESSAGE_LEN, "%s: %s", onig_err_buf, pat);
Exception::regexp_error(state, err_buf);
return NULL;
}
string_enc = source_enc;
fixed_encoding_ = true;
}
return make_managed(state, string_enc, reg);
}
cell AMX_NATIVE_CALL pawn_regex_search(AMX* amx, cell* params)
{
regex_t* RegExpr;
const char *rexp = NULL, *string = NULL;
cell* addr[2] = {NULL, NULL};
amx_GetAddr(amx, params[3], &addr[0]);
amx_GetAddr(amx, params[4], &addr[1]);
amx_StrParam(amx, params[1], string);
amx_StrParam(amx, params[2], rexp);
if(string && rexp)
{
int r=NULL;
UChar* pattern = (UChar* )rexp;
OnigErrorInfo einfo;
r = onig_new(&RegExpr, pattern, pattern + strlen((char* )pattern), ONIG_OPTION_DEFAULT, ONIG_ENCODING_ASCII, ONIG_SYNTAX_PERL, &einfo);
//logprintf("[REGEX DEBUG]: rexp %s",pattern);
if(r != ONIG_NORMAL)
{
UChar s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r, &einfo);
logprintf("[REGEX ERROR]: %s", s);
onig_free(RegExpr);
return -1;
}
UChar* str = (UChar* )string;
OnigRegion *region;
region = onig_region_new();
r = onig_search(RegExpr, str, str+strlen((char*) str), str, str+strlen((char*) str), region, ONIG_OPTION_NONE);
if(r>=0)
{
*addr[0]=r;
*addr[1]=region->end[region->num_regs-1]-1;
}
else if(r==ONIG_MISMATCH)
{
*addr[0]=-1;
*addr[1]=-1;
}
else
{
UChar s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r);
logprintf("[REGEX ERROR]: %s\n", s);
onig_region_free(region, 1);
onig_free(RegExpr);
return -1;
}
//logprintf("[REGEX DEBUG]: string %s",str);
onig_region_free(region, 1);
onig_free(RegExpr);
//logprintf("[REGEX DEBUG]: return %d",r);
return 1;
}
return -1337;
}
static int
regexp_main(char *pat0, char *str0)
{
int r;
unsigned char *start, *range, *end;
regex_t* reg;
OnigErrorInfo einfo;
OnigRegion *region;
static UChar* pattern;
static UChar* str;
int i;
pattern = (UChar* )pat0;
str = (UChar* )str0;
strcpy(data_str, str);
for (i = 0; i NAMES; i ++) {
data_rslt[i].n[0] = 0;
}
r = onig_new(®, pattern, pattern + strlen((char* )pattern),
ONIG_OPTION_DEFAULT, ONIG_ENCODING_ASCII, ONIG_SYNTAX_DEFAULT, &einfo);
if (r != ONIG_NORMAL) {
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r, &einfo);
return -1;
}
region = onig_region_new();
end = str + strlen((char* )str);
start = str;
range = end;
r = onig_search(reg, str, end, start, range, region, ONIG_OPTION_NONE);
if (r = 0) {
r = onig_foreach_name(reg, name_callback, (void* )region);
}
else if (r == ONIG_MISMATCH) {
return 1;
}
else { /* error */
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r);
return -1;
}
onig_region_free(region, 1);
/* 1:free self, 0:free contents only */
onig_free(reg);
onig_end();
return 0;
}
extern int ex(unsigned char* str, unsigned char* pattern,
OnigSyntaxType* syntax)
{
int r;
unsigned char *start, *range, *end;
regex_t* reg;
OnigErrorInfo einfo;
OnigRegion *region;
r = onig_new(®, pattern, pattern + strlen((char* )pattern),
ONIG_OPTION_DEFAULT, ONIG_ENCODING_ASCII, syntax, &einfo);
if (r != ONIG_NORMAL) {
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r, &einfo);
fprintf(stderr, "ERROR: %s\n", s);
return -1;
}
fprintf(stderr, "number of captures: %d\n", onig_number_of_captures(reg));
fprintf(stderr, "number of capture histories: %d\n",
onig_number_of_capture_histories(reg));
region = onig_region_new();
end = str + strlen((char* )str);
start = str;
range = end;
r = onig_search(reg, str, end, start, range, region, ONIG_OPTION_NONE);
if (r >= 0) {
int i;
fprintf(stderr, "match at %d\n", r);
for (i = 0; i < region->num_regs; i++) {
fprintf(stderr, "%d: (%ld-%ld)\n", i, region->beg[i], region->end[i]);
}
fprintf(stderr, "\n");
r = onig_capture_tree_traverse(region, ONIG_TRAVERSE_CALLBACK_AT_FIRST,
node_callback, (void* )0);
}
else if (r == ONIG_MISMATCH) {
fprintf(stderr, "search fail\n");
}
else { /* error */
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r);
return -1;
}
onig_region_free(region, 1 /* 1:free self, 0:free contents only */);
onig_free(reg);
return 0;
}
static int
x0(int no, char* pattern_arg, char* str_arg,
int start_offset, int expected_from, int expected_to, int backward)
{
int r;
unsigned char *start, *range, *end;
regex_t* reg;
OnigErrorInfo einfo;
OnigRegion *region;
UChar *pattern, *str;
pattern = (UChar* )pattern_arg;
str = (UChar* )str_arg;
r = onig_new(®, pattern, pattern + strlen((char* )pattern),
ONIG_OPTION_NEWLINE_CRLF, ONIG_ENCODING_UTF8, ONIG_SYNTAX_DEFAULT, &einfo);
if (r != ONIG_NORMAL) {
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r, &einfo);
fprintf(stderr, "ERROR: %s\n", s);
return -1;
}
region = onig_region_new();
end = str + strlen((char* )str);
if (backward) {
start = end + start_offset;
range = str;
}
else {
start = str + start_offset;
