Merge intl extension into core

ext/intl/CREDITS

@@ -0,0 +1,2 @@
+Internationalization
+Ed Batutis, Vladimir Iordanov, Dmitry Lakhtyuk,  Stanislav Malyshev, Vadim Savchuk, Kirti Velankar

ext/intl/TODO

@@ -0,0 +1,5 @@
+- Document U_* error constants
+- For IntlDateFormatter
+-- Accept and produce DateTime objects in 5.3 and 6
+-- Create convertor from ICU pattern to PHP pattern
+

ext/intl/collator/collator.c

@@ -0,0 +1,96 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "collator_class.h"
+#include "collator.h"
+
+#include <unicode/utypes.h>
+#include <unicode/ucol.h>
+#include <unicode/ustring.h>
+
+/* {{{ collator_register_constants
+ * Register constants common for the both (OO and procedural)
+ * APIs.
+ */
+void collator_register_constants( INIT_FUNC_ARGS )
+{
+	if( !Collator_ce_ptr )
+	{
+		zend_error( E_ERROR, "Collator class not defined" );
+		return;
+	}
+
+	#define COLLATOR_EXPOSE_CONST(x) REGISTER_LONG_CONSTANT(#x, x, CONST_CS)
+	#define COLLATOR_EXPOSE_CLASS_CONST(x) zend_declare_class_constant_long( Collator_ce_ptr, ZEND_STRS( #x ) - 1, UCOL_##x TSRMLS_CC );
+	#define COLLATOR_EXPOSE_CUSTOM_CLASS_CONST(name, value) zend_declare_class_constant_long( Collator_ce_ptr, ZEND_STRS( name ) - 1, value TSRMLS_CC );
+
+	// UColAttributeValue constants
+	COLLATOR_EXPOSE_CUSTOM_CLASS_CONST( "DEFAULT_VALUE", UCOL_DEFAULT );
+
+	COLLATOR_EXPOSE_CLASS_CONST( PRIMARY );
+	COLLATOR_EXPOSE_CLASS_CONST( SECONDARY );
+	COLLATOR_EXPOSE_CLASS_CONST( TERTIARY );
+	COLLATOR_EXPOSE_CLASS_CONST( DEFAULT_STRENGTH );
+	COLLATOR_EXPOSE_CLASS_CONST( QUATERNARY );
+	COLLATOR_EXPOSE_CLASS_CONST( IDENTICAL );
+
+	COLLATOR_EXPOSE_CLASS_CONST( OFF );
+	COLLATOR_EXPOSE_CLASS_CONST( ON );
+
+	COLLATOR_EXPOSE_CLASS_CONST( SHIFTED );
+	COLLATOR_EXPOSE_CLASS_CONST( NON_IGNORABLE );
+
+	COLLATOR_EXPOSE_CLASS_CONST( LOWER_FIRST );
+	COLLATOR_EXPOSE_CLASS_CONST( UPPER_FIRST );
+
+	// UColAttribute constants
+	COLLATOR_EXPOSE_CLASS_CONST( FRENCH_COLLATION );
+	COLLATOR_EXPOSE_CLASS_CONST( ALTERNATE_HANDLING );
+	COLLATOR_EXPOSE_CLASS_CONST( CASE_FIRST );
+	COLLATOR_EXPOSE_CLASS_CONST( CASE_LEVEL );
+	COLLATOR_EXPOSE_CLASS_CONST( NORMALIZATION_MODE );
+	COLLATOR_EXPOSE_CLASS_CONST( STRENGTH );
+	COLLATOR_EXPOSE_CLASS_CONST( HIRAGANA_QUATERNARY_MODE );
+	COLLATOR_EXPOSE_CLASS_CONST( NUMERIC_COLLATION );
+
+	// ULocDataLocaleType constants
+	COLLATOR_EXPOSE_CONST( ULOC_ACTUAL_LOCALE );
+	COLLATOR_EXPOSE_CONST( ULOC_VALID_LOCALE );
+
+	// sort flags
+	COLLATOR_EXPOSE_CUSTOM_CLASS_CONST( "SORT_REGULAR", COLLATOR_SORT_REGULAR );
+	COLLATOR_EXPOSE_CUSTOM_CLASS_CONST( "SORT_STRING",  COLLATOR_SORT_STRING  );
+	COLLATOR_EXPOSE_CUSTOM_CLASS_CONST( "SORT_NUMERIC", COLLATOR_SORT_NUMERIC );
+
+	#undef COLLATOR_EXPOSE_CUSTOM_CLASS_CONST
+	#undef COLLATOR_EXPOSE_CLASS_CONST
+	#undef COLLATOR_EXPOSE_CONST
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/collator/collator.h

@@ -0,0 +1,29 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef COLLATOR_COLLATOR_H
+#define CCOLLATOR_COLLATOR_H
+
+#include <php.h>
+
+#define COLLATOR_SORT_REGULAR   0
+#define COLLATOR_SORT_STRING    1
+#define COLLATOR_SORT_NUMERIC   2
+
+void collator_register_constants( INIT_FUNC_ARGS );
+
+#endif // COLLATOR_COLLATOR_H

ext/intl/collator/collator_attr.c

@@ -0,0 +1,157 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "collator_class.h"
+#include "collator_convert.h"
+#include "collator_attr.h"
+
+#include <unicode/ustring.h>
+
+/* {{{ proto int Collator::getAttribute( int $attr )
+ * Get collation attribute value. }}} */
+/* {{{ proto int collator_get_attribute( Collator $coll, int $attr )
+ * Get collation attribute value.
+ */
+PHP_FUNCTION( collator_get_attribute )
+{
+	long attribute, value;
+
+	COLLATOR_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Ol",
+		&object, Collator_ce_ptr, &attribute ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"collator_get_attribute: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	COLLATOR_METHOD_FETCH_OBJECT;
+
+	value = ucol_getAttribute( co->ucoll, attribute, COLLATOR_ERROR_CODE_P( co ) );
+	COLLATOR_CHECK_STATUS( co, "Error getting attribute value" );
+
+	RETURN_LONG( value );
+}
+/* }}} */
+
+/* {{{ proto bool Collator::getAttribute( int $attr )
+ * Get collation attribute value. }}} */
+/* {{{ proto bool collator_set_attribute( Collator $coll, int $attr, int $val )
+ * Set collation attribute.
+ */
+PHP_FUNCTION( collator_set_attribute )
+{
+	long attribute, value;
+	COLLATOR_METHOD_INIT_VARS
+
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Oll",
+		&object, Collator_ce_ptr, &attribute, &value ) == FAILURE)
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "collator_set_attribute: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	COLLATOR_METHOD_FETCH_OBJECT;
+
+	// Set new value for the given attribute.
+	ucol_setAttribute( co->ucoll, attribute, value, COLLATOR_ERROR_CODE_P( co ) );
+	COLLATOR_CHECK_STATUS( co, "Error setting attribute value" );
+
+	RETURN_TRUE;
+}
+/* }}} */
+
+/* {{{ proto int Collator::getStrength()
+ * Returns the current collation strength. }}} */
+/* {{{ proto int collator_get_strength(Collator coll)
+ * Returns the current collation strength.
+ */
+PHP_FUNCTION( collator_get_strength )
+{
+	COLLATOR_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O",
+		&object, Collator_ce_ptr ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "collator_get_strength: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	COLLATOR_METHOD_FETCH_OBJECT;
+
+	// Get current strength and return it.
+	RETURN_LONG( ucol_getStrength( co->ucoll ) );
+}
+/* }}} */
+
+/* {{{ proto bool Collator::setStrength(int strength)
+ * Set the collation strength. }}} */
+/* {{{ proto bool collator_set_strength(Collator coll, int strength)
+ * Set the collation strength.
+ */
+PHP_FUNCTION( collator_set_strength )
+{
+	long strength;
+
+	COLLATOR_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Ol",
+		&object, Collator_ce_ptr, &strength ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "collator_set_strength: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	COLLATOR_METHOD_FETCH_OBJECT;
+
+	// Set given strength.
+	ucol_setStrength( co->ucoll, strength );
+
+	RETURN_TRUE;
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/collator/collator_attr.h

@@ -0,0 +1,28 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef COLLATOR_ATTR_H
+#define CCOLLATOR_ATTR_H
+
+#include <php.h>
+
+PHP_FUNCTION( collator_get_attribute );
+PHP_FUNCTION( collator_set_attribute );
+PHP_FUNCTION( collator_get_strength );
+PHP_FUNCTION( collator_set_strength );
+
+#endif // COLLATOR_ATTR_H

ext/intl/collator/collator_class.c

@@ -0,0 +1,197 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#include "collator_class.h"
+#include "php_intl.h"
+#include "collator_attr.h"
+#include "collator_compare.h"
+#include "collator_sort.h"
+#include "collator_convert.h"
+#include "collator_locale.h"
+#include "collator_create.h"
+#include "collator_error.h"
+#include "intl_error.h"
+
+#include <unicode/ucol.h>
+
+zend_class_entry *Collator_ce_ptr = NULL;
+
+/////////////////////////////////////////////////////////////////////////////
+// Auxiliary functions needed by objects of 'Collator' class
+/////////////////////////////////////////////////////////////////////////////
+
+/* {{{ Collator_objects_dtor */
+static void Collator_objects_dtor(
+	void *object,
+	zend_object_handle handle TSRMLS_DC )
+{
+	zend_objects_destroy_object( object, handle TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ Collator_objects_free */
+void Collator_objects_free( zend_object *object TSRMLS_DC )
+{
+	Collator_object* co = (Collator_object*)object;
+
+	zend_object_std_dtor( &co->zo TSRMLS_CC );
+
+	collator_object_destroy( co TSRMLS_CC );
+
+	efree( co );
+}
+/* }}} */
+
+/* {{{ Collator_object_create */
+zend_object_value Collator_object_create(
+	zend_class_entry *ce TSRMLS_DC )
+{
+	zend_object_value    retval;
+	Collator_object*     intern;
+
+	intern = ecalloc( 1, sizeof(Collator_object) );
+	intl_error_init( COLLATOR_ERROR_P( intern ) TSRMLS_CC );
+	zend_object_std_init( &intern->zo, ce TSRMLS_CC );
+
+	retval.handle = zend_objects_store_put(
+		intern,
+		Collator_objects_dtor,
+		(zend_objects_free_object_storage_t)Collator_objects_free,
+		NULL TSRMLS_CC );
+
+	retval.handlers = zend_get_std_object_handlers();
+
+	return retval;
+}
+/* }}} */
+
+/////////////////////////////////////////////////////////////////////////////
+// 'Collator' class registration structures & functions
+/////////////////////////////////////////////////////////////////////////////
+
+/* {{{ Collator methods arguments info */
+// NOTE: modifying 'collator_XX_args' do not forget to
+//       modify approptiate 'collator_XX_args' for
+//       the procedural API.
+
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_0_args, 0, 0, 0 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_1_arg, 0, 0, 1 )
+	ZEND_ARG_INFO( 0, arg1 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_2_args, 0, 0, 2 )
+	ZEND_ARG_INFO( 0, arg1 )
+	ZEND_ARG_INFO( 0, arg2 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_sort_args, 0, 0, 1 )
+	ZEND_ARG_ARRAY_INFO( 1, arr, 0 )
+	ZEND_ARG_INFO( 0, flags )
+ZEND_END_ARG_INFO()
+
+/* }}} */
+
+/* {{{ Collator_class_functions
+ * Every 'Collator' class method has an entry in this table
+ */
+
+function_entry Collator_class_functions[] = {
+	PHP_ME( Collator, __construct, collator_1_arg, ZEND_ACC_PUBLIC|ZEND_ACC_CTOR )
+	ZEND_FENTRY( create, ZEND_FN( collator_create ), collator_1_arg, ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	PHP_NAMED_FE( compare, ZEND_FN( collator_compare ), collator_2_args )
+	PHP_NAMED_FE( sort, ZEND_FN( collator_sort ), collator_sort_args )
+	PHP_NAMED_FE( sortWithSortKeys, ZEND_FN( collator_sort_with_sort_keys ), collator_sort_args )
+	PHP_NAMED_FE( asort, ZEND_FN( collator_asort ), collator_sort_args )
+	PHP_NAMED_FE( getAttribute, ZEND_FN( collator_get_attribute ), collator_1_arg )
+	PHP_NAMED_FE( setAttribute, ZEND_FN( collator_set_attribute ), collator_2_args )
+	PHP_NAMED_FE( getStrength, ZEND_FN( collator_get_strength ), collator_0_args )
+	PHP_NAMED_FE( setStrength, ZEND_FN( collator_set_strength ), collator_1_arg )
+	PHP_NAMED_FE( getLocale, ZEND_FN( collator_get_locale ), collator_1_arg )
+	PHP_NAMED_FE( getErrorCode, ZEND_FN( collator_get_error_code ), collator_0_args )
+	PHP_NAMED_FE( getErrorMessage, ZEND_FN( collator_get_error_message ), collator_0_args )
+	{ NULL, NULL, NULL }
+};
+/* }}} */
+
+/* {{{ collator_register_Collator_class
+ * Initialize 'Collator' class
+ */
+void collator_register_Collator_class( TSRMLS_D )
+{
+	zend_class_entry ce;
+
+	// Create and register 'Collator' class.
+	INIT_CLASS_ENTRY( ce, "Collator", Collator_class_functions );
+	ce.create_object = Collator_object_create;
+	Collator_ce_ptr = zend_register_internal_class( &ce TSRMLS_CC );
+
+	// Declare 'Collator' class properties.
+	if( !Collator_ce_ptr )
+	{
+		zend_error( E_ERROR,
+			"Collator: attempt to create properties "
+			"on a non-registered class." );
+		return;
+	}
+}
+/* }}} */
+
+/* {{{ void collator_object_init( Collator_object* co )
+ * Initialize internals of Collator_object.
+ * Must be called before any other call to 'collator_object_...' functions.
+ */
+void collator_object_init( Collator_object* co TSRMLS_DC )
+{
+	if( !co )
+		return;
+
+	intl_error_init( COLLATOR_ERROR_P( co ) TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ void collator_object_destroy( Collator_object* co )
+ * Clean up mem allocted by internals of Collator_object
+ */
+void collator_object_destroy( Collator_object* co TSRMLS_DC )
+{
+	if( !co )
+		return;
+
+	if( co->ucoll )
+	{
+		ucol_close( co->ucoll );
+		co->ucoll = NULL;
+	}
+
+	intl_error_reset( COLLATOR_ERROR_P( co ) TSRMLS_CC );
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/collator/collator_class.h

@@ -0,0 +1,71 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef COLLATOR_CLASS_H
+#define COLLATOR_CLASS_H
+
+#include <php.h>
+
+#include "intl_common.h"
+#include "intl_error.h"
+
+#include <unicode/ucol.h>
+
+typedef struct {
+	zend_object     zo;
+
+	// ICU collator
+	UCollator*      ucoll;
+
+	// error handling
+	intl_error  err;
+
+} Collator_object;
+
+#define COLLATOR_ERROR(co) (co)->err
+#define COLLATOR_ERROR_P(co) &(COLLATOR_ERROR(co))
+
+#define COLLATOR_ERROR_CODE(co)   INTL_ERROR_CODE(COLLATOR_ERROR(co))
+#define COLLATOR_ERROR_CODE_P(co) &(INTL_ERROR_CODE(COLLATOR_ERROR(co)))
+
+void collator_register_Collator_class( TSRMLS_D );
+void collator_object_init( Collator_object* co TSRMLS_DC );
+void collator_object_destroy( Collator_object* co TSRMLS_DC );
+
+extern zend_class_entry *Collator_ce_ptr;
+
+/* Auxiliary macros */
+
+#define COLLATOR_METHOD_INIT_VARS       \
+    zval*             object  = NULL;   \
+    Collator_object*  co      = NULL;   \
+    intl_error_reset( NULL TSRMLS_CC ); \
+
+#define COLLATOR_METHOD_FETCH_OBJECT                                           \
+    co = (Collator_object *) zend_object_store_get_object( object TSRMLS_CC ); \
+    intl_error_reset( COLLATOR_ERROR_P( co ) TSRMLS_CC );                      \
+
+// Macro to check return value of a ucol_* function call.
+#define COLLATOR_CHECK_STATUS( co, msg )                                        \
+    intl_error_set_code( NULL, COLLATOR_ERROR_CODE( co ) TSRMLS_CC );           \
+    if( U_FAILURE( COLLATOR_ERROR_CODE( co ) ) )                                \
+    {                                                                           \
+        intl_errors_set_custom_msg( COLLATOR_ERROR_P( co ), msg, 0 TSRMLS_CC ); \
+        RETURN_FALSE;                                                           \
+    }                                                                           \
+
+#endif // #ifndef COLLATOR_CLASS_H

ext/intl/collator/collator_compare.c

@@ -0,0 +1,119 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "collator_class.h"
+#include "collator_compare.h"
+#include "intl_convert.h"
+
+/* {{{ proto int Collator::compare( string $str1, string $str2 )
+ * Compare two strings. }}} */
+/* {{{ proto int collator_compare( Collator $coll, string $str1, string $str2 )
+ * Compare two strings.
+ */
+PHP_FUNCTION( collator_compare )
+{
+	char*            str1      = NULL;
+	char*            str2      = NULL;
+	int              str1_len  = 0;
+	int              str2_len  = 0;
+
+	UChar*           ustr1     = NULL;
+	UChar*           ustr2     = NULL;
+	int              ustr1_len = 0;
+	int              ustr2_len = 0;
+
+	UCollationResult result;
+
+	COLLATOR_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Oss",
+		&object, Collator_ce_ptr, &str1, &str1_len, &str2, &str2_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "collator_compare: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	COLLATOR_METHOD_FETCH_OBJECT;
+
+
+	/*
+	 * Compare given strings (converting them to UTF-16 first).
+	 */
+
+	// First convert the strings to UTF-16.
+	intl_convert_utf8_to_utf16(
+		&ustr1, &ustr1_len, str1, str1_len, COLLATOR_ERROR_CODE_P( co ) );
+	if( U_FAILURE( COLLATOR_ERROR_CODE( co ) ) )
+	{
+		// Set global error code.
+		intl_error_set_code( NULL, COLLATOR_ERROR_CODE( co ) TSRMLS_CC );
+
+		// Set error messages.
+		intl_errors_set_custom_msg( COLLATOR_ERROR_P( co ),
+			"Error converting first argument to UTF-16", 0 TSRMLS_CC );
+		efree( ustr1 );
+		RETURN_FALSE;
+	}
+
+	intl_convert_utf8_to_utf16(
+		&ustr2, &ustr2_len, str2, str2_len, COLLATOR_ERROR_CODE_P( co ) );
+	if( U_FAILURE( COLLATOR_ERROR_CODE( co ) ) )
+	{
+		// Set global error code.
+		intl_error_set_code( NULL, COLLATOR_ERROR_CODE( co ) TSRMLS_CC );
+
+		// Set error messages.
+		intl_errors_set_custom_msg( COLLATOR_ERROR_P( co ),
+			"Error converting second argument to UTF-16", 0 TSRMLS_CC );
+		efree( ustr1 );
+		efree( ustr2 );
+		RETURN_FALSE;
+	}
+
+	// Then compare them.
+	result = ucol_strcoll(
+		co->ucoll,
+		ustr1, ustr1_len,
+		ustr2, ustr2_len );
+
+	if( ustr1 )
+		efree( ustr1 );
+	if( ustr2 )
+		efree( ustr2 );
+
+	// Return result of the comparison.
+	RETURN_LONG( result );
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/collator/collator_compare.h

@@ -0,0 +1,25 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef COLLATOR_COMPARE_H
+#define COLLATOR_COMPARE_H
+
+#include <php.h>
+
+PHP_FUNCTION( collator_compare );
+
+#endif // COLLATOR_COMPARE_H

ext/intl/collator/collator_convert.c

@@ -0,0 +1,482 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "collator_class.h"
+#include "collator_is_numeric.h"
+#include "collator_convert.h"
+#include "intl_convert.h"
+
+#include <unicode/ustring.h>
+#include <php.h>
+
+#if (PHP_MAJOR_VERSION == 5 && PHP_MINOR_VERSION <= 1)
+#define CAST_OBJECT_SHOULD_FREE ,0
+#else
+#define CAST_OBJECT_SHOULD_FREE
+#endif
+
+#define COLLATOR_CONVERT_RETURN_FAILED(retval) { \
+			zval_add_ref( &retval );             \
+			return retval;                       \
+	}
+
+/* {{{ collator_convert_hash_item_from_utf8_to_utf16 */
+static void collator_convert_hash_item_from_utf8_to_utf16(
+	HashTable* hash, int hashKeyType, char* hashKey, ulong hashIndex,
+	UErrorCode* status )
+{
+	const char* old_val;
+	int         old_val_len;
+	UChar*      new_val      = NULL;
+	int         new_val_len  = 0;
+	zval**      hashData     = NULL;
+	zval*       znew_val     = NULL;
+
+	// Get current hash item.
+	zend_hash_get_current_data( hash, (void**) &hashData );
+
+	// Process string values only.
+	if( Z_TYPE_P( *hashData ) != IS_STRING )
+		return;
+
+	old_val     = Z_STRVAL_P( *hashData );
+	old_val_len = Z_STRLEN_P( *hashData );
+
+	// Convert it from UTF-8 to UTF-16LE and save the result to new_val[_len].
+	intl_convert_utf8_to_utf16( &new_val, &new_val_len, old_val, old_val_len, status );
+	if( U_FAILURE( *status ) )
+		return;
+
+	// Update current hash item with the converted value.
+	MAKE_STD_ZVAL( znew_val );
+	ZVAL_STRINGL( znew_val, (char*)new_val, UBYTES(new_val_len), FALSE );
+
+	if( hashKeyType == HASH_KEY_IS_STRING )
+	{
+		zend_hash_update( hash, hashKey, strlen( hashKey ) + 1,
+			(void*) &znew_val, sizeof(zval*), NULL );
+	}
+	else // hashKeyType == HASH_KEY_IS_LONG
+	{
+		zend_hash_index_update( hash, hashIndex,
+			(void*) &znew_val, sizeof(zval*), NULL );
+	}
+}
+/* }}} */
+
+/* {{{ collator_convert_hash_item_from_utf16_to_utf8 */
+static void collator_convert_hash_item_from_utf16_to_utf8(
+	HashTable* hash, int hashKeyType, char* hashKey, ulong hashIndex,
+	UErrorCode* status )
+{
+	const char* old_val;
+	int         old_val_len;
+	char*       new_val      = NULL;
+	int         new_val_len  = 0;
+	zval**      hashData     = NULL;
+	zval*       znew_val     = NULL;
+
+	// Get current hash item.
+	zend_hash_get_current_data( hash, (void**) &hashData );
+
+	// Process string values only.
+	if( Z_TYPE_P( *hashData ) != IS_STRING )
+		return;
+
+	old_val     = Z_STRVAL_P( *hashData );
+	old_val_len = Z_STRLEN_P( *hashData );
+
+	// Convert it from UTF-16LE to UTF-8 and save the result to new_val[_len].
+	intl_convert_utf16_to_utf8( &new_val, &new_val_len,
+		(UChar*)old_val, UCHARS(old_val_len), status );
+	if( U_FAILURE( *status ) )
+		return;
+
+	// Update current hash item with the converted value.
+	MAKE_STD_ZVAL( znew_val );
+	ZVAL_STRINGL( znew_val, (char*)new_val, new_val_len, FALSE );
+
+	if( hashKeyType == HASH_KEY_IS_STRING )
+	{
+		zend_hash_update( hash, hashKey, strlen( hashKey ) + 1,
+			(void*) &znew_val, sizeof(zval*), NULL );
+	}
+	else // hashKeyType == HASH_KEY_IS_LONG
+	{
+		zend_hash_index_update( hash, hashIndex,
+			(void*) &znew_val, sizeof(zval*), NULL );
+	}
+}
+/* }}} */
+
+/* {{{ collator_convert_hash_from_utf8_to_utf16
+ *  Convert values of the given hash from UTF-8 encoding to UTF-16LE.
+ */
+void collator_convert_hash_from_utf8_to_utf16( HashTable* hash, UErrorCode* status )
+{
+	ulong    hashIndex    = 0;
+	char*    hashKey      = NULL;
+	int      hashKeyType  = 0;
+
+	zend_hash_internal_pointer_reset( hash );
+	while( ( hashKeyType = zend_hash_get_current_key( hash, &hashKey, &hashIndex, 0 ) )
+			!= HASH_KEY_NON_EXISTANT )
+	{
+		// Convert current hash item from UTF-8 to UTF-16LE.
+		collator_convert_hash_item_from_utf8_to_utf16(
+			hash, hashKeyType, hashKey, hashIndex, status );
+		if( U_FAILURE( *status ) )
+			return;
+
+		// Proceed to the next item.
+		zend_hash_move_forward( hash );
+	}
+}
+/* }}} */
+
+/* {{{ collator_convert_hash_from_utf16_to_utf8
+ * Convert values of the given hash from UTF-16LE encoding to UTF-8.
+ */
+void collator_convert_hash_from_utf16_to_utf8( HashTable* hash, UErrorCode* status )
+{
+	ulong    hashIndex    = 0;
+	char*    hashKey      = NULL;
+	int      hashKeyType  = 0;
+
+	zend_hash_internal_pointer_reset( hash );
+	while( ( hashKeyType = zend_hash_get_current_key( hash, &hashKey, &hashIndex, 0 ) )
+			!= HASH_KEY_NON_EXISTANT )
+	{
+		// Convert current hash item from UTF-16LE to UTF-8.
+		collator_convert_hash_item_from_utf16_to_utf8(
+			hash, hashKeyType, hashKey, hashIndex, status );
+		if( U_FAILURE( *status ) )
+			return;
+
+		// Proceed to the next item.
+		zend_hash_move_forward( hash );
+	}
+}
+/* }}} */
+
+/* {{{ collator_convert_zstr_utf16_to_utf8
+ *
+ * Convert string from utf16 to utf8.
+ *
+ * @param  zval* utf16_zval String to convert.
+ *
+ * @return zval* Converted string.
+ */
+zval* collator_convert_zstr_utf16_to_utf8( zval* utf16_zval )
+{
+	zval* utf8_zval   = NULL;
+	char* str         = NULL;
+	int   str_len     = 0;
+	UErrorCode status = U_ZERO_ERROR;
+
+	// Convert to utf8 then.
+	intl_convert_utf16_to_utf8( &str, &str_len,
+		(UChar*) Z_STRVAL_P(utf16_zval), UCHARS( Z_STRLEN_P(utf16_zval) ), &status );
+	if( U_FAILURE( status ) )
+		php_error( E_WARNING, "Error converting utf16 to utf8 in collator_convert_zval_utf16_to_utf8()" );
+
+	ALLOC_INIT_ZVAL( utf8_zval );
+	ZVAL_STRINGL( utf8_zval, str, str_len, FALSE );
+
+	return utf8_zval;
+}
+/* }}} */
+
+/* {{{ collator_convert_zstr_utf8_to_utf16
+ *
+ * Convert string from utf8 to utf16.
+ *
+ * @param  zval* utf8_zval String to convert.
+ *
+ * @return zval* Converted string.
+ */
+zval* collator_convert_zstr_utf8_to_utf16( zval* utf8_zval )
+{
+	zval* zstr        = NULL;
+	UChar* ustr       = NULL;
+	int    ustr_len   = 0;
+	UErrorCode status = U_ZERO_ERROR;
+
+	// Convert the string to UTF-16.
+	intl_convert_utf8_to_utf16(
+			&ustr, &ustr_len,
+			Z_STRVAL_P( utf8_zval ), Z_STRLEN_P( utf8_zval ),
+			&status );
+	if( U_FAILURE( status ) )
+		php_error( E_WARNING, "Error casting object to string in collator_convert_zstr_utf8_to_utf16()" );
+
+	// Set string.
+	ALLOC_INIT_ZVAL( zstr );
+	ZVAL_STRINGL( zstr, (char*)ustr, UBYTES(ustr_len), FALSE );
+
+	return zstr;
+}
+/* }}} */
+
+/* {{{ collator_convert_object_to_string
+ * Convert object to UTF16-encoded string.
+ */
+zval* collator_convert_object_to_string( zval* obj TSRMLS_DC )
+{
+	zval* zstr        = NULL;
+	UErrorCode status = U_ZERO_ERROR;
+	UChar* ustr       = NULL;
+	int    ustr_len   = 0;
+
+	// Bail out if it's not an object.
+	if( Z_TYPE_P( obj ) != IS_OBJECT )
+	{
+		COLLATOR_CONVERT_RETURN_FAILED( obj );
+	}
+
+	// Try object's handlers.
+	if( Z_OBJ_HT_P(obj)->get )
+	{
+		zstr = Z_OBJ_HT_P(obj)->get( obj TSRMLS_CC );
+
+		switch( Z_TYPE_P( zstr ) )
+		{
+			case IS_OBJECT:
+				{
+					// Bail out.
+					zval_ptr_dtor( &zstr );
+					COLLATOR_CONVERT_RETURN_FAILED( obj );
+				} break;
+
+			case IS_STRING:
+				break;
+
+			default:
+				{
+					convert_to_string( zstr );
+				} break;
+		}
+	}
+	else if( Z_OBJ_HT_P(obj)->cast_object )
+	{
+		ALLOC_INIT_ZVAL( zstr );
+
+		if( Z_OBJ_HT_P(obj)->cast_object( obj, zstr, IS_STRING CAST_OBJECT_SHOULD_FREE TSRMLS_CC ) == FAILURE )
+		{
+			// cast_object failed => bail out.
+			zval_ptr_dtor( &zstr );
+			COLLATOR_CONVERT_RETURN_FAILED( obj );
+		}
+	}
+
+	// Object wasn't successfuly converted => bail out.
+	if( zstr == NULL )
+	{
+		COLLATOR_CONVERT_RETURN_FAILED( obj );
+	}
+
+	// Convert the string to UTF-16.
+	intl_convert_utf8_to_utf16(
+			&ustr, &ustr_len,
+			Z_STRVAL_P( zstr ), Z_STRLEN_P( zstr ),
+			&status );
+	if( U_FAILURE( status ) )
+		php_error( E_WARNING, "Error casting object to string in collator_convert_object_to_string()" );
+
+	// Cleanup zstr to hold utf16 string.
+	zval_dtor( zstr );
+
+	// Set string.
+	ZVAL_STRINGL( zstr, (char*)ustr, UBYTES(ustr_len), FALSE );
+
+	// Don't free ustr cause it's set in zstr without copy.
+	// efree( ustr );
+
+	return zstr;
+}
+/* }}} */
+
+/* {{{ collator_convert_string_to_number
+ *
+ * Convert string to number.
+ *
+ * @param  zval* str String to convert.
+ *
+ * @return zval* Number. If str is not numeric string return number zero.
+ */
+zval* collator_convert_string_to_number( zval* str )
+{
+	zval* num = collator_convert_string_to_number_if_possible( str );
+	if( num == str )
+	{
+		// String wasn't converted => return zero.
+		zval_ptr_dtor( &num );
+
+		ALLOC_INIT_ZVAL( num );
+		ZVAL_LONG( num, 0 );
+	}
+
+	return num;
+}
+/* }}} */
+
+/* {{{ collator_convert_string_to_double
+ *
+ * Convert string to double.
+ *
+ * @param  zval* str String to convert.
+ *
+ * @return zval* Number. If str is not numeric string return number zero.
+ */
+zval* collator_convert_string_to_double( zval* str )
+{
+	zval* num = collator_convert_string_to_number( str );
+	if( Z_TYPE_P(num) == IS_LONG )
+	{
+		ZVAL_DOUBLE( num, Z_LVAL_P( num ) );
+	}
+
+	return num;
+}
+/* }}} */
+
+/* {{{ collator_convert_string_to_number_if_possible
+ *
+ * Convert string to numer.
+ *
+ * @param  zval* str String to convert.
+ *
+ * @return zval* Number if str is numeric string. Otherwise
+ *               original str param.
+ */
+zval* collator_convert_string_to_number_if_possible( zval* str )
+{
+	zval* num      = NULL;
+	int is_numeric = 0;
+	long lval      = 0;
+	double dval    = 0;
+
+	if( Z_TYPE_P( str ) != IS_STRING )
+	{
+		COLLATOR_CONVERT_RETURN_FAILED( str );
+	}
+
+	if( ( is_numeric = collator_is_numeric( (UChar*) Z_STRVAL_P(str), UCHARS( Z_STRLEN_P(str) ), &lval, &dval, 1 ) ) )
+	{
+		ALLOC_INIT_ZVAL( num );
+
+		if( is_numeric == IS_LONG )
+			Z_LVAL_P(num) = lval;
+		if( is_numeric == IS_DOUBLE )
+			Z_DVAL_P(num) = dval;
+
+		Z_TYPE_P(num) = is_numeric;
+	}
+	else
+	{
+		COLLATOR_CONVERT_RETURN_FAILED( str );
+	}
+
+	return num;
+}
+/* }}} */
+
+/* {{{ collator_make_printable_zval
+ *
+ * Returns string from input zval.
+ *
+ * @param  zval* arg zval to get string from
+ *
+ * @return zval* UTF16 string.
+ */
+zval* collator_make_printable_zval( zval* arg )
+{
+	zval arg_copy;
+	int use_copy = 0;
+	zval* str    = NULL;
+
+	if( Z_TYPE_P(arg) != IS_STRING )
+	{
+		zend_make_printable_zval(arg, &arg_copy, &use_copy);
+
+		if( use_copy )
+		{
+			str = collator_convert_zstr_utf8_to_utf16( &arg_copy );
+			zval_dtor( &arg_copy );
+		}
+		else
+		{
+			str = collator_convert_zstr_utf8_to_utf16( arg );
+		}
+	}
+	else
+	{
+		COLLATOR_CONVERT_RETURN_FAILED( arg );
+	}
+
+	return str;
+}
+/* }}} */
+
+/* {{{ collator_normalize_sort_argument
+ *
+ * Normalize argument to use in sort's compare function.
+ *
+ * @param  zval* arg Sort's argument to normalize.
+ *
+ * @return zval* Normalized copy of arg or unmodified arg
+ *               if normalization is not needed.
+ */
+zval* collator_normalize_sort_argument( zval* arg )
+{
+	zval* n_arg = NULL;
+
+	if( Z_TYPE_P( arg ) != IS_STRING )
+	{
+		// If its not a string then nothing to do.
+		// Return original arg.
+		COLLATOR_CONVERT_RETURN_FAILED( arg );
+	}
+
+	// Try convert to number.
+	n_arg = collator_convert_string_to_number_if_possible( arg );
+
+	if( n_arg == arg )
+	{
+		// Conversion to number failed.
+		zval_ptr_dtor( &n_arg );
+
+		// Convert string to utf8.
+		n_arg = collator_convert_zstr_utf16_to_utf8( arg );
+	}
+
+	return n_arg;
+}
+/* }}} */
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/collator/collator_convert.h

