Skip to content

ext/pgsql: convert regex match, using pcre jit if available and enabled. #15039

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 2 commits into from
Closed

ext/pgsql: convert regex match, using pcre jit if available and enabled. #15039

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ec1cdb0
ext/pgsql: convert regex match, using pcre jit if available and enabled.
devnexen Jul 20, 2024
985b227
changes from feedback, using cache
devnexen Jul 20, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
146 changes: 70 additions & 76 deletions ext/pgsql/pgsql.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -490,8 +490,55 @@ static PHP_GINIT_FUNCTION(pgsql)
#if defined(COMPILE_DL_PGSQL) && defined(ZTS)
ZEND_TSRMLS_CACHE_UPDATE();
#endif

size_t i = 0;
memset(pgsql_globals, 0, sizeof(zend_pgsql_globals));
zend_hash_init(&pgsql_globals->connections, 0, NULL, NULL, 1);

#define ADD_REGEX(reg) \
do { \
ZEND_ASSERT(i < PGSQL_MAX_REGEXES); \
pgsql_globals->regexes[i ++] = zend_string_init(reg, strlen(reg), true);\
} while(0)
ADD_REGEX("#^([+-]{0,1}[0-9]+)$#n");
ADD_REGEX("#^[-+]?[0-9]*\\.?[0-9]+([eE][-+]?[0-9]+)?$#n");
ADD_REGEX("#^[+-]{0,1}(inf)(inity){0,1}$#ni");
ADD_REGEX("#^[0-9]+$#n");
ADD_REGEX("#^((25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])\\.){3,3}(25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])(\\/[0-9]{1,3})?$#n");
ADD_REGEX("#^(([0-9a-fA-F]{1,4}:){7,7}[0-9a-fA-F]{1,4}|([0-9a-fA-F]{1,4}:){1,7}:|([0-9a-fA-F]{1,4}:){1,6}:[0-9a-fA-F]{1,4}|([0-9a-fA-F]{1,4}:){1,5}(:[0-9a-fA-F]{1,4}){1,2}|([0-9a-fA-F]{1,4}:){1,4}(:[0-9a-fA-F]{1,4}){1,3}|([0-9a-fA-F]{1,4}:){1,3}(:[0-9a-fA-F]{1,4}){1,4}|([0-9a-fA-F]{1,4}:){1,2}(:[0-9a-fA-F]{1,4}){1,5}|[0-9a-fA-F]{1,4}:((:[0-9a-fA-F]{1,4}){1,6})|:((:[0-9a-fA-F]{1,4}){1,7}|:)|fe80:(:[0-9a-fA-F]{0,4}){0,4}%[0-9a-zA-Z]{1,}|::(ffff(:0{1,4}){0,1}:){0,1}((25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])\\.){3,3}(25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])|([0-9a-fA-F]{1,4}:){1,4}:((25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])\\.){3,3}(25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9]))(\\/[0-9]{1,3})?$#n");
ADD_REGEX("#^([0-9]{4}[/-][0-9]{1,2}[/-][0-9]{1,2})(([ \\t]+|T)(([0-9]{1,2}:[0-9]{1,2}){1}(:[0-9]{1,2}){0,1}(\\.[0-9]+){0,1}([ \\t]*([+-][0-9]{1,4}(:[0-9]{1,2}){0,1}|[-a-zA-Z_/+]{1,50})){0,1})){0,1}$#ni");
ADD_REGEX("#^([0-9]{4}[/-][0-9]{1,2}[/-][0-9]{1,2})$#ni");
ADD_REGEX("#^(([0-9]{1,2}:[0-9]{1,2}){1}(:[0-9]{1,2}){0,1}){0,1}$#ni");
ADD_REGEX("#^(@?[ \\t]+)?("
/* Textual time units and their abbreviations: */
"(([-+]?[ \\t]+)?"
"[0-9]+(\\.[0-9]*)?[ \\t]*"
"(millenniums|millennia|millennium|mil|mils|"
"centuries|century|cent|c|"
"decades|decade|dec|decs|"
"years|year|y|"
"months|month|mon|"
"weeks|week|w|"
"days|day|d|"
"hours|hour|hr|hrs|h|"
"minutes|minute|mins|min|m|"
"seconds|second|secs|sec|s))+|"
/* Textual time units plus (dd)* hh[:mm[:ss]] */
"((([-+]?[ \\t]+)?"
"[0-9]+(\\.[0-9]*)?[ \\t]*"
"(millenniums|millennia|millennium|mil|mils|"
"centuries|century|cent|c|"
"decades|decade|dec|decs|"
"years|year|y|"
"months|month|mon|"
"weeks|week|w|"
"days|day|d))+"
"([-+]?[ \\t]+"
"([0-9]+[ \\t]+)+" /* dd */
"(([0-9]{1,2}:){0,2}[0-9]{0,2})" /* hh:[mm:[ss]] */
")?))"
"([ \\t]+ago)?$#ni");
ADD_REGEX("#^([0-9a-f]{2,2}:){5,5}[0-9a-f]{2,2}$#ni");
}

