php-src/ext/pcre/php_pcre.c

1423 lines
42 KiB
C
Raw Normal View History

/*
+----------------------------------------------------------------------+
2001-12-11 15:32:16 +00:00
| PHP Version 4 |
+----------------------------------------------------------------------+
2001-12-11 15:32:16 +00:00
| Copyright (c) 1997-2002 The PHP Group |
+----------------------------------------------------------------------+
| This source file is subject to version 2.02 of the PHP license, |
1999-07-16 13:13:16 +00:00
| that is bundled with this package in the file LICENSE, and is |
| available at through the world-wide-web at |
| http://www.php.net/license/2_02.txt. |
1999-07-16 13:13:16 +00:00
| If you did not receive a copy of the PHP license and are unable to |
| obtain it through the world-wide-web, please send a note to |
| license@php.net so we can mail you a copy immediately. |
+----------------------------------------------------------------------+
2002-02-28 08:29:35 +00:00
| Author: Andrei Zmievski <andrei@ispi.net> |
+----------------------------------------------------------------------+
*/
/* $Id$ */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "php.h"
1999-07-06 19:49:48 +00:00
#include "php_globals.h"
#include "php_pcre.h"
2000-04-05 20:42:59 +00:00
#include "ext/standard/info.h"
#include "ext/standard/php_smart_str.h"
#if HAVE_PCRE || HAVE_BUNDLED_PCRE
#include "ext/standard/php_string.h"
#define PREG_PATTERN_ORDER 0
#define PREG_SET_ORDER 1
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
#define PREG_SPLIT_NO_EMPTY (1<<0)
#define PREG_SPLIT_DELIM_CAPTURE (1<<1)
#define PREG_REPLACE_EVAL (1<<0)
#define PREG_GREP_INVERT (1<<0)
2001-07-30 15:41:11 +00:00
ZEND_DECLARE_MODULE_GLOBALS(pcre)
static void *php_pcre_malloc(size_t size)
{
return pemalloc(size, 1);
}
static void php_pcre_free(void *ptr)
{
pefree(ptr, 1);
}
static void php_free_pcre_cache(void *data)
{
pcre_cache_entry *pce = (pcre_cache_entry *) data;
pefree(pce->re, 1);
1999-07-20 20:57:04 +00:00
#if HAVE_SETLOCALE
2001-04-22 17:20:03 +00:00
if ((void*)pce->tables) pefree((void*)pce->tables, 1);
1999-07-20 20:57:04 +00:00
#endif
}
2001-07-30 15:41:11 +00:00
static void php_pcre_init_globals(zend_pcre_globals *pcre_globals TSRMLS_DC)
{
2001-07-30 15:41:11 +00:00
zend_hash_init(&pcre_globals->pcre_cache, 0, NULL, php_free_pcre_cache, 1);
}
2001-07-30 15:41:11 +00:00
#ifdef ZTS
static void php_pcre_shutdown_globals(zend_pcre_globals *pcre_globals TSRMLS_DC)
{
2001-07-30 15:41:11 +00:00
zend_hash_destroy(&pcre_globals->pcre_cache);
}
#endif
1999-09-03 21:12:10 +00:00
static /* {{{ PHP_MINFO_FUNCTION(pcre) */
1999-09-03 19:01:19 +00:00
PHP_MINFO_FUNCTION(pcre)
{
2000-04-05 20:42:59 +00:00
php_info_print_table_start();
2000-04-06 21:07:44 +00:00
php_info_print_table_row(2, "PCRE (Perl Compatible Regular Expressions) Support", "enabled" );
2000-04-05 20:42:59 +00:00
php_info_print_table_row(2, "PCRE Library Version", pcre_version() );
php_info_print_table_end();
}
/* }}} */
1999-09-03 19:01:19 +00:00
/* {{{ PHP_MINIT_FUNCTION(pcre) */
1999-09-03 21:12:10 +00:00
static PHP_MINIT_FUNCTION(pcre)
{
2001-07-30 15:41:11 +00:00
ZEND_INIT_MODULE_GLOBALS(pcre, php_pcre_init_globals, php_pcre_shutdown_globals);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
REGISTER_LONG_CONSTANT("PREG_PATTERN_ORDER", PREG_PATTERN_ORDER, CONST_CS | CONST_PERSISTENT);
REGISTER_LONG_CONSTANT("PREG_SET_ORDER", PREG_SET_ORDER, CONST_CS | CONST_PERSISTENT);
REGISTER_LONG_CONSTANT("PREG_SPLIT_NO_EMPTY", PREG_SPLIT_NO_EMPTY, CONST_CS | CONST_PERSISTENT);
REGISTER_LONG_CONSTANT("PREG_SPLIT_DELIM_CAPTURE", PREG_SPLIT_DELIM_CAPTURE, CONST_CS | CONST_PERSISTENT);
REGISTER_LONG_CONSTANT("PREG_GREP_INVERT", PREG_GREP_INVERT, CONST_CS | CONST_PERSISTENT);
return SUCCESS;
}
/* }}} */
1999-09-03 19:01:19 +00:00
/* {{{ PHP_MSHUTDOWN_FUNCTION(pcre) */
1999-09-03 21:12:10 +00:00
static PHP_MSHUTDOWN_FUNCTION(pcre)
{
return SUCCESS;
}
/* }}} */
1999-09-03 19:01:19 +00:00
/* {{{ PHP_RINIT_FUNCTION(pcre) */
1999-09-03 21:12:10 +00:00
static PHP_RINIT_FUNCTION(pcre)
{
pcre_malloc = php_pcre_malloc;
pcre_free = php_pcre_free;
return SUCCESS;
}
/* }}} */
/* {{{ pcre_get_compiled_regex
*/
PHPAPI pcre* pcre_get_compiled_regex(char *regex, pcre_extra **extra, int *preg_options) {
pcre *re = NULL;
int coptions = 0;
int soptions = 0;
const char *error;
int erroffset;
char delimiter;
char start_delimiter;
char end_delimiter;
char *p, *pp;
char *pattern;
int regex_len;
int do_study = 0;
1999-07-05 15:25:51 +00:00
int poptions = 0;
1999-07-20 20:57:04 +00:00
unsigned const char *tables = NULL;
#if HAVE_SETLOCALE
char *locale = setlocale(LC_CTYPE, NULL);
#endif
pcre_cache_entry *pce;
pcre_cache_entry new_entry;
2001-07-30 15:41:11 +00:00
TSRMLS_FETCH();
/* Try to lookup the cached regex entry, and if successful, just pass
back the compiled pattern, otherwise go on and compile it. */
regex_len = strlen(regex);
if (zend_hash_find(&PCRE_G(pcre_cache), regex, regex_len+1, (void **)&pce) == SUCCESS) {
1999-07-20 20:57:04 +00:00
#if HAVE_SETLOCALE
if (!strcmp(pce->locale, locale)) {
#endif
*extra = pce->extra;
1999-07-20 20:57:04 +00:00
*preg_options = pce->preg_options;
return pce->re;
#if HAVE_SETLOCALE
}
#endif
}
p = regex;
/* Parse through the leading whitespace, and display a warning if we
get to the end without encountering a delimiter. */
while (isspace((int)*p)) p++;
if (*p == 0) {
zend_error(E_WARNING, "Empty regular expression");
return NULL;
}
/* Get the delimiter and display a warning if it is alphanumeric
or a backslash. */
delimiter = *p++;
if (isalnum((int)delimiter) || delimiter == '\\') {
zend_error(E_WARNING, "Delimiter must not be alphanumeric or backslash");
return NULL;
}
start_delimiter = delimiter;
if ((pp = strchr("([{< )]}> )]}>", delimiter)))
delimiter = pp[5];
end_delimiter = delimiter;
if (start_delimiter == end_delimiter) {
/* We need to iterate through the pattern, searching for the ending delimiter,
but skipping the backslashed delimiters. If the ending delimiter is not
found, display a warning. */
pp = p;
while (*pp != 0) {
if (*pp == '\\' && pp[1] != 0) pp++;
else if (*pp == delimiter)
break;
pp++;
}
if (*pp == 0) {
zend_error(E_WARNING, "No ending delimiter '%c' found", delimiter);
return NULL;
}
} else {
/* We iterate through the pattern, searching for the matching ending
* delimiter. For each matching starting delimiter, we increment nesting
* level, and decrement it for each matching ending delimiter. If we
* reach the end of the pattern without matching, display a warning.