@@ -0,0 +1,38 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef COLLATOR_CONVERT_H
+#define COLLATOR_CONVERT_H
+
+#include <php.h>
+#include <unicode/utypes.h>
+
+void collator_convert_hash_from_utf8_to_utf16( HashTable* hash, UErrorCode* status );
+void collator_convert_hash_from_utf16_to_utf8( HashTable* hash, UErrorCode* status );
+
+zval* collator_convert_zstr_utf16_to_utf8( zval* utf16_zval );
+zval* collator_convert_zstr_utf8_to_utf16( zval* utf8_zval );
+
+zval* collator_normalize_sort_argument( zval* arg );
+zval* collator_convert_object_to_string( zval* obj TSRMLS_DC );
+zval* collator_convert_string_to_number( zval* arg );
+zval* collator_convert_string_to_number_if_possible( zval* str );
+zval* collator_convert_string_to_double( zval* str );
+
+zval* collator_make_printable_zval( zval* arg );
+
+#endif // COLLATOR_CONVERT_H

ext/intl/collator/collator_create.c

@@ -0,0 +1,135 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "collator_class.h"
+#include "collator_create.h"
+#include "intl_data.h"
+
+/* {{{ proto Collator collator_create( string $locale )
+ * Create collator.
+ */
+PHP_FUNCTION( collator_create )
+{
+	char*            locale;
+	int              locale_len = 0;
+	zval*            object;
+	Collator_object* co;
+
+	intl_error_reset( NULL TSRMLS_CC );
+
+	// Parse parameters.
+	if( zend_parse_parameters( ZEND_NUM_ARGS() TSRMLS_CC, "s",
+		&locale, &locale_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"collator_create: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_NULL();
+	}
+
+	INTL_CHECK_LOCALE_LEN(locale_len);
+	// Create a Collator object and save the ICU collator into it.
+	if( ( object = getThis() ) == NULL )
+		object = return_value;
+
+	if( Z_TYPE_P( object ) != IS_OBJECT )
+		object_init_ex( object, Collator_ce_ptr );
+
+	co = (Collator_object *) zend_object_store_get_object( object TSRMLS_CC );
+
+	intl_error_reset( COLLATOR_ERROR_P( co ) TSRMLS_CC );
+
+	if(locale_len == 0) {
+		locale = INTL_G(default_locale);
+	}
+
+	// Open ICU collator.
+	co->ucoll = ucol_open( locale, COLLATOR_ERROR_CODE_P( co ) );
+
+	if( U_FAILURE( COLLATOR_ERROR_CODE( co ) ) || co->ucoll == NULL )
+	{
+		intl_error_set( NULL, COLLATOR_ERROR_CODE( co ),
+			"collator_create: unable to open ICU collator", 0 TSRMLS_CC );
+
+		// Collator creation failed.
+		RETURN_NULL();
+	}
+}
+/* }}} */
+
+/* {{{ proto Collator Collator::__construct( string $locale )
+ * Collator object constructor.
+ */
+PHP_METHOD( Collator, __construct )
+{
+	char* locale     = NULL;
+	int   locale_len = 0;
+
+	COLLATOR_METHOD_INIT_VARS
+
+	object = getThis();
+	// Parse parameters.
+	if( zend_parse_parameters( ZEND_NUM_ARGS() TSRMLS_CC, "s",
+		&locale, &locale_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"__construct: unable to parse input params", 0 TSRMLS_CC );
+
+		zval_dtor(object);
+		ZVAL_NULL(object);
+		RETURN_NULL();
+	}
+
+	INTL_CHECK_LOCALE_LEN_OBJ(locale_len, object);
+	/* Fetch the object. */
+	co  = (Collator_object*) zend_object_store_get_object( object TSRMLS_CC );
+
+	intl_error_reset( COLLATOR_ERROR_P( co ) TSRMLS_CC );
+
+	if(locale_len == 0) {
+		locale = INTL_G(default_locale);
+	}
+
+	// Open ICU collator.
+	co->ucoll = ucol_open( locale, COLLATOR_ERROR_CODE_P( co ) );
+
+	if( U_FAILURE( COLLATOR_ERROR_CODE( co ) ) || co->ucoll == NULL )
+	{
+		intl_error_set( NULL, COLLATOR_ERROR_CODE( co ),
+			"__construct: unable to open ICU collator", 0 TSRMLS_CC );
+
+		zval_dtor(object);
+		ZVAL_NULL(object);
+		RETURN_NULL();
+	}
+
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/collator/collator_create.h

@@ -0,0 +1,27 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef COLLATOR_CREATE_H
+#define COLLATOR_CREATE_H
+
+#include <php.h>
+
+PHP_FUNCTION( collator_create );
+
+PHP_METHOD( Collator, __construct );
+
+#endif // COLLATOR_CREATE_H

ext/intl/collator/collator_error.c

@@ -0,0 +1,94 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "collator_class.h"
+#include "collator_error.h"
+
+/* {{{ proto int Collator::getErrorCode( Collator $coll )
+ * Get collator's last error code. }}} */
+/* {{{ proto int collator_get_error_code( Collator $coll )
+ * Get collator's last error code.
+ */
+PHP_FUNCTION( collator_get_error_code )
+{
+	COLLATOR_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O",
+		&object, Collator_ce_ptr ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"collator_get_error_code: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object (without resetting its last error code).
+	co = (Collator_object *) zend_object_store_get_object(object TSRMLS_CC);
+	if( co == NULL )
+		RETURN_FALSE;
+
+	// Return collator's last error code.
+	RETURN_LONG( COLLATOR_ERROR_CODE( co ) );
+}
+/* }}} */
+
+/* {{{ proto string Collator::getErrorMessage( Collator $coll )
+ * Get text description for collator's last error code. }}} */
+/* {{{ proto string collator_get_error_message( Collator $coll )
+ * Get text description for collator's last error code.
+ */
+PHP_FUNCTION( collator_get_error_message )
+{
+	const char* message = NULL;
+
+	COLLATOR_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O",
+		&object, Collator_ce_ptr ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"collator_get_error_message: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object (without resetting its last error code).
+	co = (Collator_object *) zend_object_store_get_object( object TSRMLS_CC );
+	if( co == NULL )
+		RETURN_FALSE;
+
+	// Return last error message.
+	message = intl_error_get_message( COLLATOR_ERROR_P( co ) TSRMLS_CC );
+	RETURN_STRING( (char*)message, FALSE );
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/collator/collator_error.h

@@ -0,0 +1,26 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef COLLATOR_ERROR_H
+#define COLLATOR_ERROR_H
+
+#include <php.h>
+
+PHP_FUNCTION( collator_get_error_code );
+PHP_FUNCTION( collator_get_error_message );
+
+#endif // COLLATOR_ERROR_H

ext/intl/collator/collator_is_numeric.c

@@ -0,0 +1,305 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#include "collator_is_numeric.h"
+
+#if ZEND_MODULE_API_NO < 20071006
+/* not 5.3 */
+#ifndef ALLOCA_FLAG
+#define ALLOCA_FLAG(use_heap)
+#endif
+#define _do_alloca(x, y) do_alloca((x))
+#define _free_alloca(x, y) free_alloca((x))
+#else
+#define _do_alloca do_alloca
+#define _free_alloca free_alloca
+#endif
+/* {{{ collator_u_strtod
+ * Taken from PHP6:zend_u_strtod()
+ */
+static double collator_u_strtod(const UChar *nptr, UChar **endptr) /* {{{ */
+{
+	const UChar *u = nptr, *nstart;
+	UChar c = *u;
+	int any = 0;
+	ALLOCA_FLAG(use_heap);
+
+	while (u_isspace(c)) {
+		c = *++u;
+	}
+	nstart = u;
+
+	if (c == 0x2D /*'-'*/ || c == 0x2B /*'+'*/) {
+		c = *++u;
+	}
+
+	while (c >= 0x30 /*'0'*/ && c <= 0x39 /*'9'*/) {
+		any = 1;
+		c = *++u;
+	}
+
+	if (c == 0x2E /*'.'*/) {
+		c = *++u;
+		while (c >= 0x30 /*'0'*/ && c <= 0x39 /*'9'*/) {
+			any = 1;
+			c = *++u;
+		}
+	}
+
+	if ((c == 0x65 /*'e'*/ || c == 0x45 /*'E'*/) && any) {
+		const UChar *e = u;
+		int any_exp = 0;
+
+		c = *++u;
+		if (c == 0x2D /*'-'*/ || c == 0x2B /*'+'*/) {
+			c = *++u;
+		}
+
+		while (c >= 0x30 /*'0'*/ && c <= 0x39 /*'9'*/) {
+			any_exp = 1;
+			c = *++u;
+		}
+
+		if (!any_exp) {
+			u = e;
+		}
+	}
+
+	if (any) {
+		char buf[64], *numbuf, *bufpos;
+		int length = u - nstart;
+		double value;
+
+		if (length < sizeof(buf)) {
+			numbuf = buf;
+		} else {
+			numbuf = (char *) _do_alloca(length + 1, use_heap);
+		}
+
+		bufpos = numbuf;
+
+		while (nstart < u) {
+			*bufpos++ = (char) *nstart++;
+		}
+
+		*bufpos = '\0';
+		value = zend_strtod(numbuf, NULL);
+
+		if (numbuf != buf) {
+			_free_alloca(numbuf, use_heap);
+		}
+
+		if (endptr != NULL) {
+			*endptr = (UChar *)u;
+		}
+
+		return value;
+	}
+
+	if (endptr != NULL) {
+		*endptr = (UChar *)nptr;
+	}
+
+	return 0;
+}
+/* }}} */
+
+/* {{{ collator_u_strtol
+ * Taken from PHP6:zend_u_strtol()
+ *
+ * Convert a Unicode string to a long integer.
+ *
+ * Ignores `locale' stuff.
+ */
+static long collator_u_strtol(nptr, endptr, base)
+	const UChar *nptr;
+	UChar **endptr;
+	register int base;
+{
+	register const UChar *s = nptr;
+	register unsigned long acc;
+	register UChar c;
+	register unsigned long cutoff;
+	register int neg = 0, any, cutlim;
+
+	if (s == NULL) {
+		errno = ERANGE;
+		if (endptr != NULL) {
+			*endptr = NULL;
+		}
+		return 0;
+	}
+
+	/*
+	 * Skip white space and pick up leading +/- sign if any.
+	 * If base is 0, allow 0x for hex and 0 for octal, else
+	 * assume decimal; if base is already 16, allow 0x.
+	 */
+	do {
+		c = *s++;
+	} while (u_isspace(c));
+	if (c == 0x2D /*'-'*/) {
+		neg = 1;
+		c = *s++;
+	} else if (c == 0x2B /*'+'*/)
+		c = *s++;
+	if ((base == 0 || base == 16) &&
+	    (c == 0x30 /*'0'*/)
+		 && (*s == 0x78 /*'x'*/ || *s == 0x58 /*'X'*/)) {
+		c = s[1];
+		s += 2;
+		base = 16;
+	}
+	if (base == 0)
+		base = (c == 0x30 /*'0'*/) ? 8 : 10;
+
+	/*
+	 * Compute the cutoff value between legal numbers and illegal
+	 * numbers.  That is the largest legal value, divided by the
+	 * base.  An input number that is greater than this value, if
+	 * followed by a legal input character, is too big.  One that
+	 * is equal to this value may be valid or not; the limit
+	 * between valid and invalid numbers is then based on the last
+	 * digit.  For instance, if the range for longs is
+	 * [-2147483648..2147483647] and the input base is 10,
+	 * cutoff will be set to 214748364 and cutlim to either
+	 * 7 (neg==0) or 8 (neg==1), meaning that if we have accumulated
+	 * a value > 214748364, or equal but the next digit is > 7 (or 8),
+	 * the number is too big, and we will return a range error.
+	 *
+	 * Set any if any `digits' consumed; make it negative to indicate
+	 * overflow.
+	 */
+	cutoff = neg ? -(unsigned long)LONG_MIN : LONG_MAX;
+	cutlim = cutoff % (unsigned long)base;
+	cutoff /= (unsigned long)base;
+	for (acc = 0, any = 0;; c = *s++) {
+		if (c >= 0x30 /*'0'*/ && c <= 0x39 /*'9'*/)
+			c -= 0x30 /*'0'*/;
+		else if (c >= 0x41 /*'A'*/ && c <= 0x5A /*'Z'*/)
+			c -= 0x41 /*'A'*/ - 10;
+		else if (c >= 0x61 /*'a'*/ && c <= 0x7A /*'z'*/)
+			c -= 0x61 /*'a'*/ - 10;
+		else
+			break;
+		if (c >= base)
+			break;
+
+		if (any < 0 || acc > cutoff || (acc == cutoff && c > cutlim))
+			any = -1;
+		else {
+			any = 1;
+			acc *= base;
+			acc += c;
+		}
+	}
+	if (any < 0) {
+		acc = neg ? LONG_MIN : LONG_MAX;
+		errno = ERANGE;
+	} else if (neg)
+		acc = -acc;
+	if (endptr != NULL)
+		*endptr = (UChar *)(any ? s - 1 : nptr);
+	return (acc);
+}
+/* }}} */
+
+
+/* {{{ collator_is_numeric]
+ * Taken from PHP6:is_numeric_unicode()
+ */
+zend_uchar collator_is_numeric( UChar *str, int length, long *lval, double *dval, int allow_errors )
+{
+	long local_lval;
+	double local_dval;
+	UChar *end_ptr_long, *end_ptr_double;
+	int conv_base=10;
+
+	if (!length) {
+		return 0;
+	}
+
+	/* handle hex numbers */
+	if (length>=2 && str[0]=='0' && (str[1]=='x' || str[1]=='X')) {
+		conv_base=16;
+	}
+
+	errno=0;
+	local_lval = collator_u_strtol(str, &end_ptr_long, conv_base);
+	if (errno != ERANGE) {
+		if (end_ptr_long == str+length) { /* integer string */
+			if (lval) {
+				*lval = local_lval;
+			}
+			return IS_LONG;
+		} else if (end_ptr_long == str && *end_ptr_long != '\0' && *str != '.' && *str != '-') { /* ignore partial string matches */
+			return 0;
+		}
+	} else {
+		end_ptr_long = NULL;
+	}
+
+	if (conv_base == 16) { /* hex string, under UNIX strtod() messes it up */
+		/* UTODO: keep compatibility with is_numeric_string() here? */
+		return 0;
+	}
+
+	local_dval = collator_u_strtod(str, &end_ptr_double);
+	if (local_dval == 0 && end_ptr_double == str) {
+		end_ptr_double = NULL;
+	} else {
+		if (end_ptr_double == str+length) { /* floating point string */
+			if (!zend_finite(local_dval)) {
+				/* "inf","nan" and maybe other weird ones */
+				return 0;
+			}
+
+			if (dval) {
+				*dval = local_dval;
+			}
+			return IS_DOUBLE;
+		}
+	}
+
+	if (!allow_errors) {
+		return 0;
+	}
+	if (allow_errors == -1) {
+		zend_error(E_NOTICE, "A non well formed numeric value encountered");
+	}
+
+	if (allow_errors) {
+		if (end_ptr_double > end_ptr_long && dval) {
+			*dval = local_dval;
+			return IS_DOUBLE;
+		} else if (end_ptr_long && lval) {
+			*lval = local_lval;
+			return IS_LONG;
+		}
+	}
+	return 0;
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/collator/collator_is_numeric.h

@@ -0,0 +1,26 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef COLLATOR_IS_NUMERIC_H
+#define COLLATOR_IS_NUMERIC_H
+
+#include <php.h>
+#include <unicode/uchar.h>
+
+zend_uchar collator_is_numeric( UChar *str, int length, long *lval, double *dval, int allow_errors );
+
+#endif // COLLATOR_IS_NUMERIC_H

ext/intl/collator/collator_locale.c

@@ -0,0 +1,71 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "collator_class.h"
+#include "collator_locale.h"
+#include "intl_convert.h"
+
+#include <zend_API.h>
+
+/* {{{ proto string Collator::getLocale( int $type )
+ * Gets the locale name of the collator. }}} */
+/* {{{ proto string collator_get_locale( Collator $coll, int $type )
+ * Gets the locale name of the collator.
+ */
+PHP_FUNCTION( collator_get_locale )
+{
+	int    type        = 0;
+	char*  locale_name = NULL;
+
+	COLLATOR_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Ol",
+		&object, Collator_ce_ptr, &type ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"collator_get_locale: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	COLLATOR_METHOD_FETCH_OBJECT;
+
+	// Get locale by specified type.
+	locale_name = (char*) ucol_getLocaleByType(
+		co->ucoll, type, COLLATOR_ERROR_CODE_P( co ) );
+	COLLATOR_CHECK_STATUS( co, "Error getting locale by type" );
+
+	// Return it.
+	RETVAL_STRINGL( locale_name, strlen(locale_name), TRUE );
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/collator/collator_locale.h

@@ -0,0 +1,25 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef COLLATOR_LOCALE_H
+#define COLLATOR_LOCALE_H
+
+#include <php.h>
+
+PHP_FUNCTION( collator_get_locale );
+
+#endif // COLLATOR_LOCALE_H

ext/intl/collator/collator_sort.c

@@ -0,0 +1,532 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "collator.h"
+#include "collator_class.h"
+#include "collator_sort.h"
+#include "collator_convert.h"
+#include "intl_convert.h"
+
+#if !defined(HAVE_PTRDIFF_T) && !defined(_PTRDIFF_T_DEFINED)
+typedef long ptrdiff_t;
+#endif
+
+/**
+ * Declare 'index' which will point to sort key in sort key
+ * buffer.
+ */
+typedef struct _collator_sort_key_index {
+	char* key;       // pointer to sort key
+	zval** zstr;     // pointer to original string(hash-item)
+} collator_sort_key_index_t;
+
+ZEND_EXTERN_MODULE_GLOBALS( intl )
+
+static const size_t DEF_SORT_KEYS_BUF_SIZE = 1048576;
+static const size_t DEF_SORT_KEYS_BUF_INCREMENT = 1048576;
+
+static const size_t DEF_SORT_KEYS_INDX_BUF_SIZE = 1048576;
+static const size_t DEF_SORT_KEYS_INDX_BUF_INCREMENT = 1048576;
+
+static const size_t DEF_UTF16_BUF_SIZE = 1024;
+
+/* {{{ collator_regular_compare_function */
+static int collator_regular_compare_function(zval *result, zval *op1, zval *op2 TSRMLS_DC)
+{
+	Collator_object* co = NULL;
+
+	int rc      = SUCCESS;
+
+	zval* str1  = collator_convert_object_to_string( op1 TSRMLS_CC );
+	zval* str2  = collator_convert_object_to_string( op2 TSRMLS_CC );
+
+	zval* num1  = NULL;
+	zval* num2  = NULL;
+	zval* norm1 = NULL;
+	zval* norm2 = NULL;
+
+	// If both args are strings AND either of args is not numeric string
+	// then use ICU-compare. Otherwise PHP-compare.
+	if( Z_TYPE_P(str1) == IS_STRING && Z_TYPE_P(str2) == IS_STRING &&
+		( str1 == ( num1 = collator_convert_string_to_number_if_possible( str1 ) ) ||
+		  str2 == ( num2 = collator_convert_string_to_number_if_possible( str2 ) ) ) )
+	{
+		// Fetch collator object.
+		co = (Collator_object *) zend_object_store_get_object( INTL_G(current_collator) TSRMLS_CC );
+
+		// Compare the strings using ICU.
+		result->value.lval = ucol_strcoll(
+				co->ucoll,
+				INTL_Z_STRVAL_P(str1), INTL_Z_STRLEN_P(str1),
+				INTL_Z_STRVAL_P(str2), INTL_Z_STRLEN_P(str2) );
+		result->type = IS_LONG;
+	}
+	else
+	{
+		// num1 is set if str1 and str2 are strings.
+		if( num1 )
+		{
+			if( num1 == str1 )
+			{
+				// str1 is string but not numeric string
+				// just convert it to utf8.
+				norm1 = collator_convert_zstr_utf16_to_utf8( str1 );
+
+				// num2 is not set but str2 is string => do normalization.
+				norm2 = collator_normalize_sort_argument( str2 );
+			}
+			else
+			{
+				// str1 is numeric strings => passthru to PHP-compare.
+				zval_add_ref( &num1 );
+				norm1 = num1;
+
+				// str2 is numeric strings => passthru to PHP-compare.
+				zval_add_ref( &num2 );
+				norm2 = num2;
+			}
+		}
+		else
+		{
+			// num1 is not set if str1 or str2 is not a string => do normalization.
+			norm1 = collator_normalize_sort_argument( str1 );
+
+			// if num1 is not set then num2 is not set as well => do normalization.
+			norm2 = collator_normalize_sort_argument( str2 );
+		}
+
+		rc = compare_function( result, norm1, norm2 TSRMLS_CC );
+
+		zval_ptr_dtor( &norm1 );
+		zval_ptr_dtor( &norm2 );
+	}
+
+	if( num1 )
+		zval_ptr_dtor( &num1 );
+
+	if( num2 )
+		zval_ptr_dtor( &num2 );
+
+	zval_ptr_dtor( &str1 );
+	zval_ptr_dtor( &str2 );
+
+	return rc;
+}
+/* }}} */
+
+/* {{{ collator_numeric_compare_function
+ * Convert input args to double and compare it.
+ */
+static int collator_numeric_compare_function(zval *result, zval *op1, zval *op2 TSRMLS_DC)
+{
+	int rc     = SUCCESS;
+	zval* num1 = NULL;
+	zval* num2 = NULL;
+
+	if( Z_TYPE_P(op1) == IS_STRING )
+	{
+		num1 = collator_convert_string_to_double( op1 );
+		op1 = num1;
+	}
+
+	if( Z_TYPE_P(op2) == IS_STRING )
+	{
+		num2 = collator_convert_string_to_double( op2 );
+		op2 = num2;
+	}
+
+	rc = numeric_compare_function( result, op1, op2 TSRMLS_CC);
+
+	if( num1 )
+		zval_ptr_dtor( &num1 );
+	if( num2 )
+		zval_ptr_dtor( &num2 );
+
+	return rc;
+}
+/* }}} */
+
+/* {{{ collator_icu_compare_function
+ * Direct use of ucol_strcoll.
+*/
+static int collator_icu_compare_function(zval *result, zval *op1, zval *op2 TSRMLS_DC)
+{
+	int rc              = SUCCESS;
+	Collator_object* co = NULL;
+	zval* str1          = NULL;
+	zval* str2          = NULL;
+
+	str1 = collator_make_printable_zval( op1 );
+	str2 = collator_make_printable_zval( op2 );
+
+	// Fetch collator object.
+	co = (Collator_object *) zend_object_store_get_object( INTL_G(current_collator) TSRMLS_CC );
+
+	// Compare the strings using ICU.
+	result->value.lval = ucol_strcoll(
+			co->ucoll,
+			INTL_Z_STRVAL_P(str1), INTL_Z_STRLEN_P(str1),
+			INTL_Z_STRVAL_P(str2), INTL_Z_STRLEN_P(str2) );
+	result->type = IS_LONG;
+
+	zval_ptr_dtor( &str1 );
+	zval_ptr_dtor( &str2 );
+
+	return rc;
+}
+/* }}} */
+
+/* {{{ collator_compare_func
+ * Taken from PHP5 source (array_data_compare).
+ */
+static int collator_compare_func( const void* a, const void* b TSRMLS_DC )
+{
+	Bucket *f;
+	Bucket *s;
+	zval result;
+	zval *first;
+	zval *second;
+
+	f = *((Bucket **) a);
+	s = *((Bucket **) b);
+
+	first = *((zval **) f->pData);
+	second = *((zval **) s->pData);
+
+	if( INTL_G(compare_func)( &result, first, second TSRMLS_CC) == FAILURE )
+		return 0;
+
+	if( Z_TYPE(result) == IS_DOUBLE )
+	{
+		if( Z_DVAL(result) < 0 )
+			return -1;
+		else if( Z_DVAL(result) > 0 )
+			return 1;
+		else
+			return 0;
+	}
+
+	convert_to_long(&result);
+
+	if( Z_LVAL(result) < 0 )
+		return -1;
+	else if( Z_LVAL(result) > 0 )
+		return 1;
+
+	return 0;
+}
+/* }}} */
+
+/* {{{ collator_cmp_sort_keys
+ * Compare sort keys
+ */
+static int collator_cmp_sort_keys( const void *p1, const void *p2 TSRMLS_DC )
+{
+	char* key1 = ((collator_sort_key_index_t*)p1)->key;
+	char* key2 = ((collator_sort_key_index_t*)p2)->key;
+
+	return strcmp( key1, key2 );
+}
+/* }}} */
+
+/* {{{ collator_get_compare_function
+ * Choose compare function according to sort flags.
+ */
+static collator_compare_func_t collator_get_compare_function( const long sort_flags )
+{
+	collator_compare_func_t func;
+
+	switch( sort_flags )
+	{
+		case COLLATOR_SORT_NUMERIC:
+			func = collator_numeric_compare_function;
+			break;
+
+		case COLLATOR_SORT_STRING:
+			func = collator_icu_compare_function;
+			break;
+
+		case COLLATOR_SORT_REGULAR:
+		default:
+			func = collator_regular_compare_function;
+			break;
+	}
+
+	return func;
+}
+/* }}} */
+
+/* {{{ collator_sort_internal
+ * Common code shared by collator_sort() and collator_asort() API functions.
+ */
+static void collator_sort_internal( int renumber, INTERNAL_FUNCTION_PARAMETERS )
+{
+	zval*          array            = NULL;
+	HashTable*     hash             = NULL;
+	zval*          saved_collator   = NULL;
+	long           sort_flags       = COLLATOR_SORT_REGULAR;
+
+	COLLATOR_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Oa|l",
+		&object, Collator_ce_ptr, &array, &sort_flags ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"collator_sort_internal: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	COLLATOR_METHOD_FETCH_OBJECT;
+
+	// Set 'compare function' according to sort flags.
+	INTL_G(compare_func) = collator_get_compare_function( sort_flags );
+
+	hash = HASH_OF( array );
+
+	// Convert strings in the specified array from UTF-8 to UTF-16.
+	collator_convert_hash_from_utf8_to_utf16( hash, COLLATOR_ERROR_CODE_P( co ) );
+	COLLATOR_CHECK_STATUS( co, "Error converting hash from UTF-8 to UTF-16" );
+
+	// Save specified collator in the request-global (?) variable.
+	saved_collator = INTL_G( current_collator );
+	INTL_G( current_collator ) = object;
+
+	// Sort specified array.
+	zend_hash_sort( hash, zend_qsort, collator_compare_func, renumber TSRMLS_CC );
+
+	// Restore saved collator.
+	INTL_G( current_collator ) = saved_collator;
+
+	// Convert strings in the specified array back to UTF-8.
+	collator_convert_hash_from_utf16_to_utf8( hash, COLLATOR_ERROR_CODE_P( co ) );
+	COLLATOR_CHECK_STATUS( co, "Error converting hash from UTF-16 to UTF-8" );
+
+	RETURN_TRUE;
+}
+/* }}} */
+
+/* {{{ proto bool Collator::sort( Collator $coll, array(string) $arr [, int $sort_flags] )
+ * Sort array using specified collator. }}} */
+/* {{{ proto bool collator_sort(  Collator $coll, array(string) $arr [, int $sort_flags] )
+ * Sort array using specified collator.
+ */
+PHP_FUNCTION( collator_sort )
+{
+	collator_sort_internal( TRUE, INTERNAL_FUNCTION_PARAM_PASSTHRU );
+}
+/* }}} */
+
+/* {{{ proto bool Collator::sortWithSortKeys( Collator $coll, array(string) $arr )
+ * Equivalent to standard PHP sort using Collator.
+ * Uses ICU ucol_getSortKey for performance. }}} */
+/* {{{ proto bool collator_sort_with_sort_keys( Collator $coll, array(string) $arr )
+ * Equivalent to standard PHP sort using Collator.
+ * Uses ICU ucol_getSortKey for performance.
+ */
+PHP_FUNCTION( collator_sort_with_sort_keys )
+{
+	zval*       array                = NULL;
+	HashTable*  hash                 = NULL;
+	zval**      hashData             = NULL;                     // currently processed item of input hash
+
+	char*       sortKeyBuf           = NULL;                     // buffer to store sort keys
+	uint32_t    sortKeyBufSize       = DEF_SORT_KEYS_BUF_SIZE;   // buffer size
+	ptrdiff_t   sortKeyBufOffset     = 0;                        // pos in buffer to store sort key
+	int32_t     sortKeyLen           = 0;                        // the length of currently processing key
+	uint32_t    bufLeft              = 0;
+	uint32_t    bufIncrement         = 0;
+
+	collator_sort_key_index_t* sortKeyIndxBuf = NULL;            // buffer to store 'indexes' which will be passed to 'qsort'
+	uint32_t    sortKeyIndxBufSize   = DEF_SORT_KEYS_INDX_BUF_SIZE;
+	uint32_t    sortKeyIndxSize      = sizeof( collator_sort_key_index_t );
+
+	uint32_t    sortKeyCount         = 0;
+	uint32_t    j                    = 0;
+
+	UChar*      utf16_buf            = NULL;                     // tmp buffer to hold current processing string in utf-16
+	int         utf16_buf_size       = DEF_UTF16_BUF_SIZE;       // the length of utf16_buf
+	int         utf16_len            = 0;                        // length of converted string
+
+	HashTable* sortedHash            = NULL;
+
+	COLLATOR_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Oa",
+		&object, Collator_ce_ptr, &array ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"collator_sort_with_sort_keys: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	COLLATOR_METHOD_FETCH_OBJECT;
+
+
+	/*
+	 * Sort specified array.
+	 */
+	hash = HASH_OF( array );
+
+	if( !hash || zend_hash_num_elements( hash ) == 0 )
+		RETURN_TRUE;
+
+	// Create bufers
+	sortKeyBuf     = ecalloc( sortKeyBufSize,     sizeof( char    ) );
+	sortKeyIndxBuf = ecalloc( sortKeyIndxBufSize, sizeof( uint8_t ) );
+	utf16_buf      = eumalloc( utf16_buf_size );
+
+	// Iterate through input hash and create a sort key for each value.
+	zend_hash_internal_pointer_reset( hash );
+	while( zend_hash_get_current_data( hash, (void**) &hashData ) == SUCCESS )
+	{
+		// Convert current hash item from UTF-8 to UTF-16LE and save the result to utf16_buf.
+
+		utf16_len = utf16_buf_size;
+
+		// Process string values only.
+		if( Z_TYPE_PP( hashData ) == IS_STRING )
+		{
+			intl_convert_utf8_to_utf16( &utf16_buf, &utf16_len, Z_STRVAL_PP( hashData ), Z_STRLEN_PP( hashData ), COLLATOR_ERROR_CODE_P( co ) );
+
+			if( U_FAILURE( COLLATOR_ERROR_CODE( co ) ) )
+			{
+				intl_error_set_code( NULL, COLLATOR_ERROR_CODE( co ) TSRMLS_CC );
+				intl_errors_set_custom_msg( COLLATOR_ERROR_P( co ), "Sort with sort keys failed", 0 TSRMLS_CC );
+
+				if( utf16_buf )
+					efree( utf16_buf );
+
+				efree( sortKeyIndxBuf );
+				efree( sortKeyBuf );
+
+				RETURN_FALSE;
+			}
+		}
+		else
+		{
+			// Set empty string
+			utf16_len = 0;
+			utf16_buf[utf16_len] = 0;
+		}
+
+		if( (utf16_len + 1) > utf16_buf_size )
+			utf16_buf_size = utf16_len + 1;
+
+		// Get sort key, reallocating the buffer if needed.
+		bufLeft = sortKeyBufSize - sortKeyBufOffset;
+
+		sortKeyLen = ucol_getSortKey( co->ucoll,
+									  utf16_buf,
+									  utf16_len,
+									  (uint8_t*)sortKeyBuf + sortKeyBufOffset,
+									  bufLeft );
+
+		// check for sortKeyBuf overflow, increasing its size of the buffer if needed
+		if( sortKeyLen > bufLeft )
+		{
+			bufIncrement = ( sortKeyLen > DEF_SORT_KEYS_BUF_INCREMENT ) ? sortKeyLen : DEF_SORT_KEYS_BUF_INCREMENT;
+
+			sortKeyBufSize += bufIncrement;
+			bufLeft += bufIncrement;
+
+			sortKeyBuf = erealloc( sortKeyBuf, sortKeyBufSize );
+
+			sortKeyLen = ucol_getSortKey( co->ucoll, utf16_buf, utf16_len, (uint8_t*)sortKeyBuf + sortKeyBufOffset, bufLeft );
+		}
+
+		// check sortKeyIndxBuf overflow, increasing its size of the buffer if needed
+		if( ( sortKeyCount + 1 ) * sortKeyIndxSize > sortKeyIndxBufSize )
+		{
+			bufIncrement = ( sortKeyIndxSize > DEF_SORT_KEYS_INDX_BUF_INCREMENT ) ? sortKeyIndxSize : DEF_SORT_KEYS_INDX_BUF_INCREMENT;
+
+			sortKeyIndxBufSize += bufIncrement;
+
+			sortKeyIndxBuf = erealloc( sortKeyIndxBuf, sortKeyIndxBufSize );
+		}
+
+		sortKeyIndxBuf[sortKeyCount].key = (char*)sortKeyBufOffset;    // remeber just offset, cause address
+		                                                               // of 'sortKeyBuf' may be changed due to realloc.
+		sortKeyIndxBuf[sortKeyCount].zstr = hashData;
+
+		sortKeyBufOffset += sortKeyLen;
+		++sortKeyCount;
+
+		zend_hash_move_forward( hash );
+	}
+
+	// update ptrs to point to valid keys.
+	for( j = 0; j < sortKeyCount; j++ )
+		sortKeyIndxBuf[j].key = sortKeyBuf + (ptrdiff_t)sortKeyIndxBuf[j].key;
+
+	// sort it
+	zend_qsort( sortKeyIndxBuf, sortKeyCount, sortKeyIndxSize, collator_cmp_sort_keys TSRMLS_CC );
+
+	// for resulting hash we'll assign new hash keys rather then reordering
+	ALLOC_HASHTABLE( sortedHash );
+	zend_hash_init( sortedHash, 0, NULL, ZVAL_PTR_DTOR, 0 );
+
+	for( j = 0; j < sortKeyCount; j++ )
+	{
+		zval_add_ref( sortKeyIndxBuf[j].zstr );
+		zend_hash_next_index_insert( sortedHash, sortKeyIndxBuf[j].zstr, sizeof(zval **), NULL );
+	}
+
+	// Save sorted hash into return variable.
+	zval_dtor( array );
+	(array)->value.ht = sortedHash;
+	(array)->type = IS_ARRAY;
+
+	if( utf16_buf )
+		efree( utf16_buf );
+
+	efree( sortKeyIndxBuf );
+	efree( sortKeyBuf );
+
+	RETURN_TRUE;
+}
+/* }}} */
+
+/* {{{ proto bool Collator::asort( Collator $coll, array(string) $arr )
+ * Sort array using specified collator, maintaining index association. }}} */
+/* {{{ proto bool collator_asort( Collator $coll, array(string) $arr )
+ * Sort array using specified collator, maintaining index association.
+ */
+PHP_FUNCTION( collator_asort )
+{
+	collator_sort_internal( FALSE, INTERNAL_FUNCTION_PARAM_PASSTHRU );
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/collator/collator_sort.h