static void php_libpq_version(char *buf, size_t len)
Expand Down Expand Up @@ -560,6 +607,9 @@ PHP_MSHUTDOWN_FUNCTION(pgsql)
UNREGISTER_INI_ENTRIES();
zend_hash_destroy(&PGG(connections));

for (size_t i = 0; i < PGSQL_MAX_REGEXES; i ++)
zend_string_release_ex(PGG(regexes[i]), true);

return SUCCESS;
}

Expand Down Expand Up @@ -4671,12 +4721,11 @@ static php_pgsql_data_type php_pgsql_get_data_type(const zend_string *type_name)
/* {{{ php_pgsql_convert_match
* test field value with regular expression specified.
*/
static int php_pgsql_convert_match(const zend_string *str, const char *regex , size_t regex_len, int icase)
static int php_pgsql_convert_match(const zend_string *str, zend_string *regex)
{
pcre_cache_entry *centry;
pcre2_code *re;
PCRE2_SIZE err_offset;
int res, errnumber;
uint32_t options = PCRE2_NO_AUTO_CAPTURE;
int res;
size_t i;
pcre2_match_data *match_data;

Expand All @@ -4689,27 +4738,21 @@ static int php_pgsql_convert_match(const zend_string *str, const char *regex , s
}
}

if (icase) {
options |= PCRE2_CASELESS;
}

re = pcre2_compile((PCRE2_SPTR)regex, regex_len, options, &errnumber, &err_offset, php_pcre_cctx());
if (NULL == re) {
PCRE2_UCHAR err_msg[128];
pcre2_get_error_message(errnumber, err_msg, sizeof(err_msg));
php_error_docref(NULL, E_WARNING, "Cannot compile regex: '%s'", err_msg);
centry = pcre_get_compiled_regex_cache(regex);
if (NULL == centry) {
return FAILURE;
}

re = php_pcre_pce_re(centry);
match_data = php_pcre_create_match_data(0, re);
if (NULL == match_data) {
pcre2_code_free(re);
php_error_docref(NULL, E_WARNING, "Cannot allocate match data");
return FAILURE;
}
php_pcre_pce_incref(centry);
res = pcre2_match(re, (PCRE2_SPTR)ZSTR_VAL(str), ZSTR_LEN(str), 0, 0, match_data, php_pcre_mctx());
php_pcre_free_match_data(match_data);
pcre2_code_free(re);
php_pcre_pce_decref(centry);

if (res == PCRE2_ERROR_NOMATCH) {
return FAILURE;
Expand Down Expand Up @@ -4890,14 +4933,12 @@ PHP_PGSQL_API zend_result php_pgsql_convert(PGconn *pg_link, const zend_string *
}
else {
/* FIXME: better regex must be used */
#define REGEX0 "^([+-]{0,1}[0-9]+)$"
if (php_pgsql_convert_match(Z_STR_P(val), REGEX0, sizeof(REGEX0)-1, 0) == FAILURE) {
if (php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[0])) == FAILURE) {
err = 1;
}
else {
ZVAL_STRINGL(&new_val, Z_STRVAL_P(val), Z_STRLEN_P(val));
}
#undef REGEX0
}
break;

Expand Down Expand Up @@ -4933,11 +4974,9 @@ PHP_PGSQL_API zend_result php_pgsql_convert(PGconn *pg_link, const zend_string *
ZVAL_STR(&new_val, ZSTR_KNOWN(ZEND_STR_NULL));
}
else {
#define REGEX0 "^[-+]?[0-9]*\\.?[0-9]+([eE][-+]?[0-9]+)?$"
#define REGEX1 "^[+-]{0,1}(inf)(inity){0,1}$"
/* better regex? */
if (php_pgsql_convert_match(Z_STR_P(val), REGEX0, sizeof(REGEX0)-1, 0) == FAILURE) {
if (php_pgsql_convert_match(Z_STR_P(val), REGEX1, sizeof(REGEX1)-1, 1) == FAILURE) {
if (php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[1])) == FAILURE) {
if (php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[2])) == FAILURE) {
err = 1;
} else {
ZVAL_STR(&new_val, php_pgsql_add_quotes(Z_STR_P(val)));
Expand All @@ -4946,8 +4985,6 @@ PHP_PGSQL_API zend_result php_pgsql_convert(PGconn *pg_link, const zend_string *
else {
ZVAL_STRING(&new_val, Z_STRVAL_P(val));
}
#undef REGEX0
#undef REGEX1
}
break;