*/
int brackets = 1; /* brackets nesting level */
pp = p;
while (*pp != 0) {
if (*pp == '\\' && pp[1] != 0) pp++;
else if (*pp == end_delimiter && --brackets <= 0)
break;
else if (*pp == start_delimiter)
brackets++;
pp++;
}
if (*pp == 0) {
zend_error(E_WARNING, "No ending matching delimiter '%c' found", end_delimiter);
return NULL;
}
}
/* Make a copy of the actual pattern. */
pattern = estrndup(p, pp-p);
/* Move on to the options */
pp++;
/* Clear out preg options */
*preg_options = 0;
/* Parse through the options, setting appropriate flags. Display
1999-07-22 14:22:32 +00:00
a warning if we encounter an unknown modifier. */
while (*pp != 0) {
switch (*pp++) {
/* Perl compatible options */
case 'i': coptions |= PCRE_CASELESS; break;
case 'm': coptions |= PCRE_MULTILINE; break;
case 's': coptions |= PCRE_DOTALL; break;
case 'x': coptions |= PCRE_EXTENDED; break;
/* PCRE specific options */
case 'A': coptions |= PCRE_ANCHORED; break;
case 'D': coptions |= PCRE_DOLLAR_ENDONLY;break;
case 'S': do_study = 1; break;
case 'U': coptions |= PCRE_UNGREEDY; break;
case 'X': coptions |= PCRE_EXTRA; break;
case 'u': coptions |= PCRE_UTF8; break;
/* Custom preg options */
case 'e': poptions |= PREG_REPLACE_EVAL; break;
case ' ':
case '\n':
break;
default:
1999-07-22 14:22:32 +00:00
zend_error(E_WARNING, "Unknown modifier '%c'", pp[-1]);
efree(pattern);
return NULL;
}
}
1999-07-20 20:57:04 +00:00
#if HAVE_SETLOCALE
if (strcmp(locale, "C"))
tables = pcre_maketables();
#endif
/* Compile pattern and display a warning if compilation failed. */
re = pcre_compile(pattern,
coptions,
&error,
&erroffset,
1999-07-20 20:57:04 +00:00
tables);
if (re == NULL) {
zend_error(E_WARNING, "Compilation failed: %s at offset %d\n", error, erroffset);
efree(pattern);
return NULL;
}
/* If study option was specified, study the pattern and
store the result in extra for passing to pcre_exec. */
if (do_study) {
*extra = pcre_study(re, soptions, &error);
if (error != NULL) {
zend_error(E_WARNING, "Error while studying pattern");
}
}
1999-07-05 15:25:51 +00:00
*preg_options = poptions;
efree(pattern);
/* Store the compiled pattern and extra info in the cache. */
new_entry.re = re;
new_entry.extra = *extra;
1999-07-05 15:25:51 +00:00
new_entry.preg_options = poptions;
1999-07-20 20:57:04 +00:00
#if HAVE_SETLOCALE
new_entry.locale = locale;
new_entry.tables = tables;
#endif
zend_hash_update(&PCRE_G(pcre_cache), regex, regex_len+1, (void *)&new_entry,
sizeof(pcre_cache_entry), NULL);
return re;
}
/* }}} */
/* {{{ php_pcre_match
*/
static void php_pcre_match(INTERNAL_FUNCTION_PARAMETERS, int global)
{
zval **regex, /* Regular expression */
**subject, /* String to match against */
**subpats = NULL, /* Array for subpatterns */
**subpats_order, /* Order of the results in the subpatterns
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
array for global match */
*result_set, /* Holds a set of subpatterns after
a global match */
**match_sets = NULL; /* An array of sets of matches for each
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
subpattern after a global match */
1999-05-22 19:46:27 +00:00
pcre *re = NULL; /* Compiled regular expression */
pcre_extra *extra = NULL; /* Holds results of studying */
int exoptions = 0; /* Execution options */
int preg_options = 0; /* Custom preg options */
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
int count = 0; /* Count of matched subpatterns */
1999-05-22 19:46:27 +00:00
int *offsets; /* Array of subpattern offsets */
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
int num_subpats; /* Number of captured subpatterns */
1999-05-22 19:46:27 +00:00
int size_offsets; /* Size of the offsets array */
2002-04-19 17:23:22 +00:00
int start_offset; /* Where the new search starts */
1999-05-22 19:46:27 +00:00
int matched; /* Has anything matched */
int i;
int subpats_order_val = 0; /* Integer value of subpats_order */
int g_notempty = 0; /* If the match should not be empty */
1999-05-22 19:46:27 +00:00
const char **stringlist; /* Used to hold list of subpatterns */
char *match; /* The current match */
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
/* Get function parameters and do error-checking. */
switch(ZEND_NUM_ARGS()) {
case 2:
2000-02-24 14:37:05 +00:00
if (global || zend_get_parameters_ex(2, &regex, &subject) == FAILURE) {
WRONG_PARAM_COUNT;
}
break;
case 3:
1999-12-18 22:40:35 +00:00
if (zend_get_parameters_ex(3, &regex, &subject, &subpats) == FAILURE) {
WRONG_PARAM_COUNT;
}
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
if (global)
subpats_order_val = PREG_PATTERN_ORDER;
break;
case 4:
1999-12-18 22:40:35 +00:00
if (zend_get_parameters_ex(4, &regex, &subject, &subpats, &subpats_order) == FAILURE) {
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
WRONG_PARAM_COUNT;
}
/* Make sure subpats_order is a number */
convert_to_long_ex(subpats_order);
2001-01-22 19:29:44 +00:00
subpats_order_val = Z_LVAL_PP(subpats_order);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
if (subpats_order_val < PREG_PATTERN_ORDER ||
subpats_order_val > PREG_SET_ORDER) {
zend_error(E_WARNING, "Wrong value for parameter 4 in call to preg_match_all()");
}
break;
default:
WRONG_PARAM_COUNT;
}
/* Make sure we're dealing with strings. */
convert_to_string_ex(regex);
convert_to_string_ex(subject);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
/* Make sure to clean up the passed array and initialize it. */
if (subpats != NULL) {
zval_dtor(*subpats);
array_init(*subpats);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
}
/* Compile regex or get it from cache. */
if ((re = pcre_get_compiled_regex(Z_STRVAL_PP(regex), &extra, &preg_options)) == NULL) {
RETURN_FALSE;
}
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
/* Calculate the size of the offsets array, and allocate memory for it. */
2001-11-10 23:45:00 +00:00
pcre_fullinfo(re, extra, PCRE_INFO_CAPTURECOUNT, &num_subpats);
num_subpats++;
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
size_offsets = num_subpats * 3;
offsets = (int *)emalloc(size_offsets * sizeof(int));
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
/* Allocate match sets array and initialize the values */
if (global && subpats_order_val == PREG_PATTERN_ORDER) {
match_sets = (zval **)emalloc(num_subpats * sizeof(zval *));
for (i=0; i<num_subpats; i++) {
1999-12-26 21:21:33 +00:00
ALLOC_ZVAL(match_sets[i]);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
array_init(match_sets[i]);
1999-07-09 20:45:55 +00:00
INIT_PZVAL(match_sets[i]);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
}
}
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
/* Start from the beginning of the string */