@@ -0,0 +1,29 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef COLLATOR_SORT_H
+#define COLLATOR_SORT_H
+
+#include <php.h>
+
+typedef int (*collator_compare_func_t)( zval *result, zval *op1, zval *op2 TSRMLS_DC );
+
+PHP_FUNCTION( collator_sort );
+PHP_FUNCTION( collator_sort_with_sort_keys );
+PHP_FUNCTION( collator_asort );
+
+#endif // COLLATOR_SORT_H

ext/intl/common/common_error.c

@@ -0,0 +1,266 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "intl_error.h"
+#include "common_error.h"
+
+/* {{{ proto int intl_get_error_code()
+ * Get code of the last occured error.
+ */
+PHP_FUNCTION( intl_get_error_code )
+{
+	RETURN_LONG( intl_error_get_code( NULL TSRMLS_CC ) );
+}
+/* }}} */
+
+/* {{{ proto string intl_get_error_message()
+ * Get text description of the last occured error.
+ */
+PHP_FUNCTION( intl_get_error_message )
+{
+	char* message = intl_error_get_message( NULL TSRMLS_CC );
+	RETURN_STRING( message, FALSE );
+}
+/* }}} */
+
+/* {{{ proto bool intl_is_failure()
+ * Check whether the given error code indicates a failure.
+ * Returns true if it does, and false if the code
+ * indicates success or a warning.
+ */
+PHP_FUNCTION( intl_is_failure )
+{
+	long err_code;
+
+	// Parse parameters.
+	if( zend_parse_parameters( ZEND_NUM_ARGS() TSRMLS_CC, "l",
+		&err_code ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"intl_is_failure: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	RETURN_BOOL( U_FAILURE( err_code ) );
+}
+
+/* {{{ proto string intl_error_name()
+ * Return a string for a given error code.
+ * The string will be the same as the name of the error code constant.
+ */
+PHP_FUNCTION( intl_error_name )
+{
+	long err_code;
+
+	// Parse parameters.
+	if( zend_parse_parameters( ZEND_NUM_ARGS() TSRMLS_CC, "l",
+		&err_code ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"intl_error_name: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	RETURN_STRING( (char*)u_errorName( err_code ), 1 );
+}
+/* }}} */
+
+/* {{{ intl_expose_icu_error_codes
+ * Expose ICU error codes
+ */
+void intl_expose_icu_error_codes( INIT_FUNC_ARGS )
+{
+	#define INTL_EXPOSE_CONST(x) REGISTER_LONG_CONSTANT(#x, x, CONST_CS)
+
+	// Warnings
+	INTL_EXPOSE_CONST( U_USING_FALLBACK_WARNING );
+	INTL_EXPOSE_CONST( U_ERROR_WARNING_START );
+	INTL_EXPOSE_CONST( U_USING_DEFAULT_WARNING );
+	INTL_EXPOSE_CONST( U_SAFECLONE_ALLOCATED_WARNING );
+	INTL_EXPOSE_CONST( U_STATE_OLD_WARNING );
+	INTL_EXPOSE_CONST( U_STRING_NOT_TERMINATED_WARNING );
+	INTL_EXPOSE_CONST( U_SORT_KEY_TOO_SHORT_WARNING );
+	INTL_EXPOSE_CONST( U_AMBIGUOUS_ALIAS_WARNING );
+	INTL_EXPOSE_CONST( U_DIFFERENT_UCA_VERSION );
+	INTL_EXPOSE_CONST( U_ERROR_WARNING_LIMIT );
+
+	// Standard errors
+	INTL_EXPOSE_CONST( U_ZERO_ERROR );
+	INTL_EXPOSE_CONST( U_ILLEGAL_ARGUMENT_ERROR );
+	INTL_EXPOSE_CONST( U_MISSING_RESOURCE_ERROR );
+	INTL_EXPOSE_CONST( U_INVALID_FORMAT_ERROR );
+	INTL_EXPOSE_CONST( U_FILE_ACCESS_ERROR );
+	INTL_EXPOSE_CONST( U_INTERNAL_PROGRAM_ERROR );
+	INTL_EXPOSE_CONST( U_MESSAGE_PARSE_ERROR );
+	INTL_EXPOSE_CONST( U_MEMORY_ALLOCATION_ERROR );
+	INTL_EXPOSE_CONST( U_INDEX_OUTOFBOUNDS_ERROR );
+	INTL_EXPOSE_CONST( U_PARSE_ERROR );
+	INTL_EXPOSE_CONST( U_INVALID_CHAR_FOUND );
+	INTL_EXPOSE_CONST( U_TRUNCATED_CHAR_FOUND );
+	INTL_EXPOSE_CONST( U_ILLEGAL_CHAR_FOUND );
+	INTL_EXPOSE_CONST( U_INVALID_TABLE_FORMAT );
+	INTL_EXPOSE_CONST( U_INVALID_TABLE_FILE );
+	INTL_EXPOSE_CONST( U_BUFFER_OVERFLOW_ERROR );
+	INTL_EXPOSE_CONST( U_UNSUPPORTED_ERROR );
+	INTL_EXPOSE_CONST( U_RESOURCE_TYPE_MISMATCH );
+	INTL_EXPOSE_CONST( U_ILLEGAL_ESCAPE_SEQUENCE );
+	INTL_EXPOSE_CONST( U_UNSUPPORTED_ESCAPE_SEQUENCE );
+	INTL_EXPOSE_CONST( U_NO_SPACE_AVAILABLE );
+	INTL_EXPOSE_CONST( U_CE_NOT_FOUND_ERROR );
+	INTL_EXPOSE_CONST( U_PRIMARY_TOO_LONG_ERROR );
+	INTL_EXPOSE_CONST( U_STATE_TOO_OLD_ERROR );
+	INTL_EXPOSE_CONST( U_TOO_MANY_ALIASES_ERROR );
+	INTL_EXPOSE_CONST( U_ENUM_OUT_OF_SYNC_ERROR );
+	INTL_EXPOSE_CONST( U_INVARIANT_CONVERSION_ERROR );
+	INTL_EXPOSE_CONST( U_INVALID_STATE_ERROR );
+	INTL_EXPOSE_CONST( U_COLLATOR_VERSION_MISMATCH );
+	INTL_EXPOSE_CONST( U_USELESS_COLLATOR_ERROR );
+	INTL_EXPOSE_CONST( U_NO_WRITE_PERMISSION );
+	INTL_EXPOSE_CONST( U_STANDARD_ERROR_LIMIT );
+
+	// The error code range 0x10000 0x10100 are reserved for Transliterator
+	INTL_EXPOSE_CONST( U_BAD_VARIABLE_DEFINITION );
+	INTL_EXPOSE_CONST( U_PARSE_ERROR_START );
+	INTL_EXPOSE_CONST( U_MALFORMED_RULE );
+	INTL_EXPOSE_CONST( U_MALFORMED_SET );
+	INTL_EXPOSE_CONST( U_MALFORMED_SYMBOL_REFERENCE );
+	INTL_EXPOSE_CONST( U_MALFORMED_UNICODE_ESCAPE );
+	INTL_EXPOSE_CONST( U_MALFORMED_VARIABLE_DEFINITION );
+	INTL_EXPOSE_CONST( U_MALFORMED_VARIABLE_REFERENCE );
+	INTL_EXPOSE_CONST( U_MISMATCHED_SEGMENT_DELIMITERS );
+	INTL_EXPOSE_CONST( U_MISPLACED_ANCHOR_START );
+	INTL_EXPOSE_CONST( U_MISPLACED_CURSOR_OFFSET );
+	INTL_EXPOSE_CONST( U_MISPLACED_QUANTIFIER );
+	INTL_EXPOSE_CONST( U_MISSING_OPERATOR );
+	INTL_EXPOSE_CONST( U_MISSING_SEGMENT_CLOSE );
+	INTL_EXPOSE_CONST( U_MULTIPLE_ANTE_CONTEXTS );
+	INTL_EXPOSE_CONST( U_MULTIPLE_CURSORS );
+	INTL_EXPOSE_CONST( U_MULTIPLE_POST_CONTEXTS );
+	INTL_EXPOSE_CONST( U_TRAILING_BACKSLASH );
+	INTL_EXPOSE_CONST( U_UNDEFINED_SEGMENT_REFERENCE );
+	INTL_EXPOSE_CONST( U_UNDEFINED_VARIABLE );
+	INTL_EXPOSE_CONST( U_UNQUOTED_SPECIAL );
+	INTL_EXPOSE_CONST( U_UNTERMINATED_QUOTE );
+	INTL_EXPOSE_CONST( U_RULE_MASK_ERROR );
+	INTL_EXPOSE_CONST( U_MISPLACED_COMPOUND_FILTER );
+	INTL_EXPOSE_CONST( U_MULTIPLE_COMPOUND_FILTERS );
+	INTL_EXPOSE_CONST( U_INVALID_RBT_SYNTAX );
+	INTL_EXPOSE_CONST( U_INVALID_PROPERTY_PATTERN );
+	INTL_EXPOSE_CONST( U_MALFORMED_PRAGMA );
+	INTL_EXPOSE_CONST( U_UNCLOSED_SEGMENT );
+	INTL_EXPOSE_CONST( U_ILLEGAL_CHAR_IN_SEGMENT );
+	INTL_EXPOSE_CONST( U_VARIABLE_RANGE_EXHAUSTED );
+	INTL_EXPOSE_CONST( U_VARIABLE_RANGE_OVERLAP );
+	INTL_EXPOSE_CONST( U_ILLEGAL_CHARACTER );
+	INTL_EXPOSE_CONST( U_INTERNAL_TRANSLITERATOR_ERROR );
+	INTL_EXPOSE_CONST( U_INVALID_ID );
+	INTL_EXPOSE_CONST( U_INVALID_FUNCTION );
+	INTL_EXPOSE_CONST( U_PARSE_ERROR_LIMIT );
+
+	// The error code range 0x10100 0x10200 are reserved for formatting API parsing error
+	INTL_EXPOSE_CONST( U_UNEXPECTED_TOKEN );
+	INTL_EXPOSE_CONST( U_FMT_PARSE_ERROR_START );
+	INTL_EXPOSE_CONST( U_MULTIPLE_DECIMAL_SEPARATORS );
+	INTL_EXPOSE_CONST( U_MULTIPLE_DECIMAL_SEPERATORS );    // Typo: kept for backward compatibility. Use U_MULTIPLE_DECIMAL_SEPARATORS
+	INTL_EXPOSE_CONST( U_MULTIPLE_EXPONENTIAL_SYMBOLS );
+	INTL_EXPOSE_CONST( U_MALFORMED_EXPONENTIAL_PATTERN );
+	INTL_EXPOSE_CONST( U_MULTIPLE_PERCENT_SYMBOLS );
+	INTL_EXPOSE_CONST( U_MULTIPLE_PERMILL_SYMBOLS );
+	INTL_EXPOSE_CONST( U_MULTIPLE_PAD_SPECIFIERS );
+	INTL_EXPOSE_CONST( U_PATTERN_SYNTAX_ERROR );
+	INTL_EXPOSE_CONST( U_ILLEGAL_PAD_POSITION );
+	INTL_EXPOSE_CONST( U_UNMATCHED_BRACES );
+	INTL_EXPOSE_CONST( U_UNSUPPORTED_PROPERTY );
+	INTL_EXPOSE_CONST( U_UNSUPPORTED_ATTRIBUTE );
+	INTL_EXPOSE_CONST( U_FMT_PARSE_ERROR_LIMIT );
+
+	// The error code range 0x10200 0x102ff are reserved for Break Iterator related error
+	INTL_EXPOSE_CONST( U_BRK_INTERNAL_ERROR );
+	INTL_EXPOSE_CONST( U_BRK_ERROR_START );
+	INTL_EXPOSE_CONST( U_BRK_HEX_DIGITS_EXPECTED );
+	INTL_EXPOSE_CONST( U_BRK_SEMICOLON_EXPECTED );
+	INTL_EXPOSE_CONST( U_BRK_RULE_SYNTAX );
+	INTL_EXPOSE_CONST( U_BRK_UNCLOSED_SET );
+	INTL_EXPOSE_CONST( U_BRK_ASSIGN_ERROR );
+	INTL_EXPOSE_CONST( U_BRK_VARIABLE_REDFINITION );
+	INTL_EXPOSE_CONST( U_BRK_MISMATCHED_PAREN );
+	INTL_EXPOSE_CONST( U_BRK_NEW_LINE_IN_QUOTED_STRING );
+	INTL_EXPOSE_CONST( U_BRK_UNDEFINED_VARIABLE );
+	INTL_EXPOSE_CONST( U_BRK_INIT_ERROR );
+	INTL_EXPOSE_CONST( U_BRK_RULE_EMPTY_SET );
+	INTL_EXPOSE_CONST( U_BRK_UNRECOGNIZED_OPTION );
+	INTL_EXPOSE_CONST( U_BRK_MALFORMED_RULE_TAG );
+	INTL_EXPOSE_CONST( U_BRK_ERROR_LIMIT );
+
+	// The error codes in the range 0x10300-0x103ff are reserved for regular expression related errrs
+	INTL_EXPOSE_CONST( U_REGEX_INTERNAL_ERROR );
+	INTL_EXPOSE_CONST( U_REGEX_ERROR_START );
+	INTL_EXPOSE_CONST( U_REGEX_RULE_SYNTAX );
+	INTL_EXPOSE_CONST( U_REGEX_INVALID_STATE );
+	INTL_EXPOSE_CONST( U_REGEX_BAD_ESCAPE_SEQUENCE );
+	INTL_EXPOSE_CONST( U_REGEX_PROPERTY_SYNTAX );
+	INTL_EXPOSE_CONST( U_REGEX_UNIMPLEMENTED );
+	INTL_EXPOSE_CONST( U_REGEX_MISMATCHED_PAREN );
+	INTL_EXPOSE_CONST( U_REGEX_NUMBER_TOO_BIG );
+	INTL_EXPOSE_CONST( U_REGEX_BAD_INTERVAL );
+	INTL_EXPOSE_CONST( U_REGEX_MAX_LT_MIN );
+	INTL_EXPOSE_CONST( U_REGEX_INVALID_BACK_REF );
+	INTL_EXPOSE_CONST( U_REGEX_INVALID_FLAG );
+	INTL_EXPOSE_CONST( U_REGEX_LOOK_BEHIND_LIMIT );
+	INTL_EXPOSE_CONST( U_REGEX_SET_CONTAINS_STRING );
+	INTL_EXPOSE_CONST( U_REGEX_ERROR_LIMIT );
+
+	// The error code in the range 0x10400-0x104ff are reserved for IDNA related error codes
+#if defined(U_IDNA_PROHIBITED_ERROR)
+	INTL_EXPOSE_CONST( U_IDNA_PROHIBITED_ERROR );
+	INTL_EXPOSE_CONST( U_IDNA_ERROR_START );
+	INTL_EXPOSE_CONST( U_IDNA_UNASSIGNED_ERROR );
+	INTL_EXPOSE_CONST( U_IDNA_CHECK_BIDI_ERROR );
+	INTL_EXPOSE_CONST( U_IDNA_STD3_ASCII_RULES_ERROR );
+	INTL_EXPOSE_CONST( U_IDNA_ACE_PREFIX_ERROR );
+	INTL_EXPOSE_CONST( U_IDNA_VERIFICATION_ERROR );
+	INTL_EXPOSE_CONST( U_IDNA_LABEL_TOO_LONG_ERROR );
+	INTL_EXPOSE_CONST( U_IDNA_ZERO_LENGTH_LABEL_ERROR );
+	INTL_EXPOSE_CONST( U_IDNA_ERROR_LIMIT );
+#endif
+
+	// Aliases for StringPrep
+	INTL_EXPOSE_CONST( U_STRINGPREP_PROHIBITED_ERROR );
+	INTL_EXPOSE_CONST( U_STRINGPREP_UNASSIGNED_ERROR );
+	INTL_EXPOSE_CONST( U_STRINGPREP_CHECK_BIDI_ERROR );
+
+	INTL_EXPOSE_CONST( U_ERROR_LIMIT );
+
+	#undef INTL_EXPOSE_CONST
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/common/common_error.h

@@ -0,0 +1,30 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef INTL_COMMON_ERROR_H
+#define INTL_COMMON_ERROR_H
+
+#include <php.h>
+
+PHP_FUNCTION( intl_get_error_code );
+PHP_FUNCTION( intl_get_error_message );
+PHP_FUNCTION( intl_is_failure );
+PHP_FUNCTION( intl_error_name );
+
+void intl_expose_icu_error_codes( INIT_FUNC_ARGS );
+
+#endif // INTL_COMMON_ERROR_H

ext/intl/config.m4

@@ -0,0 +1,66 @@
+dnl config.m4 for extension intl
+
+dnl ##########################################################################
+dnl Initialize the extension
+PHP_ARG_ENABLE(intl, whether to enable internationalization support,
+[  --enable-intl           Enable internationalization support])
+
+if test "$PHP_INTL" != "no"; then
+  PHP_SETUP_ICU(INTL_SHARED_LIBADD)
+  PHP_SUBST(INTL_SHARED_LIBADD)
+  PHP_REQUIRE_CXX()
+
+  PHP_NEW_EXTENSION(intl,
+   			php_intl.c \
+		    intl_error.c \
+		    intl_convert.c \
+		    collator/collator.c \
+		    collator/collator_class.c \
+		    collator/collator_sort.c \
+		    collator/collator_convert.c \
+		    collator/collator_locale.c \
+		    collator/collator_compare.c \
+		    collator/collator_attr.c \
+		    collator/collator_create.c \
+		    collator/collator_is_numeric.c \
+		    collator/collator_error.c \
+		    common/common_error.c \
+		    formatter/formatter.c \
+		    formatter/formatter_main.c \
+		    formatter/formatter_class.c \
+		    formatter/formatter_attr.c \
+		    formatter/formatter_data.c \
+		    formatter/formatter_format.c \
+		    formatter/formatter_parse.c \
+		    normalizer/normalizer.c \
+		    normalizer/normalizer_class.c \
+		    normalizer/normalizer_normalize.c \
+		    locale/locale.c \
+		    locale/locale_class.c \
+		    locale/locale_methods.c \
+		    dateformat/dateformat.c \
+		    dateformat/dateformat_class.c \
+		    dateformat/dateformat_attr.c \
+		    dateformat/dateformat_data.c \
+		    dateformat/dateformat_format.c \
+		    dateformat/dateformat_parse.c \
+		msgformat/msgformat.c \
+		msgformat/msgformat_attr.c \
+		msgformat/msgformat_class.c \
+		msgformat/msgformat_data.c  \
+		msgformat/msgformat_format.c \
+		msgformat/msgformat_helpers.cpp \
+		msgformat/msgformat_parse.c \
+		grapheme/grapheme_string.c \
+		grapheme/grapheme_util.c \
+		    ,$ext_shared,,$ICU_INCS)
+
+	PHP_ADD_BUILD_DIR($ext_builddir/collator)
+	PHP_ADD_BUILD_DIR($ext_builddir/common)
+	PHP_ADD_BUILD_DIR($ext_builddir/formatter)
+	PHP_ADD_BUILD_DIR($ext_builddir/normalizer)
+	PHP_ADD_BUILD_DIR($ext_builddir/dateformat)
+	PHP_ADD_BUILD_DIR($ext_builddir/locale)
+	PHP_ADD_BUILD_DIR($ext_builddir/msgformat)
+	PHP_ADD_BUILD_DIR($ext_builddir/grapheme)
+fi

ext/intl/config.w32

@@ -0,0 +1,71 @@
+// $Id$
+// vim:ft=javascript
+
+ARG_ENABLE("intl", "Enable internationalization support", "no");
+
+if (PHP_INTL != "no") {
+	if (CHECK_LIB("icuuc.lib", "intl", PHP_INTL) &&
+					CHECK_HEADER_ADD_INCLUDE("unicode/utf.h", "CFLAGS_INTL")) {
+		// always build as shared - zend_strtod.c/ICU type conflict
+		EXTENSION("intl", "php_intl.c intl_convert.c intl_error.c ", true,
+								"/I \"" + configure_module_dirname + "\"");
+		ADD_SOURCES(configure_module_dirname + "/collator", "\
+				collator.c \
+				collator_attr.c \
+				collator_class.c \
+				collator_compare.c \
+				collator_convert.c \
+				collator_create.c \
+				collator_error.c \
+				collator_is_numeric.c \
+				collator_locale.c \
+				collator_sort.c \
+				", "intl");
+		ADD_SOURCES(configure_module_dirname + "/common", "\
+				common_error.c \
+				", "intl");
+		ADD_SOURCES(configure_module_dirname + "/formatter", "\
+				formatter.c \
+				formatter_attr.c \
+				formatter_class.c \
+				formatter_data.c \
+				formatter_format.c \
+				formatter_main.c \
+				formatter_parse.c \
+				", "intl");
+		ADD_SOURCES(configure_module_dirname + "/locale", "\
+				locale.c \
+				locale_class.c \
+				locale_methods.c \
+				", "intl");
+		ADD_SOURCES(configure_module_dirname + "/msgformat", "\
+				msgformat.c \
+				msgformat_attr.c \
+				msgformat_class.c \
+				msgformat_data.c \
+				msgformat_format.c \
+				msgformat_helpers.cpp \
+				msgformat_parse.c \
+				", "intl");
+		ADD_SOURCES(configure_module_dirname + "/grapheme", "\
+                                grapheme_string.c grapheme_util.c  \
+                                ", "intl");
+		ADD_SOURCES(configure_module_dirname + "/normalizer", "\
+				normalizer.c \
+				normalizer_class.c \
+				normalizer_normalize.c \
+				", "intl");
+		ADD_SOURCES(configure_module_dirname + "/dateformat", "\
+				dateformat.c \
+				dateformat_class.c \
+				dateformat_attr.c \
+				dateformat_format.c \
+				dateformat_parse.c \
+				dateformat_data.c \
+				", "intl");
+		ADD_FLAG("LIBS_INTL", "icudt.lib icuin.lib icuio.lib icule.lib iculx.lib");
+		AC_DEFINE("HAVE_INTL", 1, "Internationalization support enabled");
+	} else {
+		WARNING("intl not enabled; libraries and/or headers not found");
+	}
+}