range = end;
}
r = onig_search(reg, str, end, start, range, region, ONIG_OPTION_NONE);
if (r >= 0) {
result(no, region->beg[0], region->end[0], expected_from, expected_to);
}
else if (r == ONIG_MISMATCH) {
result(no, r, -1, expected_from, expected_to);
}
else { /* error */
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r);
fprintf(stderr, "ERROR: %s\n", s);
return -1;
}
onig_region_free(region, 1 /* 1:free self, 0:free contents only */);
onig_free(reg);
return 0;
}
extern int main(int argc, char* argv[])
{
int r;
unsigned char *start, *range, *end;
regex_t* reg;
OnigErrorInfo einfo;
OnigRegion *region;
static UChar* pattern = (UChar* )"a(.*)b|[e-f]+";
static UChar* str = (UChar* )"zzzzaffffffffb";
r = onig_new(®, pattern, pattern + strlen((char* )pattern),
ONIG_OPTION_DEFAULT, ONIG_ENCODING_ASCII, ONIG_SYNTAX_DEFAULT, &einfo);
if (r != ONIG_NORMAL) {
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r, &einfo);
fprintf(stderr, "ERROR: %s\n", s);
return -1;
}
region = onig_region_new();
end = str + strlen((char* )str);
start = str;
range = end;
r = onig_search(reg, str, end, start, range, region, ONIG_OPTION_NONE);
if (r >= 0) {
int i;
fprintf(stderr, "match at %d\n", r);
for (i = 0; i < region->num_regs; i++) {
fprintf(stderr, "%d: (%ld-%ld)\n", i, region->beg[i], region->end[i]);
}
r = 0;
}
else if (r == ONIG_MISMATCH) {
fprintf(stderr, "search fail\n");
r = -1;
}
else { /* error */
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r);
fprintf(stderr, "ERROR: %s\n", s);
return -1;
}
onig_region_free(region, 1 /* 1:free self, 0:free contents only */);
onig_free(reg);
onig_end();
return r;
}
extern int main(int argc, char* argv[])
{
int r;
unsigned char *start, *range, *end;
regex_t* reg;
OnigErrorInfo einfo;
OnigRegion *region;
static UChar* pattern = (UChar* )"(?<foo>a*)(?<bar>b*)(?<foo>c*)";
static UChar* str = (UChar* )"aaabbbbcc";
OnigEncoding use_encs[] = { ONIG_ENCODING_ASCII };
onig_initialize(use_encs, sizeof(use_encs)/sizeof(use_encs[0]));
r = onig_new(®, pattern, pattern + strlen((char* )pattern),
ONIG_OPTION_DEFAULT, ONIG_ENCODING_ASCII, ONIG_SYNTAX_DEFAULT, &einfo);
if (r != ONIG_NORMAL) {
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r, &einfo);
fprintf(stderr, "ERROR: %s\n", s);
return -1;
}
fprintf(stderr, "number of names: %d\n", onig_number_of_names(reg));
region = onig_region_new();
end = str + strlen((char* )str);
start = str;
range = end;
r = onig_search(reg, str, end, start, range, region, ONIG_OPTION_NONE);
if (r >= 0) {
fprintf(stderr, "match at %d\n\n", r);
r = onig_foreach_name(reg, name_callback, (void* )region);
}
else if (r == ONIG_MISMATCH) {
fprintf(stderr, "search fail\n");
}
else { /* error */
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r);
return -1;
}
onig_region_free(region, 1 /* 1:free self, 0:free contents only */);
onig_free(reg);
onig_end();
return 0;
}
int NewOnigRegex(char *pattern, int pattern_length, int option,
OnigRegex *regex, OnigRegion **region, OnigEncoding *encoding, OnigErrorInfo **error_info, char **error_buffer) {
int ret = ONIG_NORMAL;
int error_msg_len = 0;
OnigUChar *pattern_start = (OnigUChar *) pattern;
OnigUChar *pattern_end = (OnigUChar *) (pattern + pattern_length);
*error_info = (OnigErrorInfo *) malloc(sizeof(OnigErrorInfo));
memset(*error_info, 0, sizeof(OnigErrorInfo));
*encoding = (void*)ONIG_ENCODING_UTF8;
*error_buffer = (char*) malloc(ONIG_MAX_ERROR_MESSAGE_LEN * sizeof(char));
memset(*error_buffer, 0, ONIG_MAX_ERROR_MESSAGE_LEN * sizeof(char));
*region = onig_region_new();
ret = onig_new(regex, pattern_start, pattern_end, (OnigOptionType)(option), *encoding, OnigDefaultSyntax, *error_info);
if (ret != ONIG_NORMAL) {
error_msg_len = onig_error_code_to_str((unsigned char*)(*error_buffer), ret, *error_info);
if (error_msg_len >= ONIG_MAX_ERROR_MESSAGE_LEN) {
error_msg_len = ONIG_MAX_ERROR_MESSAGE_LEN - 1;
}
(*error_buffer)[error_msg_len] = '\0';
}
return ret;
}
/* {{{ php_mbregex_compile_pattern */
static php_mb_regex_t *php_mbregex_compile_pattern(const char *pattern, size_t patlen, OnigOptionType options, OnigEncoding enc, OnigSyntaxType *syntax)
{
int err_code = 0;
php_mb_regex_t *retval = NULL, *rc = NULL;
OnigErrorInfo err_info;
OnigUChar err_str[ONIG_MAX_ERROR_MESSAGE_LEN];
rc = zend_hash_str_find_ptr(&MBREX(ht_rc), (char *)pattern, patlen);
if (!rc || onig_get_options(rc) != options || onig_get_encoding(rc) != enc || onig_get_syntax(rc) != syntax) {
if ((err_code = onig_new(&retval, (OnigUChar *)pattern, (OnigUChar *)(pattern + patlen), options, enc, syntax, &err_info)) != ONIG_NORMAL) {
onig_error_code_to_str(err_str, err_code, &err_info);