start_offset = 0;
match = NULL;
matched = 0;
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
do {
/* Execute the regular expression. */
2001-01-22 19:29:44 +00:00
count = pcre_exec(re, extra, Z_STRVAL_PP(subject),
Z_STRLEN_PP(subject), start_offset,
exoptions|g_notempty, offsets, size_offsets);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
/* Check for too many substrings condition. */
if (count == 0) {
zend_error(E_NOTICE, "Matched, but too many substrings\n");
count = size_offsets/3;
}
/* If something has matched */
if (count >= 0) {
matched++;
2001-01-22 19:29:44 +00:00
match = Z_STRVAL_PP(subject) + offsets[0];
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
2000-06-30 19:40:42 +00:00
/* If subpatterns array has been passed, fill it in with values. */
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
if (subpats != NULL) {
/* Try to get the list of substrings and display a warning if failed. */
2001-01-22 19:29:44 +00:00
if (pcre_get_substring_list(Z_STRVAL_PP(subject),
offsets, count, &stringlist) < 0) {
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
efree(offsets);
efree(re);
zend_error(E_WARNING, "Get subpatterns list failed");
return;
}
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
if (global) { /* global pattern matching */
if (subpats_order_val == PREG_PATTERN_ORDER) {
2001-11-10 23:45:00 +00:00
/* For each subpattern, insert it into the appropriate array. */
for (i = 0; i < count; i++) {
add_next_index_stringl(match_sets[i], (char *)stringlist[i],
offsets[(i<<1)+1] - offsets[i<<1], 1);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
}
2001-11-10 23:45:00 +00:00
/*
* If the number of captured subpatterns on this run is
* less than the total possible number, pad the result
* arrays with empty strings.
*/
if (count < num_subpats) {
for (; i < num_subpats; i++) {
add_next_index_string(match_sets[i], empty_string, 1);
}
}
} else {
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
/* Allocate the result set array */
1999-12-26 21:21:33 +00:00
ALLOC_ZVAL(result_set);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
array_init(result_set);
1999-07-09 20:45:55 +00:00
INIT_PZVAL(result_set);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
/* Add all the subpatterns to it */
2001-11-10 23:45:00 +00:00
for (i = 0; i < count; i++) {
add_next_index_stringl(result_set, (char *)stringlist[i],
offsets[(i<<1)+1] - offsets[i<<1], 1);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
}
/* And add it to the output array */
2001-01-22 19:29:44 +00:00
zend_hash_next_index_insert(Z_ARRVAL_PP(subpats), &result_set,
2001-11-10 23:45:00 +00:00
sizeof(zval *), NULL);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
}
}
else { /* single pattern matching */
/* For each subpattern, insert it into the subpatterns array. */
2001-11-10 23:45:00 +00:00
for (i = 0; i < count; i++) {
add_next_index_stringl((*subpats), (char *)stringlist[i],
offsets[(i<<1)+1] - offsets[i<<1], 1);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
}
}
1999-07-06 19:49:48 +00:00
php_pcre_free((void *) stringlist);
}
}
else { /* Failed to match */
/* If we previously set PCRE_NOTEMPTY after a null match,
this is not necessarily the end. We need to advance
the start offset, and continue. Fudge the offset values
to achieve this, unless we're already at the end of the string. */
2002-04-19 17:23:22 +00:00
if (g_notempty != 0 && start_offset < Z_STRLEN_PP(subject)) {
offsets[0] = start_offset;
offsets[1] = start_offset + 1;
} else
break;
}
/* If we have matched an empty string, mimic what Perl's /g options does.
This turns out to be rather cunning. First we set PCRE_NOTEMPTY and try
the match again at the same point. If this fails (picked up above) we
advance to the next character. */
g_notempty = (offsets[1] == offsets[0])? PCRE_NOTEMPTY | PCRE_ANCHORED : 0;
/* Advance to the position right after the last full match */
start_offset = offsets[1];
} while (global);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
/* Add the match sets to the output array and clean up */
if (global && subpats_order_val == PREG_PATTERN_ORDER) {
for (i=0; i<num_subpats; i++) {
2001-01-22 19:29:44 +00:00
zend_hash_next_index_insert(Z_ARRVAL_PP(subpats), &match_sets[i], sizeof(zval *), NULL);
}
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
efree(match_sets);
}
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
efree(offsets);
RETVAL_LONG(matched);
}
/* }}} */
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
2000-02-23 16:38:38 +00:00
/* {{{ proto int preg_match(string pattern, string subject [, array subpatterns])
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
Perform a Perl-style regular expression match */
PHP_FUNCTION(preg_match)
{
php_pcre_match(INTERNAL_FUNCTION_PARAM_PASSTHRU, 0);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
}
/* }}} */
2000-02-23 16:38:38 +00:00
/* {{{ proto int preg_match_all(string pattern, string subject, array subpatterns [, int order])
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
Perform a Perl-style global regular expression match */
PHP_FUNCTION(preg_match_all)
{
php_pcre_match(INTERNAL_FUNCTION_PARAM_PASSTHRU, 1);
A few changes here. First of all, as per extensive discussion on the list, the functions are now prefixed with "preg" instead of "pcre". Secondly, global matching is now possible using preg_match_all. Please, give suggestions on a better name if this one doesn't sit well with you. Possible names are preg_global_match and preg_gmatch. preg_match_all takes 4 arguments: a regex pattern, a subject string, the array for capturing subpatterns, and a parameter that tells how the results in the subpatterns array are arranged. Basically, preg_match_all will go through the subject string and try to capture all the matches that it finds, not just the first one like preg_match. 4th parameter can be PREG_PATTERN_ORDER (default) or PREG_SET_ORDER. Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_PATTERN_ORDER); This returns results so that $out[0] is an array of full pattern matches, $out[1] is an array of first captured subpattern matches, and so on. $out[0] -> ("<div align=left>", "</div>") $out[1] -> ("div align=left", "div") Example: preg_match_all("|</?([^>]+)>|", "<div align=left>a test</div>", $out, PREG_SET_ORDER); This returns results so that $out[0] is an array of first full pattern match and subpatterns, $out[1] is an array of second full pattern match and subpatterns. $out[0] -> ("<div align=left>", "div align=left") $out[1] -> ("</div>", "div") If anyone has a better name for these PREG_ constants and also which one should be the default, I'd like to hear it.