ext/intl/doc/Tutorial.txt

@@ -0,0 +1,239 @@
+1. Collator::getAvailableLocales().
+Return the locales available at the time of the call, including registered locales.
+If a sever error occurs (such as out of memory condition) this will return null.
+If there is no locale data, an empty enumeration will be returned. 
+Returned locales list is a strings in format of RFC4646 standart (see http://www.rfc-editor.org/rfc/rfc4646.txt).
+Examle of locales format: 'en_US', 'ru_UA', 'ua_UA' (see http://demo.icu-project.org/icu-bin/locexp).
+
+
+2. Collator::getDisplayName( $obj_locale, $disp_locale ).
+Get name of the object for the desired Locale, in the desired langauge. Both arguments 
+must be from getAvailableLocales method.
+
+     @param  string  $obj_locale   Locale to get display name for.
+     @param  string  $disp_locale  Specifies the desired locale for output
+
+Both parameters are case insensitive.
+For locale format see RFC4647 standart in ftp://ftp.rfc-editor.org/in-notes/rfc4647.txt
+
+3. Collator::getLocaleByType( $type ).
+Allow user to select whether she wants information on requested, valid or actual locale.
+Returned locale tag is a string formatted to a RFC4646 standart and normalize to normal form -
+value is a string from 
+For example, a collator for "en_US_CALIFORNIA" was requested. In the current state of ICU (2.0), 
+the requested locale is "en_US_CALIFORNIA", the valid locale is "en_US" (most specific locale 
+supported by ICU) and the actual locale is "root" (the collation data comes unmodified from the UCA) 
+The locale is considered supported by ICU if there is a core ICU bundle for that locale (although 
+it may be empty). 
+
+
+4. VariableTop
+The Variable_Top attribute is only meaningful if the Alternate attribute is not set to NonIgnorable.
+In such a case, it controls which characters count as ignorable. The string value specifies 
+the "highest" character (in UCA order) weight that is to be considered ignorable.
+Thus, for example, if a user wanted whitespace to be ignorable, but not any visible characters, 
+then s/he would use the value Variable_Top="\u0020" (space). The string should only be a 
+single character. All characters of the same primary weight are equivalent, so 
+Variable_Top="\u3000" (ideographic space) has the same effect as Variable_Top="\u0020".
+This setting (alone) has little impact on string comparison performance; setting it lower or higher
+will make sort keys slightly shorter or longer respectively.
+
+
+5. Strength
+The ICU Collation Service supports many levels of comparison (named "Levels", but also
+known as "Strengths"). Having these categories enables ICU to sort strings precisely
+according to local conventions. However, by allowing the levels to be selectively
+employed, searching for a string in text can be performed with various matching
+conditions.
+Performance optimizations have been made for ICU collation with the default level
+settings. Performance specific impacts are discussed in the Performance section below.
+Following is a list of the names for each level and an example usage:
+
+1. Primary Level: Typically, this is used to denote differences between base characters
+(for example, "a" < "b"). It is the strongest difference. For example, dictionaries are
+divided into different sections by base character. This is also called the level1
+strength.
+
+2. Secondary Level: Accents in the characters are considered secondary differences (for
+example, "as" < "as" < "at"). Other differences between letters can also be considered
+secondary differences, depending on the language. A secondary difference is ignored
+when there is a primary difference anywhere in the strings. This is also called the
+level2 strength.
+Note: In some languages (such as Danish), certain accented letters are considered to
+be separate base characters. In most languages, however, an accented letter only has a
+secondary difference from the unaccented version of that letter.
+
+3. Tertiary Level: Upper and lower case differences in characters are distinguished at the
+tertiary level (for example, "ao" < "Ao" < "ao"). In addition, a variant of a letter differs
+from the base form on the tertiary level (such as "A" and " "). Another ? example is the
+difference between large and small Kana. A tertiary difference is ignored when there is
+a primary or secondary difference anywhere in the strings. This is also called the level3
+strength.
+
+4. Quaternary Level: When punctuation is ignored (see Ignoring Punctuations ) at level
+13, an additional level can be used to distinguish words with and without punctuation
+(for example, "ab" < "a-b" < "aB"). This difference is ignored when there is a primary,
+secondary or tertiary difference. This is also known as the level4 strength. The
+quaternary level should only be used if ignoring punctuation is required or when
+processing Japanese text (see Hiragana processing).
+
+5. Identical Level: When all other levels are equal, the identical level is used as a
+tiebreaker. The Unicode code point values of the NFD form of each string are
+compared at this level, just in case there is no difference at levels 14
+. For example, Hebrew cantillation marks are only distinguished at this level. This level should be
+used sparingly, as only code point values differences between two strings is an
+extremely rare occurrence. Using this level substantially decreases the performance for
+both incremental comparison and sort key generation (as well as increasing the sort
+key length). It is also known as level 5 strength.
+
+For example, people may choose to ignore accents or ignore accents and case when searching
+for text. Almost all characters are distinguished by the first three levels, and in most
+locales the default value is thus Tertiary. However, if Alternate is set to be Shifted,
+then the Quaternary strength can be used to break ties among whitespace, punctuation, and
+symbols that would otherwise be ignored. If very fine distinctions among characters are required,
+then the Identical strength can be used (for example, Identical Strength distinguishes
+between the Mathematical Bold Small A and the Mathematical Italic Small A.). However, using
+levels higher than Tertiary the Identical strength result in significantly longer sort
+keys, and slower string comparison performance for equal strings.
+
+
+
+6. Collator::__construct( $locale ).
+The Locale attribute is typically the most important attribute for correct sorting and matching,
+according to the user expectations in different countries and regions. The default UCA
+ordering will only sort a few languages such as Dutch and Portuguese correctly ("correctly"
+meaning according to the normal expectations for users of the languages).
+Otherwise, you need to supply the locale to UCA in order to properly collate text for a
+given language. Thus a locale needs to be supplied so as to choose a collator that is correctly
+tailored for that locale. The choice of a locale will automatically preset the values for
+all of the attributes to something that is reasonable for that locale. Thus most of the time the
+other attributes do not need to be explicitly set. In some cases, the choice of locale will make a
+difference in string comparison performance and/or sort key length.
+In short attribute names, <language>_<script>_<region>_<keyword>.
+Not all the elements are required. Valid values for locale elements are general valid values
+for RFC4646 locale naming, and RFC 4647 lookup algorithm.
+Example:
+Locale="sv" (Swedish) "Kypper" < "Kopfe"
+Locale="de" (German) "Kopfe" < "Kypper"
+
+
+7. Collator::get/setAttribute.
+ICU uses UCA as a default starting point for ordering. Not all languages have sorting sequences
+that correspond with the UCA because UCA cannot simultaneously encompass the specifics of all
+the languages currently in use. Therefore, ICU provides a data-driven, flexible, and run-time
+customizable mechanism called "tailoring". Tailoring overrides the default order of code points
+and the values of the ICU Collation Service attributes.
+Collator have followed attributes:
+   - FRENCH_COLLATION, possible values are: 
+	ON
+	OFF (default)
+	DEFAULT
+
+   - CASE_FIRST, possible values are:
+	OFF (default)
+	LOWER_FIRST
+	UPPER_FIRST
+	DEFAULT
+
+   - CASE_LEVEL, possible values are:
+	OFF (default)
+	ON
+	DEFAULT
+
+   - NORMALIZATION_MODE, possible values are:
+	OFF (default)
+	ON
+	DEFAULT
+
+   - STRENGTH, possible values are:
+	PRIMARY
+	SECONDARY
+	TERTIARY (default)
+	QUATERNARY
+	IDENTICAL
+	DEFAULT
+
+   - ALTERNATE_HANDLING, possible values are:
+	NON_IGNORABLE (default)
+	SHIFTED
+	DEFAULT
+
+   - HIRAGANA_QUATERNARY_MODE, possible values are:
+	ON
+	OFF (default)
+	DEFAULT
+
+   - NUMERIC_COLLATION, possible values are:
+	ON
+	OFF (default)
+	DEFAULT
+
+Description of all of this attributes:
+
+FRENCH_COLLATION - Sort strings with different accents from the back of the string. This attribute
+is automatically set to On for the French locales and a few others. Users normally would
+not need to explicitly set this attribute. There is a string comparison performance cost when
+it is set On, but sort key length is unaffected.
+Example:
+F=X cote < cote < cote < cote
+F=O cote < cote < cote < cote
+
+CASE_FIRST - The Case_First attribute is used to control whether uppercase letters come before
+lowercase letters or vice versa, in the absence of other differences in the strings. The possible
+values are Uppercase_First (U) and Lowercase_First (L), plus the standard Default and Off.
+There is almost no difference between the Off and Lowercase_First options in terms of results,
+so typically users will not use Lowercase_First: only Off or Uppercase_First. (People interested
+in the detailed differences between X and L should consult the Collation Customization).
+Specifying either L or U won't affect string comparison performance, but will affect the sort key
+length.
+Example:
+C=X or C=L "china" < "China" < "denmark" <
+"Denmark"
+C=U "China" < "china" < "Denmark" < "denmark"
+
+CASE_LEVEL - The Case_Level attribute is used when ignoring accents but not case. In such a situation,
+set Strength to be Primary, and Case_Level to be On. In most locales, this setting is Off by default.
+There is a small string comparison performance and sort key impact if this attribute is set to be On.
+Example:
+S=1, E=X role = Role = role
+S=1, E=O role = role < Role
+
+NORMALIZATION_MODE - The Normalization setting determines whether text is thoroughly normalized
+or not in comparison. Even if the setting is off (which is the default for many locales), text as
+represented in common usage will compare correctly (for details, see UTN #5). Only if the accent
+marks are in noncanonical order will there be a problem. If the setting is On, then the best
+results are guaranteed for all possible text input. There is a medium string comparison performance
+cost if this attribute is On, depending on the frequency of sequences that require normalization.
+There is no significant effect on sort key length. If the input text is known to be in NFD or NFKD
+normalization forms, there is no need to enable this Normalization option.
+
+STRENGTH - see Collator::setStrength chapter.
+
+ALTERNATE_HANDLING - The Alternate attribute is used to control the handling of the socalled
+variable characters in the UCA: whitespace, punctuation and symbols. If Alternate is set to
+NonIgnorable (N), then differences among these characters are of the same importance as
+differences among letters. If Alternate is set to Shifted (S), then these characters are of only
+minor importance. The Shifted value is often used in combination with Strength set to Quaternary.
+In such a case, whitespace, punctuation, and symbols are considered when comparing strings,
+but only if all other aspects of the strings (base letters, accents, and case) are identical.
+If Alternate is not set to Shifted, then there is no difference between a Strength of 3 and
+a Strength of 4. For more information and examples, see
+Variable_Weighting in the UCA (http://www.unicode.org/reports/tr10/#Variable_Weighting).
+The reason the Alternate values are not simply On and Off is that additional Alternate values
+may be added in the future. The UCA option Blanked is expressed with Strength set to 3,
+and Alternate set to Shifted. The default for most locales is NonIgnorable. If Shifted is selected,
+it may be slower if there are many strings that are the same except for punctuation;
+sort key length will not be affected unless the strength level is also increased.
+Example:
+S=3, A=N di Silva < Di Silva < diSilva < U.S.A. < USA
+S=3, A=S di Silva = diSilva < Di Silva < U.S.A. = USA
+S=4, A=S di Silva < diSilva < Di Silva < U.S.A. < USA
+
+HIRAGANA_QUATERNARY_MODE - Compatibility with JIS x 4061 requires the introduction of an additional
+level to distinguish Hiragana and Katakana characters. If compatibility with that standard is required,
+then this attribute should be set On, and the strength set to Quaternary. This will affect sort key
+length and string comparison string comparison performance.
+
+NUMERIC_COLLATION - When turned on, this attribute generates a collation key for the
+numeric value of substrings of digits. This is a way to get '100' to sort AFTER '2'.
+

ext/intl/doc/collator_api.php

@@ -0,0 +1,398 @@
+<?php
+#############################################################################
+# Object-oriented API
+#############################################################################
+
+/**
+ * Collator class.
+ *
+ * This is a wrapper around ICU Collator C API (declared in ucol.h).
+ *
+ * Example:
+ * <code>
+ *
+ * </code>
+ *
+ * @see http://www.icu-project.org/apiref/icu4c/ucol_8h.html
+ * @see http://www.icu-project.org/apiref/icu4c/classCollator.html
+ *
+ */
+class Collator {
+#############################################################################
+# Common constants.
+#############################################################################
+
+/**
+ * Locale-related constants.
+ *
+ * These will be moved out of Collator when Locale class is created.
+ */
+	const ULOC_ACTUAL_LOCALE    = 0;
+	const ULOC_VALID_LOCALE     = 1;
+	const ULOC_REQUESTED_LOCALE = 2;
+
+	/*
+	 * WARNING:
+	 * The values described here are NOT the actual values in PHP code.
+	 * They are references to the ICU C definitions, so the line
+	 *    const DEFAULT_STRENGTH = 'UCOL_DEFAULT_STRENGTH';
+	 * actually means that Collator::DEFAULT_STRENGTH is the same as
+	 * UCOL_DEFAULT_STRENGTH constant in the ICU library.
+	 */
+	/**
+     * Valid attribute values.
+     *
+     * @see Collator::setAttribute()
+     * @see collator_set_attribute()
+     */
+    const DEFAULT_VALUE    = 'UCOL_DEFAULT';
+    const PRIMARY          = 'UCOL_PRIMARY';
+    const SECONDARY        = 'UCOL_SECONDARY';
+    const TERTIARY         = 'UCOL_TERTIARY';
+    const DEFAULT_STRENGTH = 'UCOL_DEFAULT_STRENGTH';
+    const QUATERNARY       = 'UCOL_QUATERNARY';
+    const IDENTICAL        = 'UCOL_IDENTICAL';
+    const OFF              = 'UCOL_OFF';
+    const ON               = 'UCOL_ON';
+    const SHIFTED          = 'UCOL_SHIFTED';
+    const NON_IGNORABLE    = 'UCOL_NON_IGNORABLE';
+    const LOWER_FIRST      = 'UCOL_LOWER_FIRST';
+    const UPPER_FIRST      = 'UCOL_UPPER_FIRST';
+
+    /**
+     * Valid attribute names.
+     *
+     * @see Collator::setAttribute()
+     * @see collator_set_attribute()
+     */
+    const FRENCH_COLLATION         = 'UCOL_FRENCH_COLLATION';
+    const ALTERNATE_HANDLING       = 'UCOL_ALTERNATE_HANDLING';
+    const CASE_FIRST               = 'UCOL_CASE_FIRST';
+    const CASE_LEVEL               = 'UCOL_CASE_LEVEL';
+    const NORMALIZATION_MODE       = 'UCOL_NORMALIZATION_MODE';
+    const STRENGTH                 = 'UCOL_STRENGTH';
+    const HIRAGANA_QUATERNARY_MODE = 'UCOL_HIRAGANA_QUATERNARY_MODE';
+    const NUMERIC_COLLATION        = 'UCOL_NUMERIC_COLLATION';
+
+    /**
+     * Create a collator
+     *
+     * @param string $locale The locale whose collation rules
+     *                       should be used. Special values for
+     *                       locales can be passed in - if null is
+     *                       passed for the locale, the default
+     *                       locale collation rules will be used. If
+     *                       empty string ("") or "root" are passed,
+     *                       UCA rules will be used.
+     *
+     * @return Collator     New instance of Collator object.
+     */
+    public function __construct( $locale ) {}
+
+    /**
+     * Create a collator
+     *
+     * Creates a new instance of Collator.
+     *
+     * This method is useful when you prefer just to get null on error,
+     * as if you called collator_create().
+     *
+     * @return Collator      Newly created Collator instance,
+     *                       or null on error.
+     *
+     * @see __construct()
+     * @see collator_create()
+     */
+    public static function create( $locale ) {}
+
+    /**
+     * Get collator's last error code.
+     *
+     * @return  int  Error code returned by the last
+     *               Collator method call.
+     */
+    public function getErrorCode() {}
+
+    /**
+     * Return error text for the last ICU operation.
+     *
+     * @return string Description of an error occured in the last
+     *                Collator method call.
+     */
+    public function getErrorMessage() {}
+
+    /**
+     * Compare two strings using PHP strcmp() semantics.
+     *
+     * Wrapper around ICU ucol_strcoll().
+     *
+     * @param string $str1  First string to compare.
+     * @param string $str2  Second string to compare.
+     *
+     * @return int   1   if $str1 is  greater than  $str2;
+     *               0   if $str1 is  equal to      $str2;
+     *               -1  if $str1 is  less than     $str2.
+     *               On error false is returned.
+     */
+    public function compare( $str1, $str2 ) {}
+
+    /**
+     * Equivalent to standard PHP sort() using Collator.
+     *
+     * @param array $arr         Array of strings to sort
+     * @param int   $sort_flags  Optional sorting type, one of the following:
+     *                           - SORT_REGULAR - compare items normally (don't change types)
+     *                           - SORT_NUMERIC - compare items numerically
+     *                           - SORT_STRING - compare items as strings
+     *                           Default sorting type is SORT_REGULAR.
+     *
+     * @return bool true on success or false on failure.
+     */
+    public function sort( $arr, $sort_flags ) {}
+
+    /**
+     * Sort array maintaining index association.
+     *
+     * Equivalent to standard PHP asort() using Collator.
+     *
+     * @param array $arr         Array of strings to sort
+     * @param int   $sort_flags  Optional sorting type
+     *
+     * @return bool true on success or false on failure.
+     *
+     * @see Collator::sort()
+     */
+    public function asort( $arr, $sort_flags ) {}
+
+    /**
+     * Equivalent to standard PHP sort() using Collator.
+     *
+     * Similar to Collator::sort().
+     * Uses ICU ucol_getSortKey() to gain more speed on large arrays.
+     *
+     * @param array $arr  Array of strings to sort
+     *
+     * @return bool       true on success or false on failure.
+     */
+    public function sortWithSortKeys( $arr ) {}
+
+    /**
+     * @todo  Do we want to support other standard PHP sort functions:  ksort, rsort, asort?
+     */
+
+    /**
+     * Get collation attribute value.
+     *
+     * Wrapper around ICU ucol_getAttribute().
+     *
+     * @param  int      $attr Attribute to get value for.
+     *
+     * @return int      Attribute value, or false on error.
+     */
+    public function getAttribute( $attr ) {}
+
+    /**
+     * Set collation attribute.
+     *
+     * Wrapper around ICU ucol_setAttribute().
+     *
+     * @param int       $attr Attribute.
+     * @param int       $val  Attribute value.
+     *
+     * @return bool     true on success, false otherwise.
+     */
+    public function setAttribute( $attr, $val ) {}
+
+    /**
+     * Get current collation strength.
+     *
+     * Wrapper around ICU ucol_getStrength().
+     *
+     * @return int     Current collation strength, or false on error.
+     */
+    public function getStrength() {}
+
+    /**
+     * Set collation strength.
+     *
+     * Wrapper around ICU ucol_setStrength().
+     *
+     * @param int      $strength Strength to set.
+     *
+     * @return bool    true on success, false otherwise.
+     */
+    public function setStrength( $strength ) {}
+
+    /**
+     * Get the locale name of the collator.
+     *
+     * Wrapper around ICU ucol_getLocaleByType().
+     *
+     * @param int      $type You can choose between requested, valid
+     *                       and actual locale
+     *                       (ULOC_REQUESTED_LOCALE,
+     *                       ULOC_VALID_LOCALE, ULOC_ACTUAL_LOCALE,
+     *                       respectively).
+     *
+     * @return string        Real locale name from which the
+     *                       collation data comes. If the collator
+     *                       was instantiated from rules or an error occured,
+     *                       returns false.
+     */
+    public function getLocale( $type ) {}
+}
+
+#############################################################################
+# Procedural API
+#############################################################################
+
+/**
+ * Create collator.
+ *
+ * @param string     $locale  The locale containing the required
+ *                            collation rules. Special values for
+ *                            locales can be passed in - if null is
+ *                            passed for the locale, the default
+ *                            locale collation rules will be used. If
+ *                            empty string ("") or "root" are passed,
+ *                            UCA rules will be used.
+ *
+ * @return Collator  New instance of Collator object, or null on error.
+ */
+function collator_create( $locale ) {}
+
+/**
+ * Compare two strings.
+ *
+ * The strings will be compared using the options already
+ * specified.
+ *
+ * @param Collator $coll Collator object.
+ * @param string   $str1 The first string to compare.
+ * @param string   $str2 The second string to compare.
+ *
+ * @return int     1   if $str1 is  greater than  $str2;
+ *                 0   if $str1 is  equal to      $str2;
+ *                 -1  if $str1 is  less than     $str2.
+ *                 On error false is returned.
+ *
+ */
+function collator_compare( $coll, $str1, $str2 ) {}
+
+/**
+ * Sort array using specified collator.
+ *
+ * @param  Collator $coll        Collator object.
+ * @param  array    $arr         Array of strings to sort.
+ * @param  int      $sort_flags  Optional sorting type, one of the following:
+ *                               - SORT_REGULAR - compare items normally (don't change types)
+ *                               - SORT_NUMERIC - compare items numerically
+ *                               - SORT_STRING - compare items as strings
+ *                               Default sorting type is SORT_REGULAR.
+ *
+ * @return bool     true on success or false on failure.
+ */
+function collator_sort( $coll, $arr, $sort_flags ) {}
+
+/**
+ * Sort array maintaining index association.
+ *
+ * @param Collator $coll        Collator object.
+ * @param array    $arr         Array of strings to sort.
+ * @param int      $sort_flags  Optional sorting type.
+ *
+ * @return bool    true on success or false on failure.
+ *
+ * @see collator_sort()
+ */
+function collator_asort( $coll, $arr, $sort_flags ) {}
+
+/**
+ * Sort array using specified collator.
+ *
+ * Similar to collator_sort().
+ * Uses ICU ucol_getSortKey() to gain more speed on large arrays.
+ *
+ * @param  Collator $coll  Collator object.
+ * @param  array    $arr   Array of strings to sort
+ *
+ * @return bool     true on success or false on failure.
+ */
+function collator_sort_with_sort_keys( $coll, $arr ) {}
+
+/**
+ * Get the locale name of the collator.
+ *
+ * @param Collator $coll Collator object.
+ * @param int      $type You can choose between valid and 
+ *                       actual locale
+ *                       (ULOC_VALID_LOCALE, ULOC_ACTUAL_LOCALE
+ *                       respectively).
+ *
+ * @return string  Real locale name from which the
+ *                 collation data comes. If the collator
+ *                 was instantiated from rules or an error occured,
+ *                 returns false.
+ */
+function collator_get_locale( $coll, $type ) {}
+
+/**
+ * Get collation attribute value.
+ *
+ * @param  Collator $coll Collator object.
+ * @param  int      $attr Attribute to get value for.
+ *
+ * @return int      Attribute value, or false on error.
+ */
+function collator_get_attribute( $coll, $attr ) {}
+
+/**
+ * Get current collation strength.
+ *
+ * @param Collator $coll Collator object.
+ *
+ * @return int     Current collation strength, or false on error.
+ */
+function collator_get_strength( $coll ) {}
+
+/**
+ * Set collation strength.
+ *
+ * @param Collator $coll      Collator object.
+ * @param int      $strength  Strength to set.
+ *
+ * @return bool    true on success, false otherwise.
+ */
+function collator_set_strength( $coll, $strength ) {}
+
+/**
+ * Set collation attribute.
+ *
+ * @param Collator  $coll  Collator object.
+ * @param int       $attr  Attribute.
+ * @param int       $val   Attribute value.
+ *
+ * @return bool     true on success, false otherwise.
+ */
+function collator_set_attribute( $coll, $attr, $val ) {}
+
+/**
+ * Get collator's last error code.
+ *
+ * @param Collator $coll    Collator object.
+ *
+ * @return int     Error code returned by the last
+ *                 Collator API function call.
+ */
+function collator_get_error_code( $coll ) {}
+
+/**
+ * Get text for collator's last error code.
+ *
+ * @param Collator $coll    Collator object.
+ *
+ * @return string  Description of an error occured in the last
+ *                 Collator API function call.
+ */
+function collator_get_error_message( $coll ) {}
+?>

ext/intl/doc/common_api.php

@@ -0,0 +1,58 @@
+<?php
+
+/**
+ * Handling of errors occured in static methods
+ * when there's no object to get error code/message from.
+ *
+ * Example #1:
+ * <code>
+ * $coll = collator_create( '<bad_param>' );
+ * if( !$coll )
+ *     handle_error( intl_get_error_code() );
+ * </code>
+ *
+ * Example #2:
+ * <code>
+ * if( Collator::getAvailableLocales() === false )
+ *     show_error( intl_get_error_message() );
+ * </code>
+ */
+
+/**
+ * Get the last error code.
+ *
+ * @return int     Error code returned by the last
+ *                 API function call.
+ */
+function intl_get_error_code() {}
+
+/**
+ * Get description of the last error.
+ *
+ * @return string  Description of an error occured in the last
+ *                 API function call.
+ */
+function intl_get_error_message() {}
+
+/**
+ * Check whether the given error code indicates failure.
+ *
+ * @param  int  $code ICU error code.
+ *
+ * @return bool true if it the code indicates some failure,
+ *              and false in case of success or a warning.
+ */
+function intl_is_failure($code) {}
+
+/**
+ * Get symbolic name for a given error code.
+ *
+ * The returned string will be the same as the name of the error code constant.
+ *
+ * @param  int       $code  ICU error code.
+ *
+ * @return string    Error code name.
+ */
+function intl_error_name($code) {}
+
+?>

ext/intl/doc/datefmt_api.php

@@ -0,0 +1,439 @@
+<?php
+
+/**
+ * Date Formatter class - locale-dependent formatting/parsing of dates using pattern strings and/or canned patterns.
+ *
+ * This class represents the ICU date formatting functionality. It allows users to
+ * display dates in a localized format or to parse strings 
+ * into PHP date values using pattern strings and/or canned patterns.
+ * 
+ * Example:
+ * <code>
+ * $datefmt = new DateFormatter("de-DE", LONG, SHORT, date_default_timezone_get());
+ * echo $formatter->format(time());
+ * </code>
+ *
+ * <code>
+ *  $datefmt = new DateFormatter("de-DE", LONG, SHORT, date_default_timezone_get() , GREGORIAN , "yyyy-MM-dd HH:mm:ss z");
+ *  echo $formatter->format(time());
+ * </code>
+ *
+ * @see http://www.icu-project.org/apiref/icu4c/udat_8h.html
+ *
+ */
+class DateFormatter {
+
+#############################################################################
+# Common constants.
+#############################################################################
+
+   /**
+    * The following constants are used to specify different formats
+    * in the constructor.
+    */
+    const NONE = -1;
+    const FULL = 0;
+    const LONG = 1;
+    const MEDIUM = 2;
+    const SHORT = 3;
+
+    /**
+     * The following int constants are used to specify the calendar. 
+     * These calendars are all based directly on the Gregorian calendar 
+     * Non-Gregorian calendars need to be specified in locale. 
+     * Examples might include locale="hi@calendar=BUDDHIST"
+     */
+    const TRADITIONAL = 0; // non-Gregorian calendar that is locale-defined, required by ICU
+    const GREGORIAN = 1 ;// Gregorian calendar
+	
+#############################################################################
+# Object-oriented API
+#############################################################################
+	/**
+	 * Create a date formatter 
+	 *
+	 * @param string  $locale     Locale to use when formatting or parsing
+	 * @param integer $datetype   Date type to use (none, short, medium, long, full)
+	 * @param integer $timetype   Time type to use (none, short, medium, long, full)
+	 * @param [String] $timezone  Time zone ID ; default is system default
+	 * @param [integer] $calendar Calendar to use for formatting or parsing; default is
+	 *                            GREGORIAN
+	 * @param [string]  $pattern  Optional pattern to use when formatting or parsing
+	 * @return DateFormatter
+	 * @see __construct
+	 * @see datefmt_create
+	 */
+	public function __construct($locale, $datetype, $timetype, $timezone = null, $calendar= null , $pattern= null) {}
+
+	/**
+	 * Create a date formatter 
+	 *
+	 * @param string  $locale     Locale to use when formatting or parsing
+	 * @param integer $datetype   Date type to use (none, short, medium, long, full)
+	 * @param integer $timetype   Time type to use (none, short, medium, long, full)
+	 * @param [string] $timezone  Time zone ID ; default is system default
+	 * @param [integer] $calendar Calendar to use for formatting or parsing; default is
+	 *                            GREGORIAN
+	 * @param [string]  $pattern  Optional pattern to use when formatting or parsing
+	 * @return DateFormatter
+	 * @see __construct
+	 * @see datefmt_create
+	 */
+	public static function create($locale, $datetype, $timetype, $timezone = null, $calendar= null , $pattern= null) {}
+
+	/**
+         * formats the time value as a string.
+	 * @param mixed $value - value to format
+         *     integer: a unix timestamp value (seconds since epoch, UTC)
+         *     array: a localtime array  - uses 24 hour clock in tm_hour field
+         * @return string  a formatted string or, if an error occurred, 'null'. 
+	 */
+        public function format($value) {}
+
+
+        /**
+         * converts string $value to an incremental time value, starting at
+	 * $parse_pos and consuming as much of the input value as possible
+	 * If no error occurs before $value is consumed, $parse_pos will contain -1
+	 * otherwise it will contain the position at which parsing ended (and the error
+	 * occurred). 
+	 * @param string  $value      string to convert to a time
+	 * @param integer $parse_pos  position at which to start the parsing in $value (zero-based)
+	 *                            This variable will contain the end position if the parse fails
+	 *                            If $parse_pos > strlen($value), the parse fails immediately.
+	 * @return integer timestamp  parsed value
+	 */
+	 public function parse($value, $parse_pos=0) {}
+
+	 
+        /**
+         * converts string $value to a field-based time value, starting at
+	 * $parse_pos and consuming as much of the input value as possible
+	 * If no error occurs before $value is consumed, $parse_pos will contain -1
+	 * otherwise it will contain the position at which parsing ended (and the error
+	 * occurred). 
+	 * @param string  $value      string to convert to a time
+	 * @param integer $parse_pos  position at which to start the parsing in $value (zero-based)
+	 *                            This variable will contain the end position if the parse fails
+	 *                            If $parse_pos > strlen($value), the parse fails immediately.
+	 * @return array localtime compatible array of integers  - uses 24 hour clock in tm_hour field
+	 */
+	 public function localtime($value, $parse_pos=0) {}
+
+
+	 /**
+	  * Gets the datetype in use 
+	  * @return integer the current 'datetype' value of the formatter
+	  */
+         public function getDateType() {}
+
+
+	 /**
+	  * Gets the timetype in use 
+	  * @return integer the current 'timetype' value of the formatter
+	  */
+         public function getTimeType() {}
+
+
+	 /**
+	  * Gets the leniency in use 
+	  * @return boolean   'true' if parser is lenient, 'false' if parser is strict
+	  *                   default value for parser is 'false'.
+	  */
+         public function isLenient() {}
+
+
+	 /**
+	  * Sets the leniency to use 
+	  * @param boolean $lenient  sets whether the parser is lenient or not, default is 'false'
+          *                          'true' sets the parser to accept otherwise flawed date or 
+	  *                          time patterns, parsing as much as possible to obtain a value.
+          *                          'false' sets the parser to strictly parse strings into dates. 
+	  *                          Extra space, unrecognized tokens, or invalid values 
+	  *                          ("Feburary 30th") are not accepted.
+	  *
+          * @return boolean          'true' if successful; 'false' if an error occurred. 
+	  */
+	 public function setLenient($lenient) {}
+
+
+	 /**
+	  * Gets the locale in use 
+	  * @param  [integer]  which locale should be returned? 
+	  *                    values may include ULOC_ACTUAL_LOCALE,
+	  *                    ULOC_VALID_LOCALE. By default the actual
+	  *                    locale is returned.
+	  * @return string     the locale of this formatter  or 'false' if error
+	 */
+
+	 public function getLocale($type = ULOC_ACTUAL_LOCALE) {}
+
+
+         /**
+	  * @return string ID string for the time zone used by this formatter
+	  */
+	 public function getTimeZoneId() {}
+
+
+	/**
+	 * sets the time zone to use
+	 * @param string $zone zone ID string of the time zone to use.
+	 *                     if null or the empty string, the default time zone for
+	 *                     the runtime is used.
+	 * @return boolean 'true' on successful setting of the time zone, 'false'
+	 *                 if an error occurred (such as the time zone wasn't recognized).
+	 */
+        public function setTimeZoneId($zone) {}
+
+
+        /**
+         * Sets the calendar used to the appropriate calendar, which must be 
+	 * one of the constants defined above. Some examples include:
+         *   - Gregorian calendar
+         *   - Traditional
+         * Default value is GREGORIAN
+	 * @param integer $which the calendar (an enumerated constant) to use.
+         * @return boolean 'true' if successful, 'false' if an error occurred or if the calendar was not recognized 
+	 */
+         public function setCalendar($which) {}
+
+
+        /**
+	 * Gets the Calendar in use 
+	 * @return integer the calendar being used by the formatter
+	 */
+         public function getCalendar() {}
+
+
+        /**
+	 * Gets the pattern in use 
+    	 * @return string the pattern string being used to format/parse 
+	 */
+	public function getPattern() {}
+
+
+        /**
+	 * Sets the pattern to  use 
+         * @param  string $pattern new pattern string to use
+         * @return boolean 'true' if successful, 'false' if an error occured. Bad format
+         *                 strings are usually the cause of the latter.
+         */
+	public function setPattern($pattern) {}
+
+
+	/**
+	 * Get the error code from last operation
+	 *
+	 * Returns error code from the last number formatting operation.
+	 *
+	 * @return integer the error code, one of UErrorCode values. Initial value is U_ZERO_ERROR.
+	 */
+	public function getErrorCode() {}
+
+
+	/**
+	 * Get the error text from the last operation.
+	 *
+	 * @return string Description of the last error.
+	 */
+	public function getErrorMessage() {}
+
+
+}
+
+#############################################################################
+# Procedural API
+#############################################################################
+
+
+	/**
+	 * Create a date formatter 
+	 *
+	 * @param string  $locale     Locale to use when formatting or parsing
+	 * @param integer $datetype   Date type to use (none, short, medium, long, full)
+	 * @param integer $timetype   Time type to use (none, short, medium, long, full)
+	 * @param [string] $timezone  Time zone ID ; default is system default
+	 * @param [integer] $calendar Calendar to use for formatting or parsing; default is
+	 *                            GREGORIAN
+	 * @param [string]  $pattern  Optional pattern to use when formatting or parsing
+	 * @return DateFormatter
+	 * @see datefmt_create
+	 */
+	function datefmt_create($locale, $datetype, $timetype, $timezone = null, $calendar= null ,$pattern=null ) {}
+
+
+	/**
+         * formats the time value as a string.
+         * @param DateFormatter $fmt The date formatter resource
+	 * @param mixed $value - value to format
+         *     integer: a unix timestamp value (seconds since epoch, UTC)
+         *     array: a localtime array   - uses 24 hour clock in tm_hour field
+         * @return string  a formatted string or, if an error occurred, 'null'. 
+	 */
+        function datefmt_format($fmt , $value) {}
+
+
+        /**
+         * converts string $value to an incremental time value, starting at
+	 * $parse_pos and consuming as much of the input value as possible
+	 * If no error occurs before $value is consumed, $parse_pos will contain -1
+	 * otherwise it will contain the position at which parsing ended (and the error
+	 * occurred). 
+         * @param DateFormatter $fmt The date formatter resource
+	 * @param string  $value      string to convert to a time
+	 * @param integer $parse_pos  position at which to start the parsing in $value (zero-based)
+	 *                            This variable will contain the end position if the parse fails
+	 *                            If $parse_pos > strlen($value), the parse fails immediately.
+	 * @return integer timestamp  parsed value
+	 */
+	 function datefmt_parse($fmt , $value, $parse_pos=0) {}
+
+	 
+        /**
+         * converts string $value to a field-based time value, starting at
+	 * $parse_pos and consuming as much of the input value as possible
+	 * If no error occurs before $value is consumed, $parse_pos will contain -1
+	 * otherwise it will contain the position at which parsing ended (and the error
+	 * occurred). 
+         * @param DateFormatter $fmt The date formatter resource
+	 * @param string  $value      string to convert to a time
+	 * @param integer $parse_pos  position at which to start the parsing in $value (zero-based)
+	 *                            This variable will contain the end position if the parse fails
+	 *                            If $parse_pos > strlen($value), the parse fails immediately.
+	 * @return array localtime compatible array of integers  - uses 24 hour clock in tm_hour field
+	 */
+	 function datefmt_localtime($fmt , $value, $parse_pos=0) {}
+
+
+	 /**
+	  * Gets the  Datetype in use
+          * @param DateFormatter $fmt The date formatter resource
+	  * @return integer the current 'datetype' value of the formatter
+	  */
+         function datefmt_get_datetype($fmt ) {}
+
+
+	 /**
+	  * Gets the  timetype in use
+          * @param DateFormatter $fmt The date formatter resource
+	  * @return integer the current 'timetype' value of the formatter
+	  */
+         function datefmt_get_timetype($fmt) {}
+
+
+	 /**
+	  * Gets the leniency of the formatter 
+          * @param DateFormatter $fmt The date formatter resource
+	  * @return boolean   'true' if parser is lenient, 'false' if parser is strict
+	  *                   default value for parser is 'false'.
+	  */
+         function datefmt_is_lenient($fmt) {}
+
+
+	 /**
+	  * Sets the leniency of the formatter 
+          * @param DateFormatter $fmt The date formatter resource
+	  * @param boolean $lenient  sets whether the parser is lenient or not, default is 'false'
+          *                          'true' sets the parser to accept otherwise flawed date or 
+	  *                          time patterns, parsing as much as possible to obtain a value.
+          *                          'false' sets the parser to strictly parse strings into dates. 
+	  *                          Extra space, unrecognized tokens, or invalid values 
+	  *                          ("Feburary 30th") are not accepted.
+	  *
+          * @return boolean          'true' if successful; 'false' if an error occurred. 
+	  */
+	 function datefmt_set_lenient($fmt , $lenient) {}
+
+
+	 /**
+	  * Gets the locale in  use
+          * @param DateFormatter $fmt The date formatter resource
+	  * @param  [integer]  which locale should be returned? 
+	  *                    values may include ULOC_ACTUAL_LOCALE,
+	  *                    ULOC_VALID_LOCALE. By default the actual
+	  *                    locale is returned.
+	  * @return string     the locale of this formatter  or 'false' if error
+	 */
+	 function datefmt_get_locale($fmt , $type = ULOC_ACTUAL_LOCALE) {}
+
+
+         /**
+	  * Gets the time zone id in  use
+          * @param DateFormatter $fmt The date formatter resource
+	  * @return string ID string for the time zone used by this formatter
+	  */
+	 function datefmt_get_timezone_id($fmt) {}
+
+
+	/**
+	 * Sets the time zone to use
+         * @param DateFormatter $fmt The date formatter resource
+	 * @param string $zone zone ID string of the time zone to use.
+	 *                     if null or the empty string, the default time zone for
+	 *                     the runtime is used.
+	 * @return boolean 'true' on successful setting of the time zone, 'false'
+	 *                 if an error occurred (such as the time zone wasn't recognized).
+	 */
+        function datefmt_set_timezone_id($fmt , $zone) {}
+
+
+        /**
+         * Sets the calendar used to the appropriate calendar, which must be 
+	 * one of the constants defined above. Some examples include:
+         *   - Gregorian calendar
+         *   - Traditional
+         * Default value is GREGORIAN
+         * @param DateFormatter $fmt The date formatter resource
+	 * @param integer $which the calendar (an enumerated constant) to use.
+         * @return boolean 'true' if successful, 'false' if an error occurred or if the calendar was not recognized 
+	 */
+         function datefmt_set_calendar($fmt , $which) {}
+
+
+        /**
+	 * Gets the calendar in use
+         * @param DateFormatter $fmt The date formatter resource
+	 * @return integer the calendar being used by the formatter
+	 */
+         function datefmt_get_calendar($fmt) {}
+
+
+        /**
+	 * Gets the pattern in use
+         * @param DateFormatter $fmt The date formatter resource
+         * @return string the pattern string being used to format/parse
+         */
+        function  datefmt_get_pattern($fmt) {}
+
+
+        /**
+	 * Sets the pattern to use
+         * @param DateFormatter $fmt The date formatter resource
+         * @param  string $pattern new pattern string to use
+         * @return boolean 'true' if successful, 'false' if an error occured. Bad format
+         *                 strings are usually the cause of the latter.
+         */
+        function datefmt_set_pattern($fmt , $pattern) {}
+
+
+	/**
+	 * Get the error code from last operation
+	 *
+         * @param DateFormatter $fmt The date formatter resource
+	 * Returns error code from the last number formatting operation.
+	 *
+	 * @return integer the error code, one of UErrorCode values. Initial value is U_ZERO_ERROR.
+	 */
+	function datefmt_get_error_code($fmt) {}
+
+
+	/**
+	 * Get the error text from the last operation.
+	 *
+         * @param DateFormatter $fmt The date formatter resource
+	 * @return string Description of the last error.
+	 */
+	function datefmt_get_error_message($fmt) {}
+
+
+?>