php_error_docref(NULL, E_WARNING, "mbregex compile err: %s", err_str);
retval = NULL;
goto out;
}
if (rc == MBREX(search_re)) {
/* reuse the new rc? see bug #72399 */
MBREX(search_re) = NULL;
}
zend_hash_str_update_ptr(&MBREX(ht_rc), (char *)pattern, patlen, retval);
} else {
retval = rc;
}
out:
return retval;
}
static void throw_onig_error (int err_code, OnigErrorInfo *einfo)
{
UChar err_buf[ONIG_MAX_ERROR_MESSAGE_LEN];
(void) onig_error_code_to_str (err_buf, err_code, einfo);
SLang_verror (slOnig_Error, "%s", err_buf);
}
cell AMX_NATIVE_CALL pawn_regex_match(AMX* amx, cell* params)
{
regex_t* RegExpr;
const char *rexp = NULL, *string = NULL;
amx_StrParam(amx, params[1], string);
amx_StrParam(amx, params[2], rexp);
if(string && rexp)
{
int r=NULL;
UChar* pattern = (UChar* )rexp;
OnigErrorInfo einfo;
r = onig_new(&RegExpr, pattern, pattern + strlen((char* )pattern), ONIG_OPTION_DEFAULT, ONIG_ENCODING_ASCII, ONIG_SYNTAX_PERL, &einfo);
//logprintf("[REGEX DEBUG]: rexp %s",pattern);
if(r != ONIG_NORMAL)
{
UChar s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r, &einfo);
logprintf("[REGEX ERROR]: %s", s);
onig_free(RegExpr);
return -1;
}
UChar* str = (UChar* )string;
OnigRegion *region;
region = onig_region_new();
r = onig_match(RegExpr, str, str+strlen((char*) str), str, region, ONIG_OPTION_NONE);
//logprintf("[REGEX DEBUG]: string %s",str);
onig_region_free(region, 1);
onig_free(RegExpr);
//logprintf("[REGEX DEBUG]: return %d",r);
return r;
}
return -1337;
}
int onig_err_to_bregexp_msg(OnigPosition err_code, OnigErrorInfo* err_info, TCHAR *msg)
{
char err_str[ONIG_MAX_ERROR_MESSAGE_LEN];
int ret = onig_error_code_to_str((UChar*) err_str, err_code, err_info);
err_str[BREGEXP_MAX_ERROR_MESSAGE_LEN-1] = '\0';
asc2tcs(msg, err_str, BREGEXP_MAX_ERROR_MESSAGE_LEN);
return ret;
}
/*
* This is a primitive so #initialize_copy can work.
*/
Regexp* Regexp::initialize(STATE, String* pattern, Fixnum* options) {
const UChar *pat;
const UChar *end;
OnigErrorInfo err_info;
OnigOptionType opts;
OnigEncoding enc;
int err, num_names, kcode;
pat = (UChar*)pattern->byte_address();
end = pat + pattern->size();
opts = options->to_native();
kcode = opts & KCODE_MASK;
opts &= OPTION_MASK;
if(kcode == 0) {
enc = current_encoding(state);
} else {
// Don't attempt to fix the encoding later, it's been specified by the
// user.
enc = get_enc_from_kcode(kcode);
forced_encoding_ = true;
}
thread::Mutex::LockGuard lg(state->shared().onig_lock());
err = onig_new(&this->onig_data, pat, end, opts, enc, ONIG_SYNTAX_RUBY, &err_info);
if(err != ONIG_NORMAL) {
UChar onig_err_buf[ONIG_MAX_ERROR_MESSAGE_LEN];
char err_buf[1024];
onig_error_code_to_str(onig_err_buf, err, &err_info);
snprintf(err_buf, 1024, "%s: %s", onig_err_buf, pat);
Exception::regexp_error(state, err_buf);
return 0;
}
this->source(state, pattern);
num_names = onig_number_of_names(this->onig_data);
if(num_names == 0) {
this->names(state, nil<LookupTable>());
} else {
struct _gather_data gd;
gd.state = state;
LookupTable* tbl = LookupTable::create(state);
gd.tbl = tbl;
onig_foreach_name(this->onig_data, (int (*)(const OnigUChar*, const OnigUChar*,int,int*,OnigRegex,void*))_gather_names, (void*)&gd);
this->names(state, tbl);
}
make_managed(state);
return this;
}
static OnigRegex
regexp_compile(grn_ctx *ctx,
const char *pattern,
unsigned int pattern_len,
OnigSyntaxType *syntax)
{
OnigRegex regex;
OnigEncoding onig_encoding;
int onig_result;
OnigErrorInfo onig_error_info;
if (ctx->encoding == GRN_ENC_NONE) {
return NULL;
}
switch (ctx->encoding) {
case GRN_ENC_EUC_JP :
onig_encoding = ONIG_ENCODING_EUC_JP;
break;
case GRN_ENC_UTF8 :
onig_encoding = ONIG_ENCODING_UTF8;
break;
case GRN_ENC_SJIS :
onig_encoding = ONIG_ENCODING_CP932;
break;
case GRN_ENC_LATIN1 :
onig_encoding = ONIG_ENCODING_ISO_8859_1;
break;
case GRN_ENC_KOI8R :
onig_encoding = ONIG_ENCODING_KOI8_R;
break;
default :
return NULL;
}
onig_result = onig_new(®ex,
pattern,
pattern + pattern_len,
ONIG_OPTION_ASCII_RANGE |
ONIG_OPTION_MULTILINE,
onig_encoding,
syntax,
&onig_error_info);
if (onig_result != ONIG_NORMAL) {
char message[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(message, onig_result, onig_error_info);
ERR(GRN_INVALID_ARGUMENT,
"[operator][regexp] "
"failed to create regular expression object: <%.*s>: %s",
pattern_len, pattern,
message);
return NULL;
}
return regex;
}
static void get_thread(un2ch_t *get, unarray_t *tl, int board_no, MYSQL *mysql)
{
unstr_t *data = 0;
unstr_t *moto = 0;
nich_t *nich = 0;
size_t i = 0;
regex_t *reg;
OnigErrorInfo einfo;
int r;
if(get == NULL || tl == NULL){
return;
}
pthread_mutex_lock(&g_onig_mutex);
r = onig_new(®, (UChar *)g_pattern, (UChar *)(g_pattern + strlen(g_pattern)),