1999-05-26 15:22:02 +00:00
}
/* }}} */
/* {{{ preg_get_backref
*/
static inline int preg_get_backref(const char *walk, int *backref)
{
1999-06-09 15:02:57 +00:00
if (*walk && *walk >= '0' && *walk <= '9')
*backref = *walk - '0';
1999-06-09 15:02:57 +00:00
else
return 0;
1999-06-09 15:02:57 +00:00
if (walk[1] && walk[1] >= '0' && walk[1] <= '9')
*backref = *backref * 10 + walk[1] - '0';
1999-06-09 15:02:57 +00:00
return 1;
}
/* }}} */
/* {{{ preg_do_repl_func
*/
static int preg_do_repl_func(zval *function, char *subject, int *offsets, int count, char **result)
{
zval *retval_ptr; /* Function return value */
2000-11-24 00:12:19 +00:00
zval **args[1]; /* Argument to pass to function */
zval *subpats; /* Captured subpatterns */
int result_len; /* Return value length */
int i;
TSRMLS_FETCH();
MAKE_STD_ZVAL(subpats);
array_init(subpats);
for (i = 0; i < count; i++)
add_next_index_stringl(subpats, &subject[offsets[i<<1]], offsets[(i<<1)+1] - offsets[i<<1], 1);
args[0] = &subpats;
2001-07-30 08:24:42 +00:00
if (call_user_function_ex(EG(function_table), NULL, function, &retval_ptr, 1, args, 0, NULL TSRMLS_CC) == SUCCESS && retval_ptr) {
convert_to_string_ex(&retval_ptr);
*result = estrndup(Z_STRVAL_P(retval_ptr), Z_STRLEN_P(retval_ptr));
result_len = Z_STRLEN_P(retval_ptr);
zval_ptr_dtor(&retval_ptr);
} else {
php_error(E_WARNING, "Unable to call custom replacement function");
result_len = offsets[1] - offsets[0];
*result = estrndup(&subject[offsets[0]], result_len);
}
zval_dtor(subpats);
FREE_ZVAL(subpats);
return result_len;
}
/* }}} */
/* {{{ preg_do_eval
*/
static int preg_do_eval(char *eval_str, int eval_str_len, char *subject,
2001-08-06 03:50:52 +00:00
int *offsets, int count, char **result TSRMLS_DC)
{
zval retval; /* Return value from evaluation */
char *eval_str_end, /* End of eval string */
*match, /* Current match for a backref */
*esc_match, /* Quote-escaped match */
*walk, /* Used to walk the code string */
*segment, /* Start of segment to append while walking */
walk_last; /* Last walked character */
int match_len; /* Length of the match */
int esc_match_len; /* Length of the quote-escaped match */
int result_len; /* Length of the result of the evaluation */
int backref; /* Current backref */
char *compiled_string_description;
smart_str code = {0};
eval_str_end = eval_str + eval_str_len;
walk = segment = eval_str;
walk_last = 0;
while (walk < eval_str_end) {
/* If found a backreference.. */
if ('\\' == *walk || '$' == *walk) {
smart_str_appendl(&code, segment, walk - segment);
if (walk_last == '\\') {
code.c[code.len-1] = *walk++;
segment = walk;
walk_last = 0;
continue;
}
segment = walk;
if (preg_get_backref(walk+1, &backref)) {
if (backref < count) {
/* Find the corresponding string match and substitute it
in instead of the backref */
match = subject + offsets[backref<<1];
match_len = offsets[(backref<<1)+1] - offsets[backref<<1];
if (match_len)
2001-08-06 03:50:52 +00:00
esc_match = php_addslashes(match, match_len, &esc_match_len, 0 TSRMLS_CC);
else {
esc_match = match;
esc_match_len = 0;
}
} else {
esc_match = empty_string;
esc_match_len = 0;
match_len = 0;
}
smart_str_appendl(&code, esc_match, esc_match_len);
/* Adjust the walk pointer */
walk += (backref > 9 ? 3 : 2);
segment = walk;
/* Clean up and reassign */
if (esc_match_len)
efree(esc_match);
continue;
}
}
walk++;
walk_last = walk[-1];
}
smart_str_appendl(&code, segment, walk - segment);
smart_str_0(&code);
1999-07-08 21:42:29 +00:00
compiled_string_description = zend_make_compiled_string_description("regexp code" TSRMLS_CC);
/* Run the code */
if (zend_eval_string(code.c, &retval, compiled_string_description TSRMLS_CC) == FAILURE) {
efree(compiled_string_description);
zend_error(E_ERROR, "Failed evaluating code:\n%s\n", code);
/* zend_error() does not return in this case */
}
efree(compiled_string_description);
convert_to_string(&retval);
/* Save the return value and its length */
2001-01-22 19:29:44 +00:00
*result = estrndup(Z_STRVAL(retval), Z_STRLEN(retval));
result_len = Z_STRLEN(retval);
/* Clean up */
zval_dtor(&retval);
smart_str_free(&code);
return result_len;
}
/* }}} */
/* {{{ php_pcre_replace
*/
PHPAPI char *php_pcre_replace(char *regex, int regex_len,
char *subject, int subject_len,
zval *replace_val, int is_callable_replace,
int *result_len, int limit TSRMLS_DC)
{
1999-05-22 19:46:27 +00:00
pcre *re = NULL; /* Compiled regular expression */
pcre_extra *extra = NULL; /* Holds results of studying */
int exoptions = 0; /* Execution options */
int preg_options = 0; /* Custom preg options */
1999-05-22 19:46:27 +00:00
int count = 0; /* Count of matched subpatterns */
int *offsets; /* Array of subpattern offsets */
int size_offsets; /* Size of the offsets array */
int new_len; /* Length of needed storage */
int alloc_len; /* Actual allocated length */
int eval_result_len=0; /* Length of the eval'ed or
function-returned string */
int match_len; /* Length of the current match */
1999-05-22 19:46:27 +00:00
int backref; /* Backreference number */
int eval; /* If the replacement string should be eval'ed */
2002-04-19 17:23:22 +00:00
int start_offset; /* Where the new search starts */
