/ext/standard/string.c

Large files files are truncated, but you can click here to view the full file

/*
   +----------------------------------------------------------------------+
   | Copyright (c) The PHP Group                                          |
   +----------------------------------------------------------------------+
   | This source file is subject to version 3.01 of the PHP license,      |
   | that is bundled with this package in the file LICENSE, and is        |
   | available through the world-wide-web at the following url:           |
   | https://www.php.net/license/3_01.txt                                 |
   | 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.               |
   +----------------------------------------------------------------------+
   | Authors: Rasmus Lerdorf <rasmus@php.net>                             |
   |          Stig Sæther Bakken <ssb@php.net>                          |
   |          Zeev Suraski <zeev@php.net>                                 |
   +----------------------------------------------------------------------+
 */

#include <stdio.h>
#include "php.h"
#include "php_rand.h"
#include "php_string.h"
#include "php_variables.h"
#include <locale.h>
#ifdef HAVE_LANGINFO_H
# include <langinfo.h>
#endif

#ifdef HAVE_LIBINTL
# include <libintl.h> /* For LC_MESSAGES */
#endif

#include "scanf.h"
#include "zend_API.h"
#include "zend_execute.h"
#include "php_globals.h"
#include "basic_functions.h"
#include "zend_smart_str.h"
#include <Zend/zend_exceptions.h>
#ifdef ZTS
#include "TSRM.h"
#endif

/* For str_getcsv() support */
#include "ext/standard/file.h"
/* For php_next_utf8_char() */
#include "ext/standard/html.h"

#ifdef __SSE2__
#include <emmintrin.h>
#endif

#define STR_PAD_LEFT			0
#define STR_PAD_RIGHT			1
#define STR_PAD_BOTH			2
#define PHP_PATHINFO_DIRNAME 	1
#define PHP_PATHINFO_BASENAME 	2
#define PHP_PATHINFO_EXTENSION 	4
#define PHP_PATHINFO_FILENAME 	8
#define PHP_PATHINFO_ALL	(PHP_PATHINFO_DIRNAME | PHP_PATHINFO_BASENAME | PHP_PATHINFO_EXTENSION | PHP_PATHINFO_FILENAME)

#define STR_STRSPN				0
#define STR_STRCSPN				1

/* {{{ register_string_constants */
void register_string_constants(INIT_FUNC_ARGS)
{
	REGISTER_LONG_CONSTANT("STR_PAD_LEFT", STR_PAD_LEFT, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("STR_PAD_RIGHT", STR_PAD_RIGHT, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("STR_PAD_BOTH", STR_PAD_BOTH, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("PATHINFO_DIRNAME", PHP_PATHINFO_DIRNAME, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("PATHINFO_BASENAME", PHP_PATHINFO_BASENAME, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("PATHINFO_EXTENSION", PHP_PATHINFO_EXTENSION, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("PATHINFO_FILENAME", PHP_PATHINFO_FILENAME, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("PATHINFO_ALL", PHP_PATHINFO_ALL, CONST_CS | CONST_PERSISTENT);

	/* If last members of struct lconv equal CHAR_MAX, no grouping is done */
	REGISTER_LONG_CONSTANT("CHAR_MAX", CHAR_MAX, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("LC_CTYPE", LC_CTYPE, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("LC_NUMERIC", LC_NUMERIC, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("LC_TIME", LC_TIME, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("LC_COLLATE", LC_COLLATE, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("LC_MONETARY", LC_MONETARY, CONST_CS | CONST_PERSISTENT);
	REGISTER_LONG_CONSTANT("LC_ALL", LC_ALL, CONST_CS | CONST_PERSISTENT);
# ifdef LC_MESSAGES
	REGISTER_LONG_CONSTANT("LC_MESSAGES", LC_MESSAGES, CONST_CS | CONST_PERSISTENT);
# endif

}
/* }}} */

int php_tag_find(char *tag, size_t len, const char *set);

/* this is read-only, so it's ok */
ZEND_SET_ALIGNED(16, static const char hexconvtab[]) = "0123456789abcdef";

/* localeconv mutex */
#ifdef ZTS
static MUTEX_T locale_mutex = NULL;
#endif

/* {{{ php_bin2hex */
static zend_string *php_bin2hex(const unsigned char *old, const size_t oldlen)
{
	zend_string *result;
	size_t i, j;

	result = zend_string_safe_alloc(oldlen, 2 * sizeof(char), 0, 0);

	for (i = j = 0; i < oldlen; i++) {
		ZSTR_VAL(result)[j++] = hexconvtab[old[i] >> 4];
		ZSTR_VAL(result)[j++] = hexconvtab[old[i] & 15];
	}
	ZSTR_VAL(result)[j] = '\0';

	return result;
}
/* }}} */

/* {{{ php_hex2bin */
static zend_string *php_hex2bin(const unsigned char *old, const size_t oldlen)
{
	size_t target_length = oldlen >> 1;
	zend_string *str = zend_string_alloc(target_length, 0);
	unsigned char *ret = (unsigned char *)ZSTR_VAL(str);
	size_t i, j;

	for (i = j = 0; i < target_length; i++) {
		unsigned char c = old[j++];
		unsigned char l = c & ~0x20;
		int is_letter = ((unsigned int) ((l - 'A') ^ (l - 'F' - 1))) >> (8 * sizeof(unsigned int) - 1);
		unsigned char d;

		/* basically (c >= '0' && c <= '9') || (l >= 'A' && l <= 'F') */
		if (EXPECTED((((c ^ '0') - 10) >> (8 * sizeof(unsigned int) - 1)) | is_letter)) {
			d = (l - 0x10 - 0x27 * is_letter) << 4;
		} else {
			zend_string_efree(str);
			return NULL;
		}
		c = old[j++];
		l = c & ~0x20;
		is_letter = ((unsigned int) ((l - 'A') ^ (l - 'F' - 1))) >> (8 * sizeof(unsigned int) - 1);
		if (EXPECTED((((c ^ '0') - 10) >> (8 * sizeof(unsigned int) - 1)) | is_letter)) {
			d |= l - 0x10 - 0x27 * is_letter;
		} else {
			zend_string_efree(str);
			return NULL;
		}
		ret[i] = d;
	}
	ret[i] = '\0';

	return str;
}
/* }}} */

/* {{{ localeconv_r
 * glibc's localeconv is not reentrant, so lets make it so ... sorta */
PHPAPI struct lconv *localeconv_r(struct lconv *out)
{

#ifdef ZTS
	tsrm_mutex_lock( locale_mutex );
#endif

/*  cur->locinfo is struct __crt_locale_info which implementation is
	hidden in vc14. TODO revisit this and check if a workaround available
	and needed. */
#if defined(PHP_WIN32) && _MSC_VER < 1900 && defined(ZTS)
	{
		/* Even with the enabled per thread locale, localeconv
			won't check any locale change in the master thread. */
		_locale_t cur = _get_current_locale();
		*out = *cur->locinfo->lconv;
		_free_locale(cur);
	}
#else
	/* localeconv doesn't return an error condition */
	*out = *localeconv();
#endif

#ifdef ZTS
	tsrm_mutex_unlock( locale_mutex );
#endif

	return out;
}
/* }}} */

#ifdef ZTS
/* {{{ PHP_MINIT_FUNCTION */
PHP_MINIT_FUNCTION(localeconv)
{
	locale_mutex = tsrm_mutex_alloc();
	return SUCCESS;
}
/* }}} */

/* {{{ PHP_MSHUTDOWN_FUNCTION */
PHP_MSHUTDOWN_FUNCTION(localeconv)
{
	tsrm_mutex_free( locale_mutex );
	locale_mutex = NULL;
	return SUCCESS;
}
/* }}} */
#endif

/* {{{ Converts the binary representation of data to hex */
PHP_FUNCTION(bin2hex)
{
	zend_string *result;
	zend_string *data;

	ZEND_PARSE_PARAMETERS_START(1, 1)
		Z_PARAM_STR(data)
	ZEND_PARSE_PARAMETERS_END();

	result = php_bin2hex((unsigned char *)ZSTR_VAL(data), ZSTR_LEN(data));

	RETURN_STR(result);
}
/* }}} */

/* {{{ Converts the hex representation of data to binary */
PHP_FUNCTION(hex2bin)
{
	zend_string *result, *data;

	ZEND_PARSE_PARAMETERS_START(1, 1)
		Z_PARAM_STR(data)
	ZEND_PARSE_PARAMETERS_END();

	if (ZSTR_LEN(data) % 2 != 0) {
		php_error_docref(NULL, E_WARNING, "Hexadecimal input string must have an even length");
		RETURN_FALSE;
	}

	result = php_hex2bin((unsigned char *)ZSTR_VAL(data), ZSTR_LEN(data));

	if (!result) {
		php_error_docref(NULL, E_WARNING, "Input string must be hexadecimal string");
		RETURN_FALSE;
	}