ext/intl/doc/formatter_api.php

@@ -0,0 +1,502 @@
+<?php
+
+/**
+ * Number formatter class - locale-dependent number formatting/parsing.
+ *
+ * This class represents the ICU number formatting functionality. It allows to display
+ * number according to the localized format or given pattern or set of rules, and to
+ * parse strings into numbers according to the above patterns.
+ *
+ * Example:
+ * <code>
+ * $value = 1234567;
+ * $formatter = new NumberFormatter("de_DE", NumberFormatter::DECIMAL);
+ * echo $formatter->format($value);
+ * </code>
+ *
+ * @see http://www.icu-project.org/apiref/icu4c/unum_8h.html
+ * @see http://www.icu-project.org/apiref/icu4c/classNumberFormat.html
+ *
+ * The class would also contain all the constants listed in the following enums:
+ * UNumberFormatStyle, UNumberFormatRoundingMode, UNumberFormatPadPosition,
+ * UNumberFormatAttribute, UNumberFormatTextAttribute, UNumberFormatSymbol.
+ */
+class NumberFormatter {
+#############################################################################
+# Common constants.
+#############################################################################
+
+	/*
+	 * WARNING:
+	 * The values described here are NOT the actual values in PHP code.
+	 * They are references to the ICU C definitions, so the line
+	 *    const PATTERN_DECIMAL = 'UNUM_PATTERN_DECIMAL';
+	 * actually means that NumberFormatter::PATTERN_DECIMAL is the same as
+	 * UNUM_PATTERN_DECIMAL constant in the ICU library.
+	 */
+
+	/*
+	 * These constants define formatter/parser argument type - integer, floating point or currency.
+	 */
+	const TYPE_DEFAULT     = 'FORMAT_TYPE_DEFAULT';
+	const TYPE_INT32       = 'FORMAT_TYPE_INT32';
+	const TYPE_INT64       = 'FORMAT_TYPE_INT64';
+	const TYPE_DOUBLE      = 'FORMAT_TYPE_DOUBLE';
+	const TYPE_CURRENCY    = 'FORMAT_TYPE_CURRENCY';
+
+	/*
+	 * UNumberFormatStyle constants
+	 */
+	const PATTERN_DECIMAL   = 'UNUM_PATTERN_DECIMAL';
+	const DECIMAL           = 'UNUM_DECIMAL';
+	const CURRENCY          = 'UNUM_CURRENCY';
+	const PERCENT           = 'UNUM_PERCENT';
+	const SCIENTIFIC        = 'UNUM_SCIENTIFIC';
+	const SPELLOUT          = 'UNUM_SPELLOUT';
+	const ORDINAL           = 'UNUM_ORDINAL';
+	const DURATION          = 'UNUM_DURATION';
+	const PATTERN_RULEBASED = 'UNUM_PATTERN_RULEBASED';
+	const DEFAULT           = 'UNUM_DEFAULT';
+	const IGNORE            = 'UNUM_IGNORE';
+
+	/*
+	 * UNumberFormatRoundingMode
+	 */
+	const ROUND_CEILING  = 'UNUM_ROUND_CEILING';
+	const ROUND_FLOOR    = 'UNUM_ROUND_FLOOR';
+	const ROUND_DOWN     = 'UNUM_ROUND_DOWN';
+	const ROUND_UP       = 'UNUM_ROUND_UP';
+	const ROUND_HALFEVEN = 'UNUM_ROUND_HALFEVEN';
+	const ROUND_HALFDOWN = 'UNUM_ROUND_HALFDOWN';
+	const ROUND_HALFUP   = 'UNUM_ROUND_HALFUP';
+
+	/*
+	 * UNumberFormatPadPosition
+	 */
+	const PAD_BEFORE_PREFIX = 'UNUM_PAD_BEFORE_PREFIX';
+	const PAD_AFTER_PREFIX  = 'UNUM_PAD_AFTER_PREFIX';
+	const PAD_BEFORE_SUFFIX = 'UNUM_PAD_BEFORE_SUFFIX';
+	const PAD_AFTER_SUFFIX  = 'UNUM_PAD_AFTER_SUFFIX';
+
+	/*
+	 * UNumberFormatAttribute
+	 */
+	const PARSE_INT_ONLY          = 'UNUM_PARSE_INT_ONLY';
+	const GROUPING_USED           = 'UNUM_GROUPING_USED';
+	const DECIMAL_ALWAYS_SHOWN    = 'UNUM_DECIMAL_ALWAYS_SHOWN';
+	const MAX_INTEGER_DIGITS      = 'UNUM_MAX_INTEGER_DIGITS';
+	const MIN_INTEGER_DIGITS      = 'UNUM_MIN_INTEGER_DIGITS';
+	const INTEGER_DIGITS          = 'UNUM_INTEGER_DIGITS';
+	const MAX_FRACTION_DIGITS     = 'UNUM_MAX_FRACTION_DIGITS';
+	const MIN_FRACTION_DIGITS     = 'UNUM_MIN_FRACTION_DIGITS';
+	const FRACTION_DIGITS         = 'UNUM_FRACTION_DIGITS';
+	const MULTIPLIER              = 'UNUM_MULTIPLIER';
+	const GROUPING_SIZE           = 'UNUM_GROUPING_SIZE';
+	const ROUNDING_MODE           = 'UNUM_ROUNDING_MODE';
+	const ROUNDING_INCREMENT      = 'UNUM_ROUNDING_INCREMENT';
+	const FORMAT_WIDTH            = 'UNUM_FORMAT_WIDTH';
+	const PADDING_POSITION        = 'UNUM_PADDING_POSITION';
+	const SECONDARY_GROUPING_SIZE = 'UNUM_SECONDARY_GROUPING_SIZE';
+	const SIGNIFICANT_DIGITS_USED = 'UNUM_SIGNIFICANT_DIGITS_USED';
+	const MIN_SIGNIFICANT_DIGITS  = 'UNUM_MIN_SIGNIFICANT_DIGITS';
+	const MAX_SIGNIFICANT_DIGITS  = 'UNUM_MAX_SIGNIFICANT_DIGITS';
+	const LENIENT_PARSE           = 'UNUM_LENIENT_PARSE';
+
+	/*
+	 * UNumberFormatTextAttribute
+	 */
+	const POSITIVE_PREFIX   = 'UNUM_POSITIVE_PREFIX';
+	const POSITIVE_SUFFIX   = 'UNUM_POSITIVE_SUFFIX';
+	const NEGATIVE_PREFIX   = 'UNUM_NEGATIVE_PREFIX';
+	const NEGATIVE_SUFFIX   = 'UNUM_NEGATIVE_SUFFIX';
+	const PADDING_CHARACTER = 'UNUM_PADDING_CHARACTER';
+	const CURRENCY_CODE     = 'UNUM_CURRENCY_CODE';
+	const DEFAULT_RULESET   = 'UNUM_DEFAULT_RULESET';
+	const PUBLIC_RULESETS   = 'UNUM_PUBLIC_RULESETS';
+
+	/*
+	 * UNumberFormatSymbol
+	 */
+	const DECIMAL_SEPARATOR_SYMBOL           = 'UNUM_DECIMAL_SEPARATOR_SYMBOL';
+	const GROUPING_SEPARATOR_SYMBOL          = 'UNUM_GROUPING_SEPARATOR_SYMBOL';
+	const PATTERN_SEPARATOR_SYMBOL           = 'UNUM_PATTERN_SEPARATOR_SYMBOL';
+	const PERCENT_SYMBOL                     = 'UNUM_PERCENT_SYMBOL';
+	const ZERO_DIGIT_SYMBOL                  = 'UNUM_ZERO_DIGIT_SYMBOL';
+	const DIGIT_SYMBOL                       = 'UNUM_DIGIT_SYMBOL';
+	const MINUS_SIGN_SYMBOL                  = 'UNUM_MINUS_SIGN_SYMBOL';
+	const PLUS_SIGN_SYMBOL                   = 'UNUM_PLUS_SIGN_SYMBOL';
+	const CURRENCY_SYMBOL                    = 'UNUM_CURRENCY_SYMBOL';
+	const INTL_CURRENCY_SYMBOL               = 'UNUM_INTL_CURRENCY_SYMBOL';
+	const MONETARY_SEPARATOR_SYMBOL          = 'UNUM_MONETARY_SEPARATOR_SYMBOL';
+	const EXPONENTIAL_SYMBOL                 = 'UNUM_EXPONENTIAL_SYMBOL';
+	const PERMILL_SYMBOL                     = 'UNUM_PERMILL_SYMBOL';
+	const PAD_ESCAPE_SYMBOL                  = 'UNUM_PAD_ESCAPE_SYMBOL';
+	const INFINITY_SYMBOL                    = 'UNUM_INFINITY_SYMBOL';
+	const NAN_SYMBOL                         = 'UNUM_NAN_SYMBOL';
+	const SIGNIFICANT_DIGIT_SYMBOL           = 'UNUM_SIGNIFICANT_DIGIT_SYMBOL';
+	const MONETARY_GROUPING_SEPARATOR_SYMBOL = 'UNUM_MONETARY_GROUPING_SEPARATOR_SYMBOL';
+
+	/**
+	 * Create a number formatter
+	 *
+	 * Creates a number formatter from locale and pattern. This formatter would be used to
+	 * format or parse numbers.
+	 *
+	 * @param integer $style     Style of the formatting, one of the UNumberFormatStyle constants
+	 * @param string $locale     Locale in which the number would be formatted
+	 * @param [string] $pattern  Pattern string in case chose style requires pattern
+	 * @return NumberFormatter
+	 */
+	public function __construct($locale, $style, $pattern = null) {}
+
+	/**
+	 * Create a number formatter
+	 *
+	 * Creates a number formatter from locale and pattern. This formatter would be used to
+	 * format or parse numbers.
+	 *
+	 * This method is useful when you prefer just to get null on error,
+	 * as if you called numfmt_create().
+	 *
+	 * @param integer  $style    Style of the formatting, one of the UNumberFormatStyle constants
+	 * @param string   $locale   Locale in which the number would be formatted
+	 * @param [string] $pattern  Pattern string in case chose style requires pattern
+	 * @return NumberFormatter
+	 * @see __construct
+	 * @see numfmt_create
+	 */
+	public static function create($locale, $style, $pattern = null) {}
+
+	/**
+	 * Format a number according to current formatting rules.
+	 *
+	 * If the type is not specified, the type is derived from the $number parameter. I.e., if it's
+	 * integer then INT32 would be chosen on 32-bit, INT64 on 64-bit, if it's double, DOUBLE would be
+	 * chosen. It is possible to format 64-bit number on 32-bit machine by passing it as double and using
+	 * TYPE_INT64.
+	 * When formatting currency, default formatter's currency code is used.
+	 *
+	 * @param integer|double $number Number to format
+	 * @param [integer]      $type   Type of the formatting - one of TYPE constants. If not specified, default for the type.
+	 * @return string formatted number
+	 */
+	public function format($number, $type = 0) {}
+
+	/**
+	 * Parse a number according to current formatting rules.
+	 *
+	 * @param string     $string   String to parse
+	 * @param [integer]  $type     Type of the formatting - one of TYPE constants.
+	 *                             TYPE_DOUBLE is used by default.
+	 * @param [integer]  $position On input, the position to start parsing, default is 0;
+	 *                             on output, moved to after the last successfully parse character;
+	 *                             on parse failure, does not change.
+	 * @return integer|double|false  Parsed number, false if parsing failed
+	 */
+	public function parse($string, $type, &$position) {}
+
+	/**
+	 * Format number as currency.
+	 *
+	 * Uses user-defined currency string.
+	 *
+	 * @param double $number    Number to format
+	 * @param string $currency  3-letter currency code (ISO 4217) to use in format
+	 */
+	public function formatCurrency($number, $currency) {}
+
+	/**
+	 * Parse currency string
+	 *
+	 * This parser would use parseCurrency API string to parse currency string. The format is defined by the
+	 * formatter, returns both number and currency code.
+	 *
+	 * @param string    $string    String to parse
+	 * @param string    $currency  Parameter to return parsed currency code
+	 * @param [integer] $position  On input, the position within text to match, default is 0;
+	 *                             on output, the position after the last matched character;
+	 *                             on parse failure, does not change.
+	 * @return double currency number
+	 */
+	public function parseCurrency($string, &$currency, &$position) {}
+
+	/**
+	 * Set formatter attribute.
+	 *
+	 * This function is used to set any of the formatter attributes. Example:
+	 *
+	 * $formatter->setAttribute(NumberFormat::FORMAT_WIDTH, 10);
+ 	 *
+	 * @param integer        $attr  One of UNumberFormatAttribute constants
+	 * @param integer|double $value Value of the attribute
+	 * @return false if attribute is unknown or can not be set, true otherwise
+	 */
+	public function setAttribute($attr, $value) {}
+	/**
+	 * Set formatter attribute.
+	 *
+	 * This function is used to set any of the formatter attributes. Example:
+	 *
+	 * $formatter->setTextAttribute(NumberFormat::POSITIVE_PREFIX, "+");
+	 *
+	 * @param integer $attr  One of UNumberFormatTextAttribute constants
+	 * @param string  $value Value of the attribute
+	 * @return false if attribute is unknown or can not be set, true otherwise
+	 */
+	public function setTextAttribute($attr, $value) {}
+	/**
+	 * Set formatting symbol.
+	 *
+	 * Example:
+	 *
+	 * $formatter->setSymbol(NumberFormat::EXPONENTIAL_SYMBOL, "E");
+	 *
+	 * @param integer|array $attr  One of UNumberFormatSymbol constants or array of symbols, indexed by
+	 * 								these constants
+	 * @param string        $value Value of the symbol
+	 */
+	public function setSymbol($attr, $value) {}
+	/**
+	 * Set pattern used by the formatter
+	 *
+	 * Valid only if the formatter is using pattern and is not rule-based.
+	 * @see http://www.icu-project.org/apiref/icu4c/classDecimalFormat.html
+	 * Localized patterns are not currently supported.
+	 *
+	 * @param string $pattern  The pattern to be used.
+	 * @return boolean         false if formatter pattern could not be set, true otherwise
+	 */
+	public function setPattern($pattern) {}
+	/**
+	 * Get value of the formatter attribute
+	 *
+	 * @param integer $attr One of UNumberFormatAttribute constants
+	 * @return integer|double value of the attribute or false if the value can not be obtained
+	 */
+	public function getAttribute($attr) {}
+	/**
+	 * Get value of the formatter attribute
+	 *
+	 * @param integer $attr One of UNumberFormatTextAttribute constants
+	 * @return string value of the attribute or false if the value can not be obtained
+	 */
+	public function getTextAttribute($attr) {}
+	/**
+	 * Get value of the formatter symbol
+	 *
+	 * @param integer $attr One of UNumberFormatSymbol constants specifying the symbol
+	 * @return string|false The symbol value, or false if the value can not be obtained
+	 */
+	public function getSymbol($attr) {}
+	/**
+	 * Get pattern used by the formatter.
+	 *
+	 * Gets current state of the formatter as a pattern.
+	 * Localized patterns are not currently supported.
+	 *
+	 * Valid only if the formatter is UNUM_PATTERN_DECIMAL
+	 * @return string|false The pattern used by the formatter or false if formatter is of a type
+	 *                      that does not support patterns.
+	 */
+	public function getPattern() {}
+	/**
+	 * Get the locale for which the formatter was created.
+	 *
+	 * @param [integer] $type One of  ULocDataLocaleType  values
+	 * @return string locale name
+	 */
+	public function getLocale($type = 0) {}
+	/**
+	 * Get the error code from last operation
+	 *
+	 * Returns error code from the last number formatting operation.
+	 *
+	 * @return integer the error code, one of UErrorCode values. Initial value is U_ZERO_ERROR.
+	 */
+	public function getErrorCode() {}
+	/**
+	 * Get the error text from the last operation.
+	 *
+	 * @return string Description of the last occured error.
+	 */
+	public public function getErrorMessage() {}
+
+}
+
+/** Now the same as procedural API */
+
+/**
+ * Create a number formatter
+ *
+ * Creates a number formatter from locale and pattern. This formatter would be used to
+ * format or parse numbers.
+ *
+ * @param string   $locale   Locale in which the number would be formatted
+ * @param integer  $style    Style of the formatting, one of the UNumberFormatStyle constants
+ * @param [string] $pattern  Pattern string in case chose style requires pattern
+ * @return Numberformatter resource NumberFormatter
+ */
+function numfmt_create($locale, $style, $pattern = null) {}
+/**
+ * Format a number according to current formatting rules.
+ *
+ * If the type is not specified, the type is derived from the $number parameter. I.e., if it's
+ * integer then INT32 would be chosen on 32-bit, INT64 on 64-bit, if it's double, DOUBLE would be
+ * chosen. It is possible to format 64-bit number on 32-bit machine by passing it as double and using
+ * TYPE_INT64.
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param integer|double  $number	 Number to format
+ * @param [integer]       $type		 Type of the formatting - one of TYPE constants. If not specified, default for the type.
+ * @return string formatted number
+ */
+function numfmt_format($formatter, $number, $type = null) {}
+/**
+ * Parse a number according to current formatting rules.
+ *
+ * This parser uses DOUBLE type by default. When parsing currency,
+ * default currency definitions are used.
+ *
+ * @param NumberFormatter $formatter       The formatter resource
+ * @param string                 $string   String to parse
+ * @param [integer]              $type     Type of the formatting - one of TYPE constants.
+ * @param [integer]              $position String position after the end of parsed data.
+ * @return integer|double|false  Parsed number, false if parsing failed
+ */
+function numfmt_parse($formatter, $string, $type, &$position) {}
+/**
+ * Format number as currency.
+ *
+ * Uses user-defined currency string.
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param double          $number    Number to format
+ * @param string          $currency  3-letter currency code (ISO 4217) to use in format
+ */
+function numfmt_format_currency($formatter, $number, $currency) {}
+/**
+ * Parse currency string
+ *
+ * This parser would use parseCurrency API string to parse currency string. The format is defined by the
+ * formatter, returns both number and currency code.
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param string          $string    String to parse
+ * @param string          $currency  Parameter to return parsed currency code
+ * @param [integer]       $position  String position after the end of parsed data.
+ * @return double currency number
+ */
+function numfmt_parse_currency($formatter, $string, &$currency, &$position) {}
+/**
+ * Set formatter attribute.
+ *
+ * This function is used to set any of the formatter attributes. Example:
+ *
+ * numfmt_format_set_attribute($formatter, NumberFormat::FORMAT_WIDTH, 10);
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param integer         $attr      One of UNumberFormatAttribute constants
+ * @param integer|double  $value     Value of the attribute
+ * @return false if attribute is unknown or can not be set, true otherwise
+ */
+function numfmt_set_attribute($formatter, $attribute, $value) {}
+/**
+ * Set formatter attribute.
+ *
+ * This function is used to set any of the formatter attributes. Example:
+ *
+ * numfmt_format_set_text_attribute($formatter, NumberFormat::POSITIVE_PREFIX, "+");
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param integer         $attr      One of UNumberFormatTextAttribute constants
+ * @param string          $value     Value of the attribute
+ * @return false if attribute is unknown or can not be set, true otherwise
+ */
+function numfmt_set_text_attribute($formatter, $attribute, $value) {}
+/**
+ * Set formatting symbol.
+ *
+ * Example:
+ *
+ * $formatter->setSymbol(NumberFormat::EXPONENTIAL_SYMBOL, "E");
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param integer|array   $attr      One of UNumberFormatSymbol constants or array of symbols,
+ *                                   indexed by these constants
+ * @param string $value Value of the symbol
+ */
+function numfmt_set_symbol($formatter, $attribute, $value) {}
+/**
+ * Set pattern used by the formatter
+ *
+ * Valid only if the formatter is using pattern and is not rule-based.
+ * @see http://www.icu-project.org/apiref/icu4c/classDecimalFormat.html
+ * Localized patterns are not currently supported.
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param string          $pattern   The pattern to be used.
+ * @return boolean false if formatter pattern could not be set, true otherwise
+ */
+function numfmt_set_pattern($formatter, $pattern) {}
+/**
+ * Get value of the formatter attribute
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param integer         $attribute One of UNumberFormatAttribute constants
+ * @return integer|double value of the attribute or false if the value can not be obtained
+ */
+function numfmt_get_attribute($formatter, $attribute) {}
+/**
+ * Get value of the formatter attribute
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param integer         $attribute One of UNumberFormatTextAttribute constants
+ * @return string value of the attribute or false if the value can not be obtained
+ */
+function numfmt_get_text_attribute($formatter, $attribute) {}
+/**
+ * Get value of the formatter symbol
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param integer         $attribute One of UNumberFormatSymbol constants specifying the symbol
+ * @return string|false The symbol value, or false if the value can not be obtained
+ */
+function numfmt_get_symbol($formatter, $attribute) {}
+/**
+ * Get pattern used by the formatter.
+ *
+ * Gets current state of the formatter as a pattern.
+ * Localized patterns are not currently supported.
+ *
+ * Valid only if the formatter is   UNUM_PATTERN_DECIMAL
+ * @param NumberFormatter $formatter The formatter resource
+ * @return string|false The pattern used by the formatter or false if formatter is of a type
+ *                      that does not support patterns.
+ */
+function numfmt_get_pattern($formatter) {}
+/**
+ * Get the locale for which the formatter was created.
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @param [integer]       $type      One of ULocDataLocaleType values
+ * @return string locale name
+ */
+function numfmt_get_locale($formatter, $type = 0) {}
+/**
+ * Get the error code from last operation
+ *
+ * Returns error code from the last number formatting operation.
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @return integer the error code, one of UErrorCode values. Initial value is U_ZERO_ERROR.
+ */
+function numfmt_get_error_code($formatter) {}
+/**
+ * Get the error text from the last operation.
+ *
+ * @param NumberFormatter $formatter The formatter resource
+ * @return string Description of the last occured error.
+ */
+function numfmt_get_error_message($formatter) {}
+
+?>