ONIG_OPTION_NONE, ONIG_ENCODING_SJIS, ONIG_SYNTAX_PERL, &einfo);
pthread_mutex_unlock(&g_onig_mutex);
if(r != ONIG_NORMAL){
UChar s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r, &einfo);
printf("ERROR: %s\n", s);
return;
}
for(i = 0; i < tl->length; i++){
nich = unarray_at(tl, i);
un2ch_set_info(get, nich->server, nich->board, nich->thread);
moto = unstr_file_get_contents(get->logfile);
data = un2ch_get_data(get);
if(unstr_empty(data)){
printf("error %ld %s/%s/%s\n", get->code, nich->server->data,
nich->board->data, nich->thread->data);
} else {
printf("code:%ld OK %s/%s/%s\n", get->code, nich->server->data,
nich->board->data, nich->thread->data);
if((board_no >= 0) && (mysql != NULL)){
nich->reg = reg;
nich->board_no = board_no;
set_mysql_res(data, moto, nich, mysql);
}
}
nich = NULL;
unstr_delete(2, moto, data);
/* 4秒止める */
sleep(4);
}
pthread_mutex_lock(&g_onig_mutex);
onig_free(reg);
pthread_mutex_unlock(&g_onig_mutex);
}
extern int
re_compile_pattern(const char* pattern, int size, regex_t* reg, char* ebuf)
{
int r;
OnigErrorInfo einfo;
r = onig_compile(reg, (UChar* )pattern, (UChar* )(pattern + size), &einfo);
if (r != ONIG_NORMAL) {
if (IS_NOT_NULL(ebuf))
(void )onig_error_code_to_str((UChar* )ebuf, r, &einfo);
}
return r;
}
pone_val* pone_regex_new(pone_world* world, const char* str, size_t len) {
pone_universe* universe = world->universe;
pone_val* obj = pone_obj_new(world, universe->class_regex);
OnigErrorInfo errinfo;
regex_t* re = pone_malloc(world, sizeof(regex_t));
int err_code;
if ((err_code = onig_new_without_alloc(re, (const OnigUChar*)str, (const OnigUChar*)str + len, ONIG_OPTION_NONE, ONIG_ENCODING_UTF8, ONIG_SYNTAX_PERL, &errinfo)) != ONIG_NORMAL) {
char buf[ONIG_MAX_ERROR_MESSAGE_LEN];
int errsize = onig_error_code_to_str((OnigUChar*)buf, err_code, errinfo);
pone_throw(world, pone_str_new_strdup(world, buf, errsize));
// TODO create X::Regexp
}
pone_obj_set_ivar(world, obj, "$!re", pone_opaque_new(world, universe->class_onig, re, re_finalizer));
pone_obj_set_ivar(world, obj, "$!str", pone_str_new_strdup(world, str, len));
return obj;
}
int SearchOnigRegex(void *str, int str_length, int offset, int option,
OnigRegex regex, OnigRegion *region, OnigErrorInfo *error_info, char *error_buffer, int *captures, int *numCaptures) {
int ret = ONIG_MISMATCH;
int error_msg_len = 0;
#ifdef BENCHMARK_CHELP
struct timeval tim1, tim2;
long t;
#endif
OnigUChar *str_start = (OnigUChar *) str;
OnigUChar *str_end = (OnigUChar *) (str_start + str_length);
OnigUChar *search_start = (OnigUChar *)(str_start + offset);
OnigUChar *search_end = str_end;
#ifdef BENCHMARK_CHELP
gettimeofday(&tim1, NULL);
#endif
ret = onig_search(regex, str_start, str_end, search_start, search_end, region, option);
if (ret < 0 && error_buffer != NULL) {
error_msg_len = onig_error_code_to_str((unsigned char*)(error_buffer), ret, error_info);
if (error_msg_len >= ONIG_MAX_ERROR_MESSAGE_LEN) {
error_msg_len = ONIG_MAX_ERROR_MESSAGE_LEN - 1;
}
error_buffer[error_msg_len] = '\0';
}
else if (captures != NULL) {
int i;
int count = 0;
for (i = 0; i < region->num_regs; i++) {
captures[2*count] = region->beg[i];
captures[2*count+1] = region->end[i];
count ++;
}
*numCaptures = count;
}
#ifdef BENCHMARK_CHELP
gettimeofday(&tim2, NULL);
t = (tim2.tv_sec - tim1.tv_sec) * 1000000 + tim2.tv_usec - tim1.tv_usec;
printf("%ld microseconds elapsed\n", t);
#endif
return ret;
}
// Called with the onig_lock held.
void Regexp::maybe_recompile(STATE, String* string) {
const UChar *pat;
const UChar *end;
OnigEncoding enc;
OnigErrorInfo err_info;
int err;
if(fixed_encoding_) return;
enc = string->get_encoding_kcode_fallback(state);
if(enc == onig_data->enc) return;
pat = (UChar*)source()->byte_address();
end = pat + source()->byte_size();
int options = onig_data->options;
OnigEncoding orig_enc = onig_data->enc;
err = onig_new(&this->onig_data, pat, end, options,
enc, ONIG_SYNTAX_RUBY, &err_info);
// If it doesn't work out, then abort and reset the encoding back
// and say that it's forced.
if(err != ONIG_NORMAL) {
err = onig_new(&this->onig_data, pat, end, options,
orig_enc, ONIG_SYNTAX_RUBY, &err_info);
// Ok, wtf. Well, no way to proceed now.
if(err != ONIG_NORMAL) {
OnigUChar buf[1024];
onig_error_code_to_str((UChar*)buf, err, &err_info);
std::cout << "Fatal ONIG error: " << buf << "\n";
assert(err == ONIG_NORMAL);
}
fixed_encoding_ = true;
}
make_managed(state);
}
extern int
re_recompile_pattern(const char* pattern, int size, regex_t* reg, char* ebuf)
{
int r;
OnigErrorInfo einfo;
OnigEncoding enc;
/* I think encoding and options should be arguments of this function.