	RETVAL_STR(result);
}
/* }}} */

static void php_spn_common_handler(INTERNAL_FUNCTION_PARAMETERS, int behavior) /* {{{ */
{
	zend_string *s11, *s22;
	zend_long start = 0, len = 0;
	bool len_is_null = 1;

	ZEND_PARSE_PARAMETERS_START(2, 4)
		Z_PARAM_STR(s11)
		Z_PARAM_STR(s22)
		Z_PARAM_OPTIONAL
		Z_PARAM_LONG(start)
		Z_PARAM_LONG_OR_NULL(len, len_is_null)
	ZEND_PARSE_PARAMETERS_END();

	size_t remain_len = ZSTR_LEN(s11);
	if (start < 0) {
		start += remain_len;
		if (start < 0) {
			start = 0;
		}
	} else if ((size_t) start > remain_len) {
		start = remain_len;
	}

	remain_len -= start;
	if (!len_is_null) {
		if (len < 0) {
			len += remain_len;
			if (len < 0) {
				len = 0;
			}
		} else if ((size_t) len > remain_len) {
			len = remain_len;
		}
	} else {
		len = remain_len;
	}

	if (len == 0) {
		RETURN_LONG(0);
	}

	if (behavior == STR_STRSPN) {
		RETURN_LONG(php_strspn(ZSTR_VAL(s11) + start /*str1_start*/,
						ZSTR_VAL(s22) /*str2_start*/,
						ZSTR_VAL(s11) + start + len /*str1_end*/,
						ZSTR_VAL(s22) + ZSTR_LEN(s22) /*str2_end*/));
	} else {
		ZEND_ASSERT(behavior == STR_STRCSPN);
		RETURN_LONG(php_strcspn(ZSTR_VAL(s11) + start /*str1_start*/,
						ZSTR_VAL(s22) /*str2_start*/,
						ZSTR_VAL(s11) + start + len /*str1_end*/,
						ZSTR_VAL(s22) + ZSTR_LEN(s22) /*str2_end*/));
	}
}
/* }}} */

/* {{{ Finds length of initial segment consisting entirely of characters found in mask. If start or/and length is provided works like strspn(substr($s,$start,$len),$good_chars) */
PHP_FUNCTION(strspn)
{
	php_spn_common_handler(INTERNAL_FUNCTION_PARAM_PASSTHRU, STR_STRSPN);
}
/* }}} */

/* {{{ Finds length of initial segment consisting entirely of characters not found in mask. If start or/and length is provide works like strcspn(substr($s,$start,$len),$bad_chars) */
PHP_FUNCTION(strcspn)
{
	php_spn_common_handler(INTERNAL_FUNCTION_PARAM_PASSTHRU, STR_STRCSPN);
}
/* }}} */

/* {{{ PHP_MINIT_FUNCTION(nl_langinfo) */
#if HAVE_NL_LANGINFO
PHP_MINIT_FUNCTION(nl_langinfo)
{
#define REGISTER_NL_LANGINFO_CONSTANT(x)	REGISTER_LONG_CONSTANT(#x, x, CONST_CS | CONST_PERSISTENT)
#ifdef ABDAY_1
	REGISTER_NL_LANGINFO_CONSTANT(ABDAY_1);
	REGISTER_NL_LANGINFO_CONSTANT(ABDAY_2);
	REGISTER_NL_LANGINFO_CONSTANT(ABDAY_3);
	REGISTER_NL_LANGINFO_CONSTANT(ABDAY_4);
	REGISTER_NL_LANGINFO_CONSTANT(ABDAY_5);
	REGISTER_NL_LANGINFO_CONSTANT(ABDAY_6);
	REGISTER_NL_LANGINFO_CONSTANT(ABDAY_7);
#endif
#ifdef DAY_1
	REGISTER_NL_LANGINFO_CONSTANT(DAY_1);
	REGISTER_NL_LANGINFO_CONSTANT(DAY_2);
	REGISTER_NL_LANGINFO_CONSTANT(DAY_3);
	REGISTER_NL_LANGINFO_CONSTANT(DAY_4);
	REGISTER_NL_LANGINFO_CONSTANT(DAY_5);
	REGISTER_NL_LANGINFO_CONSTANT(DAY_6);
	REGISTER_NL_LANGINFO_CONSTANT(DAY_7);
#endif
#ifdef ABMON_1
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_1);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_2);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_3);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_4);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_5);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_6);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_7);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_8);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_9);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_10);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_11);
	REGISTER_NL_LANGINFO_CONSTANT(ABMON_12);
#endif
#ifdef MON_1
	REGISTER_NL_LANGINFO_CONSTANT(MON_1);
	REGISTER_NL_LANGINFO_CONSTANT(MON_2);
	REGISTER_NL_LANGINFO_CONSTANT(MON_3);
	REGISTER_NL_LANGINFO_CONSTANT(MON_4);
	REGISTER_NL_LANGINFO_CONSTANT(MON_5);
	REGISTER_NL_LANGINFO_CONSTANT(MON_6);
	REGISTER_NL_LANGINFO_CONSTANT(MON_7);
	REGISTER_NL_LANGINFO_CONSTANT(MON_8);
	REGISTER_NL_LANGINFO_CONSTANT(MON_9);
	REGISTER_NL_LANGINFO_CONSTANT(MON_10);
	REGISTER_NL_LANGINFO_CONSTANT(MON_11);
	REGISTER_NL_LANGINFO_CONSTANT(MON_12);
#endif
#ifdef AM_STR
	REGISTER_NL_LANGINFO_CONSTANT(AM_STR);
#endif
#ifdef PM_STR
	REGISTER_NL_LANGINFO_CONSTANT(PM_STR);
#endif
#ifdef D_T_FMT
	REGISTER_NL_LANGINFO_CONSTANT(D_T_FMT);
#endif
#ifdef D_FMT
	REGISTER_NL_LANGINFO_CONSTANT(D_FMT);
#endif
#ifdef T_FMT
	REGISTER_NL_LANGINFO_CONSTANT(T_FMT);
#endif
#ifdef T_FMT_AMPM
	REGISTER_NL_LANGINFO_CONSTANT(T_FMT_AMPM);
#endif
#ifdef ERA
	REGISTER_NL_LANGINFO_CONSTANT(ERA);
#endif
#ifdef ERA_YEAR
	REGISTER_NL_LANGINFO_CONSTANT(ERA_YEAR);
#endif
#ifdef ERA_D_T_FMT
	REGISTER_NL_LANGINFO_CONSTANT(ERA_D_T_FMT);
#endif
#ifdef ERA_D_FMT
	REGISTER_NL_LANGINFO_CONSTANT(ERA_D_FMT);
#endif
#ifdef ERA_T_FMT
	REGISTER_NL_LANGINFO_CONSTANT(ERA_T_FMT);
#endif
#ifdef ALT_DIGITS
	REGISTER_NL_LANGINFO_CONSTANT(ALT_DIGITS);
#endif
#ifdef INT_CURR_SYMBOL
	REGISTER_NL_LANGINFO_CONSTANT(INT_CURR_SYMBOL);
#endif
#ifdef CURRENCY_SYMBOL
	REGISTER_NL_LANGINFO_CONSTANT(CURRENCY_SYMBOL);
#endif
#ifdef CRNCYSTR
	REGISTER_NL_LANGINFO_CONSTANT(CRNCYSTR);
#endif
#ifdef MON_DECIMAL_POINT
	REGISTER_NL_LANGINFO_CONSTANT(MON_DECIMAL_POINT);
#endif
#ifdef MON_THOUSANDS_SEP
	REGISTER_NL_LANGINFO_CONSTANT(MON_THOUSANDS_SEP);
#endif
#ifdef MON_GROUPING
	REGISTER_NL_LANGINFO_CONSTANT(MON_GROUPING);
#endif
#ifdef POSITIVE_SIGN
	REGISTER_NL_LANGINFO_CONSTANT(POSITIVE_SIGN);
#endif
#ifdef NEGATIVE_SIGN
	REGISTER_NL_LANGINFO_CONSTANT(NEGATIVE_SIGN);
#endif
#ifdef INT_FRAC_DIGITS
	REGISTER_NL_LANGINFO_CONSTANT(INT_FRAC_DIGITS);
#endif
#ifdef FRAC_DIGITS
	REGISTER_NL_LANGINFO_CONSTANT(FRAC_DIGITS);
#endif
#ifdef P_CS_PRECEDES
	REGISTER_NL_LANGINFO_CONSTANT(P_CS_PRECEDES);
#endif
#ifdef P_SEP_BY_SPACE
	REGISTER_NL_LANGINFO_CONSTANT(P_SEP_BY_SPACE);
#endif
#ifdef N_CS_PRECEDES
	REGISTER_NL_LANGINFO_CONSTANT(N_CS_PRECEDES);
#endif
#ifdef N_SEP_BY_SPACE
	REGISTER_NL_LANGINFO_CONSTANT(N_SEP_BY_SPACE);
#endif
#ifdef P_SIGN_POSN
	REGISTER_NL_LANGINFO_CONSTANT(P_SIGN_POSN);
#endif
#ifdef N_SIGN_POSN
	REGISTER_NL_LANGINFO_CONSTANT(N_SIGN_POSN);
#endif
#ifdef DECIMAL_POINT
	REGISTER_NL_LANGINFO_CONSTANT(DECIMAL_POINT);
#endif
#ifdef RADIXCHAR
	REGISTER_NL_LANGINFO_CONSTANT(RADIXCHAR);
#endif
#ifdef THOUSANDS_SEP
	REGISTER_NL_LANGINFO_CONSTANT(THOUSANDS_SEP);
#endif
#ifdef THOUSEP
	REGISTER_NL_LANGINFO_CONSTANT(THOUSEP);
#endif
#ifdef GROUPING
	REGISTER_NL_LANGINFO_CONSTANT(GROUPING);
#endif
#ifdef YESEXPR
	REGISTER_NL_LANGINFO_CONSTANT(YESEXPR);
#endif
#ifdef NOEXPR
	REGISTER_NL_LANGINFO_CONSTANT(NOEXPR);
#endif
#ifdef YESSTR
	REGISTER_NL_LANGINFO_CONSTANT(YESSTR);
#endif
#ifdef NOSTR
	REGISTER_NL_LANGINFO_CONSTANT(NOSTR);
#endif
#ifdef CODESET
	REGISTER_NL_LANGINFO_CONSTANT(CODESET);
#endif
#undef REGISTER_NL_LANGINFO_CONSTANT
	return SUCCESS;
}
/* }}} */