int g_notempty = 0; /* If the match should not be empty */
int replace_len; /* Length of replacement string */
1999-05-22 19:46:27 +00:00
char *result, /* Result of replacement */
*replace, /* Replacement string */
1999-05-22 19:46:27 +00:00
*new_buf, /* Temporary buffer for re-allocation */
*walkbuf, /* Location of current replacement in the result */
*walk, /* Used to walk the replacement string */
*match, /* The current match */
*piece, /* The current piece of subject */
*replace_end, /* End of replacement string */
*eval_result, /* Result of eval or custom function */
walk_last; /* Last walked character */
/* Compile regex or get it from cache. */
if ((re = pcre_get_compiled_regex(regex, &extra, &preg_options)) == NULL) {
return NULL;
}
eval = preg_options & PREG_REPLACE_EVAL;
if (is_callable_replace) {
if (eval) {
php_error(E_WARNING, "/e modifier cannot be used with replacement callback");
return NULL;
}
} else {
replace = Z_STRVAL_P(replace_val);
replace_len = Z_STRLEN_P(replace_val);
replace_end = replace + replace_len;
}
/* Calculate the size of the offsets array, and allocate memory for it. */
size_offsets = (pcre_info(re, NULL, NULL) + 1) * 3;
offsets = (int *)emalloc(size_offsets * sizeof(int));
alloc_len = 2 * subject_len + 1;
result = emalloc(alloc_len * sizeof(char));
/* Initialize */
match = NULL;
*result_len = 0;
start_offset = 0;
while (1) {
/* Execute the regular expression. */
count = pcre_exec(re, extra, subject, subject_len, start_offset,
exoptions|g_notempty, offsets, size_offsets);
/* Check for too many substrings condition. */
if (count == 0) {
zend_error(E_NOTICE, "Matched, but too many substrings\n");
count = size_offsets/3;
}
piece = subject + start_offset;
if (count > 0 && (limit == -1 || limit > 0)) {
/* Set the match location in subject */
match = subject + offsets[0];
new_len = *result_len + offsets[0] - start_offset; /* part before the match */
/* If evaluating, do it and add the return string's length */
if (eval) {
eval_result_len = preg_do_eval(replace, replace_len, subject,
2001-08-06 03:50:52 +00:00
offsets, count, &eval_result TSRMLS_CC);
new_len += eval_result_len;
} else if (is_callable_replace) {
/* Use custom function to get replacement string and its length. */
eval_result_len = preg_do_repl_func(replace_val, subject, offsets,
count, &eval_result);
new_len += eval_result_len;
} else { /* do regular substitution */
walk = replace;
walk_last = 0;
while (walk < replace_end) {
if ('\\' == *walk || '$' == *walk) {
if (walk_last == '\\') {
walk++;
walk_last = 0;
continue;
}
if (preg_get_backref(walk+1, &backref)) {
if (backref < count)
new_len += offsets[(backref<<1)+1] - offsets[backref<<1];
walk += (backref > 9) ? 3 : 2;
continue;
}
}
new_len++;
walk++;
walk_last = walk[-1];
}
}
if (new_len + 1 > alloc_len) {
alloc_len = 1 + alloc_len + 2 * new_len;
new_buf = emalloc(alloc_len);
memcpy(new_buf, result, *result_len);
efree(result);
result = new_buf;
}
/* copy the part of the string before the match */
memcpy(&result[*result_len], piece, match-piece);
*result_len += match-piece;
/* copy replacement and backrefs */
walkbuf = result + *result_len;
/* If evaluating or using custom function, copy result to the buffer
* and clean up. */
if (eval || is_callable_replace) {
memcpy(walkbuf, eval_result, eval_result_len);
*result_len += eval_result_len;
STR_FREE(eval_result);
} else { /* do regular backreference copying */
walk = replace;
walk_last = 0;
while (walk < replace_end) {
if ('\\' == *walk || '$' == *walk) {
if (walk_last == '\\') {
*(walkbuf-1) = *walk++;
walk_last = 0;
continue;
}
if (preg_get_backref(walk+1, &backref)) {
if (backref < count) {
match_len = offsets[(backref<<1)+1] - offsets[backref<<1];
memcpy(walkbuf, subject + offsets[backref<<1], match_len);
walkbuf += match_len;
}
walk += (backref > 9) ? 3 : 2;
continue;
}
}
*walkbuf++ = *walk++;
walk_last = walk[-1];
}
2000-06-10 19:43:20 +00:00
*walkbuf = '\0';
/* increment the result length by how much we've added to the string */
*result_len += walkbuf - (result + *result_len);
}
if (limit != -1)
limit--;
} else { /* Failed to match */
/* If we previously set PCRE_NOTEMPTY after a null match,
this is not necessarily the end. We need to advance
the start offset, and continue. Fudge the offset values
to achieve this, unless we're already at the end of the string. */
2002-04-19 17:23:22 +00:00
if (g_notempty != 0 && start_offset < subject_len) {
offsets[0] = start_offset;
offsets[1] = start_offset + 1;
memcpy(&result[*result_len], piece, 1);
(*result_len)++;
} else {
new_len = *result_len + subject_len - start_offset;
if (new_len + 1 > alloc_len) {
alloc_len = new_len + 1; /* now we know exactly how long it is */
new_buf = emalloc(alloc_len * sizeof(char));
memcpy(new_buf, result, *result_len);
efree(result);
result = new_buf;
}
/* stick that last bit of string on our output */
memcpy(&result[*result_len], piece, subject_len - start_offset);
*result_len += subject_len - start_offset;
2000-06-10 19:44:34 +00:00
result[*result_len] = '\0';
break;
}
}
/* If we have matched an empty string, mimic what Perl's /g options does.