ext/intl/doc/grapheme_api.php

@@ -0,0 +1,132 @@
+<?php
+#############################################################################
+# Grapheme constants.
+#############################################################################
+
+	/**
+	 * grapheme_extract extract_type
+	 *
+	*/
+	/** Extract the given number of whole grapheme clusters from the string: */
+	define('GRAPHEME_EXTR_COUNT', 0);
+	/** Extract as many whole grapheme clusters as will fit into the given number of bytes: */
+	define('GRAPHEME_EXTR_MAXBYTES', 1);
+	/** Extract whole grapheme clusters up to a maximum number of UTF-8 characters: */
+	define('GRAPHEME_EXTR_MAXCHARS', 2);
+
+
+#############################################################################
+# Grapheme API
+#############################################################################
+
+	/**
+	 * Get string length in grapheme units
+	 * @param  string	$input		The string being measured for length.
+	 * @return int		The length of the string on success, and 0 if the string is empty.
+	*/
+	function grapheme_strlen($input) {}
+
+	/**
+	 * Find position (in grapheme units) of first occurrence of a string
+	 * @param string	$haystack	The string to look in
+	 * @param string	$needle		The string to look for
+	 * @param [int]		$offset		The optional offset parameter allows you to specify 
+						which character in haystack  to start searching. The position 
+						returned is still relative to the beginning of haystack.
+	 * @return int		Returns the position as an integer. If needle is not found, strpos() will return boolean FALSE.
+	*/
+	function grapheme_strpos($haystack, $needle, $offset = 0) {}
+
+	
+	/**
+         * Find position (in grapheme units) of first occurrence of a case-insensitive string
+         * @param string        $haystack       The string to look in
+         * @param string        $needle         The string to look for
+         * @param [int]         $offset         The optional offset parameter allows you to specify
+                                                which character in haystack  to start searching. The position
+                                                returned is still relative to the beginning of haystack.
+         * @return int          Returns the position as an integer. If needle is not found, grapheme_stripos() will return boolean FALSE.
+        */
+        function grapheme_stripos($haystack, $needle, $offset = 0) {}
+ 
+
+	/**
+         * Find position (in grapheme units) of last occurrence of a string
+         * @param string        $haystack       The string to look in
+         * @param string        $needle         The string to look for
+         * @param [int]         $offset         The optional offset parameter allows you to specify
+                                                which character in haystack  to start searching. The position
+                                                returned is still relative to the beginning of haystack.
+         * @return int          Returns the position as an integer. If needle is not found, grapheme_strrpos() will return boolean FALSE.
+        */
+        function grapheme_strrpos($haystack, $needle, $offset = 0) {}
+
+
+	/**
+         * Find position (in grapheme units) of last occurrence of a case-insensitive string
+         * @param string        $haystack       The string to look in
+         * @param string        $needle         The string to look for
+         * @param [int]         $offset         The optional offset parameter allows you to specify
+                                                which character in haystack  to start searching. The position
+                                                returned is still relative to the beginning of haystack.
+         * @return int          Returns the position as an integer. If needle is not found, grapheme_strripos() will return boolean FALSE.
+        */
+        function grapheme_strripos($haystack, $needle, $offset = 0) {}
+
+
+	/**
+	 * Return part of a string
+	 * @param string	$string		The input string.
+	 * @param int		$start		If start  is non-negative, the returned string will start at the 
+						start'th position in string, counting from zero. If start is negative,
+						the returned string will start at the start'th character from the 
+						end of string.
+	 * @param [int]		$length		If length  is given and is positive, the string returned will contain
+						at most length characters beginning from start (depending on the 
+						length of string). If string is less than or equal to start characters
+						long, FALSE  will be returned. If length is given and is negative, then
+						that many characters will be omitted from the end of string (after the
+						start position has been calculated when a start is negative). If start
+						denotes a position beyond this truncation, an empty string will be returned.
+	 * @return int		Returns the extracted part of string.
+	*/
+	function grapheme_substr($string, $start, $length = -1) {}
+
+
+	/**
+	 * Returns part of haystack string from the first occurrence of needle to the end of haystack.
+	 * @param string	$haystack	The input string.
+	 * @param string	$needle		The string to look for.
+	 * @param [boolean]	$before_needle	If TRUE (the default is FALSE), grapheme_strstr() returns the part of the
+						haystack before the first occurence of the needle.
+	 * @return string	Returns the portion of string, or FALSE if needle is not found.
+	*/
+	function grapheme_strstr($haystack, $needle, $before_needle = FALSE) {}
+
+
+	/**
+	 * Returns part of haystack string from the first occurrence of case-insensitive needle to the end of haystack.
+         * @param string        $haystack       The input string.
+         * @param string        $needle         The string to look for.
+         * @param [boolean]     $before_needle  If TRUE (the default is FALSE), grapheme_strstr() returns the part of the
+                                                haystack before the first occurence of the needle.
+         * @return string       Returns the portion of string, or FALSE if needle is not found.
+        */
+        function grapheme_stristr($haystack, $needle, $before_needle = FALSE) {}
+
+
+	/**
+	 * Function to extract a sequence of default grapheme clusters from a text buffer, which must be encoded in UTF-8.
+	 * @param string 	$haystack	string to search
+	 * @param int		$size		maximum number of units - based on the $extract_type - to return
+	 * @param [int]		$extract_type	one of GRAPHEME_EXTR_COUNT (default), GRAPHEME_EXTR_MAXBYTES, or GRAPHEME_EXTR_MAXCHARS
+	 * @param [int]		$start		starting position in $haystack in bytes
+	 * @param [&int]	$next		set to next starting position in bytes
+	 * @return string	A string starting at offset $start containing no more than $size grapheme clusters 
+				and ending on a default grapheme cluster boundary.
+	*/
+	function grapheme_extract($haystack, $size, $extract_type = GRAPHEME_EXTR_COUNT, $start = 0, &$next) {}
+
+?>
+
+

ext/intl/doc/locale_api.php

@@ -0,0 +1,432 @@
+<?php
+
+/**
+ * A "Locale" is an identifier used to get language, culture, or regionally-specific 
+ * behavior from an API. PHP locales are organized and identified the same
+ * way that the CLDR locales used by ICU (and many vendors of Unix-like operating 
+ * systems, the Mac, Java, and so forth) use. Locales are identified using
+ * RFC 4646 language tags (which use hyphen, not underscore) in addition to the
+ * more traditional underscore-using identifiers. Unless otherwise noted
+ * the functions in this class are tolerant of both formats. 
+ *
+ * Examples of identifiers include:
+ *
+ *  * en-US (English, United States)
+ *  * zh-Hant-TW (Chinese, Traditional Script, Taiwan)
+ *  * fr-CA, fr-FR (French for Canada and France respectively) 
+ *
+ * The Locale class (and related procedural functions) are used to interact 
+ * with locale identifiers--to verify that an ID is well-formed, valid, 
+ * etc. The extensions used by CLDR in UAX #35 (and inherited by ICU) are 
+ * valid and used wherever they would be in ICU normally.
+ *
+ * Locales cannot be instantiated as objects. All of the functions/methods 
+ * provided are static.
+ *
+ *   * The null or empty string obtains the "root" locale. 
+ *     The "root" locale is equivalent to "en_US_POSIX" in CLDR.
+ *   * Language tags (and thus locale identifiers) are case insensitive. There
+ *     exists a canonicalization function to make case match the specification.
+ *
+ * @see http://www.icu-project.org/apiref/icu4c/uloc_8h.html
+ * @see http://www.unicode.org/reports/tr35/
+ *
+ */
+class Locale {
+
+#############################################################################
+# Common constants.
+#############################################################################
+
+   	/**
+     	 * The following static members are used with the getLocale methods of
+     	 * the various locale affected classes, such as numfmt.
+     	 */
+  	const DEFAULT_LOCALE 			= default_locale;
+
+   	/**
+	 * identifiers for the actual locale, valid locale
+ 	 * WARNING:
+	 * The values described here are NOT the actual values in PHP code.
+	 * They are references to the ICU C definitions, so the line
+	 *    const ACTUAL_LOCALE = 'ULOC_ACTUAL_LOCALE';
+	 * actually means that Locale::ACTUAL_LOCALE is the same as
+	 * ULOC_ACTUAL_LOCALE constant in the ICU library.
+	 */
+	const ACTUAL_LOCALE 			= 'ULOC_ACTUAL_LOCALE';
+   	const VALID_LOCALE 			= 'ULOC_VALID_LOCALE';
+
+	/** 
+	 * Valid locale tag and subtag values
+	 */
+	LANG_TAG 				= "language";
+	EXTLANG_TAG 				= "extlang";
+	SCRIPT_TAG 				= "script";
+	REGION_TAG 				= "region";
+	VARIANT_TAG 				= "variant";
+	GRANDFATHERED_LANG_TAG 			= "grandfathered";
+	PRIVATE_TAG 				= "private";
+	
+
+#############################################################################
+# Object-oriented API
+#############################################################################
+
+	/**
+	 * Gets the default locale value from the INTL global 'default_locale'
+	 * At the PHP initilaization (MINIT) this value is set to 
+	 * 'intl.default_locale' value from php.ini if that value exists 
+	 * or from ICU's function  uloc_getDefault() 
+	 * Then onwards picks up changes from setDefault() calls also
+	 *
+     	 * @return string the current runtime locale 
+     	 */
+	public static function getDefault() {}
+
+	/**
+     	 * sets the default runtime locale to $locale
+ 	 * This changes the value of INTL global 'default_locale'
+     	 *
+     	 * @param 	string	 $locale 	is a BCP 47 compliant language tag containing the 
+     	 *					locale identifier. UAX #35 extensions are accepted.
+     	 * @return	boolean 		'true' if okay, 'false' if an error
+     	 */
+    	public static function setDefault($locale) {}
+
+
+    	/**
+	 * Gets the primary language for the input locale
+	 *
+     	 * @param	string	$locale 	the locale to extract the primary language code from
+     	 * @return 	string 			the language code associated with the language 
+	 *					or null in case of error. 
+     	 */
+    	public static function getPrimaryLanguage($locale) {}
+
+
+    	/**
+	 * Gets the script for the input locale
+	 *
+     	 * @param 	string 	$locale 	the locale to extract the script code from
+     	 * @return 	string 			the script subtag for the locale or null if not present
+     	 */
+    	public static function getScript($locale) {}
+
+
+    	/**
+	 * Gets the region for the input locale
+	 *
+     	 * @param 	string 	$locale 	the locale to extract the region code from
+     	 * @return 	string 			the region subtag for the locale or null if not present
+     	 */
+    	public static function getRegion($locale) {}
+
+
+    	/**
+	 * Gets the variants for the input locale
+	 *
+    	 * @param 	string 	$locale 	the locale to extract the variants from
+     	 * @return 	array 			the array containing the list of all variants 
+	 *					subtag for the locale or null if not present
+     	 */
+	public static function getAllVariants($locale) {}
+
+
+	/**
+	 * Gets the keywords for the input locale
+	 *
+     	 * @param 	string 	$locale 	the locale to extract the keywords from
+     	 * @return 	array 			associative array containing the keyword-value pairs for this locale 
+     	 */
+    	public static function getKeywords($locale) {}
+
+
+    	/**
+	 * Returns an appropriately localized display name for the input locale
+	 *
+    	 * @param 	string 	 	$locale 	the locale to return a displayname for
+	 * @param 	[string] 	$in_locale  	optional format locale
+	 * 						If is 'null' then the default locale is used. 
+	 * @return 	string  			display name of the locale in the format
+	 * 						appropriate for $in_locale. 
+     	 */
+    	public static function getDisplayName($locale, $in_locale = null) {}
+
+
+    	/**
+	 * Returns an appropriately localized display name for language of the input locale
+	 *
+     	 * @param 	string 	 	$locale 	the locale to return a display language for
+	 * @param 	[string] 	$in_locale  	optional format locale to use to display the language name
+	 * 						If is 'null' then the default locale is used. 
+	 * @return 	string  			display name of the language for the $locale in the format
+	 * 						appropriate for $in_locale. 
+     	 */
+	public static function getDisplayLanguage($lang, $in_locale = null) {}
+
+    	/**
+	 * Returns an appropriately localized display name for script of the input locale
+	 *
+    	 * @param 	string 	 	$locale 	the locale to return a display script for
+	 * @param 	[string] 	$in_locale  	optional format locale to use to display the script name
+	 * 						If is 'null' then the default locale is used. 
+	 * @return 	string  			display name of the script for the $locale in the format
+	 * 						appropriate for $in_locale. 
+     	 */
+	public static function getDisplayScript($script, $in_locale = null) {}
+
+
+    	/**
+	 * Returns an appropriately localized display name for region of the input locale
+	 *
+    	 * @param 	string 	 	$locale 	the locale to return a display region for
+	 * @param 	[string] 	$in_locale  	optional format locale to use to display the region name
+	 * 						If is 'null' then the default locale is used. 
+	 * @return 	string  			display name of the region for the $locale in the format
+	 * 						appropriate for $in_locale. 
+     	 */
+	public static function getDisplayRegion($region, $in_locale = null) {}
+
+
+    	/**
+	 * Returns an appropriately localized display name for variants of the input locale
+	 *
+    	 * @param 	string 	 	$locale 	the locale to return a display variant for
+	 * @param 	[string] 	$in_locale  	optional format locale to use to display the variant name
+	 * 						If is 'null' then the default locale is used. 
+	 * @return 	string  			display name of the variant for the $locale in the format
+	 * 						appropriate for $in_locale. 
+     	 */
+	public static function getDisplayVariant($variant, $in_locale = null) {}
+
+
+    	/**
+	 * Checks if a $langtag  filter matches with $locale according to 
+	 * RFC 4647's basic filtering algorithm
+	 *
+     	 * @param 	string 		$langtag	the language tag to check
+	 * @param 	string 		$locale		the language range to check against
+     	 * @return 	boolean 			'true' if $locale matches $langtag 'false' otherwise
+	 */
+    	public static function filterMatches($langtag, $locale) {}
+
+	/**
+     	 * Searchs the items in $langtag for the best match to the language
+	 * range specified in $locale according to RFC 4647's lookup algorithm. 
+	 *
+	 * @param 	array 		$langtag  	an array containing a list of language tags to compare
+	 *            					to $locale
+	 * @param 	string 		$locale   	the locale to use as the language range when matching
+	 * @param 	string 		$default  	the locale to use if no match is found
+	 * @return 	string 				closest matching language tag, $default, 
+	 *						or empty string
+	 */
+    	public static function lookup(array $langtag, $locale, $default = null) {}
+
+
+	/**
+     	 * Returns a correctly ordered and delimited locale ID 
+	 *
+	 * @param 	array 		$subtags 	an array containing a list of key-value pairs, where 
+	 * 						the keys identify the particular locale ID subtags, 	
+	 *						and the values are the associated subtag values. 
+	 *
+	 * @return 	string 				the corresponding locale identifier.
+	 */
+    	public static function composeLocale(array $subtags) {}
+
+
+	/**
+	 * Returns a key-value array of locale ID subtag elements.
+	 *
+	 * @param 	string		$locale		the locale to extract the subtag array from 					
+	 *
+	 * @return 	array 		$subtags 	an array containing a list of key-value pairs, where 
+	 * 						the keys identify the particular locale ID subtags, 	
+	 *						and the values are the associated subtag values. 
+	 */
+    	public static function parseLocale($locale) {}
+
+}
+
+#############################################################################
+# Procedural API
+#############################################################################
+
+
+	/**
+	 * Gets the default locale value from the INTL global 'default_locale'
+	 * At the PHP initilaization (MINIT) this value is set to 
+	 * 'intl.default_locale' value from php.ini if that value exists 
+	 * or from ICU's function  uloc_getDefault() 
+	 * Then onwards picks up changes from setDefault() calls also
+	 *
+     	 * @return string the current runtime locale 
+     	 */
+	public static function locale_get_default() {}
+
+	/**
+     	 * sets the default runtime locale to $locale
+ 	 * This changes the value of INTL global 'default_locale'
+     	 *
+     	 * @param 	string	 $locale 	is a BCP 47 compliant language tag containing the 
+     	 *					locale identifier. UAX #35 extensions are accepted.
+     	 * @return	boolean 		'true' if okay, 'false' if an error
+     	 */
+    	public static function locale_set_default($locale) {}
+
+
+    	/**
+	 * Gets the primary language for the input locale
+	 *
+     	 * @param	string	$locale 	the locale to extract the primary language code from
+     	 * @return 	string 			the language code associated with the language 
+	 *					or null in case of error. 
+     	 */
+    	public static function locale_get_primary_language($locale) {}
+
+
+    	/**
+	 * Gets the script for the input locale
+	 *
+     	 * @param 	string 	$locale 	the locale to extract the script code from
+     	 * @return 	string 			the script subtag for the locale or null if not present
+     	 */
+    	public static function locale_get_script($locale) {}
+
+
+    	/**
+	 * Gets the region for the input locale
+	 *
+     	 * @param 	string 	$locale 	the locale to extract the region code from
+     	 * @return 	string 			the region subtag for the locale or null if not present
+     	 */
+    	public static function locale_get_region($locale) {}
+
+
+    	/**
+	 * Gets the variants for the input locale
+	 *
+    	 * @param 	string 	$locale 	the locale to extract the variants from
+     	 * @return 	array 			the array containing the list of all variants 
+	 *					subtag for the locale or null if not present
+     	 */
+	public static function locale_get_all_variants($locale) {}
+
+
+	/**
+	 * Gets the keywords for the input locale
+	 *
+     	 * @param 	string 	$locale 	the locale to extract the keywords from
+     	 * @return 	array 			associative array containing the keyword-value pairs for this locale 
+     	 */
+    	public static function locale_get_keywords($locale) {}
+
+
+    	/**
+	 * Returns an appropriately localized display name for the input locale
+	 *
+    	 * @param 	string 	 	$locale 	the locale to return a displayname for
+	 * @param 	[string] 	$in_locale  	optional format locale
+	 * 						If is 'null' then the default locale is used. 
+	 * @return 	string  			display name of the locale in the format
+	 * 						appropriate for $in_locale. 
+     	 */
+    	public static function locale_get_display_name($locale, $in_locale = null) {}
+
+
+    	/**
+	 * Returns an appropriately localized display name for language of the input locale
+	 *
+     	 * @param 	string 	 	$locale 	the locale to return a display language for
+	 * @param 	[string] 	$in_locale  	optional format locale to use to display the language name
+	 * 						If is 'null' then the default locale is used. 
+	 * @return 	string  			display name of the language for the $locale in the format
+	 * 						appropriate for $in_locale. 
+     	 */
+	public static function locale_get_display_language($lang, $in_locale = null) {}
+
+    	/**
+	 * Returns an appropriately localized display name for script of the input locale
+	 *
+    	 * @param 	string 	 	$locale 	the locale to return a display script for
+	 * @param 	[string] 	$in_locale  	optional format locale to use to display the script name
+	 * 						If is 'null' then the default locale is used. 
+	 * @return 	string  			display name of the script for the $locale in the format
+	 * 						appropriate for $in_locale. 
+     	 */
+	public static function locale_get_display_script($script, $in_locale = null) {}
+
+
+    	/**
+	 * Returns an appropriately localized display name for region of the input locale
+	 *
+    	 * @param 	string 	 	$locale 	the locale to return a display region for
+	 * @param 	[string] 	$in_locale  	optional format locale to use to display the region name
+	 * 						If is 'null' then the default locale is used. 
+	 * @return 	string  			display name of the region for the $locale in the format
+	 * 						appropriate for $in_locale. 
+     	 */
+	public static function locale_get_display_region($region, $in_locale = null) {}
+
+
+    	/**
+	 * Returns an appropriately localized display name for variants of the input locale
+	 *
+    	 * @param 	string 	 	$locale 	the locale to return a display variant for
+	 * @param 	[string] 	$in_locale  	optional format locale to use to display the variant name
+	 * 						If is 'null' then the default locale is used. 
+	 * @return 	string  			display name of the variant for the $locale in the format
+	 * 						appropriate for $in_locale. 
+     	 */
+	public static function locale_get_display_variant($variant, $in_locale = null) {}
+
+
+    	/**
+	 * Checks if a $langtag  filter matches with $locale according to 
+	 * RFC 4647's basic filtering algorithm
+	 *
+     	 * @param 	string 		$langtag	the language tag to check
+	 * @param 	string 		$locale		the language range to check against
+     	 * @return 	boolean 			'true' if $locale matches $langtag 'false' otherwise
+	 */
+    	public static function locale_filter_matches($langtag, $locale) {}
+
+	/**
+     	 * Searchs the items in $langtag for the best match to the language
+	 * range specified in $locale according to RFC 4647's lookup algorithm. 
+	 *
+	 * @param 	array 		$langtag  	an array containing a list of language tags to compare
+	 *            					to $locale
+	 * @param 	string 		$locale   	the locale to use as the language range when matching
+	 * @param 	string 		$default  	the locale to use if no match is found
+	 * @return 	string 				closest matching language tag, $default, 
+	 *						or empty string
+	 */
+    	public static function locale_lookup(array $langtag, $locale, $default = null) {}
+
+
+	/**
+     	 * Returns a correctly ordered and delimited locale ID 
+	 *
+	 * @param 	array 		$subtags 	an array containing a list of key-value pairs, where 
+	 * 						the keys identify the particular locale ID subtags, 	
+	 *						and the values are the associated subtag values. 
+	 *
+	 * @return 	string 				the corresponding locale identifier.
+	 */
+    	public static function locale_compose_locale(array $subtags) {}
+
+
+	/**
+	 * Returns a key-value array of locale ID subtag elements.
+	 *
+	 * @param 	string		$locale		the locale to extract the subtag array from 					
+	 *
+	 * @return 	array 		$subtags 	an array containing a list of key-value pairs, where 
+	 * 						the keys identify the particular locale ID subtags, 	
+	 *						and the values are the associated subtag values. 
+	 */
+    	public static function locale_parse_locale($locale) {}
+
+?>

ext/intl/doc/msgfmt_api.php

@@ -0,0 +1,209 @@
+<?php
+
+/**
+ * Message formatter class.
+ *
+ * Message Format provides for runtime formatting of messages in a manner
+ * somewhat similar to sprintf. The pattern string has its component parts
+ * replaced in a locale-sensitive manner using items in the arguments array.
+ *
+ * @see http://www.icu-project.org/apiref/icu4c/umsg_8h.html
+ *
+ */
+class MessageFormatter {
+
+	/**
+	 * Constructs a new Message Formatter
+	 * 
+	 * @param  string   $locale  the locale to use when formatting arguments
+	 * @param  string   $pattern the pattern string to stick arguments into
+	 */
+	public function __construct($locale, $pattern) {}
+
+	/**
+	 * Constructs a new Message Formatter
+	 * 
+	 * @param  string   $locale  the locale to use when formatting arguments
+	 * @param  string   $pattern the pattern string to stick arguments into
+	 */
+	public static function create($locale, $pattern) {}
+
+	/**
+	 * Format the message
+	 * @param   array   $args   arguments to insert into the pattern string
+	 * @return  string  the formatted string, or false if an error ocurred
+	 */
+	public function format($args) {}
+
+	/**
+	 * Parse input string and returns any extracted items as an array
+	 *
+	 * $error will contain any error code. If an error occurs, $parse_pos contains
+	 * the position of the error.
+	 *
+	 * @param  string  $value      string to parse for items
+	 * @return array               array containing items extracted
+	 *
+	 */
+	public function parse($value) {}
+
+	/**
+	 * Inserts the items in $args into $pattern, formatting them
+	 * according to $locale. This is the static implementation.
+	 *
+	 * @param string    $locale   the locale to use when formatting numbers and dates and suchlike
+	 * @param string    $pattern  the pattern string to insert things into
+	 * @param array     $args     the array of values to insert into $pattern
+	 * @return string             the formatted pattern string or false if an error occured
+	 */
+	public static function formatMessage($locale, $pattern, $args) {}
+
+	/**
+	 * parses input string and returns any extracted items as an array
+	 *
+	 * $error will contain any error code. If an error occurs, $parse_pos contains
+	 * the position of the error.
+	 *
+	 * @param  string  $locale     the locale to use when formatting numbers and dates and suchlike
+	 * @param  string  $value      string to parse for items
+	 * @return array               array containing items extracted
+	 *
+	 */
+	public static function parseMessage($locale, $value) {}
+
+	/**
+	 * Get the pattern used by the formatter
+	 *
+	 * @return  string  the pattern string for this message formatter
+	 */
+	public function getPattern() {}
+
+	/**
+	 * Set the pattern used by the formatter
+	 *
+	 * @param  string  $pattern  the pattern string to use in this message formatter
+	 * @return boolean 'true' if successful, 'false' if an error
+	 */
+	public function setPattern($pattern) {}
+
+	/**
+	 * Get the error code from last operation
+	 *
+	 * Returns error code from the last number formatting operation.
+	 *
+	 * @return integer the error code, one of UErrorCode values. Initial value is U_ZERO_ERROR.
+	 */
+	public function getErrorCode() {}
+	/**
+	 * Get the error text from the last operation.
+	 *
+	 * @return string Description of the last error.
+	 */
+	public function getErrorMessage() {}
+	/**
+	 * Get the locale for which the formatter was created.
+	 *
+	 * @return string locale name
+	 */
+	public function getLocale() {}
+}
+
+	/** Now the same as procedural API */
+
+
+	/**
+	 * Constructs a new Message Formatter
+	 * 
+	 * @param  string   $locale  the locale to use when formatting arguments
+	 * @param  string   $pattern the pattern string to stick arguments into
+	 * @return MessageFormatter formatter object
+	 */
+	function msgfmt_create($locale, $pattern) {}
+
+	/**
+	 * Format the message
+	 * @param MessageFormatter $fmt The message formatter 
+	 * @param   array   $args   arguments to insert into the pattern string
+	 * @return  string  the formatted string, or false if an error ocurred
+	 */
+	function msgfmt_format($fmt, $args) {}
+
+	/**
+	 * parses input string and returns any extracted items as an array
+	 *
+	 * $error will contain any error code. If an error occurs, $parse_pos contains
+	 * the position of the error.
+	 *
+	 * @param MessageFormatter $fmt The message formatter 
+	 * @param  string  $value      string to parse for items
+	 * @return array               array containing items extracted
+	 *
+	 */
+	function msgfmt_parse($fmt, $value) {}
+
+	/**
+	 * Inserts the items in $args into $pattern, formatting them
+	 * according to $locale. This is the static implementation.
+	 *
+	 * @param string    $locale   the locale to use when formatting numbers and dates and suchlike
+	 * @param string    $pattern  the pattern string to insert things into
+	 * @param array     $args     the array of values to insert into $pattern
+	 * @return string             the formatted pattern string or false if an error occured
+	 */
+	function msgfmt_format_message($locale, $pattern, $args) {}
+
+	/**
+	 * parses input string and returns any extracted items as an array
+	 *
+	 * $error will contain any error code. If an error occurs, $parse_pos contains
+	 * the position of the error.
+	 *
+	 * @param  string  $locale     the locale to use when formatting numbers and dates and suchlike
+	 * @param  string  $value      string to parse for items
+	 * @return array               array containing items extracted
+	 *
+	 */
+	function msgfmt_parse_message($locale, $value) {}
+
+	/**
+	 * Get the pattern used by the formatter
+	 *
+	 * @param MessageFormatter $fmt The message formatter 
+	 * @return  string  the pattern string for this message formatter
+	 */
+	function msgfmt_get_pattern($fmt) {}
+
+	/**
+	 * Set the pattern used by the formatter
+	 *
+	 * @param MessageFormatter $fmt The message formatter 
+	 * @param  string  $pattern  the pattern string to use in this message formatter
+	 * @return boolean 'true' if successful, 'false' if an error
+	 */
+	function msgfmt_set_pattern($fmt, $pattern) {}
+
+	/**
+	 * Get the error code from last operation
+	 *
+	 * Returns error code from the last number formatting operation.
+	 *
+	 * @param MessageFormatter $fmt The message formatter 
+	 * @return integer the error code, one of UErrorCode values. Initial value is U_ZERO_ERROR.
+	 */
+	function msgfmt_get_error_code($fmt) {}
+	/**
+	 * Get the error text from the last operation.
+	 *
+	 * @param MessageFormatter $fmt The message formatter 
+	 * @return string Description of the last error.
+	 */
+	function msgfmt_get_error_message($fmt) {}
+
+	/**
+	 * Get the locale for which the formatter was created.
+	 *
+	 * @param NumberFormatter $formatter The formatter resource
+	 * @return string locale name
+	 */
+	function msgfmt_get_locale($formatter) {}
+?>

ext/intl/doc/normalizer_api.php

@@ -0,0 +1,92 @@
+<?php
+
+#############################################################################
+# Object-oriented API
+#############################################################################
+
+/**
+ * Normalizer class.
+ *
+ * Normalizer provides access to Unicode normalization of strings. This class consists
+ * only of static methods. The iterator interface to normalizer is rarely used, so is
+ * not provided here.
+ *
+ * Example:
+ * <code>
+ *
+ * </code>
+ *
+ * @see http://www.icu-project.org/apiref/icu4c/unorm_8h.html
+ * @see http://www.icu-project.org/apiref/icu4c/classNormalizer.html
+ *
+ */
+class Normalizer {
+#############################################################################
+# Common constants.
+#############################################################################
+
+	/**
+	 * Valid normalization form values.
+	 *
+	 * @see Normalizer::normalize()
+	 * @see Normalizer::isNormalize()
+	 * @see normalizer_normalize()
+	 * @see normalizer_is_normalized()
+	*/
+	const NONE    = 1;
+	/** Canonical decomposition. */
+	const NFD = 2;
+	const FORM_D  = NFD;
+	/** Compatibility decomposition. */
+	const NFKD = 3;
+	const FORM_KD = NFKD;
+	/** Canonical decomposition followed by canonical composition. */
+	const NFC = 4;
+	const FORM_C  = NFC;
+	/** Compatibility decomposition followed by canonical composition. */
+	const NFKC =5;
+	const FORM_KC = NFKC;
+
+
+	/**
+	 * Normalizes the input provided and returns the normalized string
+	 * @param   string     $input    The input string to normalize
+	 * @param   [int]      $form     One of the normalization forms
+	 * @return  string               The normalized string or null if an error occurred.
+	*/
+	public static function normalize($input, $form = Normalizer::FORM_C) {}
+
+	/**
+	 * Checks if the provided string is already in the specified normalization form.
+	 * @param  string    $input    The input string to normalize
+	 * @param  [int]     $form     One of the normalization forms
+	 * @return boolean             True if normalized, false otherwise or if there is an error
+	*/
+	public static function isNormalized($input, $form = Normalizer::FORM_C) {}
+
+}
+
+#############################################################################
+# Procedural API
+#############################################################################
+
+	/**
+	 * Normalizes the input provided and returns the normalized string
+	 * @param  string    $input    The input string to normalize
+	 * @param  [int]     $form     One of the normalization forms
+	 * @return string              The normalized string or null if an error occurred.
+	*/
+	function normalizer_normalize($input, $form = Normalizer::FORM_C) {}
+
+	/**
+	 * Checks if the provided string is already in the specified normalization form.
+	 * @param  string    $input    The input string to normalize
+	 * @param  [int]     $form     One of the normalization forms
+	 * @return boolean             True if normalized, false otherwise or if there an error
+	*/
+	function normalizer_is_normalized($input, $form = Normalizer::FORM_C) {}
+
+
+?>
+
+