/* {{{ Query language and locale information */
PHP_FUNCTION(nl_langinfo)
{
	zend_long item;
	char *value;

	ZEND_PARSE_PARAMETERS_START(1, 1)
		Z_PARAM_LONG(item)
	ZEND_PARSE_PARAMETERS_END();

	switch(item) { /* {{{ */
#ifdef ABDAY_1
		case ABDAY_1:
		case ABDAY_2:
		case ABDAY_3:
		case ABDAY_4:
		case ABDAY_5:
		case ABDAY_6:
		case ABDAY_7:
#endif
#ifdef DAY_1
		case DAY_1:
		case DAY_2:
		case DAY_3:
		case DAY_4:
		case DAY_5:
		case DAY_6:
		case DAY_7:
#endif
#ifdef ABMON_1
		case ABMON_1:
		case ABMON_2:
		case ABMON_3:
		case ABMON_4:
		case ABMON_5:
		case ABMON_6:
		case ABMON_7:
		case ABMON_8:
		case ABMON_9:
		case ABMON_10:
		case ABMON_11:
		case ABMON_12:
#endif
#ifdef MON_1
		case MON_1:
		case MON_2:
		case MON_3:
		case MON_4:
		case MON_5:
		case MON_6:
		case MON_7:
		case MON_8:
		case MON_9:
		case MON_10:
		case MON_11:
		case MON_12:
#endif
#ifdef AM_STR
		case AM_STR:
#endif
#ifdef PM_STR
		case PM_STR:
#endif
#ifdef D_T_FMT
		case D_T_FMT:
#endif
#ifdef D_FMT
		case D_FMT:
#endif
#ifdef T_FMT
		case T_FMT:
#endif
#ifdef T_FMT_AMPM
		case T_FMT_AMPM:
#endif
#ifdef ERA
		case ERA:
#endif
#ifdef ERA_YEAR
		case ERA_YEAR:
#endif
#ifdef ERA_D_T_FMT
		case ERA_D_T_FMT:
#endif
#ifdef ERA_D_FMT
		case ERA_D_FMT:
#endif
#ifdef ERA_T_FMT
		case ERA_T_FMT:
#endif
#ifdef ALT_DIGITS
		case ALT_DIGITS:
#endif
#ifdef INT_CURR_SYMBOL
		case INT_CURR_SYMBOL:
#endif
#ifdef CURRENCY_SYMBOL
		case CURRENCY_SYMBOL:
#endif
#ifdef CRNCYSTR
		case CRNCYSTR:
#endif
#ifdef MON_DECIMAL_POINT
		case MON_DECIMAL_POINT:
#endif
#ifdef MON_THOUSANDS_SEP
		case MON_THOUSANDS_SEP:
#endif
#ifdef MON_GROUPING
		case MON_GROUPING:
#endif
#ifdef POSITIVE_SIGN
		case POSITIVE_SIGN:
#endif
#ifdef NEGATIVE_SIGN
		case NEGATIVE_SIGN:
#endif
#ifdef INT_FRAC_DIGITS
		case INT_FRAC_DIGITS:
#endif
#ifdef FRAC_DIGITS
		case FRAC_DIGITS:
#endif
#ifdef P_CS_PRECEDES
		case P_CS_PRECEDES:
#endif
#ifdef P_SEP_BY_SPACE
		case P_SEP_BY_SPACE:
#endif
#ifdef N_CS_PRECEDES
		case N_CS_PRECEDES:
#endif
#ifdef N_SEP_BY_SPACE
		case N_SEP_BY_SPACE:
#endif
#ifdef P_SIGN_POSN
		case P_SIGN_POSN:
#endif
#ifdef N_SIGN_POSN
		case N_SIGN_POSN:
#endif
#ifdef DECIMAL_POINT
		case DECIMAL_POINT:
#elif defined(RADIXCHAR)
		case RADIXCHAR:
#endif
#ifdef THOUSANDS_SEP
		case THOUSANDS_SEP:
#elif defined(THOUSEP)
		case THOUSEP:
#endif
#ifdef GROUPING
		case GROUPING:
#endif
#ifdef YESEXPR
		case YESEXPR:
#endif
#ifdef NOEXPR
		case NOEXPR:
#endif
#ifdef YESSTR
		case YESSTR:
#endif
#ifdef NOSTR
		case NOSTR:
#endif
#ifdef CODESET
		case CODESET:
#endif
			break;
		default:
			php_error_docref(NULL, E_WARNING, "Item '" ZEND_LONG_FMT "' is not valid", item);
			RETURN_FALSE;
	}
	/* }}} */

	value = nl_langinfo(item);
	if (value == NULL) {
		RETURN_FALSE;
	} else {
		RETURN_STRING(value);
	}
}
#endif
/* }}} */

/* {{{ Compares two strings using the current locale */
PHP_FUNCTION(strcoll)
{
	zend_string *s1, *s2;

	ZEND_PARSE_PARAMETERS_START(2, 2)
		Z_PARAM_STR(s1)
		Z_PARAM_STR(s2)
	ZEND_PARSE_PARAMETERS_END();

	RETURN_LONG(strcoll((const char *) ZSTR_VAL(s1),
	                    (const char *) ZSTR_VAL(s2)));
}
/* }}} */

/* {{{ php_charmask
 * Fills a 256-byte bytemask with input. You can specify a range like 'a..z',
 * it needs to be incrementing.
 * Returns: FAILURE/SUCCESS whether the input was correct (i.e. no range errors)
 */
static inline int php_charmask(const unsigned char *input, size_t len, char *mask)
{
	const unsigned char *end;
	unsigned char c;
	int result = SUCCESS;

	memset(mask, 0, 256);
	for (end = input+len; input < end; input++) {
		c=*input;
		if ((input+3 < end) && input[1] == '.' && input[2] == '.'
				&& input[3] >= c) {
			memset(mask+c, 1, input[3] - c + 1);
			input+=3;
		} else if ((input+1 < end) && input[0] == '.' && input[1] == '.') {
			/* Error, try to be as helpful as possible:
			   (a range ending/starting with '.' won't be captured here) */
			if (end-len >= input) { /* there was no 'left' char */
				php_error_docref(NULL, E_WARNING, "Invalid '..'-range, no character to the left of '..'");
				result = FAILURE;
				continue;
			}
			if (input+2 >= end) { /* there is no 'right' char */
				php_error_docref(NULL, E_WARNING, "Invalid '..'-range, no character to the right of '..'");
				result = FAILURE;
				continue;
			}
			if (input[-1] > input[2]) { /* wrong order */
				php_error_docref(NULL, E_WARNING, "Invalid '..'-range, '..'-range needs to be incrementing");
				result = FAILURE;
				continue;
			}
			/* FIXME: better error (a..b..c is the only left possibility?) */
			php_error_docref(NULL, E_WARNING, "Invalid '..'-range");
			result = FAILURE;
			continue;
		} else {
			mask[c]=1;
		}
	}
	return result;
}
/* }}} */