This turns out to be rather cunning. First we set PCRE_NOTEMPTY and try
the match again at the same point. If this fails (picked up above) we
advance to the next character. */
g_notempty = (offsets[1] == offsets[0])? PCRE_NOTEMPTY | PCRE_ANCHORED : 0;
/* Advance to the next piece. */
start_offset = offsets[1];
}
efree(offsets);
return result;
}
/* }}} */
/* {{{ php_replace_in_subject
*/
2001-08-06 03:50:52 +00:00
static char *php_replace_in_subject(zval *regex, zval *replace, zval **subject, int *result_len, int limit, zend_bool is_callable_replace TSRMLS_DC)
{
zval **regex_entry,
**replace_entry = NULL,
*replace_value,
empty_replace;
char *subject_value,
*result;
int subject_len;
1999-05-22 19:46:27 +00:00
/* Make sure we're dealing with strings. */
convert_to_string_ex(subject);
ZVAL_STRINGL(&empty_replace, empty_string, 0, 0);
1999-05-22 19:46:27 +00:00
/* If regex is an array */
2001-01-22 19:29:44 +00:00
if (Z_TYPE_P(regex) == IS_ARRAY) {
/* Duplicate subject string for repeated replacement */
2001-01-22 19:29:44 +00:00
subject_value = estrndup(Z_STRVAL_PP(subject), Z_STRLEN_PP(subject));
subject_len = Z_STRLEN_PP(subject);
*result_len = subject_len;
1999-05-22 19:46:27 +00:00
2001-01-22 19:29:44 +00:00
zend_hash_internal_pointer_reset(Z_ARRVAL_P(regex));
replace_value = replace;
if (Z_TYPE_P(replace) == IS_ARRAY && !is_callable_replace)
2001-01-22 19:29:44 +00:00
zend_hash_internal_pointer_reset(Z_ARRVAL_P(replace));
1999-05-22 19:46:27 +00:00
/* For each entry in the regex array, get the entry */
while (zend_hash_get_current_data(Z_ARRVAL_P(regex), (void **)&regex_entry) == SUCCESS) {
1999-05-22 19:46:27 +00:00
/* Make sure we're dealing with strings. */
convert_to_string_ex(regex_entry);
/* If replace is an array and not a callable construct */
if (Z_TYPE_P(replace) == IS_ARRAY && !is_callable_replace) {
1999-05-22 19:46:27 +00:00
/* Get current entry */
2001-01-22 19:29:44 +00:00
if (zend_hash_get_current_data(Z_ARRVAL_P(replace), (void **)&replace_entry) == SUCCESS) {
if (!is_callable_replace) {
convert_to_string_ex(replace_entry);
}
replace_value = *replace_entry;
2001-01-22 19:29:44 +00:00
zend_hash_move_forward(Z_ARRVAL_P(replace));
} else {
1999-05-22 19:46:27 +00:00
/* We've run out of replacement strings, so use an empty one */
replace_value = &empty_replace;
}
}
1999-05-22 19:46:27 +00:00
/* Do the actual replacement and put the result back into subject_value
for further replacements. */
2001-01-22 19:29:44 +00:00
if ((result = php_pcre_replace(Z_STRVAL_PP(regex_entry),
Z_STRLEN_PP(regex_entry),
subject_value,
subject_len,
replace_value,
is_callable_replace,
result_len,
2001-08-06 03:50:52 +00:00
limit TSRMLS_CC)) != NULL) {
efree(subject_value);
subject_value = result;
subject_len = *result_len;
}
2001-01-22 19:29:44 +00:00
zend_hash_move_forward(Z_ARRVAL_P(regex));
}
return subject_value;
} else {
2001-01-22 19:29:44 +00:00
result = php_pcre_replace(Z_STRVAL_P(regex),
Z_STRLEN_P(regex),
Z_STRVAL_PP(subject),
Z_STRLEN_PP(subject),
replace,
is_callable_replace,
result_len,
2001-08-06 03:50:52 +00:00
limit TSRMLS_CC);
return result;
}
}
/* }}} */
/* {{{ preg_replace_impl
*/
2001-03-13 16:39:47 +00:00
static void preg_replace_impl(INTERNAL_FUNCTION_PARAMETERS, zend_bool is_callable_replace)
{
zval **regex,
**replace,
**subject,
**limit,
**subject_entry;
char *result;
int result_len;
int limit_val = -1;
char *string_key;
ulong num_key;
2001-05-12 00:45:55 +00:00
char *callback_name = NULL;
/* Get function parameters and do error-checking. */
if (ZEND_NUM_ARGS() < 3 || ZEND_NUM_ARGS() > 4 ||
zend_get_parameters_ex(ZEND_NUM_ARGS(), &regex, &replace, &subject, &limit) == FAILURE) {
WRONG_PARAM_COUNT;
}
SEPARATE_ZVAL(replace);
if (Z_TYPE_PP(replace) != IS_ARRAY)
convert_to_string_ex(replace);
if (is_callable_replace) {
if (!zend_is_callable(*replace, 0, &callback_name)) {
php_error(E_WARNING, "%s() requires argument 2, '%s', to be a valid callback",
2001-07-30 08:24:42 +00:00
get_active_function_name(TSRMLS_C), callback_name);
efree(callback_name);
*return_value = **subject;
zval_copy_ctor(return_value);
return;
}
efree(callback_name);
}
SEPARATE_ZVAL(regex);
SEPARATE_ZVAL(subject);
if (ZEND_NUM_ARGS() > 3) {
convert_to_long_ex(limit);
limit_val = Z_LVAL_PP(limit);
}
if (Z_TYPE_PP(regex) != IS_ARRAY)
convert_to_string_ex(regex);
1999-05-22 19:46:27 +00:00
/* if subject is an array */
2001-01-22 19:29:44 +00:00
if (Z_TYPE_PP(subject) == IS_ARRAY) {
array_init(return_value);
2001-01-22 19:29:44 +00:00
zend_hash_internal_pointer_reset(Z_ARRVAL_PP(subject));
1999-05-22 19:46:27 +00:00
/* For each subject entry, convert it to string, then perform replacement
and add the result to the return_value array. */
while (zend_hash_get_current_data(Z_ARRVAL_PP(subject), (void **)&subject_entry) == SUCCESS) {
2001-08-06 03:50:52 +00:00
if ((result = php_replace_in_subject(*regex, *replace, subject_entry, &result_len, limit_val, is_callable_replace TSRMLS_CC)) != NULL) {
/* Add to return array */
2001-01-22 19:29:44 +00:00
switch(zend_hash_get_current_key(Z_ARRVAL_PP(subject), &string_key, &num_key, 0))
{
case HASH_KEY_IS_STRING:
add_assoc_stringl(return_value, string_key, result, result_len, 0);
break;
case HASH_KEY_IS_LONG:
add_index_stringl(return_value, num_key, result, result_len, 0);
break;
}
}
1999-05-22 19:46:27 +00:00
2001-01-22 19:29:44 +00:00
zend_hash_move_forward(Z_ARRVAL_PP(subject));
}
}
1999-05-22 19:46:27 +00:00
else { /* if subject is not an array */
2001-08-06 03:50:52 +00:00
if ((result = php_replace_in_subject(*regex, *replace, subject, &result_len, limit_val, is_callable_replace TSRMLS_CC)) != NULL) {
RETVAL_STRINGL(result, result_len, 0);
}
1999-05-22 19:46:27 +00:00
}
}
/* }}} */
/* {{{ proto string preg_replace(mixed regex, mixed replace, mixed subject [, int limit])