But this is adapted to present re.c. (2002/11/29)
*/
enc = OnigEncDefaultCharEncoding;
r = onig_recompile(reg, (UChar* )pattern, (UChar* )(pattern + size),
reg->options, enc, OnigDefaultSyntax, &einfo);
if (r != ONIG_NORMAL) {
if (IS_NOT_NULL(ebuf))
(void )onig_error_code_to_str((UChar* )ebuf, r, &einfo);
}
return r;
}
cell AMX_NATIVE_CALL pawn_regex_exsearch(AMX* amx, cell* params)
{
const char *string = NULL;
cell* addr[2] = {NULL, NULL};
amx_GetAddr(amx, params[3], &addr[0]);
amx_GetAddr(amx, params[4], &addr[1]);
amx_StrParam(amx, params[1], string);
if(string)
{
int id=(int)params[2];
if(id>=0 && id<TotalExpressions)
{
int r=NULL;
UChar* str = (UChar* )string;
onig_region_clear(rexpression[id].zreg);
r = onig_search(rexpression[id].RegExpr, str, str+strlen(string), str, str+strlen(string), rexpression[id].zreg, ONIG_OPTION_NONE);
if(r>=0)
{
*addr[0]=r;
*addr[1]=rexpression[id].zreg->end[rexpression[id].zreg->num_regs-1]-1;
}
else if(r==ONIG_MISMATCH)
{
*addr[0]=-1;
*addr[1]=-1;
}
else
{
UChar s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r);
logprintf("[REGEX ERROR]: %s\n", s);
return -1;
}
return 1;
}
logprintf("[REGEX ERROR]: Call regex_exsearch with undefined parameter at index %d", id);
return -1;
}
return -1337;
}
/* {{{ php_mbregex_compile_pattern */
static php_mb_regex_t *php_mbregex_compile_pattern(const char *pattern, int patlen, OnigOptionType options, OnigEncoding enc, OnigSyntaxType *syntax)
{
int err_code = 0;
php_mb_regex_t *retval = NULL, *rc = NULL;
OnigErrorInfo err_info;
OnigUChar err_str[ONIG_MAX_ERROR_MESSAGE_LEN];
rc = zend_hash_str_find_ptr(&MBREX(ht_rc), (char *)pattern, patlen);
if (!rc || rc->options != options || rc->enc != enc || rc->syntax != syntax) {
if ((err_code = onig_new(&retval, (OnigUChar *)pattern, (OnigUChar *)(pattern + patlen), options, enc, syntax, &err_info)) != ONIG_NORMAL) {
onig_error_code_to_str(err_str, err_code, err_info);
php_error_docref(NULL, E_WARNING, "mbregex compile err: %s", err_str);
retval = NULL;
goto out;
}
zend_hash_str_update_ptr(&MBREX(ht_rc), (char *)pattern, patlen, retval);
} else if (rc) {
retval = rc;
}
out:
return retval;
}
cell AMX_NATIVE_CALL pawn_regex_build(AMX* amx, cell* params)
{
const char *rexp = NULL;
amx_StrParam(amx, params[1], rexp);
if(rexp)
{
UChar* pattern = (UChar* )rexp;
OnigErrorInfo einfo;
rexpression[TotalExpressions].id = onig_new(&rexpression[TotalExpressions].RegExpr, pattern, pattern + strlen(rexp), ONIG_OPTION_NONE, ONIG_ENCODING_UTF8, ONIG_SYNTAX_DEFAULT, &einfo);
if(rexpression[TotalExpressions].id != ONIG_NORMAL)
{
UChar s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, rexpression[TotalExpressions].id, &einfo);
logprintf("[REGEX ERROR]: %s | at index %d. regex_exbuild aborted.", s, TotalExpressions);
onig_free(rexpression[TotalExpressions].RegExpr);
return -1;
}
rexpression[TotalExpressions].zreg = onig_region_new();
return TotalExpressions++;
}
return -1337;
}
cell AMX_NATIVE_CALL pawn_regex_exreplace(AMX* amx, cell* params)
{
const char *string = NULL, *replace = NULL;
cell* addr = NULL;
amx_GetAddr(amx, params[1], &addr);
amx_StrParam(amx, params[1], string);
amx_StrParam(amx, params[3], replace);
if(string)
{
int id=(int)params[2];
if(id>=0 && id<TotalExpressions)
{
int r=NULL;
UChar* str = (UChar* )string;
onig_region_clear(rexpression[id].zreg);
r = onig_search(rexpression[id].RegExpr, str, str+strlen(string), str, str+strlen(string), rexpression[id].zreg, ONIG_OPTION_NONE);
if(r>=0)
{
std::string asd = std::string(string);
asd.replace(asd.begin()+r, asd.begin()+rexpression[id].zreg->end[rexpression[id].zreg->num_regs-1], replace);
amx_SetString(addr, asd.c_str(), 0, 0, params[1]);
}
else if(r<ONIG_MISMATCH)
{
UChar s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r);
logprintf("[REGEX ERROR]: %s\n", s);
return -1;
}
return r;
}
logprintf("[REGEX ERROR]: Call regex_exreplace with undefined parameter at index %d", id);
return -1;
}
return -1337;
}
extern int main(int argc, char* argv[])
{
static OnigSyntaxType SQLSyntax;
int r;
unsigned char *start, *range, *end;
regex_t* reg;
OnigErrorInfo einfo;
OnigRegion *region;
static UChar* pattern = (UChar* )"\\_%\\\\__zz";
static UChar* str = (UChar* )"a_abcabcabc\\ppzz";
onig_set_syntax_op (&SQLSyntax, ONIG_SYN_OP_VARIABLE_META_CHARACTERS);
onig_set_syntax_op2 (&SQLSyntax, 0);
onig_set_syntax_behavior(&SQLSyntax, 0);
onig_set_syntax_options (&SQLSyntax, ONIG_OPTION_MULTILINE);
onig_set_meta_char(&SQLSyntax, ONIG_META_CHAR_ESCAPE, (OnigCodePoint )'\\');
onig_set_meta_char(&SQLSyntax, ONIG_META_CHAR_ANYCHAR, (OnigCodePoint )'_');
onig_set_meta_char(&SQLSyntax, ONIG_META_CHAR_ANYTIME,
ONIG_INEFFECTIVE_META_CHAR);
onig_set_meta_char(&SQLSyntax, ONIG_META_CHAR_ZERO_OR_ONE_TIME,
ONIG_INEFFECTIVE_META_CHAR);
onig_set_meta_char(&SQLSyntax, ONIG_META_CHAR_ONE_OR_MORE_TIME,
ONIG_INEFFECTIVE_META_CHAR);
onig_set_meta_char(&SQLSyntax, ONIG_META_CHAR_ANYCHAR_ANYTIME,