/* {{{ php_trim_int()
 * mode 1 : trim left
 * mode 2 : trim right
 * mode 3 : trim left and right
 * what indicates which chars are to be trimmed. NULL->default (' \t\n\r\v\0')
 */
static zend_always_inline zend_string *php_trim_int(zend_string *str, const char *what, size_t what_len, int mode)
{
	const char *start = ZSTR_VAL(str);
	const char *end = start + ZSTR_LEN(str);
	char mask[256];

	if (what) {
		if (what_len == 1) {
			char p = *what;
			if (mode & 1) {
				while (start != end) {
					if (*start == p) {
						start++;
					} else {
						break;
					}
				}
			}
			if (mode & 2) {
				while (start != end) {
					if (*(end-1) == p) {
						end--;
					} else {
						break;
					}
				}
			}
		} else {
			php_charmask((const unsigned char *) what, what_len, mask);

			if (mode & 1) {
				while (start != end) {
					if (mask[(unsigned char)*start]) {
						start++;
					} else {
						break;
					}
				}
			}
			if (mode & 2) {
				while (start != end) {
					if (mask[(unsigned char)*(end-1)]) {
						end--;
					} else {
						break;
					}
				}
			}
		}
	} else {
		if (mode & 1) {
			while (start != end) {
				unsigned char c = (unsigned char)*start;

				if (c <= ' ' &&
				    (c == ' ' || c == '\n' || c == '\r' || c == '\t' || c == '\v' || c == '\0')) {
					start++;
				} else {
					break;
				}
			}
		}
		if (mode & 2) {
			while (start != end) {
				unsigned char c = (unsigned char)*(end-1);

				if (c <= ' ' &&
				    (c == ' ' || c == '\n' || c == '\r' || c == '\t' || c == '\v' || c == '\0')) {
					end--;
				} else {
					break;
				}
			}
		}
	}

	if (ZSTR_LEN(str) == end - start) {
		return zend_string_copy(str);
	} else if (end - start == 0) {
		return ZSTR_EMPTY_ALLOC();
	} else {
		return zend_string_init(start, end - start, 0);
	}
}
/* }}} */

/* {{{ php_trim_int()
 * mode 1 : trim left
 * mode 2 : trim right
 * mode 3 : trim left and right
 * what indicates which chars are to be trimmed. NULL->default (' \t\n\r\v\0')
 */
PHPAPI zend_string *php_trim(zend_string *str, const char *what, size_t what_len, int mode)
{
	return php_trim_int(str, what, what_len, mode);
}
/* }}} */

/* {{{ php_do_trim
 * Base for trim(), rtrim() and ltrim() functions.
 */
static zend_always_inline void php_do_trim(INTERNAL_FUNCTION_PARAMETERS, int mode)
{
	zend_string *str;
	zend_string *what = NULL;

	ZEND_PARSE_PARAMETERS_START(1, 2)
		Z_PARAM_STR(str)
		Z_PARAM_OPTIONAL
		Z_PARAM_STR(what)
	ZEND_PARSE_PARAMETERS_END();

	ZVAL_STR(return_value, php_trim_int(str, (what ? ZSTR_VAL(what) : NULL), (what ? ZSTR_LEN(what) : 0), mode));
}
/* }}} */

/* {{{ Strips whitespace from the beginning and end of a string */
PHP_FUNCTION(trim)
{
	php_do_trim(INTERNAL_FUNCTION_PARAM_PASSTHRU, 3);
}
/* }}} */

/* {{{ Removes trailing whitespace */
PHP_FUNCTION(rtrim)
{
	php_do_trim(INTERNAL_FUNCTION_PARAM_PASSTHRU, 2);
}
/* }}} */

/* {{{ Strips whitespace from the beginning of a string */
PHP_FUNCTION(ltrim)
{
	php_do_trim(INTERNAL_FUNCTION_PARAM_PASSTHRU, 1);
}
/* }}} */

/* {{{ Wraps buffer to selected number of characters using string break char */
PHP_FUNCTION(wordwrap)
{
	zend_string *text;
	char *breakchar = "\n";
	size_t newtextlen, chk, breakchar_len = 1;
	size_t alloced;
	zend_long current = 0, laststart = 0, lastspace = 0;
	zend_long linelength = 75;
	bool docut = 0;
	zend_string *newtext;

	ZEND_PARSE_PARAMETERS_START(1, 4)
		Z_PARAM_STR(text)
		Z_PARAM_OPTIONAL
		Z_PARAM_LONG(linelength)
		Z_PARAM_STRING(breakchar, breakchar_len)
		Z_PARAM_BOOL(docut)
	ZEND_PARSE_PARAMETERS_END();

	if (ZSTR_LEN(text) == 0) {
		RETURN_EMPTY_STRING();
	}

	if (breakchar_len == 0) {
		zend_argument_value_error(3, "cannot be empty");
		RETURN_THROWS();
	}

	if (linelength == 0 && docut) {
		zend_argument_value_error(4, "cannot be true when argument #2 ($width) is 0");
		RETURN_THROWS();
	}

	/* Special case for a single-character break as it needs no
	   additional storage space */
	if (breakchar_len == 1 && !docut) {
		newtext = zend_string_init(ZSTR_VAL(text), ZSTR_LEN(text), 0);

		laststart = lastspace = 0;
		for (current = 0; current < (zend_long)ZSTR_LEN(text); current++) {
			if (ZSTR_VAL(text)[current] == breakchar[0]) {
				laststart = lastspace = current + 1;
			} else if (ZSTR_VAL(text)[current] == ' ') {
				if (current - laststart >= linelength) {
					ZSTR_VAL(newtext)[current] = breakchar[0];
					laststart = current + 1;
				}
				lastspace = current;
			} else if (current - laststart >= linelength && laststart != lastspace) {
				ZSTR_VAL(newtext)[lastspace] = breakchar[0];
				laststart = lastspace + 1;
			}
		}

		RETURN_NEW_STR(newtext);
	} else {
		/* Multiple character line break or forced cut */
		if (linelength > 0) {
			chk = (size_t)(ZSTR_LEN(text)/linelength + 1);
			newtext = zend_string_safe_alloc(chk, breakchar_len, ZSTR_LEN(text), 0);
			alloced = ZSTR_LEN(text) + chk * breakchar_len + 1;
		} else {
			chk = ZSTR_LEN(text);
			alloced = ZSTR_LEN(text) * (breakchar_len + 1) + 1;
			newtext = zend_string_safe_alloc(ZSTR_LEN(text), breakchar_len + 1, 0, 0);
		}

		/* now keep track of the actual new text length */
		newtextlen = 0;

		laststart = lastspace = 0;
		for (current = 0; current < (zend_long)ZSTR_LEN(text); current++) {
			if (chk == 0) {
				alloced += (size_t) (((ZSTR_LEN(text) - current + 1)/linelength + 1) * breakchar_len) + 1;
				newtext = zend_string_extend(newtext, alloced, 0);
				chk = (size_t) ((ZSTR_LEN(text) - current)/linelength) + 1;
			}
			/* when we hit an existing break, copy to new buffer, and
			 * fix up laststart and lastspace */
			if (ZSTR_VAL(text)[current] == breakchar[0]
				&& current + breakchar_len < ZSTR_LEN(text)
				&& !strncmp(ZSTR_VAL(text) + current, breakchar, breakchar_len)) {
				memcpy(ZSTR_VAL(newtext) + newtextlen, ZSTR_VAL(text) + laststart, current - laststart + breakchar_len);
				newtextlen += current - laststart + breakchar_len;
				current += breakchar_len - 1;
				laststart = lastspace = current + 1;
				chk--;
			}
			/* if it is a space, check if it is at the line boundary,
			 * copy and insert a break, or just keep track of it */
			else if (ZSTR_VAL(text)[current] == ' ') {
				if (current - laststart >= linelength) {
					memcpy(ZSTR_VAL(newtext) + newtextlen, ZSTR_VAL(text) + laststart, current - laststart);
					newtextlen += current - laststart;
					memcpy(ZSTR_VAL(newtext) + newtextlen, breakchar, breakchar_len);
					newtextlen += breakchar_len;
					laststart = current + 1;
					chk--;
				}
				lastspace = current;
			}
			/* if we are cutting, and we've accumulated enough
			 * characters, and we haven't see a space for this line,
			 * copy and insert a break. */
			else if (current - laststart >= linelength
					&& docut && laststart >= lastspace) {
				memcpy(ZSTR_VAL(newtext) + newtextlen, ZSTR_VAL(text) + laststart, current - laststart);
				newtextlen += current - laststart;
				memcpy(ZSTR_VAL(newtext) + newtextlen, breakchar, breakchar_len);
				newtextlen += breakchar_len;
				laststart = lastspace = current;
				chk--;
			}
			/* if the current word puts us over the linelength, copy
			 * back up until the last space, insert a break, and move
			 * up the laststart */
			else if (current - laststart >= linelength
					&& laststart < lastspace) {
				memcpy(ZSTR_VAL(newtext) + newtextlen, ZSTR_VAL(text) + laststart, lastspace - laststart);
				newtextlen += lastspace - laststart;
				memcpy(ZSTR_VAL(newtext) + newtextlen, breakchar, breakchar_len);
				newtextlen += breakchar_len;
				laststart = lastspace = lastspace + 1;
				chk--;
			}
		}