Perform Perl-style regular expression replacement. */
PHP_FUNCTION(preg_replace)
{
preg_replace_impl(INTERNAL_FUNCTION_PARAM_PASSTHRU, 0);
}
/* }}} */
/* {{{ proto string preg_replace_callback(mixed regex, mixed callback, mixed subject [, int limit])
Perform Perl-style regular expression replacement using replacement callback. */
PHP_FUNCTION(preg_replace_callback)
{
preg_replace_impl(INTERNAL_FUNCTION_PARAM_PASSTHRU, 1);
}
/* }}} */
/* {{{ proto array preg_split(string pattern, string subject [, int limit [, int flags]])
2000-02-23 16:38:38 +00:00
Split string into an array using a perl-style regular expression as a delimiter */
PHP_FUNCTION(preg_split)
{
zval **regex, /* Regular expression to split by */
**subject, /* Subject string to split */
**limit, /* Number of pieces to return */
**flags;
pcre *re = NULL; /* Compiled regular expression */
pcre_extra *extra = NULL; /* Holds results of studying */
int *offsets; /* Array of subpattern offsets */
int size_offsets; /* Size of the offsets array */
int exoptions = 0; /* Execution options */
int preg_options = 0; /* Custom preg options */
int argc; /* Argument count */
int limit_val = -1; /* Integer value of limit */
int no_empty = 0; /* If NO_EMPTY flag is set */
int delim_capture = 0; /* If delimiters should be captured */
int count = 0; /* Count of matched subpatterns */
2002-04-19 17:23:22 +00:00
int start_offset; /* Where the new search starts */
int g_notempty = 0; /* If the match should not be empty */
1999-05-29 20:45:00 +00:00
char *match, /* The current match */
*last_match; /* Location of last match */
/* Get function parameters and do error checking */
argc = ZEND_NUM_ARGS();
2000-12-27 13:59:40 +00:00
if (argc < 2 || argc > 4 || zend_get_parameters_ex(argc, &regex, &subject, &limit, &flags) == FAILURE) {
WRONG_PARAM_COUNT;
}
if (argc > 2) {
convert_to_long_ex(limit);
2001-01-22 19:29:44 +00:00
limit_val = Z_LVAL_PP(limit);
if (limit_val == 0)
limit_val = -1;
if (argc > 3) {
convert_to_long_ex(flags);
no_empty = Z_LVAL_PP(flags) & PREG_SPLIT_NO_EMPTY;
delim_capture = Z_LVAL_PP(flags) & PREG_SPLIT_DELIM_CAPTURE;
}
}
/* Make sure we're dealing with strings */
convert_to_string_ex(regex);
convert_to_string_ex(subject);
/* Compile regex or get it from cache. */
if ((re = pcre_get_compiled_regex(Z_STRVAL_PP(regex), &extra, &preg_options)) == NULL) {
RETURN_FALSE;
}
/* Initialize return value */
array_init(return_value);
/* Calculate the size of the offsets array, and allocate memory for it. */
size_offsets = (pcre_info(re, NULL, NULL) + 1) * 3;
offsets = (int *)emalloc(size_offsets * sizeof(int));
/* Start at the beginning of the string */
start_offset = 0;
2001-01-22 19:29:44 +00:00
last_match = Z_STRVAL_PP(subject);
1999-05-29 20:45:00 +00:00
match = NULL;
/* Get next piece if no limit or limit not yet reached and something matched*/
while ((limit_val == -1 || limit_val > 1)) {
2001-01-22 19:29:44 +00:00
count = pcre_exec(re, extra, Z_STRVAL_PP(subject),
Z_STRLEN_PP(subject), start_offset,
exoptions|g_notempty, offsets, size_offsets);
/* Check for too many substrings condition. */
if (count == 0) {
zend_error(E_NOTICE, "Matched, but too many substrings\n");
count = size_offsets/3;
}
/* If something matched */
if (count > 0) {
2001-01-22 19:29:44 +00:00
match = Z_STRVAL_PP(subject) + offsets[0];
if (!no_empty || &Z_STRVAL_PP(subject)[offsets[0]] != last_match) {
/* Add the piece to the return value */
add_next_index_stringl(return_value, last_match,
2001-01-22 19:29:44 +00:00
&Z_STRVAL_PP(subject)[offsets[0]]-last_match, 1);
/* One less left to do */
if (limit_val != -1)
limit_val--;
}
2001-01-22 19:29:44 +00:00
last_match = &Z_STRVAL_PP(subject)[offsets[1]];
if (delim_capture) {
int i, match_len;
for (i = 1; i < count; i++) {
match_len = offsets[(i<<1)+1] - offsets[i<<1];
if (!no_empty || match_len > 0)
add_next_index_stringl(return_value,
&Z_STRVAL_PP(subject)[offsets[i<<1]],
match_len, 1);
}
}
} else { /* Failed to match */
/* If we previously set PCRE_NOTEMPTY after a null match,
this is not necessarily the end. We need to advance
the start offset, and continue. Fudge the offset values
to achieve this, unless we're already at the end of the string. */
2002-04-19 17:23:22 +00:00
if (g_notempty != 0 && start_offset < Z_STRLEN_PP(subject)) {
offsets[0] = start_offset;
offsets[1] = start_offset + 1;
} else
break;
}
/* If we have matched an empty string, mimic what Perl's /g options does.
This turns out to be rather cunning. First we set PCRE_NOTEMPTY and try
the match again at the same point. If this fails (picked up above) we
advance to the next character. */
g_notempty = (offsets[1] == offsets[0])? PCRE_NOTEMPTY | PCRE_ANCHORED : 0;
/* Advance to the position right after the last full match */
start_offset = offsets[1];
}
2002-04-19 17:23:22 +00:00
if (!no_empty || start_offset != Z_STRLEN_PP(subject))
/* Add the last piece to the return value */
add_next_index_string(return_value,
2001-01-22 19:29:44 +00:00
&Z_STRVAL_PP(subject)[start_offset], 1);
/* Clean up */
efree(offsets);
}
/* }}} */
2000-05-25 21:07:07 +00:00
/* {{{ proto string preg_quote(string str, string delim_char)
Quote regular expression characters plus an optional character */
1999-06-04 13:56:23 +00:00
PHP_FUNCTION(preg_quote)
{
zval **in_str_arg; /* Input string argument */
zval **delim; /* Additional delimiter argument */
1999-06-04 13:56:23 +00:00
char *in_str, /* Input string */
*in_str_end, /* End of the input string */
1999-06-04 13:56:23 +00:00
*out_str, /* Output string with quoted characters */
*p, /* Iterator for input string */
*q, /* Iterator for output string */
2000-11-27 13:28:26 +00:00
delim_char=0, /* Delimiter character to be quoted */
1999-06-04 13:56:23 +00:00
c; /* Current character */