ext/intl/formatter/formatter.c

@@ -0,0 +1,152 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <unicode/utypes.h>
+#include <unicode/unum.h>
+#include <unicode/ustring.h>
+
+#include "php_intl.h"
+#include "formatter_class.h"
+#include "formatter_format.h"
+
+#if U_ICU_VERSION_MAJOR_NUM == 3 && U_ICU_VERSION_MINOR_NUM <= 4
+#define UNUM_MONETARY_GROUPING_SEPARATOR_SYMBOL 17
+#endif
+
+/* {{{ formatter_register_constants
+ * Register constants common for the both (OO and procedural)
+ * APIs.
+ */
+void formatter_register_constants( INIT_FUNC_ARGS )
+{
+	if( NumberFormatter_ce_ptr == NULL) {
+		zend_error(E_ERROR, "NumberFormatter class not defined");
+	}
+
+	#define FORMATTER_EXPOSE_CONST(x) REGISTER_LONG_CONSTANT(#x, x, CONST_CS)
+	#define FORMATTER_EXPOSE_CLASS_CONST(x) zend_declare_class_constant_long( NumberFormatter_ce_ptr, ZEND_STRS( #x ) - 1, UNUM_##x TSRMLS_CC );
+	#define FORMATTER_EXPOSE_CUSTOM_CLASS_CONST(name, value) zend_declare_class_constant_long( NumberFormatter_ce_ptr, ZEND_STRS( name ) - 1, value TSRMLS_CC );
+
+	// UNumberFormatStyle constants
+	FORMATTER_EXPOSE_CLASS_CONST( PATTERN_DECIMAL );
+	FORMATTER_EXPOSE_CLASS_CONST( DECIMAL );
+	FORMATTER_EXPOSE_CLASS_CONST( CURRENCY );
+	FORMATTER_EXPOSE_CLASS_CONST( PERCENT );
+	FORMATTER_EXPOSE_CLASS_CONST( SCIENTIFIC );
+	FORMATTER_EXPOSE_CLASS_CONST( SPELLOUT );
+	FORMATTER_EXPOSE_CLASS_CONST( ORDINAL );
+	FORMATTER_EXPOSE_CLASS_CONST( DURATION );
+	FORMATTER_EXPOSE_CLASS_CONST( PATTERN_RULEBASED );
+	FORMATTER_EXPOSE_CLASS_CONST( IGNORE );
+
+	FORMATTER_EXPOSE_CUSTOM_CLASS_CONST( "DEFAULT_STYLE",  UNUM_DEFAULT );
+
+/* workaround for ICU bug */
+#if U_ICU_VERSION_MAJOR_NUM == 3 && U_ICU_VERSION_MINOR_NUM < 8
+#define UNUM_ROUND_HALFEVEN UNUM_FOUND_HALFEVEN
+#endif
+
+	// UNumberFormatRoundingMode 
+	FORMATTER_EXPOSE_CLASS_CONST( ROUND_CEILING );
+	FORMATTER_EXPOSE_CLASS_CONST( ROUND_FLOOR );
+	FORMATTER_EXPOSE_CLASS_CONST( ROUND_DOWN );
+	FORMATTER_EXPOSE_CLASS_CONST( ROUND_UP );
+	FORMATTER_EXPOSE_CLASS_CONST( ROUND_HALFEVEN );
+	FORMATTER_EXPOSE_CLASS_CONST( ROUND_HALFDOWN );
+	FORMATTER_EXPOSE_CLASS_CONST( ROUND_HALFUP );
+
+	// UNumberFormatPadPosition
+	FORMATTER_EXPOSE_CLASS_CONST( PAD_BEFORE_PREFIX ); 
+	FORMATTER_EXPOSE_CLASS_CONST( PAD_AFTER_PREFIX ); 
+	FORMATTER_EXPOSE_CLASS_CONST( PAD_BEFORE_SUFFIX ); 
+	FORMATTER_EXPOSE_CLASS_CONST( PAD_AFTER_SUFFIX );
+
+	// UNumberFormatAttribute
+	FORMATTER_EXPOSE_CLASS_CONST( PARSE_INT_ONLY ); 
+	FORMATTER_EXPOSE_CLASS_CONST( GROUPING_USED ); 
+	FORMATTER_EXPOSE_CLASS_CONST( DECIMAL_ALWAYS_SHOWN ); 
+	FORMATTER_EXPOSE_CLASS_CONST( MAX_INTEGER_DIGITS );
+	FORMATTER_EXPOSE_CLASS_CONST( MIN_INTEGER_DIGITS ); 
+	FORMATTER_EXPOSE_CLASS_CONST( INTEGER_DIGITS ); 
+	FORMATTER_EXPOSE_CLASS_CONST( MAX_FRACTION_DIGITS ); 
+	FORMATTER_EXPOSE_CLASS_CONST( MIN_FRACTION_DIGITS );
+	FORMATTER_EXPOSE_CLASS_CONST( FRACTION_DIGITS ); 
+	FORMATTER_EXPOSE_CLASS_CONST( MULTIPLIER ); 
+	FORMATTER_EXPOSE_CLASS_CONST( GROUPING_SIZE ); 
+	FORMATTER_EXPOSE_CLASS_CONST( ROUNDING_MODE );
+	FORMATTER_EXPOSE_CLASS_CONST( ROUNDING_INCREMENT ); 
+	FORMATTER_EXPOSE_CLASS_CONST( FORMAT_WIDTH ); 
+	FORMATTER_EXPOSE_CLASS_CONST( PADDING_POSITION ); 
+	FORMATTER_EXPOSE_CLASS_CONST( SECONDARY_GROUPING_SIZE );
+	FORMATTER_EXPOSE_CLASS_CONST( SIGNIFICANT_DIGITS_USED ); 
+	FORMATTER_EXPOSE_CLASS_CONST( MIN_SIGNIFICANT_DIGITS ); 
+	FORMATTER_EXPOSE_CLASS_CONST( MAX_SIGNIFICANT_DIGITS ); 
+	FORMATTER_EXPOSE_CLASS_CONST( LENIENT_PARSE ); 
+
+	// UNumberFormatTextAttribute 
+	FORMATTER_EXPOSE_CLASS_CONST( POSITIVE_PREFIX ); 
+	FORMATTER_EXPOSE_CLASS_CONST( POSITIVE_SUFFIX ); 
+	FORMATTER_EXPOSE_CLASS_CONST( NEGATIVE_PREFIX ); 
+	FORMATTER_EXPOSE_CLASS_CONST( NEGATIVE_SUFFIX );
+	FORMATTER_EXPOSE_CLASS_CONST( PADDING_CHARACTER ); 
+	FORMATTER_EXPOSE_CLASS_CONST( CURRENCY_CODE ); 
+	FORMATTER_EXPOSE_CLASS_CONST( DEFAULT_RULESET ); 
+	FORMATTER_EXPOSE_CLASS_CONST( PUBLIC_RULESETS );
+
+	// UNumberFormatSymbol
+	FORMATTER_EXPOSE_CLASS_CONST( DECIMAL_SEPARATOR_SYMBOL ); 
+	FORMATTER_EXPOSE_CLASS_CONST( GROUPING_SEPARATOR_SYMBOL ); 
+	FORMATTER_EXPOSE_CLASS_CONST( PATTERN_SEPARATOR_SYMBOL ); 
+	FORMATTER_EXPOSE_CLASS_CONST( PERCENT_SYMBOL );
+	FORMATTER_EXPOSE_CLASS_CONST( ZERO_DIGIT_SYMBOL ); 
+	FORMATTER_EXPOSE_CLASS_CONST( DIGIT_SYMBOL ); 
+	FORMATTER_EXPOSE_CLASS_CONST( MINUS_SIGN_SYMBOL ); 
+	FORMATTER_EXPOSE_CLASS_CONST( PLUS_SIGN_SYMBOL );
+	FORMATTER_EXPOSE_CLASS_CONST( CURRENCY_SYMBOL ); 
+	FORMATTER_EXPOSE_CLASS_CONST( INTL_CURRENCY_SYMBOL );
+	FORMATTER_EXPOSE_CLASS_CONST( MONETARY_SEPARATOR_SYMBOL ); 
+	FORMATTER_EXPOSE_CLASS_CONST( EXPONENTIAL_SYMBOL );
+	FORMATTER_EXPOSE_CLASS_CONST( PERMILL_SYMBOL ); 
+	FORMATTER_EXPOSE_CLASS_CONST( PAD_ESCAPE_SYMBOL );
+	FORMATTER_EXPOSE_CLASS_CONST( INFINITY_SYMBOL );
+	FORMATTER_EXPOSE_CLASS_CONST( NAN_SYMBOL );
+	FORMATTER_EXPOSE_CLASS_CONST( SIGNIFICANT_DIGIT_SYMBOL );
+	FORMATTER_EXPOSE_CLASS_CONST( MONETARY_GROUPING_SEPARATOR_SYMBOL ); 
+
+	FORMATTER_EXPOSE_CUSTOM_CLASS_CONST( "TYPE_DEFAULT", FORMAT_TYPE_DEFAULT );
+	FORMATTER_EXPOSE_CUSTOM_CLASS_CONST( "TYPE_INT32", FORMAT_TYPE_INT32 );
+	FORMATTER_EXPOSE_CUSTOM_CLASS_CONST( "TYPE_INT64", FORMAT_TYPE_INT64 );
+	FORMATTER_EXPOSE_CUSTOM_CLASS_CONST( "TYPE_DOUBLE", FORMAT_TYPE_DOUBLE );
+	FORMATTER_EXPOSE_CUSTOM_CLASS_CONST( "TYPE_CURRENCY", FORMAT_TYPE_CURRENCY );
+
+	#undef FORMATTER_EXPOSE_CUSTOM_CLASS_CONST
+	#undef FORMATTER_EXPOSE_CLASS_CONST
+	#undef FORMATTER_EXPOSE_CONST
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/formatter/formatter.h

@@ -0,0 +1,24 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef FORMATTER_FORMATTER_H
+#define FORMATTER_FORMATTER_H
+
+#include <php.h>
+
+void formatter_register_constants( INIT_FUNC_ARGS );
+
+#endif // FORMATTER_FORMATTER_H

ext/intl/formatter/formatter_attr.c

@@ -0,0 +1,443 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "formatter_class.h"
+#include "formatter_attr.h"
+#include "intl_convert.h"
+
+#include <unicode/ustring.h>
+
+/* {{{ proto mixed NumberFormatter::getAttribute( int $attr )
+ * Get formatter attribute value. }}} */
+/* {{{ proto mixed numfmt_get_attribute( NumberFormatter $nf, int $attr )
+ * Get formatter attribute value.
+ */
+PHP_FUNCTION( numfmt_get_attribute )
+{
+	long attribute, value;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Ol",
+		&object, NumberFormatter_ce_ptr, &attribute ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_get_attribute: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	switch(attribute) {
+		case UNUM_PARSE_INT_ONLY:
+		case UNUM_GROUPING_USED:
+		case UNUM_DECIMAL_ALWAYS_SHOWN:
+		case UNUM_MAX_INTEGER_DIGITS:
+		case UNUM_MIN_INTEGER_DIGITS:
+		case UNUM_INTEGER_DIGITS:
+		case UNUM_MAX_FRACTION_DIGITS:
+		case UNUM_MIN_FRACTION_DIGITS:
+		case UNUM_FRACTION_DIGITS:
+		case UNUM_MULTIPLIER:
+		case UNUM_GROUPING_SIZE:
+		case UNUM_ROUNDING_MODE:
+		case UNUM_FORMAT_WIDTH:
+		case UNUM_PADDING_POSITION:
+		case UNUM_SECONDARY_GROUPING_SIZE:
+		case UNUM_SIGNIFICANT_DIGITS_USED:
+		case UNUM_MIN_SIGNIFICANT_DIGITS:
+		case UNUM_MAX_SIGNIFICANT_DIGITS:
+		case UNUM_LENIENT_PARSE:
+			value = unum_getAttribute(FORMATTER_OBJECT(nfo), attribute);
+			if(value == -1) {
+				INTL_DATA_ERROR_CODE(nfo) = U_UNSUPPORTED_ERROR;
+			} else {
+				RETVAL_LONG(value);
+			}
+			break;
+		case UNUM_ROUNDING_INCREMENT:
+		{
+			double value = unum_getDoubleAttribute(FORMATTER_OBJECT(nfo), attribute);
+			if(value == -1) {
+				INTL_DATA_ERROR_CODE(nfo) = U_UNSUPPORTED_ERROR;
+			} else {
+				RETVAL_DOUBLE(value);
+			}
+		}
+			break;
+		default:
+			INTL_DATA_ERROR_CODE(nfo) = U_UNSUPPORTED_ERROR;
+			break;
+	}
+
+	INTL_METHOD_CHECK_STATUS( nfo, "Error getting attribute value" );
+}
+/* }}} */
+
+/* {{{ proto string NumberFormatter::getTextAttribute( int $attr )
+ * Get formatter attribute value. }}} */
+/* {{{ proto string numfmt_get_text_attribute( NumberFormatter $nf, int $attr )
+ * Get formatter attribute value.
+ */
+PHP_FUNCTION( numfmt_get_text_attribute )
+{
+	long   attribute;
+	UChar  value_buf[64];
+	int    value_buf_size = USIZE( value_buf );
+	UChar* value  = value_buf;
+	int    length = 0;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Ol",
+		&object, NumberFormatter_ce_ptr, &attribute ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_get_text_attribute: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	length = unum_getTextAttribute( FORMATTER_OBJECT(nfo), attribute, value, value_buf_size, &INTL_DATA_ERROR_CODE(nfo) );
+	if(INTL_DATA_ERROR_CODE(nfo) == U_BUFFER_OVERFLOW_ERROR && length >= value_buf_size) {
+		++length; // to avoid U_STRING_NOT_TERMINATED_WARNING
+		INTL_DATA_ERROR_CODE(nfo) = U_ZERO_ERROR;
+		value = eumalloc(length);
+		length = unum_getTextAttribute( FORMATTER_OBJECT(nfo), attribute, value, length, &INTL_DATA_ERROR_CODE(nfo) );
+		if(U_FAILURE(INTL_DATA_ERROR_CODE(nfo))) {
+			efree(value);
+			value = value_buf;
+		}
+	}
+	INTL_METHOD_CHECK_STATUS( nfo, "Error getting attribute value" );
+
+	INTL_METHOD_RETVAL_UTF8( nfo, value, length, ( value != value_buf ) );
+}
+/* }}} */
+
+/* {{{ proto bool NumberFormatter::setAttribute( int $attr, mixed $value )
+ * Get formatter attribute value. }}} */
+/* {{{ proto bool numfmt_set_attribute( NumberFormatter $nf, int $attr, mixed $value )
+ * Get formatter attribute value.
+ */
+PHP_FUNCTION( numfmt_set_attribute )
+{
+	long attribute;
+	zval **value;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "OlZ",
+		&object, NumberFormatter_ce_ptr, &attribute, &value ) == FAILURE)
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_set_attribute: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	switch(attribute) {
+		case UNUM_PARSE_INT_ONLY:
+		case UNUM_GROUPING_USED:
+		case UNUM_DECIMAL_ALWAYS_SHOWN:
+		case UNUM_MAX_INTEGER_DIGITS:
+		case UNUM_MIN_INTEGER_DIGITS:
+		case UNUM_INTEGER_DIGITS:
+		case UNUM_MAX_FRACTION_DIGITS:
+		case UNUM_MIN_FRACTION_DIGITS:
+		case UNUM_FRACTION_DIGITS:
+		case UNUM_MULTIPLIER:
+		case UNUM_GROUPING_SIZE:
+		case UNUM_ROUNDING_MODE:
+		case UNUM_FORMAT_WIDTH:
+		case UNUM_PADDING_POSITION:
+		case UNUM_SECONDARY_GROUPING_SIZE:
+		case UNUM_SIGNIFICANT_DIGITS_USED:
+		case UNUM_MIN_SIGNIFICANT_DIGITS:
+		case UNUM_MAX_SIGNIFICANT_DIGITS:
+		case UNUM_LENIENT_PARSE:
+			convert_to_long_ex(value);
+			unum_setAttribute(FORMATTER_OBJECT(nfo), attribute, Z_LVAL_PP(value));
+			break;
+		case UNUM_ROUNDING_INCREMENT:
+			convert_to_double_ex(value);
+			unum_setDoubleAttribute(FORMATTER_OBJECT(nfo), attribute, Z_DVAL_PP(value));
+			break;
+		default:
+			INTL_DATA_ERROR_CODE(nfo) = U_UNSUPPORTED_ERROR;
+			break;
+	}
+
+	INTL_METHOD_CHECK_STATUS( nfo, "Error setting attribute value" );
+
+	RETURN_TRUE;
+}
+/* }}} */
+
+/* {{{ proto bool NumberFormatter::setTextAttribute( int $attr, string $value )
+ * Get formatter attribute value. }}} */
+/* {{{ proto bool numfmt_set_text_attribute( NumberFormatter $nf, int $attr, string $value )
+ * Get formatter attribute value.
+ */
+PHP_FUNCTION( numfmt_set_text_attribute )
+{
+	int slength = 0;
+	UChar *svalue = NULL;
+	long attribute;
+	char *value;
+	int len;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Ols",
+		&object, NumberFormatter_ce_ptr, &attribute, &value, &len ) == FAILURE)
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_set_text_attribute: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	// Convert given attribute value to UTF-16.
+	intl_convert_utf8_to_utf16(&svalue, &slength, value, len, &INTL_DATA_ERROR_CODE(nfo));
+	INTL_METHOD_CHECK_STATUS( nfo, "Error converting attribute value to UTF-16" );
+
+	// Actually set new attribute value.
+	unum_setTextAttribute(FORMATTER_OBJECT(nfo), attribute, svalue, slength, &INTL_DATA_ERROR_CODE(nfo));
+	efree(svalue);
+	INTL_METHOD_CHECK_STATUS( nfo, "Error setting text attribute" );
+
+	RETURN_TRUE;
+}
+/* }}} */
+
+/* {{{ proto string NumberFormatter::getSymbol( int $attr )
+ * Get formatter symbol value. }}} */
+/* {{{ proto string numfmt_get_symbol( NumberFormatter $nf, int $attr )
+ * Get formatter symbol value.
+ */
+PHP_FUNCTION( numfmt_get_symbol )
+{
+	long symbol;
+	UChar value_buf[4];
+	UChar *value = value_buf;
+	int length = USIZE(value);
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Ol",
+		&object, NumberFormatter_ce_ptr, &symbol ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_get_symbol: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	length = unum_getSymbol(FORMATTER_OBJECT(nfo), symbol, value_buf, length, &INTL_DATA_ERROR_CODE(nfo));
+	if(INTL_DATA_ERROR_CODE(nfo) == U_BUFFER_OVERFLOW_ERROR && length >= USIZE( value )) {
+		++length; // to avoid U_STRING_NOT_TERMINATED_WARNING
+		INTL_DATA_ERROR_CODE(nfo) = U_ZERO_ERROR;
+		value = eumalloc(length);
+		length = unum_getSymbol(FORMATTER_OBJECT(nfo), symbol, value, length, &INTL_DATA_ERROR_CODE(nfo));
+		if(U_FAILURE(INTL_DATA_ERROR_CODE(nfo))) {
+			efree(value);
+			value = value_buf;
+		}
+	}
+	INTL_METHOD_CHECK_STATUS( nfo, "Error getting symbol value" );
+
+	INTL_METHOD_RETVAL_UTF8( nfo, value, length, ( value_buf != value ) );
+}
+/* }}} */
+
+/* {{{ proto bool NumberFormatter::setSymbol( int $attr, string $symbol )
+ * Set formatter symbol value. }}} */
+/* {{{ proto bool numfmt_set_symbol( NumberFormatter $nf, int $attr, string $symbol )
+ * Set formatter symbol value.
+ */
+PHP_FUNCTION( numfmt_set_symbol )
+{
+	long       symbol;
+	char*      value     = NULL;
+	int        value_len = 0;
+	UChar*     svalue  = 0;
+	int        slength = 0;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Ols",
+		&object, NumberFormatter_ce_ptr, &symbol, &value, &value_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_set_symbol: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	// Convert given symbol to UTF-16.
+	intl_convert_utf8_to_utf16(&svalue, &slength, value, value_len, &INTL_DATA_ERROR_CODE(nfo));
+	INTL_METHOD_CHECK_STATUS( nfo, "Error converting symbol value to UTF-16" );
+
+	// Actually set the symbol.
+	unum_setSymbol(FORMATTER_OBJECT(nfo), symbol, svalue, slength, &INTL_DATA_ERROR_CODE(nfo));
+	efree(svalue);
+	INTL_METHOD_CHECK_STATUS( nfo, "Error setting symbol value" );
+
+	RETURN_TRUE;
+}
+/* }}} */
+
+/* {{{ proto string NumberFormatter::getPattern( )
+ * Get formatter pattern. }}} */
+/* {{{ proto string numfmt_get_pattern( NumberFormatter $nf )
+ * Get formatter pattern.
+ */
+PHP_FUNCTION( numfmt_get_pattern )
+{
+	UChar  value_buf[64];
+	int    length = USIZE( value_buf );
+	UChar* value  = value_buf;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O",
+		&object, NumberFormatter_ce_ptr ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_get_pattern: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	length = unum_toPattern(FORMATTER_OBJECT(nfo), 0, value, length, &INTL_DATA_ERROR_CODE(nfo));
+	if(INTL_DATA_ERROR_CODE(nfo) == U_BUFFER_OVERFLOW_ERROR && length >= USIZE( value_buf )) {
+		++length; // to avoid U_STRING_NOT_TERMINATED_WARNING
+		INTL_DATA_ERROR_CODE(nfo) = U_ZERO_ERROR;
+		value = eumalloc(length);
+		length = unum_toPattern( FORMATTER_OBJECT(nfo), 0, value, length, &INTL_DATA_ERROR_CODE(nfo) );
+		if(U_FAILURE(INTL_DATA_ERROR_CODE(nfo))) {
+			efree(value);
+			value = value_buf;
+		}
+	}
+	INTL_METHOD_CHECK_STATUS( nfo, "Error getting formatter pattern" );
+
+	INTL_METHOD_RETVAL_UTF8( nfo, value, length, ( value != value_buf ) );
+}
+/* }}} */
+
+/* {{{ proto bool NumberFormatter::setPattern( string $pattern )
+ * Set formatter pattern. }}} */
+/* {{{ proto bool numfmt_set_pattern( NumberFormatter $nf, string $pattern )
+ * Set formatter pattern.
+ */
+PHP_FUNCTION( numfmt_set_pattern )
+{
+	char*       value = NULL;
+	int         value_len = 0;
+	int         slength = 0;
+	UChar*	    svalue  = NULL;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Os",
+		&object, NumberFormatter_ce_ptr, &value, &value_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_set_pattern: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	// Convert given pattern to UTF-16.
+	intl_convert_utf8_to_utf16(&svalue, &slength, value, value_len, &INTL_DATA_ERROR_CODE(nfo));
+	INTL_METHOD_CHECK_STATUS( nfo, "Error converting pattern to UTF-16" );
+
+	// TODO: add parse error information
+	unum_applyPattern(FORMATTER_OBJECT(nfo), 0, svalue, slength, NULL, &INTL_DATA_ERROR_CODE(nfo));
+	efree(svalue);
+	INTL_METHOD_CHECK_STATUS( nfo, "Error setting pattern value" );
+
+	RETURN_TRUE;
+}
+/* }}} */
+
+/* {{{ proto string NumberFormatter::getLocale([int type])
+ * Get formatter locale. }}} */
+/* {{{ proto string numfmt_get_locale( NumberFormatter $nf[, int type] )
+ * Get formatter locale.
+ */
+PHP_FUNCTION( numfmt_get_locale )
+{
+	long type = ULOC_ACTUAL_LOCALE;
+	char* loc;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O|l",
+		&object, NumberFormatter_ce_ptr, &type ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_get_locale: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	loc = (char *)unum_getLocaleByType(FORMATTER_OBJECT(nfo), type, &INTL_DATA_ERROR_CODE(nfo));
+	INTL_METHOD_CHECK_STATUS( nfo, "Error getting locale" );
+	RETURN_STRING(loc, 1);
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/formatter/formatter_attr.h

@@ -0,0 +1,32 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef FORMATTER_ATTR_H
+#define FORMATTER_ATTR_H
+
+#include <php.h>
+
+PHP_FUNCTION( numfmt_set_attribute ); 
+PHP_FUNCTION( numfmt_get_attribute );
+PHP_FUNCTION( numfmt_set_text_attribute );
+PHP_FUNCTION( numfmt_get_text_attribute );
+PHP_FUNCTION( numfmt_set_symbol );
+PHP_FUNCTION( numfmt_get_symbol );
+PHP_FUNCTION( numfmt_set_pattern );
+PHP_FUNCTION( numfmt_get_pattern );
+PHP_FUNCTION( numfmt_get_locale );
+
+#endif // FORMATTER_ATTR_H

ext/intl/formatter/formatter_class.c

@@ -0,0 +1,147 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#include <unicode/unum.h>
+
+#include "formatter_class.h"
+#include "php_intl.h"
+#include "formatter_data.h"
+#include "formatter_format.h"
+#include "formatter_parse.h"
+#include "formatter_main.h"
+#include "formatter_attr.h"
+
+zend_class_entry *NumberFormatter_ce_ptr = NULL;
+
+/////////////////////////////////////////////////////////////////////////////
+// Auxiliary functions needed by objects of 'NumberFormatter' class
+/////////////////////////////////////////////////////////////////////////////
+
+/* {{{ NumberFormatter_objects_dtor */
+static void NumberFormatter_object_dtor(
+	void *object,
+	zend_object_handle handle TSRMLS_DC )
+{
+	zend_objects_destroy_object( object, handle TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ NumberFormatter_objects_free */
+void NumberFormatter_object_free( zend_object *object TSRMLS_DC )
+{
+	NumberFormatter_object* nfo = (NumberFormatter_object*)object;
+
+	zend_object_std_dtor( &nfo->zo TSRMLS_CC );
+
+	formatter_data_free( &nfo->nf_data TSRMLS_CC );
+
+	efree( nfo );
+}
+/* }}} */
+
+/* {{{ NumberFormatter_object_create */
+zend_object_value NumberFormatter_object_create(
+	zend_class_entry *ce TSRMLS_DC )
+{
+	zend_object_value    retval;
+	NumberFormatter_object*     intern;
+
+	intern = ecalloc( 1, sizeof(NumberFormatter_object) );
+	formatter_data_init( &intern->nf_data TSRMLS_CC );
+	zend_object_std_init( &intern->zo, ce TSRMLS_CC );
+
+	retval.handle = zend_objects_store_put(
+		intern,
+		NumberFormatter_object_dtor,
+		(zend_objects_free_object_storage_t)NumberFormatter_object_free,
+		NULL TSRMLS_CC );
+
+	retval.handlers = zend_get_std_object_handlers();
+
+	return retval;
+}
+/* }}} */
+
+/////////////////////////////////////////////////////////////////////////////
+// 'NumberFormatter' class registration structures & functions
+/////////////////////////////////////////////////////////////////////////////
+
+/* {{{ NumberFormatter_class_functions
+ * Every 'NumberFormatter' class method has an entry in this table
+ */
+static ZEND_BEGIN_ARG_INFO_EX( number_parse_arginfo, 0, 0, 1 )
+	ZEND_ARG_INFO( 0, string )
+	ZEND_ARG_INFO( 0, type )
+	ZEND_ARG_INFO( 1, position )
+ZEND_END_ARG_INFO()
+
+static ZEND_BEGIN_ARG_INFO_EX( number_parse_currency_arginfo, 0, 0, 2 )
+	ZEND_ARG_INFO( 0, string )
+	ZEND_ARG_INFO( 1, currency )
+	ZEND_ARG_INFO( 1, position )
+ZEND_END_ARG_INFO()
+
+static function_entry NumberFormatter_class_functions[] = {
+	PHP_ME( NumberFormatter, __construct, NULL, ZEND_ACC_PUBLIC|ZEND_ACC_CTOR )
+	ZEND_FENTRY( create, ZEND_FN( numfmt_create ), NULL, ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	PHP_NAMED_FE( format, ZEND_FN( numfmt_format ), NULL )
+	PHP_NAMED_FE( parse, ZEND_FN( numfmt_parse ), number_parse_arginfo )
+	PHP_NAMED_FE( formatCurrency, ZEND_FN( numfmt_format_currency ), NULL )
+	PHP_NAMED_FE( parseCurrency, ZEND_FN( numfmt_parse_currency ), number_parse_currency_arginfo )
+	PHP_NAMED_FE( setAttribute, ZEND_FN( numfmt_set_attribute ), NULL )
+	PHP_NAMED_FE( getAttribute, ZEND_FN( numfmt_get_attribute ), NULL )
+	PHP_NAMED_FE( setTextAttribute, ZEND_FN( numfmt_set_text_attribute ), NULL )
+	PHP_NAMED_FE( getTextAttribute, ZEND_FN( numfmt_get_text_attribute ), NULL )
+	PHP_NAMED_FE( setSymbol, ZEND_FN( numfmt_set_symbol ), NULL )
+	PHP_NAMED_FE( getSymbol, ZEND_FN( numfmt_get_symbol ), NULL )
+	PHP_NAMED_FE( setPattern, ZEND_FN( numfmt_set_pattern ), NULL )
+	PHP_NAMED_FE( getPattern, ZEND_FN( numfmt_get_pattern ), NULL )
+	PHP_NAMED_FE( getLocale, ZEND_FN( numfmt_get_locale ), NULL )
+	PHP_NAMED_FE( getErrorCode, ZEND_FN( numfmt_get_error_code ), NULL )
+	PHP_NAMED_FE( getErrorMessage, ZEND_FN( numfmt_get_error_message ), NULL )
+	{ NULL, NULL, NULL }
+};
+/* }}} */
+
+/* {{{ formatter_register_class
+ * Initialize 'NumberFormatter' class
+ */
+void formatter_register_class( TSRMLS_D )
+{
+	zend_class_entry ce;
+
+	// Create and register 'NumberFormatter' class.
+	INIT_CLASS_ENTRY( ce, "NumberFormatter", NumberFormatter_class_functions );
+	ce.create_object = NumberFormatter_object_create;
+	NumberFormatter_ce_ptr = zend_register_internal_class( &ce TSRMLS_CC );
+
+	// Declare 'NumberFormatter' class properties.
+	if( !NumberFormatter_ce_ptr )
+	{
+		zend_error(E_ERROR, "Failed to register NumberFormatter class");
+		return;
+	}
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/formatter/formatter_class.h

@@ -0,0 +1,41 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef FORMATTER_CLASS_H
+#define FORMATTER_CLASS_H
+
+#include <php.h>
+
+#include "intl_common.h"
+#include "intl_error.h"
+#include "intl_data.h"
+#include "formatter_data.h"
+
+typedef struct {
+	zend_object     zo;
+	formatter_data  nf_data;
+} NumberFormatter_object;
+
+void formatter_register_class( TSRMLS_D );
+extern zend_class_entry *NumberFormatter_ce_ptr;
+
+/* Auxiliary macros */
+
+#define FORMATTER_METHOD_INIT_VARS		INTL_METHOD_INIT_VARS(NumberFormatter, nfo)
+#define FORMATTER_METHOD_FETCH_OBJECT	INTL_METHOD_FETCH_OBJECT(NumberFormatter, nfo)                                                
+#define FORMATTER_OBJECT(nfo)			(nfo)->nf_data.unum
+
+#endif // #ifndef FORMATTER_CLASS_H

ext/intl/formatter/formatter_data.c

@@ -0,0 +1,72 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "formatter_data.h"
+
+/* {{{ void formatter_data_init( formatter_data* nf_data )
+ * Initialize internals of formatter_data.
+ */
+void formatter_data_init( formatter_data* nf_data TSRMLS_DC )
+{
+	if( !nf_data )
+		return;
+
+	nf_data->unum                = NULL;
+	intl_error_reset( &nf_data->error TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ void formatter_data_free( formatter_data* nf_data )
+ * Clean up mem allocted by internals of formatter_data
+ */
+void formatter_data_free( formatter_data* nf_data TSRMLS_DC )
+{
+	if( !nf_data )
+		return;
+
+	if( nf_data->unum )
+		unum_close( nf_data->unum );
+
+	nf_data->unum = NULL;
+	intl_error_reset( &nf_data->error TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ formatter_data* formatter_data_create()
+ * Alloc mem for formatter_data and initialize it with default values.
+ */
+formatter_data* formatter_data_create( TSRMLS_D )
+{
+	formatter_data* nf_data = ecalloc( 1, sizeof(formatter_data) );
+
+	formatter_data_init( nf_data TSRMLS_CC );
+
+	return nf_data;
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/formatter/formatter_data.h