		/* copy over any stragglers */
		if (laststart != current) {
			memcpy(ZSTR_VAL(newtext) + newtextlen, ZSTR_VAL(text) + laststart, current - laststart);
			newtextlen += current - laststart;
		}

		ZSTR_VAL(newtext)[newtextlen] = '\0';
		/* free unused memory */
		newtext = zend_string_truncate(newtext, newtextlen, 0);

		RETURN_NEW_STR(newtext);
	}
}
/* }}} */

/* {{{ php_explode */
PHPAPI void php_explode(const zend_string *delim, zend_string *str, zval *return_value, zend_long limit)
{
	const char *p1 = ZSTR_VAL(str);
	const char *endp = ZSTR_VAL(str) + ZSTR_LEN(str);
	const char *p2 = php_memnstr(ZSTR_VAL(str), ZSTR_VAL(delim), ZSTR_LEN(delim), endp);
	zval  tmp;

	if (p2 == NULL) {
		ZVAL_STR_COPY(&tmp, str);
		zend_hash_next_index_insert_new(Z_ARRVAL_P(return_value), &tmp);
	} else {
		zend_hash_real_init_packed(Z_ARRVAL_P(return_value));
		ZEND_HASH_FILL_PACKED(Z_ARRVAL_P(return_value)) {
			do {
				ZEND_HASH_FILL_GROW();
				ZEND_HASH_FILL_SET_STR(zend_string_init_fast(p1, p2 - p1));
				ZEND_HASH_FILL_NEXT();
				p1 = p2 + ZSTR_LEN(delim);
				p2 = php_memnstr(p1, ZSTR_VAL(delim), ZSTR_LEN(delim), endp);
			} while (p2 != NULL && --limit > 1);

			if (p1 <= endp) {
				ZEND_HASH_FILL_GROW();
				ZEND_HASH_FILL_SET_STR(zend_string_init_fast(p1, endp - p1));
				ZEND_HASH_FILL_NEXT();
			}
		} ZEND_HASH_FILL_END();
	}
}
/* }}} */

/* {{{ php_explode_negative_limit */
PHPAPI void php_explode_negative_limit(const zend_string *delim, zend_string *str, zval *return_value, zend_long limit)
{
#define EXPLODE_ALLOC_STEP 64
	const char *p1 = ZSTR_VAL(str);
	const char *endp = ZSTR_VAL(str) + ZSTR_LEN(str);
	const char *p2 = php_memnstr(ZSTR_VAL(str), ZSTR_VAL(delim), ZSTR_LEN(delim), endp);
	zval  tmp;

	if (p2 == NULL) {
		/*
		do nothing since limit <= -1, thus if only one chunk - 1 + (limit) <= 0
		by doing nothing we return empty array
		*/
	} else {
		size_t allocated = EXPLODE_ALLOC_STEP, found = 0;
		zend_long i, to_return;
		const char **positions = emalloc(allocated * sizeof(char *));

		positions[found++] = p1;
		do {
			if (found >= allocated) {
				allocated = found + EXPLODE_ALLOC_STEP;/* make sure we have enough memory */
				positions = erealloc(ZEND_VOIDP(positions), allocated*sizeof(char *));
			}
			positions[found++] = p1 = p2 + ZSTR_LEN(delim);
			p2 = php_memnstr(p1, ZSTR_VAL(delim), ZSTR_LEN(delim), endp);
		} while (p2 != NULL);

		to_return = limit + found;
		/* limit is at least -1 therefore no need of bounds checking : i will be always less than found */
		for (i = 0; i < to_return; i++) { /* this checks also for to_return > 0 */
			ZVAL_STRINGL(&tmp, positions[i], (positions[i+1] - ZSTR_LEN(delim)) - positions[i]);
			zend_hash_next_index_insert_new(Z_ARRVAL_P(return_value), &tmp);
		}
		efree((void *)positions);
	}
#undef EXPLODE_ALLOC_STEP
}
/* }}} */

/* {{{ Splits a string on string separator and return array of components. If limit is positive only limit number of components is returned. If limit is negative all components except the last abs(limit) are returned. */
PHP_FUNCTION(explode)
{
	zend_string *str, *delim;
	zend_long limit = ZEND_LONG_MAX; /* No limit */
	zval tmp;

	ZEND_PARSE_PARAMETERS_START(2, 3)
		Z_PARAM_STR(delim)
		Z_PARAM_STR(str)
		Z_PARAM_OPTIONAL
		Z_PARAM_LONG(limit)
	ZEND_PARSE_PARAMETERS_END();

	if (ZSTR_LEN(delim) == 0) {
		zend_argument_value_error(1, "cannot be empty");
		RETURN_THROWS();
	}

	array_init(return_value);

	if (ZSTR_LEN(str) == 0) {
		if (limit >= 0) {
			ZVAL_EMPTY_STRING(&tmp);
			zend_hash_index_add_new(Z_ARRVAL_P(return_value), 0, &tmp);
		}
		return;
	}

	if (limit > 1) {
		php_explode(delim, str, return_value, limit);
	} else if (limit < 0) {
		php_explode_negative_limit(delim, str, return_value, limit);
	} else {
		ZVAL_STR_COPY(&tmp, str);
		zend_hash_index_add_new(Z_ARRVAL_P(return_value), 0, &tmp);
	}
}
/* }}} */

/* {{{ An alias for implode */
/* }}} */

/* {{{ php_implode */
PHPAPI void php_implode(const zend_string *glue, HashTable *pieces, zval *return_value)
{
	zval         *tmp;
	int           numelems;
	zend_string  *str;
	char         *cptr;
	size_t        len = 0;
	struct {
		zend_string *str;
		zend_long    lval;
	} *strings, *ptr;
	ALLOCA_FLAG(use_heap)

	numelems = zend_hash_num_elements(pieces);

	if (numelems == 0) {
		RETURN_EMPTY_STRING();
	} else if (numelems == 1) {
		/* loop to search the first not undefined element... */
		ZEND_HASH_FOREACH_VAL(pieces, tmp) {
			RETURN_STR(zval_get_string(tmp));
		} ZEND_HASH_FOREACH_END();
	}

	ptr = strings = do_alloca((sizeof(*strings)) * numelems, use_heap);

	ZEND_HASH_FOREACH_VAL(pieces, tmp) {
		if (EXPECTED(Z_TYPE_P(tmp) == IS_STRING)) {
			ptr->str = Z_STR_P(tmp);
			len += ZSTR_LEN(ptr->str);
			ptr->lval = 0;
			ptr++;
		} else if (UNEXPECTED(Z_TYPE_P(tmp) == IS_LONG)) {
			zend_long val = Z_LVAL_P(tmp);

			ptr->str = NULL;
			ptr->lval = val;
			ptr++;
			if (val <= 0) {
				len++;
			}
			while (val) {
				val /= 10;
				len++;
			}
		} else {
			ptr->str = zval_get_string_func(tmp);
			len += ZSTR_LEN(ptr->str);
			ptr->lval = 1;
			ptr++;
		}
	} ZEND_HASH_FOREACH_END();