(OnigCodePoint )'%');
r = onig_new(®, pattern, pattern + strlen((char* )pattern),
ONIG_OPTION_DEFAULT, ONIG_ENCODING_ASCII, &SQLSyntax, &einfo);
if (r != ONIG_NORMAL) {
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r, &einfo);
fprintf(stderr, "ERROR: %s\n", s);
return -1;
}
region = onig_region_new();
end = str + strlen((char* )str);
start = str;
range = end;
r = onig_search(reg, str, end, start, range, region, ONIG_OPTION_NONE);
if (r >= 0) {
int i;
fprintf(stderr, "match at %d\n", r);
for (i = 0; i < region->num_regs; i++) {
fprintf(stderr, "%d: (%ld-%ld)\n", i, region->beg[i], region->end[i]);
}
}
else if (r == ONIG_MISMATCH) {
fprintf(stderr, "search fail\n");
}
else { /* error */
char s[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(s, r);
fprintf(stderr, "ERROR: %s\n", s);
return -1;
}
onig_region_free(region, 1 /* 1:free self, 0:free contents only */);
onig_free(reg);
onig_end();
return 0;
}
/* {{{ _php_mb_regex_ereg_replace_exec */
static void _php_mb_regex_ereg_replace_exec(INTERNAL_FUNCTION_PARAMETERS, OnigOptionType options, int is_callable)
{
zval *arg_pattern_zval;
char *arg_pattern;
size_t arg_pattern_len;
char *replace;
size_t replace_len;
zend_fcall_info arg_replace_fci;
zend_fcall_info_cache arg_replace_fci_cache;
char *string;
size_t string_len;
char *p;
php_mb_regex_t *re;
OnigSyntaxType *syntax;
OnigRegion *regs = NULL;
smart_str out_buf = {0};
smart_str eval_buf = {0};
smart_str *pbuf;
size_t i;
int err, eval, n;
OnigUChar *pos;
OnigUChar *string_lim;
char *description = NULL;
char pat_buf[6];
const mbfl_encoding *enc;
{
const char *current_enc_name;
current_enc_name = _php_mb_regex_mbctype2name(MBREX(current_mbctype));
if (current_enc_name == NULL ||
(enc = mbfl_name2encoding(current_enc_name)) == NULL) {
php_error_docref(NULL, E_WARNING, "Unknown error");
RETURN_FALSE;
}
}
eval = 0;
{
char *option_str = NULL;
size_t option_str_len = 0;
if (!is_callable) {
if (zend_parse_parameters(ZEND_NUM_ARGS(), "zss|s",
&arg_pattern_zval,
&replace, &replace_len,
&string, &string_len,
&option_str, &option_str_len) == FAILURE) {
RETURN_FALSE;
}
} else {
if (zend_parse_parameters(ZEND_NUM_ARGS(), "zfs|s",
&arg_pattern_zval,
&arg_replace_fci, &arg_replace_fci_cache,
&string, &string_len,
&option_str, &option_str_len) == FAILURE) {
RETURN_FALSE;
}
}
if (!php_mb_check_encoding(
string,
string_len,
_php_mb_regex_mbctype2name(MBREX(current_mbctype))
)) {
RETURN_NULL();
}
if (option_str != NULL) {
_php_mb_regex_init_options(option_str, option_str_len, &options, &syntax, &eval);
} else {
options |= MBREX(regex_default_options);
syntax = MBREX(regex_default_syntax);
}
}
if (eval && !is_callable) {
php_error_docref(NULL, E_DEPRECATED, "The 'e' option is deprecated, use mb_ereg_replace_callback instead");
}
if (Z_TYPE_P(arg_pattern_zval) == IS_STRING) {
arg_pattern = Z_STRVAL_P(arg_pattern_zval);
arg_pattern_len = Z_STRLEN_P(arg_pattern_zval);
} else {
/* FIXME: this code is not multibyte aware! */
convert_to_long_ex(arg_pattern_zval);
pat_buf[0] = (char)Z_LVAL_P(arg_pattern_zval);
pat_buf[1] = '\0';
pat_buf[2] = '\0';
pat_buf[3] = '\0';
pat_buf[4] = '\0';
pat_buf[5] = '\0';
arg_pattern = pat_buf;
arg_pattern_len = 1;
}
/* create regex pattern buffer */
re = php_mbregex_compile_pattern(arg_pattern, arg_pattern_len, options, MBREX(current_mbctype), syntax);
if (re == NULL) {
RETURN_FALSE;
}
if (eval || is_callable) {
pbuf = &eval_buf;
description = zend_make_compiled_string_description("mbregex replace");
} else {
pbuf = &out_buf;
description = NULL;
}
if (is_callable) {
if (eval) {
php_error_docref(NULL, E_WARNING, "Option 'e' cannot be used with replacement callback");
RETURN_FALSE;
}
}
/* do the actual work */
err = 0;
pos = (OnigUChar *)string;
string_lim = (OnigUChar*)(string + string_len);
regs = onig_region_new();
while (err >= 0) {
err = onig_search(re, (OnigUChar *)string, (OnigUChar *)string_lim, pos, (OnigUChar *)string_lim, regs, 0);
if (err <= -2) {
OnigUChar err_str[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(err_str, err);
php_error_docref(NULL, E_WARNING, "mbregex search failure in php_mbereg_replace_exec(): %s", err_str);
break;
}
if (err >= 0) {
#if moriyoshi_0
if (regs->beg[0] == regs->end[0]) {
php_error_docref(NULL, E_WARNING, "Empty regular expression");
break;
}
#endif
/* copy the part of the string before the match */
smart_str_appendl(&out_buf, (char *)pos, (size_t)((OnigUChar *)(string + regs->beg[0]) - pos));
if (!is_callable) {
/* copy replacement and backrefs */
i = 0;
p = replace;
while (i < replace_len) {
int fwd = (int) php_mb_mbchar_bytes_ex(p, enc);
n = -1;
if ((replace_len - i) >= 2 && fwd == 1 &&
p[0] == '\\' && p[1] >= '0' && p[1] <= '9') {
n = p[1] - '0';
}
if (n >= 0 && n < regs->num_regs) {
if (regs->beg[n] >= 0 && regs->beg[n] < regs->end[n] && (size_t)regs->end[n] <= string_len) {
smart_str_appendl(pbuf, string + regs->beg[n], regs->end[n] - regs->beg[n]);
}
p += 2;
i += 2;