Expand Down Expand Up @@ -5043,7 +5080,7 @@ PHP_PGSQL_API zend_result php_pgsql_convert(PGconn *pg_link, const zend_string *
}
else {
/* better regex? */
if (php_pgsql_convert_match(Z_STR_P(val), "^[0-9]+$", sizeof("^[0-9]+$")-1, 0) == FAILURE) {
if (php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[3])) == FAILURE) {
err = 1;
}
else {
Expand Down Expand Up @@ -5083,20 +5120,16 @@ PHP_PGSQL_API zend_result php_pgsql_convert(PGconn *pg_link, const zend_string *
ZVAL_STR(&new_val, ZSTR_KNOWN(ZEND_STR_NULL));
}
else {
#define REGEX0 "^((25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])\\.){3,3}(25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])(\\/[0-9]{1,3})?$"
#define REGEX1 "^(([0-9a-fA-F]{1,4}:){7,7}[0-9a-fA-F]{1,4}|([0-9a-fA-F]{1,4}:){1,7}:|([0-9a-fA-F]{1,4}:){1,6}:[0-9a-fA-F]{1,4}|([0-9a-fA-F]{1,4}:){1,5}(:[0-9a-fA-F]{1,4}){1,2}|([0-9a-fA-F]{1,4}:){1,4}(:[0-9a-fA-F]{1,4}){1,3}|([0-9a-fA-F]{1,4}:){1,3}(:[0-9a-fA-F]{1,4}){1,4}|([0-9a-fA-F]{1,4}:){1,2}(:[0-9a-fA-F]{1,4}){1,5}|[0-9a-fA-F]{1,4}:((:[0-9a-fA-F]{1,4}){1,6})|:((:[0-9a-fA-F]{1,4}){1,7}|:)|fe80:(:[0-9a-fA-F]{0,4}){0,4}%[0-9a-zA-Z]{1,}|::(ffff(:0{1,4}){0,1}:){0,1}((25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])\\.){3,3}(25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])|([0-9a-fA-F]{1,4}:){1,4}:((25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])\\.){3,3}(25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9]))(\\/[0-9]{1,3})?$"
/* The inet type holds an IPv4 or IPv6 host address, and optionally its subnet, all in one field. See more in the doc.
The regex might still be not perfect, but catches the most of IP variants. We might decide to remove the regex
at all though and let the server side to handle it.*/
if (php_pgsql_convert_match(Z_STR_P(val), REGEX0, sizeof(REGEX0)-1, 0) == FAILURE
&& php_pgsql_convert_match(Z_STR_P(val), REGEX1, sizeof(REGEX1)-1, 0) == FAILURE) {
if (php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[4])) == FAILURE
&& php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[5])) == FAILURE) {
err = 2;
}
else {
ZVAL_STR(&new_val, php_pgsql_add_quotes(Z_STR_P(val)));
}
#undef REGEX0
#undef REGEX1
}
break;

Expand Down Expand Up @@ -5127,14 +5160,12 @@ PHP_PGSQL_API zend_result php_pgsql_convert(PGconn *pg_link, const zend_string *
} else if (zend_string_equals_literal_ci(Z_STR_P(val), "now()")) {
ZVAL_STRINGL(&new_val, "NOW()", sizeof("NOW()")-1);
} else {
#define REGEX0 "^([0-9]{4}[/-][0-9]{1,2}[/-][0-9]{1,2})(([ \\t]+|T)(([0-9]{1,2}:[0-9]{1,2}){1}(:[0-9]{1,2}){0,1}(\\.[0-9]+){0,1}([ \\t]*([+-][0-9]{1,4}(:[0-9]{1,2}){0,1}|[-a-zA-Z_/+]{1,50})){0,1})){0,1}$"
/* better regex? */
if (php_pgsql_convert_match(Z_STR_P(val), REGEX0, sizeof(REGEX0)-1, 1) == FAILURE) {
if (php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[6])) == FAILURE) {
err = 1;
} else {
ZVAL_STR(&new_val, php_pgsql_add_quotes(Z_STR_P(val)));
}
#undef REGEX0
}
break;

Expand All @@ -5158,15 +5189,13 @@ PHP_PGSQL_API zend_result php_pgsql_convert(PGconn *pg_link, const zend_string *
ZVAL_STR(&new_val, ZSTR_KNOWN(ZEND_STR_NULL));
}
else {
#define REGEX0 "^([0-9]{4}[/-][0-9]{1,2}[/-][0-9]{1,2})$"
/* FIXME: better regex must be used */
if (php_pgsql_convert_match(Z_STR_P(val), REGEX0, sizeof(REGEX0)-1, 1) == FAILURE) {
if (php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[7])) == FAILURE) {
err = 1;
}
else {
ZVAL_STR(&new_val, php_pgsql_add_quotes(Z_STR_P(val)));
}
#undef REGEX0
}
break;