	/* numelems cannot be 0, we checked above */
	str = zend_string_safe_alloc(numelems - 1, ZSTR_LEN(glue), len, 0);
	cptr = ZSTR_VAL(str) + ZSTR_LEN(str);
	*cptr = 0;

	while (1) {
		ptr--;
		if (EXPECTED(ptr->str)) {
			cptr -= ZSTR_LEN(ptr->str);
			memcpy(cptr, ZSTR_VAL(ptr->str), ZSTR_LEN(ptr->str));
			if (ptr->lval) {
				zend_string_release_ex(ptr->str, 0);
			}
		} else {
			char *oldPtr = cptr;
			char oldVal = *cptr;
			cptr = zend_print_long_to_buf(cptr, ptr->lval);
			*oldPtr = oldVal;
		}

		if (ptr == strings) {
			break;
		}

		cptr -= ZSTR_LEN(glue);
		memcpy(cptr, ZSTR_VAL(glue), ZSTR_LEN(glue));
	}

	free_alloca(strings, use_heap);
	RETURN_NEW_STR(str);
}
/* }}} */

/* {{{ Joins array elements placing glue string between items and return one string */
PHP_FUNCTION(implode)
{
	zend_string *arg1_str = NULL;
	HashTable *arg1_array = NULL;
	zend_array *pieces = NULL;

	ZEND_PARSE_PARAMETERS_START(1, 2)
		Z_PARAM_ARRAY_HT_OR_STR(arg1_array, arg1_str)
		Z_PARAM_OPTIONAL
		Z_PARAM_ARRAY_HT_OR_NULL(pieces)
	ZEND_PARSE_PARAMETERS_END();

	if (pieces == NULL) {
		if (arg1_array == NULL) {
			zend_type_error("%s(): Argument #1 ($pieces) must be of type array, string given", get_active_function_name());
			RETURN_THROWS();
		}

		arg1_str = ZSTR_EMPTY_ALLOC();
		pieces = arg1_array;
	} else {
		if (arg1_str == NULL) {
			zend_argument_type_error(1, "must be of type string, array given");
			RETURN_THROWS();
		}
	}

	php_implode(arg1_str, pieces, return_value);
}
/* }}} */

#define STRTOK_TABLE(p) BG(strtok_table)[(unsigned char) *p]

/* {{{ Tokenize a string */
PHP_FUNCTION(strtok)
{
	zend_string *str, *tok = NULL;
	char *token;
	char *token_end;
	char *p;
	char *pe;
	size_t skipped = 0;

	ZEND_PARSE_PARAMETERS_START(1, 2)
		Z_PARAM_STR(str)
		Z_PARAM_OPTIONAL
		Z_PARAM_STR_OR_NULL(tok)
	ZEND_PARSE_PARAMETERS_END();

	if (!tok) {
		tok = str;
	} else {
		if (BG(strtok_string)) {
			zend_string_release(BG(strtok_string));
		}
		BG(strtok_string) = zend_string_copy(str);
		BG(strtok_last) = ZSTR_VAL(str);
		BG(strtok_len) = ZSTR_LEN(str);
	}

	if (!BG(strtok_string)) {
		/* String to tokenize not set. */
		// TODO: Should this warn?
		RETURN_FALSE;
	}

	p = BG(strtok_last); /* Where we start to search */
	pe = ZSTR_VAL(BG(strtok_string)) + BG(strtok_len);
	if (p >= pe) {
		/* Reached the end of the string. */
		RETURN_FALSE;
	}

	token = ZSTR_VAL(tok);
	token_end = token + ZSTR_LEN(tok);

	while (token < token_end) {
		STRTOK_TABLE(token++) = 1;
	}

	/* Skip leading delimiters */
	while (STRTOK_TABLE(p)) {
		if (++p >= pe) {
			/* no other chars left */
			goto return_false;
		}
		skipped++;
	}

	/* We know at this place that *p is no delimiter, so skip it */
	while (++p < pe) {
		if (STRTOK_TABLE(p)) {
			goto return_token;
		}
	}

	if (p - BG(strtok_last)) {
return_token:
		RETVAL_STRINGL(BG(strtok_last) + skipped, (p - BG(strtok_last)) - skipped);
		BG(strtok_last) = p + 1;
	} else {
return_false:
		RETVAL_FALSE;
		zend_string_release(BG(strtok_string));
		BG(strtok_string) = NULL;
	}

	/* Restore table -- usually faster then memset'ing the table on every invocation */
	token = ZSTR_VAL(tok);
	while (token < token_end) {
		STRTOK_TABLE(token++) = 0;
	}
}
/* }}} */

/* {{{ php_strtoupper */
PHPAPI char *php_strtoupper(char *s, size_t len)
{
	unsigned char *c;
	const unsigned char *e;

	c = (unsigned char *)s;
	e = (unsigned char *)c+len;

	while (c < e) {
		*c = toupper(*c);
		c++;
	}
	return s;
}
/* }}} */

/* {{{ php_string_toupper */
PHPAPI zend_string *php_string_toupper(zend_string *s)
{
	unsigned char *c;
	const unsigned char *e;

	if (EXPECTED(!BG(ctype_string))) {
		return zend_string_toupper(s);
	}
	c = (unsigned char *)ZSTR_VAL(s);
	e = c + ZSTR_LEN(s);

	while (c < e) {
		if (islower(*c)) {
			unsigned char *r;
			zend_string *res = zend_string_alloc(ZSTR_LEN(s), 0);

			if (c != (unsigned char*)ZSTR_VAL(s)) {
				memcpy(ZSTR_VAL(res), ZSTR_VAL(s), c - (unsigned char*)ZSTR_VAL(s));
			}
			r = c + (ZSTR_VAL(res) - ZSTR_VAL(s));
			while (c < e) {
				*r = toupper(*c);
				r++;
				c++;
			}
			*r = '\0';
			return res;
		}
		c++;
	}
	return zend_string_copy(s);
}
/* }}} */

/* {{{ Makes a string uppercase */
PHP_FUNCTION(strtoupper)
{
	zend_string *arg;

	ZEND_PARSE_PARAMETERS_START(1, 1)
		Z_PARAM_STR(arg)
	ZEND_PARSE_PARAMETERS_END();

	RETURN_STR(php_string_toupper(arg));
}
/* }}} */

/* {{{ php_strtolower */
PHPAPI char *php_strtolower(char *s, size_t len)
{
	unsigned char *c;
	const unsigned char *e;

	c = (unsigned char *)s;
	e = c+len;

	while (c < e) {
		*c = tolower(*c);
		c++;
	}
	return s;
}
/* }}} */

/* {{{ php_string_tolower */
PHPAPI zend_string *php_string_tolower(zend_string *s)
{
	if (EXPECTED(!BG(ctype_string))) {
		return zend_string_tolower(s);
	}

	unsigned char *c = (unsigned char *)ZSTR_VAL(s);
	const unsigned char *e = c + ZSTR_LEN(s);
	while (c < e) {
		if (isupper(*c)) {
			unsigned char *r;
			zend_string *res = zend_string_alloc(ZSTR_LEN(s), 0);

			if (c != (unsigned char*)ZSTR_VAL(s)) {
				memcpy(ZSTR_VAL(res), ZSTR_VAL(s), c - (unsigned char*)ZSTR_VAL(s));
			}
			r = c + (ZSTR_VAL(res) - ZSTR_VAL(s));
			while (c < e) {
				*r = tolower(*c);
				r++;
				c++;
			}
			*r = '\0';
			return res;
		}
		c++;
	}
	return zend_string_copy(s);
}
/* }}} */

/* {{{ Makes a string lowercase */
PHP_FUNCTION(strtolower)
{
	zend_string *str;

	ZEND_PARSE_PARAMETERS_START(1, 1)
		Z_PARAM_STR(str)
	ZEND_PARSE_PARAMETERS_END();

	RETURN_STR(php_string_tolower(str));
}
/* }}} */

#if defined(PHP_WIN32)
static bool _is_basename_start(const char *start, const char *pos)
{
	if (pos - start >= 1
	    && *(pos-1) != '/'
	    && *(pos-1) != '\\') {
		if (pos - start == 1) {
			return 1;
		} else if (*(pos-2) == '/' || *(pos-2) == '\\') {
			return 1;
		} else if (*(pos-2) == ':'
			&& _is_basename_start(start, pos - 2)) {
			return 1;
		}
	}
	return 0;
}
#endif

/* {{{ php_basename */
PHPAPI zend_string *php_basename(const char *s, size_t len, const char *suffix, size_t suffix_len)
{
	const char *basename_start;
	const char *basename_end;

	if (CG(ascii_compatible_locale)) {
		basename_end = s + len - 1;

		/* Strip trailing slashes */
		while (basename_end >= s
#if defined(PHP_WIN32)
			&& (*basename_end == '/'
				|| *basename_end == '\\'
				|| (*basename_end == ':'
					&& _is_basename_start(s, basename_end)))) {
#else
			&& *basename_end == '/') {
#endif
			basename_end--;
		}
		if (basename_end < s) {
			return ZSTR_EMPTY_ALLOC();
		}