} else {
smart_str_appendl(pbuf, p, fwd);
p += fwd;
i += fwd;
}
}
}
if (eval) {
zval v;
zend_string *eval_str;
/* null terminate buffer */
smart_str_0(&eval_buf);
if (eval_buf.s) {
eval_str = eval_buf.s;
} else {
eval_str = ZSTR_EMPTY_ALLOC();
}
/* do eval */
if (zend_eval_stringl(ZSTR_VAL(eval_str), ZSTR_LEN(eval_str), &v, description) == FAILURE) {
efree(description);
zend_throw_error(NULL, "Failed evaluating code: %s%s", PHP_EOL, ZSTR_VAL(eval_str));
onig_region_free(regs, 0);
smart_str_free(&out_buf);
smart_str_free(&eval_buf);
RETURN_FALSE;
}
/* result of eval */
convert_to_string(&v);
smart_str_appendl(&out_buf, Z_STRVAL(v), Z_STRLEN(v));
/* Clean up */
smart_str_free(&eval_buf);
zval_dtor(&v);
} else if (is_callable) {
zval args[1];
zval subpats, retval;
int i;
array_init(&subpats);
for (i = 0; i < regs->num_regs; i++) {
add_next_index_stringl(&subpats, string + regs->beg[i], regs->end[i] - regs->beg[i]);
}
ZVAL_COPY_VALUE(&args[0], &subpats);
/* null terminate buffer */
smart_str_0(&eval_buf);
arg_replace_fci.param_count = 1;
arg_replace_fci.params = args;
arg_replace_fci.retval = &retval;
if (zend_call_function(&arg_replace_fci, &arg_replace_fci_cache) == SUCCESS &&
!Z_ISUNDEF(retval)) {
convert_to_string_ex(&retval);
smart_str_appendl(&out_buf, Z_STRVAL(retval), Z_STRLEN(retval));
smart_str_free(&eval_buf);
zval_ptr_dtor(&retval);
} else {
if (!EG(exception)) {
php_error_docref(NULL, E_WARNING, "Unable to call custom replacement function");
}
}
zval_ptr_dtor(&subpats);
}
n = regs->end[0];
if ((pos - (OnigUChar *)string) < n) {
pos = (OnigUChar *)string + n;
} else {
if (pos < string_lim) {
smart_str_appendl(&out_buf, (char *)pos, 1);
}
pos++;
}
} else { /* nomatch */
/* stick that last bit of string on our output */
if (string_lim - pos > 0) {
smart_str_appendl(&out_buf, (char *)pos, string_lim - pos);
}
}
onig_region_free(regs, 0);
}
if (description) {
efree(description);
}
if (regs != NULL) {
onig_region_free(regs, 1);
}
smart_str_free(&eval_buf);
if (err <= -2) {
smart_str_free(&out_buf);
RETVAL_FALSE;
} else if (out_buf.s) {
smart_str_0(&out_buf);
RETVAL_STR(out_buf.s);
} else {
RETVAL_EMPTY_STRING();
}
}
/* {{{ _php_mb_regex_ereg_search_exec */
static void
_php_mb_regex_ereg_search_exec(INTERNAL_FUNCTION_PARAMETERS, int mode)
{
char *arg_pattern = NULL, *arg_options = NULL;
size_t arg_pattern_len, arg_options_len;
int err;
size_t n, i, pos, len, beg, end;
OnigOptionType option;
OnigUChar *str;
OnigSyntaxType *syntax;
if (zend_parse_parameters(ZEND_NUM_ARGS(), "|ss", &arg_pattern, &arg_pattern_len, &arg_options, &arg_options_len) == FAILURE) {
return;
}
option = MBREX(regex_default_options);
if (arg_options) {
option = 0;
_php_mb_regex_init_options(arg_options, arg_options_len, &option, &syntax, NULL);
}
if (arg_pattern) {
/* create regex pattern buffer */
if ((MBREX(search_re) = php_mbregex_compile_pattern(arg_pattern, arg_pattern_len, option, MBREX(current_mbctype), MBREX(regex_default_syntax))) == NULL) {
RETURN_FALSE;
}
}
pos = MBREX(search_pos);
str = NULL;
len = 0;
if (!Z_ISUNDEF(MBREX(search_str)) && Z_TYPE(MBREX(search_str)) == IS_STRING){
str = (OnigUChar *)Z_STRVAL(MBREX(search_str));
len = Z_STRLEN(MBREX(search_str));
}
if (MBREX(search_re) == NULL) {
php_error_docref(NULL, E_WARNING, "No regex given");
RETURN_FALSE;
}
if (str == NULL) {
php_error_docref(NULL, E_WARNING, "No string given");
RETURN_FALSE;
}
if (MBREX(search_regs)) {
onig_region_free(MBREX(search_regs), 1);
}
MBREX(search_regs) = onig_region_new();
err = onig_search(MBREX(search_re), str, str + len, str + pos, str + len, MBREX(search_regs), 0);
if (err == ONIG_MISMATCH) {
MBREX(search_pos) = len;
RETVAL_FALSE;
} else if (err <= -2) {
OnigUChar err_str[ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(err_str, err);
php_error_docref(NULL, E_WARNING, "mbregex search failure in mbregex_search(): %s", err_str);
RETVAL_FALSE;
} else {
switch (mode) {
case 1:
array_init(return_value);
beg = MBREX(search_regs)->beg[0];
end = MBREX(search_regs)->end[0];
add_next_index_long(return_value, beg);
add_next_index_long(return_value, end - beg);
break;
case 2:
array_init(return_value);
n = MBREX(search_regs)->num_regs;
for (i = 0; i < n; i++) {
beg = MBREX(search_regs)->beg[i];
end = MBREX(search_regs)->end[i];
if (beg >= 0 && beg <= end && end <= len) {
add_index_stringl(return_value, i, (char *)&str[beg], end - beg);
} else {
add_index_bool(return_value, i, 0);
}
}
break;
default:
RETVAL_TRUE;
break;
}
end = MBREX(search_regs)->end[0];
if (pos <= end) {
MBREX(search_pos) = end;
} else {
MBREX(search_pos) = pos + 1;
}
}
if (err < 0) {
onig_region_free(MBREX(search_regs), 1);
MBREX(search_regs) = (OnigRegion *)NULL;
}
}
static int generate_error (lua_State *L, const TOnig *ud, int errcode) {
char buf [ONIG_MAX_ERROR_MESSAGE_LEN];
onig_error_code_to_str(buf, errcode, ud->einfo);
return luaL_error(L, buf);
}
/**
Call the Oniguruma regex match API.