Expand All @@ -5190,15 +5219,13 @@ PHP_PGSQL_API zend_result php_pgsql_convert(PGconn *pg_link, const zend_string *
ZVAL_STR(&new_val, ZSTR_KNOWN(ZEND_STR_NULL));
}
else {
#define REGEX0 "^(([0-9]{1,2}:[0-9]{1,2}){1}(:[0-9]{1,2}){0,1}){0,1}$"
/* FIXME: better regex must be used */
if (php_pgsql_convert_match(Z_STR_P(val), REGEX0, sizeof(REGEX0)-1, 1) == FAILURE) {
if (php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[8])) == FAILURE) {
err = 1;
}
else {
ZVAL_STR(&new_val, php_pgsql_add_quotes(Z_STR_P(val)));
}
#undef REGEX0
}
break;

Expand Down Expand Up @@ -5239,44 +5266,13 @@ PHP_PGSQL_API zend_result php_pgsql_convert(PGconn *pg_link, const zend_string *
unit markings. For example, '1 12:59:10' is read the same as '1 day 12 hours 59 min 10
sec'.
*/
#define REGEX0 \
"^(@?[ \\t]+)?(" \
/* Textual time units and their abbreviations: */ \
"(([-+]?[ \\t]+)?" \
"[0-9]+(\\.[0-9]*)?[ \\t]*" \
"(millenniums|millennia|millennium|mil|mils|" \
"centuries|century|cent|c|" \
"decades|decade|dec|decs|" \
"years|year|y|" \
"months|month|mon|" \
"weeks|week|w|" \
"days|day|d|" \
"hours|hour|hr|hrs|h|" \
"minutes|minute|mins|min|m|" \
"seconds|second|secs|sec|s))+|" \
/* Textual time units plus (dd)* hh[:mm[:ss]] */ \
"((([-+]?[ \\t]+)?" \
"[0-9]+(\\.[0-9]*)?[ \\t]*" \
"(millenniums|millennia|millennium|mil|mils|" \
"centuries|century|cent|c|" \
"decades|decade|dec|decs|" \
"years|year|y|" \
"months|month|mon|" \
"weeks|week|w|" \
"days|day|d))+" \
"([-+]?[ \\t]+" \
"([0-9]+[ \\t]+)+" /* dd */ \
"(([0-9]{1,2}:){0,2}[0-9]{0,2})" /* hh:[mm:[ss]] */ \
")?))" \
"([ \\t]+ago)?$"

if (php_pgsql_convert_match(Z_STR_P(val), REGEX0, sizeof(REGEX0)-1, 1) == FAILURE) {

if (php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[9])) == FAILURE) {
err = 1;
}
else {
ZVAL_STR(&new_val, php_pgsql_add_quotes(Z_STR_P(val)));
}
#undef REGEX0
}
break;

Expand Down Expand Up @@ -5341,14 +5337,12 @@ PHP_PGSQL_API zend_result php_pgsql_convert(PGconn *pg_link, const zend_string *
ZVAL_STR(&new_val, ZSTR_KNOWN(ZEND_STR_NULL));
}
else {
#define REGEX0 "^([0-9a-f]{2,2}:){5,5}[0-9a-f]{2,2}$"
if (php_pgsql_convert_match(Z_STR_P(val), REGEX0, sizeof(REGEX0)-1, 1) == FAILURE) {
if (php_pgsql_convert_match(Z_STR_P(val), PGG(regexes[10])) == FAILURE) {
err = 1;
}
else {
ZVAL_STR(&new_val, php_pgsql_add_quotes(Z_STR_P(val)));
}
#undef REGEX0
}
break;

Expand Down
5 changes: 4 additions & 1 deletion ext/pgsql/php_pgsql.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -175,14 +175,17 @@ static const php_stream_ops php_stream_pgsql_fd_ops = {
php_pgsql_fd_set_option
};

#define PGSQL_MAX_REGEXES 11

ZEND_BEGIN_MODULE_GLOBALS(pgsql)
zend_long num_links,num_persistent;
zend_long max_links,max_persistent;
bool allow_persistent;
int ignore_notices;
int ignore_notices;
zend_long auto_reset_persistent;
int log_notices;
zend_object *default_link; /* default link when connection is omitted */
zend_string *regexes[PGSQL_MAX_REGEXES];
HashTable field_oids;
HashTable table_oids;
HashTable connections;
Expand Down
Loading