		/* Extract filename */
		basename_start = basename_end;
		basename_end++;
		while (basename_start > s
#if defined(PHP_WIN32)
			&& *(basename_start-1) != '/'
			&& *(basename_start-1) != '\\') {

			if (*(basename_start-1) == ':' &&
				_is_basename_start(s, basename_start - 1)) {
				break;
			}
#else
			&& *(basename_start-1) != '/') {
#endif
			basename_start--;
		}
	} else {
		/* State 0 is directly after a directory separator (or at the start of the string).
		 * State 1 is everything else. */
		int state = 0;

		basename_start = s;
		basename_end = s;
		while (len > 0) {
			int inc_len = (*s == '\0' ? 1 : php_mblen(s, len));

			switch (inc_len) {
				case 0:
					goto quit_loop;
				case 1:
#if defined(PHP_WIN32)
					if (*s == '/' || *s == '\\') {
#else
					if (*s == '/') {
#endif
						if (state == 1) {
							state = 0;
							basename_end = s;
						}
#if defined(PHP_WIN32)
					/* Catch relative paths in c:file.txt style. They're not to confuse
					   with the NTFS streams. This part ensures also, that no drive
					   letter traversing happens. */
					} else if ((*s == ':' && (s - basename_start == 1))) {
						if (state == 0) {
							basename_start = s;
							state = 1;
						} else {
							basename_end = s;
							state = 0;
						}
#endif
					} else {
						if (state == 0) {
							basename_start = s;
							state = 1;
						}
					}
					break;
				default:
					if (inc_len < 0) {
						/* If character is invalid, treat it like other non-significant characters. */
						inc_len = 1;
						php_mb_reset();
					}
					if (state == 0) {
						basename_start = s;
						state = 1;
					}
					break;
			}
			s += inc_len;
			len -= inc_len;
		}

quit_loop:
		if (state == 1) {
			basename_end = s;
		}
	}

	if (suffix != NULL && suffix_len < (size_t)(basename_end - basename_start) &&
			memcmp(basename_end - suffix_len, suffix, suffix_len) == 0) {
		basename_end -= suffix_len;
	}

	return zend_string_init(basename_start, basename_end - basename_start, 0);
}
/* }}} */

/* {{{ Returns the filename component of the path */
PHP_FUNCTION(basename)
{
	char *string, *suffix = NULL;
	size_t   string_len, suffix_len = 0;

	ZEND_PARSE_PARAMETERS_START(1, 2)
		Z_PARAM_STRING(string, string_len)
		Z_PARAM_OPTIONAL
		Z_PARAM_STRING(suffix, suffix_len)
	ZEND_PARSE_PARAMETERS_END();

	RETURN_STR(php_basename(string, string_len, suffix, suffix_len));
}
/* }}} */

/* {{{ php_dirname
   Returns directory name component of path */
PHPAPI size_t php_dirname(char *path, size_t len)
{
	return zend_dirname(path, len);
}
/* }}} */

/* {{{ Returns the directory name component of the path */
PHP_FUNCTION(dirname)
{
	char *str;
	size_t str_len;
	zend_string *ret;
	zend_long levels = 1;

	ZEND_PARSE_PARAMETERS_START(1, 2)
		Z_PARAM_STRING(str, str_len)
		Z_PARAM_OPTIONAL
		Z_PARAM_LONG(levels)
	ZEND_PARSE_PARAMETERS_END();

	ret = zend_string_init(str, str_len, 0);

	if (levels == 1) {
		/* Default case */
#ifdef PHP_WIN32
		ZSTR_LEN(ret) = php_win32_ioutil_dirname(ZSTR_VAL(ret), str_len);
#else
		ZSTR_LEN(ret) = zend_dirname(ZSTR_VAL(ret), str_len);
#endif
	} else if (levels < 1) {
		zend_argument_value_error(2, "must be greater than or equal to 1");
		zend_string_efree(ret);
		RETURN_THROWS();
	} else {
		/* Some levels up */
		do {
#ifdef PHP_WIN32
			ZSTR_LEN(ret) = php_win32_ioutil_dirname(ZSTR_VAL(ret), str_len = ZSTR_LEN(ret));
#else
			ZSTR_LEN(ret) = zend_dirname(ZSTR_VAL(ret), str_len = ZSTR_LEN(ret));
#endif
		} while (ZSTR_LEN(ret) < str_len && --levels);
	}

	RETURN_NEW_STR(ret);
}
/* }}} */

/* {{{ Returns information about a certain string */
PHP_FUNCTION(pathinfo)
{
	zval tmp;
	char *path, *dirname;
	size_t path_len;
	int have_basename;
	zend_long opt = PHP_PATHINFO_ALL;
	zend_string *ret = NULL;

	ZEND_PARSE_PARAMETERS_START(1, 2)
		Z_PARAM_STRING(path, path_len)
		Z_PARAM_OPTIONAL
		Z_PARAM_LONG(opt)
	ZEND_PARSE_PARAMETERS_END();

	have_basename = ((opt & PHP_PATHINFO_BASENAME) == PHP_PATHINFO_BASENAME);

	array_init(&tmp);

	if ((opt & PHP_PATHINFO_DIRNAME) == PHP_PATHINFO_DIRNAME) {
		dirname = estrndup(path, path_len);
		php_dirname(dirname, path_len);
		if (*dirname) {
			add_assoc_string(&tmp, "dirname", dirname);
		}
		efree(dirname);
	}

	if (have_basename) {
		ret = php_basename(path, path_len, NULL, 0);
		add_assoc_str(&tmp, "basename", zend_string_copy(ret));
	}

	if ((opt & PHP_PATHINFO_EXTENSION) == PHP_PATHINFO_EXTENSION) {
		const char *p;
		ptrdiff_t idx;

		if (!have_basename) {
			ret = php_basename(path, path_len, NULL, 0);
		}

		p = zend_memrchr(ZSTR_VAL(ret), '.', ZSTR_LEN(ret));

		if (p) {
			idx = p - ZSTR_VAL(ret);
			add_assoc_stringl(&tmp, "extension", ZSTR_VAL(ret) + idx + 1, ZSTR_LEN(ret) - idx - 1);
		}
	}

	if ((opt & PHP_PATHINFO_FILENAME) == PHP_PATHINFO_FILENAME) {
		const char *p;
		ptrdiff_t idx;

		/* Have we already looked up the basename? */
		if (!have_basename && !ret) {
			ret = php_basename(path, path_len, NULL, 0);
		}

		p = zend_memrchr(ZSTR_VAL(ret), '.', ZSTR_LEN(ret));

		idx = p ? (p - ZSTR_VAL(ret)) : (ptrdiff_t)ZSTR_LEN(ret);
		add_assoc_stringl(&tmp, "filename", ZSTR_VAL(ret), idx);
	}

	if (ret) {
		zend_string_release_ex(ret, 0);
	}

	if (opt == PHP_PATHINFO_ALL) {
		RETURN_COPY_VALUE(&tmp);
	} else {
		zval *element;
		if ((element = zend_hash_get_current_data(Z_ARRVAL(tmp))) != NULL) {
			RETVAL_COPY_DEREF(element);
		} else {
			RETVAL_EMPTY_STRING();
		}
		zval_ptr_dtor(&tmp);
	}
}
/* }}} */

/* {{{ php_stristr
   case insensitive strstr */
PHPAPI char *php_stristr(char *s, char *t, size_t s_len, size_t t_len)
{
	php_strtolower(s, s_len);
	php_strtolower(t, t_len);
	return (char*)php_memnstr(s, t, t_len, s + s_len);
}
/* }}} */

/* {{{ php_strspn */
PHPAPI size_t php_strspn(const char *s1, const char *s2, const char *s1_end, const char *s2_end)
{
	const char *p = s1, *spanp;
	char c = *p;

cont:
	for (spanp = s2; p != s1_end && spanp != s2_end;) {
		if (*spanp++ == c) {
			c = *(++p);
			goto cont;
		}
	}
	return (p - s1);
}
/* }}} */

/* {{{ php_strcspn */
PHPAPI size_t php_strcspn(const char *s1, const char *s2, const char *s1_end, const char *s2_end)
{
	const char *p, *spanp;
	char c = *s1;

	for (p = s1;;) {
		spanp = s2;
		do {
			if (*spanp == c || p == s1_end) {
				return p - s1;
			}
		} while (spanp++ < (s2_end - 1));
		c = *++p;
	}
	/* NOTREACHED */
}
/* }}} */

/* {{{ Finds first occurrence of a string within another, case insensitive */
PHP_FUNCTION(stristr)
{
	zend_string *haystack, *needle;
	const char *found = NULL;
	size_t  found_offset;
	char *haystack_dup;
	char *orig_needle;
	bool part = 0;