Same parameters as RegularExpressionMatch, except SyntaxType is required.
@param String A pointer to a NULL terminated string to match against the
regular expression string specified by Pattern.
@param Pattern A pointer to a NULL terminated string that represents the
regular expression.
@param SyntaxType A pointer to the EFI_REGEX_SYNTAX_TYPE that identifies the
regular expression syntax type to use. May be NULL in which
case the function will use its default regular expression
syntax type.
@param Result On return, points to TRUE if String fully matches against
the regular expression Pattern using the regular expression
SyntaxType. Otherwise, points to FALSE.
@param Captures A Pointer to an array of EFI_REGEX_CAPTURE objects to receive
the captured groups in the event of a match. The full
sub-string match is put in Captures[0], and the results of N
capturing groups are put in Captures[1:N]. If Captures is
NULL, then this function doesn't allocate the memory for the
array and does not build up the elements. It only returns the
number of matching patterns in CapturesCount. If Captures is
not NULL, this function returns a pointer to an array and
builds up the elements in the array. CapturesCount is also
updated to the number of matching patterns found. It is the
caller's responsibility to free the memory pool in Captures
and in each CapturePtr in the array elements.
@param CapturesCount On output, CapturesCount is the number of matching patterns
found in String. Zero means no matching patterns were found
in the string.
@retval EFI_SUCCESS Regex compilation and match completed successfully.
@retval EFI_DEVICE_ERROR Regex compilation failed.
**/
STATIC
EFI_STATUS
OnigurumaMatch (
IN CHAR16 *String,
IN CHAR16 *Pattern,
IN EFI_REGEX_SYNTAX_TYPE *SyntaxType,
OUT BOOLEAN *Result,
OUT EFI_REGEX_CAPTURE **Captures, OPTIONAL
OUT UINTN *CapturesCount
)
{
regex_t *OnigRegex;
OnigSyntaxType *OnigSyntax;
OnigRegion *Region;
INT32 OnigResult;
OnigErrorInfo ErrorInfo;
CHAR8 ErrorMessage[ONIG_MAX_ERROR_MESSAGE_LEN];
UINT32 Index;
OnigUChar *Start;
EFI_STATUS Status;
Status = EFI_SUCCESS;
//
// Detemine the internal syntax type
//
OnigSyntax = ONIG_SYNTAX_DEFAULT;
if (CompareGuid (SyntaxType, &gEfiRegexSyntaxTypePosixExtendedGuid)) {
OnigSyntax = ONIG_SYNTAX_POSIX_EXTENDED;
} else if (CompareGuid (SyntaxType, &gEfiRegexSyntaxTypePerlGuid)) {
OnigSyntax = ONIG_SYNTAX_PERL;
} else {
DEBUG ((DEBUG_ERROR, "Unsupported regex syntax - using default\n"));
return EFI_UNSUPPORTED;
}
//
// Compile pattern
//
Start = (OnigUChar*)Pattern;
OnigResult = onig_new (
&OnigRegex,
Start,
Start + onigenc_str_bytelen_null (CHAR16_ENCODING, Start),
ONIG_OPTION_DEFAULT,
CHAR16_ENCODING,
OnigSyntax,
&ErrorInfo
);
if (OnigResult != ONIG_NORMAL) {
onig_error_code_to_str (ErrorMessage, OnigResult, &ErrorInfo);
DEBUG ((DEBUG_ERROR, "Regex compilation failed: %a\n", ErrorMessage));
return EFI_DEVICE_ERROR;
}
//
// Try to match
//
Start = (OnigUChar*)String;
Region = onig_region_new ();
if (Region == NULL) {
onig_free (OnigRegex);
return EFI_OUT_OF_RESOURCES;
}
OnigResult = onig_search (
OnigRegex,
Start,
Start + onigenc_str_bytelen_null (CHAR16_ENCODING, Start),
Start,
Start + onigenc_str_bytelen_null (CHAR16_ENCODING, Start),
Region,
ONIG_OPTION_NONE
);
if (OnigResult >= 0) {
*Result = TRUE;
} else {
*Result = FALSE;
if (OnigResult != ONIG_MISMATCH) {
onig_error_code_to_str (ErrorMessage, OnigResult);
DEBUG ((DEBUG_ERROR, "Regex match failed: %a\n", ErrorMessage));
onig_region_free (Region, 1);
onig_free (OnigRegex);
return EFI_DEVICE_ERROR;
}
}
//
// If successful, copy out the region (capture) information
//
if (*Result && Captures != NULL) {
*CapturesCount = Region->num_regs;
*Captures = AllocateZeroPool (*CapturesCount * sizeof(**Captures));
if (*Captures != NULL) {
for (Index = 0; Index < *CapturesCount; ++Index) {
//
// Region beg/end values represent bytes, not characters
//
(*Captures)[Index].Length = (Region->end[Index] - Region->beg[Index]) / sizeof(CHAR16);
(*Captures)[Index].CapturePtr = AllocateCopyPool (
((*Captures)[Index].Length) * sizeof (CHAR16),
(CHAR16*)((UINTN)String + Region->beg[Index])
);
if ((*Captures)[Index].CapturePtr == NULL) {
Status = EFI_OUT_OF_RESOURCES;
break;
}
}
if (EFI_ERROR (Status)) {
for (Index = 0; Index < *CapturesCount; ++Index) {
if ((*Captures)[Index].CapturePtr != NULL) {
FreePool ((CHAR16*)(*Captures)[Index].CapturePtr);
}
}
FreePool (*Captures);
}
}
}
onig_region_free (Region, 1);
onig_free (OnigRegex);
return Status;
}