zend_bool quote_delim = 0; /* Whether to quote additional delim char */
1999-06-04 13:56:23 +00:00
/* Get the arguments and check for errors */
if (ZEND_NUM_ARGS() < 1 || ZEND_NUM_ARGS() > 2 ||
zend_get_parameters_ex(ZEND_NUM_ARGS(), &in_str_arg, &delim) == FAILURE) {
1999-06-04 13:56:23 +00:00
WRONG_PARAM_COUNT;
}
/* Make sure we're working with strings */
convert_to_string_ex(in_str_arg);
2001-01-22 19:29:44 +00:00
in_str = Z_STRVAL_PP(in_str_arg);
in_str_end = Z_STRVAL_PP(in_str_arg) + Z_STRLEN_PP(in_str_arg);
1999-06-04 13:56:23 +00:00
/* Nothing to do if we got an empty string */
if (in_str == in_str_end) {
RETVAL_STRINGL(empty_string, 0, 0);
1999-06-04 13:56:23 +00:00
}
if (ZEND_NUM_ARGS() == 2) {
convert_to_string_ex(delim);
if (Z_STRLEN_PP(delim) > 0) {
delim_char = Z_STRVAL_PP(delim)[0];
quote_delim = 1;
}
}
1999-06-04 13:56:23 +00:00
/* Allocate enough memory so that even if each character
is quoted, we won't run out of room */
2001-01-22 19:29:44 +00:00
out_str = emalloc(2 * Z_STRLEN_PP(in_str_arg) + 1);
1999-06-04 13:56:23 +00:00
/* Go through the string and quote necessary characters */
for(p = in_str, q = out_str; p != in_str_end; p++) {
c = *p;
1999-06-04 13:56:23 +00:00
switch(c) {
case '.':
case '\\':
case '+':
case '*':
case '?':
case '[':
case '^':
case ']':
case '$':
case '(':
case ')':
case '{':
case '}':
case '=':
case '!':
case '>':
case '<':
case '|':
case ':':
*q++ = '\\';
*q++ = c;
break;
1999-06-04 13:56:23 +00:00
default:
if (quote_delim && c == delim_char)
*q++ = '\\';
1999-06-04 13:56:23 +00:00
*q++ = c;
break;
1999-06-04 13:56:23 +00:00
}
}
*q = '\0';
/* Reallocate string and return it */
RETVAL_STRINGL(erealloc(out_str, q - out_str + 1), q - out_str, 0);
1999-06-04 13:56:23 +00:00
}
/* }}} */
1999-06-10 14:41:38 +00:00
/* {{{ proto array preg_grep(string regex, array input)
Searches array and returns entries which match regex */
PHP_FUNCTION(preg_grep)
{
zval **regex, /* Regular expression */
**input, /* Input array */
**flags,
**entry; /* An entry in the input array */
1999-06-10 14:41:38 +00:00
pcre *re = NULL; /* Compiled regular expression */
pcre_extra *extra = NULL; /* Holds results of studying */
int preg_options = 0; /* Custom preg options */
1999-06-10 14:41:38 +00:00
int *offsets; /* Array of subpattern offsets */
int size_offsets; /* Size of the offsets array */
int count = 0; /* Count of matched subpatterns */
char *string_key;
ulong num_key;
zend_bool invert = 0; /* Whether to return non-matching
entries */
1999-06-10 14:41:38 +00:00
/* Get arguments and do error checking */
if (ZEND_NUM_ARGS() < 2 || ZEND_NUM_ARGS() > 3 ||
zend_get_parameters_ex(ZEND_NUM_ARGS(), &regex, &input, &flags) == FAILURE) {
1999-06-10 14:41:38 +00:00
WRONG_PARAM_COUNT;
}
2001-01-22 19:29:44 +00:00
if (Z_TYPE_PP(input) != IS_ARRAY) {
1999-06-10 14:41:38 +00:00
zend_error(E_WARNING, "Secong argument to preg_grep() should be an array");
return;
}
SEPARATE_ZVAL(input);
1999-06-10 14:41:38 +00:00
/* Make sure regex is a string */
convert_to_string_ex(regex);
if (ZEND_NUM_ARGS() > 2) {
convert_to_long_ex(flags);
invert = (Z_LVAL_PP(flags) & PREG_GREP_INVERT) ? 1 : 0;
}
1999-06-10 14:41:38 +00:00
/* Compile regex or get it from cache. */
if ((re = pcre_get_compiled_regex(Z_STRVAL_PP(regex), &extra, &preg_options)) == NULL) {
1999-06-10 14:41:38 +00:00
RETURN_FALSE;
}
/* Calculate the size of the offsets array, and allocate memory for it. */
size_offsets = (pcre_info(re, NULL, NULL) + 1) * 3;
offsets = (int *)emalloc(size_offsets * sizeof(int));
/* Initialize return array */
array_init(return_value);
1999-06-10 14:41:38 +00:00
/* Go through the input array */
2001-01-22 19:29:44 +00:00
zend_hash_internal_pointer_reset(Z_ARRVAL_PP(input));
while(zend_hash_get_current_data(Z_ARRVAL_PP(input), (void **)&entry) == SUCCESS) {
convert_to_string_ex(entry);
1999-06-10 14:41:38 +00:00
/* Perform the match */
2001-01-22 19:29:44 +00:00
count = pcre_exec(re, extra, Z_STRVAL_PP(entry),
Z_STRLEN_PP(entry), 0,
0, offsets, size_offsets);
/* Check for too many substrings condition. */
if (count == 0) {
zend_error(E_NOTICE, "Matched, but too many substrings\n");
count = size_offsets/3;
}
/* If the entry fits our requirements */
if ((count > 0 && !invert) ||
(count < 0 && invert)) {
(*entry)->refcount++;
/* Add to return array */
2001-01-22 19:29:44 +00:00
switch(zend_hash_get_current_key(Z_ARRVAL_PP(input), &string_key, &num_key, 0))
{
case HASH_KEY_IS_STRING:
2001-01-22 19:29:44 +00:00
zend_hash_update(Z_ARRVAL_P(return_value), string_key,
strlen(string_key)+1, entry, sizeof(zval *), NULL);
break;
case HASH_KEY_IS_LONG:
2001-01-22 19:29:44 +00:00
zend_hash_index_update(Z_ARRVAL_P(return_value), num_key, entry,
sizeof(zval *), NULL);
break;
1999-06-10 14:41:38 +00:00
}
}
2001-01-22 19:29:44 +00:00
zend_hash_move_forward(Z_ARRVAL_PP(input));
1999-06-10 14:41:38 +00:00
}
/* Clean up */
1999-06-10 14:41:38 +00:00
efree(offsets);
}
/* }}} */
1999-09-03 21:12:10 +00:00
/* {{{ module definition structures */
function_entry pcre_functions[] = {
PHP_FE(preg_match, third_arg_force_ref)
PHP_FE(preg_match_all, third_arg_force_ref)
PHP_FE(preg_replace, NULL)
PHP_FE(preg_replace_callback, NULL)
PHP_FE(preg_split, NULL)
PHP_FE(preg_quote, NULL)
PHP_FE(preg_grep, NULL)
{NULL, NULL, NULL}
1999-09-03 21:12:10 +00:00
};
zend_module_entry pcre_module_entry = {
STANDARD_MODULE_HEADER,
"pcre",
pcre_functions,
PHP_MINIT(pcre),
PHP_MSHUTDOWN(pcre),
PHP_RINIT(pcre),
NULL,
PHP_MINFO(pcre),
NO_VERSION_YET,
STANDARD_MODULE_PROPERTIES
1999-09-03 21:12:10 +00:00
};
2000-05-23 09:33:51 +00:00
#ifdef COMPILE_DL_PCRE
ZEND_GET_MODULE(pcre)
#endif
1999-09-03 21:12:10 +00:00
/* }}} */
#endif /* HAVE_PCRE || HAVE_BUNDLED_PCRE */
/*
* Local variables:
* tab-width: 4
* c-basic-offset: 4
* End:
* vim600: sw=4 ts=4 fdm=marker
* vim<600: sw=4 ts=4
*/