	ZEND_PARSE_PARAMETERS_START(2, 3)
		Z_PARAM_STR(haystack)
		Z_PARAM_STR(needle)
		Z_PARAM_OPTIONAL
		Z_PARAM_BOOL(part)
	ZEND_PARSE_PARAMETERS_END();

	haystack_dup = estrndup(ZSTR_VAL(haystack), ZSTR_LEN(haystack));
	orig_needle = estrndup(ZSTR_VAL(needle), ZSTR_LEN(needle));
	found = php_stristr(haystack_dup, orig_needle, ZSTR_LEN(haystack), ZSTR_LEN(needle));
	efree(orig_needle);

	if (found) {
		found_offset = found - haystack_dup;
		if (part) {
			RETVAL_STRINGL(ZSTR_VAL(haystack), found_offset);
		} else {
			RETVAL_STRINGL(ZSTR_VAL(haystack) + found_offset, ZSTR_LEN(haystack) - found_offset);
		}
	} else {
		RETVAL_FALSE;
	}

	efree(haystack_dup);
}
/* }}} */

/* {{{ Finds first occurrence of a string within another */
PHP_FUNCTION(strstr)
{
	zend_string *haystack, *needle;
	const char *found = NULL;
	zend_long found_offset;
	bool part = 0;

	ZEND_PARSE_PARAMETERS_START(2, 3)
		Z_PARAM_STR(haystack)
		Z_PARAM_STR(needle)
		Z_PARAM_OPTIONAL
		Z_PARAM_BOOL(part)
	ZEND_PARSE_PARAMETERS_END();

	found = php_memnstr(ZSTR_VAL(haystack), ZSTR_VAL(needle), ZSTR_LEN(needle), ZSTR_VAL(haystack) + ZSTR_LEN(haystack));

	if (found) {
		found_offset = found - ZSTR_VAL(haystack);
		if (part) {
			RETURN_STRINGL(ZSTR_VAL(haystack), found_offset);
		} else {
			RETURN_STRINGL(found, ZSTR_LEN(haystack) - found_offset);
		}
	}
	RETURN_FALSE;
}
/* }}} */

/* {{{ Checks if a string contains another */
PHP_FUNCTION(str_contains)
{
	zend_string *haystack, *needle;

	ZEND_PARSE_PARAMETERS_START(2, 2)
		Z_PARAM_STR(haystack)
		Z_PARAM_STR(needle)
	ZEND_PARSE_PARAMETERS_END();

	RETURN_BOOL(php_memnstr(ZSTR_VAL(haystack), ZSTR_VAL(needle), ZSTR_LEN(needle), ZSTR_VAL(haystack) + ZSTR_LEN(haystack)));
}
/* }}} */

/* {{{ Checks if haystack starts with needle */
PHP_FUNCTION(str_starts_with)
{
	zend_string *haystack, *needle;

	ZEND_PARSE_PARAMETERS_START(2, 2)
		Z_PARAM_STR(haystack)
		Z_PARAM_STR(needle)
	ZEND_PARSE_PARAMETERS_END();

	if (ZSTR_LEN(needle) > ZSTR_LEN(haystack)) {
		RETURN_FALSE;
	}

	RETURN_BOOL(memcmp(ZSTR_VAL(haystack), ZSTR_VAL(needle), ZSTR_LEN(needle)) == 0);
}
/* }}} */

/* {{{ Checks if haystack ends with needle */
PHP_FUNCTION(str_ends_with)
{
	zend_string *haystack, *needle;

	ZEND_PARSE_PARAMETERS_START(2, 2)
		Z_PARAM_STR(haystack)
		Z_PARAM_STR(needle)
	ZEND_PARSE_PARAMETERS_END();

	if (ZSTR_LEN(needle) > ZSTR_LEN(haystack)) {
		RETURN_FALSE;
	}

	RETURN_BOOL(memcmp(
		ZSTR_VAL(haystack) + ZSTR_LEN(haystack) - ZSTR_LEN(needle),
		ZSTR_VAL(needle), ZSTR_LEN(needle)) == 0);
}
/* }}} */

/* {{{ An alias for strstr */
/* }}} */

/* {{{ Finds position of first occurrence of a string within another */
PHP_FUNCTION(strpos)
{
	zend_string *haystack, *needle;
	const char *found = NULL;
	zend_long offset = 0;

	ZEND_PARSE_PARAMETERS_START(2, 3)
		Z_PARAM_STR(haystack)
		Z_PARAM_STR(needle)
		Z_PARAM_OPTIONAL
		Z_PARAM_LONG(offset)
	ZEND_PARSE_PARAMETERS_END();

	if (offset < 0) {
		offset += (zend_long)ZSTR_LEN(haystack);
	}
	if (offset < 0 || (size_t)offset > ZSTR_LEN(haystack)) {
		zend_argument_value_error(3, "must be contained in argument #1 ($haystack)");
		RETURN_THROWS();
	}

	found = (char*)php_memnstr(ZSTR_VAL(haystack) + offset,
						ZSTR_VAL(needle), ZSTR_LEN(needle),
						ZSTR_VAL(haystack) + ZSTR_LEN(haystack));

	if (found) {
		RETURN_LONG(found - ZSTR_VAL(haystack));
	} else {
		RETURN_FALSE;
	}
}
/* }}} */

/* {{{ Finds position of first occurrence of a string within another, case insensitive */
PHP_FUNCTION(stripos)
{
	const char *found = NULL;
	zend_string *haystack, *needle;
	zend_long offset = 0;
	zend_string *needle_dup = NULL, *haystack_dup;

	ZEND_PARSE_PARAMETERS_START(2, 3)
		Z_PARAM_STR(haystack)
		Z_PARAM_STR(needle)
		Z_PARAM_OPTIONAL
		Z_PARAM_LONG(offset)
	ZEND_PARSE_PARAMETERS_END();

	if (offset < 0) {
		offset += (zend_long)ZSTR_LEN(haystack);
	}
	if (offset < 0 || (size_t)offset > ZSTR_LEN(haystack)) {
		zend_argument_value_error(3, "must be contained in argument #1 ($haystack)");
		RETURN_THROWS();
	}

	if (ZSTR_LEN(needle) > ZSTR_LEN(haystack)) {
		RETURN_FALSE;
	}

	haystack_dup = php_string_tolower(haystack);
	needle_dup = php_string_tolower(needle);
	found = (char*)php_memnstr(ZSTR_VAL(haystack_dup) + offset,
			ZSTR_VAL(needle_dup), ZSTR_LEN(needle_dup), ZSTR_VAL(haystack_dup) + ZSTR_LEN(haystack));

	if (found) {
		RETVAL_LONG(found - ZSTR_VAL(haystack_dup));
	} else {
		RETVAL_FALSE;
	}

	zend_string_release_ex(haystack_dup, 0);
	zend_string_release_ex(needle_dup, 0);
}
/* }}} */

/* {{{ Finds position of last occurrence of a string within another string */
PHP_FUNCTION(strrpos)
{
	zend_string *needle;
	zend_string *haystack;
	zend_long offset = 0;
	const char *p, *e, *found;

	ZEND_PARSE_PARAMETERS_START(2, 3)
		Z_PARAM_STR(haystack)
		Z_PARAM_STR(needle)
		Z_PARAM_OPTIONAL
		Z_PARAM_LONG(offset)
	ZEND_PARSE_PARAMETERS_END();

	if (offset >= 0) {
		if ((size_t)offset > ZSTR_LEN(haystack)) {
			zend_argument_value_error(3, "must be contained in argument #1 ($haystack)");
			RETURN_THROWS();
		}
		p = ZSTR_VAL(haystack) + (size_t)offset;
		e = ZSTR_VAL(haystack) + ZSTR_LEN(haystack);
	} else {
		if (offset < -ZEND_LONG_MAX || (size_t)(-offset) > ZSTR_LEN(haystack)) {
			zend_argument_value_error(3, "must be contained in argument #1 ($haystack)");
			RETURN_THROWS();
		}

		p = ZSTR_VAL(haystack);
		if ((size_t)-offset < ZSTR_LEN(needle)) {
			e = ZSTR_VAL(haystack) + ZSTR_LEN(haystack);
		} else {
			e = ZSTR_VAL(haystack) + ZSTR_LEN(haystack) + offset + ZSTR_LEN(needle);
		}
	}

	if ((found = zend_memnrstr(p, ZSTR_VAL(needle), ZSTR_LEN(needle), e))) {
		RETURN_LONG(found - ZSTR_VAL(haystack));…