@@ -0,0 +1,38 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef FORMATTER_DATA_H
+#define FORMATTER_DATA_H
+
+#include <php.h>
+
+#include <unicode/unum.h>
+
+#include "intl_error.h"
+
+typedef struct {
+	// error hangling
+	intl_error      error;
+
+	// formatter handling
+	UNumberFormat*  unum;
+} formatter_data;
+
+formatter_data* formatter_data_create( TSRMLS_D );
+void formatter_data_init( formatter_data* nf_data TSRMLS_DC );
+void formatter_data_free( formatter_data* nf_data TSRMLS_DC );
+
+#endif // FORMATTER_DATA_H

ext/intl/formatter/formatter_format.c

@@ -0,0 +1,207 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <unicode/ustring.h>
+
+#include "php_intl.h"
+#include "formatter_class.h"
+#include "formatter_format.h"
+#include "intl_convert.h"
+
+/* {{{ proto mixed NumberFormatter::format( mixed $num[, int $type] )
+ * Format a number. }}} */
+/* {{{ proto mixed numfmt_format( NumberFormatter $nf, mixed $num[, int type] )
+ * Format a number.
+ */
+PHP_FUNCTION( numfmt_format )
+{
+	zval **number;
+	long type = FORMAT_TYPE_DEFAULT;
+	UChar format_buf[32];
+	UChar* formatted = format_buf;
+	int formatted_len = USIZE(format_buf);
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "OZ|l",
+		&object, NumberFormatter_ce_ptr,  &number, &type ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_format: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	if(type == FORMAT_TYPE_DEFAULT) {
+		if(Z_TYPE_PP(number) == IS_STRING) {
+			SEPARATE_ZVAL_IF_NOT_REF(number);
+			if ((Z_TYPE_PP(number)=is_numeric_string(Z_STRVAL_PP(number), Z_STRLEN_PP(number), 
+				&Z_LVAL_PP(number), &Z_DVAL_PP(number), 1)) == 0) {
+				ZVAL_LONG(*number, 0);
+			}
+		}
+
+		if(Z_TYPE_PP(number) == IS_LONG) {
+			// take INT32 on 32-bit, int64 on 64-bit
+			type = (sizeof(long) == 8)?FORMAT_TYPE_INT64:FORMAT_TYPE_INT32;
+		} else if(Z_TYPE_PP(number) == IS_DOUBLE) {
+			type = FORMAT_TYPE_DOUBLE;
+		} else {
+			type = FORMAT_TYPE_INT32;
+		}
+	}
+
+	if(Z_TYPE_PP(number) != IS_DOUBLE && Z_TYPE_PP(number) != IS_LONG) {
+		SEPARATE_ZVAL_IF_NOT_REF(number);
+		convert_scalar_to_number( *number TSRMLS_CC );
+	}
+
+	switch(type) {
+		case FORMAT_TYPE_INT32:
+			convert_to_long_ex(number);
+			formatted_len = unum_format(FORMATTER_OBJECT(nfo), (int32_t)Z_LVAL_PP(number), 
+				formatted, formatted_len, NULL, &INTL_DATA_ERROR_CODE(nfo));
+			if (INTL_DATA_ERROR_CODE(nfo) == U_BUFFER_OVERFLOW_ERROR) {
+				intl_error_reset(INTL_DATA_ERROR_P(nfo) TSRMLS_CC); 
+				formatted = eumalloc(formatted_len);
+				formatted_len = unum_format(FORMATTER_OBJECT(nfo), (int32_t)Z_LVAL_PP(number), 
+					formatted, formatted_len, NULL, &INTL_DATA_ERROR_CODE(nfo));
+				if (U_FAILURE( INTL_DATA_ERROR_CODE(nfo) ) ) {
+					efree(formatted);
+				}
+			}
+			INTL_METHOD_CHECK_STATUS( nfo, "Number formatting failed" );
+			break;
+
+		case FORMAT_TYPE_INT64:
+		{
+			int64_t value = (Z_TYPE_PP(number) == IS_DOUBLE)?(int64_t)Z_DVAL_PP(number):Z_LVAL_PP(number);
+			formatted_len = unum_formatInt64(FORMATTER_OBJECT(nfo), value, formatted, formatted_len, NULL, &INTL_DATA_ERROR_CODE(nfo));
+			if (INTL_DATA_ERROR_CODE(nfo) == U_BUFFER_OVERFLOW_ERROR) {
+				intl_error_reset(INTL_DATA_ERROR_P(nfo) TSRMLS_CC); 
+				formatted = eumalloc(formatted_len);
+				formatted_len = unum_formatInt64(FORMATTER_OBJECT(nfo), value, formatted, formatted_len, NULL, &INTL_DATA_ERROR_CODE(nfo));
+				if (U_FAILURE( INTL_DATA_ERROR_CODE(nfo) ) ) {
+					efree(formatted);
+				}
+			}
+			INTL_METHOD_CHECK_STATUS( nfo, "Number formatting failed" );
+		}
+			break;
+
+		case FORMAT_TYPE_DOUBLE:
+			convert_to_double_ex(number);
+			formatted_len = unum_formatDouble(FORMATTER_OBJECT(nfo), Z_DVAL_PP(number), formatted, formatted_len, NULL, &INTL_DATA_ERROR_CODE(nfo));
+			if (INTL_DATA_ERROR_CODE(nfo) == U_BUFFER_OVERFLOW_ERROR) {
+				intl_error_reset(INTL_DATA_ERROR_P(nfo) TSRMLS_CC); 
+				formatted = eumalloc(formatted_len);
+				unum_formatDouble(FORMATTER_OBJECT(nfo), Z_DVAL_PP(number), formatted, formatted_len, NULL, &INTL_DATA_ERROR_CODE(nfo));
+				if (U_FAILURE( INTL_DATA_ERROR_CODE(nfo) ) ) {
+					efree(formatted);
+				}
+			}
+			INTL_METHOD_CHECK_STATUS( nfo, "Number formatting failed" );
+			break;
+
+		default:
+			php_error_docref(NULL TSRMLS_CC, E_WARNING, "Unsupported format type %ld", type);
+			RETURN_FALSE;
+			break;
+	}
+
+	INTL_METHOD_RETVAL_UTF8( nfo, formatted, formatted_len, ( formatted != format_buf ) );
+}
+/* }}} */
+
+/* {{{ proto mixed NumberFormatter::formatCurrency( double $num, string $currency )
+ * Format a number as currency. }}} */
+/* {{{ proto mixed numfmt_format_currency( NumberFormatter $nf, double $num, string $currency )
+ * Format a number as currency.
+ */
+PHP_FUNCTION( numfmt_format_currency )
+{
+	double     number;
+	UChar      format_buf[32];
+	UChar*     formatted     = format_buf;
+	int        formatted_len = USIZE(format_buf);
+	char*      currency      = NULL;
+	int        currency_len  = 0;
+	UChar*     scurrency     = NULL;
+	int        scurrency_len = 0;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Ods",
+		&object, NumberFormatter_ce_ptr,  &number, &currency, &currency_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_format_currency: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	// Convert currency to UTF-16.
+	intl_convert_utf8_to_utf16(&scurrency, &scurrency_len, currency, currency_len, &INTL_DATA_ERROR_CODE(nfo));
+	INTL_METHOD_CHECK_STATUS( nfo, "Currency conversion to UTF-16 failed" );
+
+	// Format the number using a fixed-length buffer.
+	formatted_len = unum_formatDoubleCurrency(FORMATTER_OBJECT(nfo), number, scurrency, formatted, formatted_len, NULL, &INTL_DATA_ERROR_CODE(nfo));
+
+	// If the buffer turned out to be too small
+	// then allocate another buffer dynamically
+	// and use it to format the number.
+	if (INTL_DATA_ERROR_CODE(nfo) == U_BUFFER_OVERFLOW_ERROR) {
+		intl_error_reset(INTL_DATA_ERROR_P(nfo) TSRMLS_CC); 
+		formatted = eumalloc(formatted_len);
+		unum_formatDoubleCurrency(FORMATTER_OBJECT(nfo), number, scurrency, formatted, formatted_len, NULL, &INTL_DATA_ERROR_CODE(nfo));
+	}
+
+	if( U_FAILURE( INTL_DATA_ERROR_CODE((nfo)) ) ) {
+		intl_error_set_code( NULL, INTL_DATA_ERROR_CODE((nfo)) TSRMLS_CC );
+		intl_errors_set_custom_msg( INTL_DATA_ERROR_P(nfo), "Number formatting failed", 0 TSRMLS_CC );
+		RETVAL_FALSE;
+		if (formatted != format_buf) {
+			efree(formatted);
+		}
+	} else {
+		INTL_METHOD_RETVAL_UTF8( nfo, formatted, formatted_len, ( formatted != format_buf ) );
+	}
+
+	if(scurrency) {
+		efree(scurrency);
+	}
+}
+
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/formatter/formatter_format.h

@@ -0,0 +1,31 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef FORMATTER_FORMAT_H
+#define FORMATTER_FORMAT_H
+
+#include <php.h>
+
+PHP_FUNCTION( numfmt_format );
+PHP_FUNCTION( numfmt_format_currency );
+
+#define FORMAT_TYPE_DEFAULT	0
+#define FORMAT_TYPE_INT32	1
+#define FORMAT_TYPE_INT64	2 
+#define FORMAT_TYPE_DOUBLE	3
+#define FORMAT_TYPE_CURRENCY	4
+
+#endif // FORMATTER_FORMAT_H

ext/intl/formatter/formatter_main.c

@@ -0,0 +1,220 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <unicode/ustring.h>
+
+#include "php_intl.h"
+#include "formatter_class.h"
+#include "intl_convert.h"
+
+/* {{{ proto NumberFormatter NumberFormatter::create( string $locale, int style[, string $pattern ] )
+ * Create formatter. }}} */
+/* {{{ proto NumberFormatter numfmt_create( string $locale, int style[, string $pattern ] )
+ * Create formatter.
+ */
+PHP_FUNCTION( numfmt_create )
+{
+	char*       locale;
+	char*       pattern = NULL;
+	int         locale_len = 0, pattern_len = 0;
+	long        style;
+	UChar*      spattern     = NULL;
+	int         spattern_len = 0;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_parameters( ZEND_NUM_ARGS() TSRMLS_CC, "sl|s",
+		&locale, &locale_len, &style, &pattern, &pattern_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_create: unable to parse input parameters", 0 TSRMLS_CC );
+
+		RETURN_NULL();
+	}
+
+	INTL_CHECK_LOCALE_LEN(locale_len);
+	// Create a NumberFormatter object and save the ICU formatter into it.
+	if( ( object = getThis() ) == NULL )
+		object = return_value;
+
+	if( Z_TYPE_P( object ) != IS_OBJECT )
+	object_init_ex( object, NumberFormatter_ce_ptr );
+
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	// Convert pattern (if specified) to UTF-16.
+	if(pattern && pattern_len) {
+		intl_convert_utf8_to_utf16(&spattern, &spattern_len, pattern, pattern_len, &INTL_DATA_ERROR_CODE(nfo));
+		if( U_FAILURE( INTL_DATA_ERROR_CODE((nfo)) ) )
+		{
+			intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+				"numfmt_create: error converting pattern to UTF-16", 0 TSRMLS_CC );
+			zval_dtor(return_value);
+			RETURN_NULL();
+		}
+	}
+
+	if(locale_len == 0) {
+		locale = INTL_G(default_locale);
+	}
+
+	// Create an ICU number formatter.
+	FORMATTER_OBJECT(nfo) = unum_open(style, spattern, spattern_len, locale, NULL, &INTL_DATA_ERROR_CODE(nfo));
+
+	if(spattern) {
+		efree(spattern);
+	}
+
+	if( U_FAILURE( INTL_DATA_ERROR_CODE((nfo)) ) )
+	{
+		intl_error_set( NULL, INTL_DATA_ERROR_CODE( nfo ),
+			"numfmt_create: number formatter creation failed", 0 TSRMLS_CC );
+		zval_dtor(return_value);
+		RETURN_NULL();
+	}
+}
+/* }}} */
+
+/* {{{ proto void NumberFormatter::__construct( string $locale, int style[, string $pattern ] )
+ * NumberFormatter object constructor.
+ */
+PHP_METHOD( NumberFormatter, __construct )
+{
+	char*       locale;
+	char*       pattern = NULL;
+	int         locale_len = 0, pattern_len = 0;
+	long        style;
+	UChar*      spattern     = NULL;
+	int         spattern_len = 0;
+	FORMATTER_METHOD_INIT_VARS;
+
+	object = getThis();
+
+	// Parse parameters.
+	if( zend_parse_parameters( ZEND_NUM_ARGS() TSRMLS_CC, "sl|s",
+		&locale, &locale_len, &style, &pattern, &pattern_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"__construct: unable to parse input params", 0 TSRMLS_CC );
+		zval_dtor(object);
+		ZVAL_NULL(object);
+		RETURN_NULL();
+	}
+
+	INTL_CHECK_LOCALE_LEN_OBJ(locale_len, object);
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	// Convert pattern (if specified) to UTF-16.
+	if(pattern && pattern_len) {
+		intl_convert_utf8_to_utf16(&spattern, &spattern_len, pattern, pattern_len, &INTL_DATA_ERROR_CODE(nfo));
+		if( U_FAILURE( INTL_DATA_ERROR_CODE((nfo)) ) )
+		{
+			intl_error_set( NULL, INTL_DATA_ERROR_CODE( nfo ),
+				"__construct: Error converting pattern to UTF-16", 0 TSRMLS_CC );
+			zval_dtor(object);
+			ZVAL_NULL(object);
+			RETURN_NULL();
+		}
+	}
+
+	if(locale_len == 0) {
+		locale = INTL_G(default_locale);
+	}
+
+	// Create an ICU number formatter.
+	FORMATTER_OBJECT(nfo) = unum_open(style, spattern, spattern_len, locale, NULL, &INTL_DATA_ERROR_CODE(nfo));
+
+	if(spattern) {
+		efree(spattern);
+	}
+
+	if( U_FAILURE( INTL_DATA_ERROR_CODE((nfo)) ) )
+	{
+		intl_error_set( NULL, INTL_DATA_ERROR_CODE( nfo ),
+			"__construct: number formatter creation failed", 0 TSRMLS_CC );
+		zval_dtor(object);
+		ZVAL_NULL(object);
+		RETURN_NULL();
+	}
+}
+/* }}} */
+
+/* {{{ proto int NumberFormatter::getErrorCode()
+ * Get formatter's last error code. }}} */
+/* {{{ proto int numfmt_get_error_code( NumberFormatter $nf )
+ * Get formatter's last error code.
+ */
+PHP_FUNCTION( numfmt_get_error_code )
+{
+	FORMATTER_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O",
+		&object, NumberFormatter_ce_ptr ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_get_error_code: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	nfo = (NumberFormatter_object *) zend_object_store_get_object( object TSRMLS_CC );
+
+	// Return formatter's last error code.
+	RETURN_LONG( INTL_DATA_ERROR_CODE(nfo) );
+}
+/* }}} */
+
+/* {{{ proto string NumberFormatter::getErrorMessage( )
+ * Get text description for formatter's last error code. }}} */
+/* {{{ proto string numfmt_get_error_message( NumberFormatter $nf )
+ * Get text description for formatter's last error code.
+ */
+PHP_FUNCTION( numfmt_get_error_message )
+{
+	char*                    message = NULL;
+	FORMATTER_METHOD_INIT_VARS
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O",
+		&object, NumberFormatter_ce_ptr ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"numfmt_get_error_message: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	nfo = (NumberFormatter_object *) zend_object_store_get_object( object TSRMLS_CC );
+
+	// Return last error message.
+	message = intl_error_get_message( &INTL_DATA_ERROR(nfo) TSRMLS_CC );
+	RETURN_STRING( message, 0);
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/formatter/formatter_main.h

@@ -0,0 +1,27 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef FORMATTER_MAIN_H
+#define FORMATTER_MAIN_H
+
+#include <php.h>
+
+PHP_FUNCTION( numfmt_create );
+PHP_FUNCTION( numfmt_get_error_code );
+PHP_FUNCTION( numfmt_get_error_message );
+PHP_METHOD( NumberFormatter, __construct );
+
+#endif // FORMATTER_FORMAT_H

ext/intl/formatter/formatter_parse.c

@@ -0,0 +1,173 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <unicode/ustring.h>
+
+#include "php_intl.h"
+#include "formatter_class.h"
+#include "formatter_format.h"
+#include "formatter_parse.h"
+#include "intl_convert.h"
+
+/* {{{ proto mixed NumberFormatter::parse( string $str[, int $type, int &$position ])
+ * Parse a number. }}} */
+/* {{{ proto mixed numfmt_parse( NumberFormatter $nf, string $str[, int $type, int &$position ])
+ * Parse a number.
+ */
+PHP_FUNCTION( numfmt_parse )
+{
+	long type = FORMAT_TYPE_DOUBLE;
+	UChar* sstr = NULL;
+	int sstr_len = 0;
+	char* str = NULL;
+	int str_len;
+	int32_t val32, position = 0;
+	int64_t val64;
+	double val_double;
+	int32_t* position_p = NULL;
+	zval *zposition = NULL;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Os|lz!",
+		&object, NumberFormatter_ce_ptr,  &str, &str_len, &type, &zposition ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"number_parse: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	// Convert given string to UTF-16.
+	intl_convert_utf8_to_utf16(&sstr, &sstr_len, str, str_len, &INTL_DATA_ERROR_CODE(nfo));
+	INTL_METHOD_CHECK_STATUS( nfo, "String conversion to UTF-16 failed" );
+
+	if(zposition) {
+		convert_to_long(zposition);
+		position = (int32_t)Z_LVAL_P( zposition );
+		position_p = &position;
+	}
+
+	switch(type) {
+		case FORMAT_TYPE_INT32:
+			val32 = unum_parse(FORMATTER_OBJECT(nfo), sstr, sstr_len, position_p, &INTL_DATA_ERROR_CODE(nfo));
+			RETVAL_LONG(val32);
+			break;
+		case FORMAT_TYPE_INT64:
+			val64 = unum_parseInt64(FORMATTER_OBJECT(nfo), sstr, sstr_len, position_p, &INTL_DATA_ERROR_CODE(nfo));
+			if(val64 > LONG_MAX || val64 < -LONG_MAX) {
+				RETVAL_DOUBLE(val64);
+			} else {
+				val32 = (int32_t)val64;
+				RETVAL_LONG(val32);
+			}
+			break;
+		case FORMAT_TYPE_DOUBLE:
+			val_double = unum_parseDouble(FORMATTER_OBJECT(nfo), sstr, sstr_len, position_p, &INTL_DATA_ERROR_CODE(nfo));
+			RETVAL_DOUBLE(val_double);
+			break;
+		default:
+			php_error_docref(NULL TSRMLS_CC, E_WARNING, "Unsupported format type %ld", type);
+			RETVAL_FALSE;
+			break;
+	}
+	if(zposition) {
+		zval_dtor(zposition);
+		ZVAL_LONG(zposition, position);
+	}
+
+	efree(sstr);
+
+	INTL_METHOD_CHECK_STATUS( nfo, "Number parsing failed" );
+}
+/* }}} */
+
+/* {{{ proto double NumberFormatter::parseCurrency( string $str, string $&currency[, int $&position] )
+ * Parse a number as currency. }}} */
+/* {{{ proto double numfmt_parse_currency( NumberFormatter $nf, string $str, string $&currency[, int $&position] )
+ * Parse a number as currency.
+ */
+PHP_FUNCTION( numfmt_parse_currency )
+{
+	double number;
+	UChar currency[5] = {0};
+	UChar* sstr = NULL;
+	int sstr_len = 0;
+	char *currency_str = NULL;
+	int currency_len = 0;
+	char *str;
+	int str_len;
+	int32_t* position_p = NULL;
+	int32_t position = 0;
+	zval *zcurrency, *zposition = NULL;
+	FORMATTER_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Osz|z!",
+		&object, NumberFormatter_ce_ptr,  &str, &str_len, &zcurrency, &zposition ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"number_parse_currency: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	FORMATTER_METHOD_FETCH_OBJECT;
+
+	// Convert given string to UTF-16.
+	intl_convert_utf8_to_utf16(&sstr, &sstr_len, str, str_len, &INTL_DATA_ERROR_CODE(nfo));
+	INTL_METHOD_CHECK_STATUS( nfo, "String conversion to UTF-16 failed" );
+
+	if(zposition) {
+		convert_to_long(zposition);
+		position = (int32_t)Z_LVAL_P( zposition );
+		position_p = &position;
+	}
+
+	number = unum_parseDoubleCurrency(FORMATTER_OBJECT(nfo), sstr, sstr_len, position_p, currency, &INTL_DATA_ERROR_CODE(nfo));
+	if(zposition) {
+		zval_dtor(zposition);
+		ZVAL_LONG(zposition, position);
+	}
+	efree(sstr);
+	INTL_METHOD_CHECK_STATUS( nfo, "Number parsing failed" );
+
+	// Convert parsed currency to UTF-8 and pass it back to caller.
+	intl_convert_utf16_to_utf8(&currency_str, &currency_len, currency, u_strlen(currency), &INTL_DATA_ERROR_CODE(nfo));
+	INTL_METHOD_CHECK_STATUS( nfo, "Currency conversion to UTF-8 failed" );
+	zval_dtor( zcurrency );
+	ZVAL_STRINGL(zcurrency, currency_str, currency_len, 0);
+
+	RETVAL_DOUBLE( number );
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/formatter/formatter_parse.h

@@ -0,0 +1,25 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef FORMATTER_PARSE_H
+#define FORMATTER_PARSE_H
+
+#include <php.h>
+
+PHP_FUNCTION( numfmt_parse );
+PHP_FUNCTION( numfmt_parse_currency );
+
+#endif // FORMATTER_PARSE_H

ext/intl/grapheme/grapheme.h

@@ -0,0 +1,37 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Ed Batutis <ed@batutis.com>                                 |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef GRAPHEME_GRAPHEME_H
+#define GRAPHEME_GRAPHEME_H
+
+#include <php.h>
+#include <unicode/utypes.h>
+#include <unicode/ubrk.h>
+
+PHP_FUNCTION(grapheme_strlen);
+PHP_FUNCTION(grapheme_strpos);
+PHP_FUNCTION(grapheme_stripos);
+PHP_FUNCTION(grapheme_strrpos);
+PHP_FUNCTION(grapheme_strripos);
+PHP_FUNCTION(grapheme_substr);
+PHP_FUNCTION(grapheme_strstr);
+PHP_FUNCTION(grapheme_stristr);
+PHP_FUNCTION(grapheme_extract);
+
+void grapheme_register_constants( INIT_FUNC_ARGS );
+void grapheme_close_global_iterator( TSRMLS_D );
+
+#endif // GRAPHEME_GRAPHEME_H

ext/intl/grapheme/grapheme_string.c

@@ -0,0 +1,913 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5														  |
+   +----------------------------------------------------------------------+
+   | 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:			  |
+   | http://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.				  |
+   +----------------------------------------------------------------------+
+   | Author: Ed Batutis <ed@batutis.com>								  |
+   +----------------------------------------------------------------------+
+ */
+
+/* {{{ includes */
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <php.h>
+#include "grapheme.h"
+#include "grapheme_util.h"
+
+#include <unicode/utypes.h>
+#include <unicode/ucol.h>
+#include <unicode/ustring.h>
+#include <unicode/ubrk.h>
+
+#include "ext/standard/php_string.h"
+
+/* }}} */
+
+#define GRAPHEME_EXTRACT_TYPE_COUNT		0
+#define GRAPHEME_EXTRACT_TYPE_MAXBYTES	1
+#define GRAPHEME_EXTRACT_TYPE_MAXCHARS	2
+#define GRAPHEME_EXTRACT_TYPE_MIN	GRAPHEME_EXTRACT_TYPE_COUNT
+#define GRAPHEME_EXTRACT_TYPE_MAX	GRAPHEME_EXTRACT_TYPE_MAXCHARS
+
+
+/* {{{ grapheme_register_constants
+ * Register API constants
+ */
+void grapheme_register_constants( INIT_FUNC_ARGS )
+{
+	REGISTER_LONG_CONSTANT("GRAPHEME_EXTR_COUNT", GRAPHEME_EXTRACT_TYPE_COUNT, CONST_CS | CONST_PERSISTENT);
+	REGISTER_LONG_CONSTANT("GRAPHEME_EXTR_MAXBYTES", GRAPHEME_EXTRACT_TYPE_MAXBYTES, CONST_CS | CONST_PERSISTENT);
+	REGISTER_LONG_CONSTANT("GRAPHEME_EXTR_MAXCHARS", GRAPHEME_EXTRACT_TYPE_MAXCHARS, CONST_CS | CONST_PERSISTENT);
+}
+/* }}} */
+
+/* {{{ proto int grapheme_strlen(string str)
+   Get number of graphemes in a string */
+PHP_FUNCTION(grapheme_strlen)
+{
+	unsigned char* string;
+	int string_len;
+	UChar* ustring = NULL;
+	int ustring_len = 0;
+	int ret_len;
+	UErrorCode status;
+
+	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "s", (char **)&string, &string_len) == FAILURE) {
+
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "grapheme_strlen: unable to parse input param", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	ret_len = grapheme_ascii_check(string, string_len);
+	
+	if ( ret_len >= 0 ) 
+		RETURN_LONG(ret_len);
+
+	/* convert the string to UTF-16. */
+	status = U_ZERO_ERROR;
+	intl_convert_utf8_to_utf16(&ustring, &ustring_len, (char*) string, string_len, &status );
+
+	if ( U_FAILURE( status ) ) {
+		/* Set global error code. */
+		intl_error_set_code( NULL, status TSRMLS_CC );
+
+		/* Set error messages. */
+		intl_error_set_custom_msg( NULL, "Error converting input string to UTF-16", 1 TSRMLS_CC );
+		efree( ustring );
+		RETURN_NULL();
+	}
+	
+	ret_len = grapheme_split_string(ustring, ustring_len, NULL, 0 TSRMLS_CC );
+
+	efree( ustring );
+
+	if (ret_len >= 0) {
+		RETVAL_LONG(ret_len);
+	} else {
+		RETVAL_FALSE;
+	}
+}
+/* }}} */
+
+/* {{{ proto int grapheme_strpos(string haystack, string needle [, int offset ])
+   Find position of first occurrence of a string within another */
+PHP_FUNCTION(grapheme_strpos)
+{
+	unsigned char *haystack, *needle;
+	int haystack_len, needle_len;
+	unsigned char *found;
+	long loffset = 0;
+	int32_t offset = 0;
+	int ret_pos, uchar_pos;
+	
+	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "ss|l", (char **)&haystack, &haystack_len, (char **)&needle, &needle_len, &loffset) == FAILURE) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "grapheme_strpos: unable to parse input param", 0 TSRMLS_CC );
+			 
+		RETURN_FALSE;
+	}
+
+	if ( OUTSIDE_STRING(loffset, haystack_len) ) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_strpos: Offset not contained in string", 1 TSRMLS_CC );
+		
+		RETURN_FALSE;
+	}
+
+	/* we checked that it will fit: */	
+	offset = (int32_t) loffset;
+
+	/* the offset is 'grapheme count offset' so it still might be invalid - we'll check it later */
+
+	if (needle_len == 0) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_strpos: Empty delimiter", 1 TSRMLS_CC );
+		
+		RETURN_FALSE;
+	}
+
+
+	/* quick check to see if the string might be there
+	 * I realize that 'offset' is 'grapheme count offset' but will work in spite of that 
+	*/
+	found = (unsigned char *)php_memnstr((char *)haystack + offset, (char *)needle, needle_len, (char *)haystack + haystack_len);
+
+	/* if it isn't there the we are done */
+	if (!found) {
+		RETURN_FALSE;
+	}
+
+	/* if it is there, and if the haystack is ascii, we are all done */
+	if ( grapheme_ascii_check(haystack, haystack_len) >= 0 ) {
+
+		RETURN_LONG(found - haystack);
+	}
+
+	/* do utf16 part of the strpos */
+	ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, offset, &uchar_pos, 0 /* fIgnoreCase */ TSRMLS_CC );
+
+	if ( ret_pos >= 0 ) {
+		RETURN_LONG(ret_pos + offset);
+	} else {
+		RETURN_FALSE;
+	}
+
+}
+/* }}} */
+
+/* {{{ proto int grapheme_stripos(string haystack, string needle [, int offset ])
+   Find position of first occurrence of a string within another, ignoring case differences */
+PHP_FUNCTION(grapheme_stripos)
+{
+	unsigned char *haystack, *needle, *haystack_dup, *needle_dup;
+	int haystack_len, needle_len;
+	unsigned char *found;
+	long loffset = 0;
+	int32_t offset = 0;
+	int ret_pos, uchar_pos;
+	int is_ascii;
+	
+	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "ss|l", (char **)&haystack, &haystack_len, (char **)&needle, &needle_len, &loffset) == FAILURE) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "grapheme_stripos: unable to parse input param", 0 TSRMLS_CC );
+			 
+		RETURN_FALSE;
+	}
+
+	if ( OUTSIDE_STRING(loffset, haystack_len) ) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_stripos: Offset not contained in string", 1 TSRMLS_CC );
+		
+		RETURN_FALSE;
+	}
+	
+	/* we checked that it will fit: */
+	offset = (int32_t) loffset;
+
+	/* the offset is 'grapheme count offset' so it still might be invalid - we'll check it later */
+
+	if (needle_len == 0) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_stripos: Empty delimiter", 1 TSRMLS_CC );
+		
+		RETURN_FALSE;
+	}
+
+
+	is_ascii = ( grapheme_ascii_check(haystack, haystack_len) >= 0 );
+
+	if ( is_ascii ) {
+		needle_dup = (unsigned char *)estrndup((char *)needle, needle_len);
+		php_strtolower((char *)needle_dup, needle_len);
+		haystack_dup = (unsigned char *)estrndup((char *)haystack, haystack_len);
+		php_strtolower((char *)haystack_dup, haystack_len);
+
+		found = (unsigned char*) php_memnstr((char *)haystack_dup + offset, (char *)needle_dup, needle_len, (char *)haystack_dup + haystack_len);
+
+		efree(haystack_dup);
+		efree(needle_dup);
+
+		if (found) {
+			RETURN_LONG(found - haystack_dup);
+		}
+
+		/* if needle was ascii too, we are all done, otherwise we need to try using Unicode to see what we get */
+		if ( grapheme_ascii_check(needle, needle_len) >= 0 ) {
+			RETURN_FALSE;
+		}
+	}
+
+	/* do utf16 part of the strpos */
+	ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, offset, &uchar_pos, 1 /* fIgnoreCase */ TSRMLS_CC );
+
+	if ( ret_pos >= 0 ) {
+		RETURN_LONG(ret_pos + offset);
+	} else {
+		RETURN_FALSE;
+	}
+
+}
+/* }}} */
+
+/* {{{ proto int grapheme_strrpos(string haystack, string needle [, int offset])
+   Find position of last occurrence of a string within another */
+PHP_FUNCTION(grapheme_strrpos)
+{
+	unsigned char *haystack, *needle;
+	int haystack_len, needle_len;
+	long loffset = 0;
+	int32_t offset = 0;
+	int32_t ret_pos;
+	int is_ascii;
+	
+	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "ss|l", (char **)&haystack, &haystack_len, (char **)&needle, &needle_len, &loffset) == FAILURE) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "grapheme_strrpos: unable to parse input param", 0 TSRMLS_CC );
+			 
+		RETURN_FALSE;
+	}
+
+	if ( OUTSIDE_STRING(loffset, haystack_len) ) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_strpos: Offset not contained in string", 1 TSRMLS_CC );
+		
+		RETURN_FALSE;
+	}
+	
+	/* we checked that it will fit: */
+	offset = (int32_t) loffset;
+
+	/* the offset is 'grapheme count offset' so it still might be invalid - we'll check it later */
+
+	if (needle_len == 0) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_strpos: Empty delimiter", 1 TSRMLS_CC );
+		
+		RETURN_FALSE;
+	}
+
+	is_ascii = grapheme_ascii_check(haystack, haystack_len) >= 0;
+
+	if ( is_ascii ) {
+	
+		ret_pos = grapheme_strrpos_ascii(haystack, haystack_len, needle, needle_len, offset);
+		
+
+		if ( ret_pos >= 0 ) {
+			RETURN_LONG(ret_pos);
+		}
+
+		/* if the needle was ascii too, we are done */
+
+		if (  grapheme_ascii_check(needle, needle_len) >= 0 ) {
+			RETURN_FALSE;
+		}
+
+		/* else we need to continue via utf16 */
+	}
+
+	ret_pos = grapheme_strrpos_utf16(haystack, haystack_len, needle, needle_len, offset, 0 /* f_ignore_case */ TSRMLS_CC);
+
+	if ( ret_pos >= 0 ) {
+		RETURN_LONG(ret_pos);
+	} else {
+		RETURN_FALSE;
+	}
+	
+
+}
+/* }}} */
+
+/* {{{ proto int grapheme_strripos(string haystack, string needle [, int offset])
+   Find position of last occurrence of a string within another, ignoring case */
+PHP_FUNCTION(grapheme_strripos)
+{
+	unsigned char *haystack, *needle;
+	int haystack_len, needle_len;
+	long loffset = 0;
+	int32_t offset = 0;
+	int32_t ret_pos;
+	int is_ascii;
+	
+	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "ss|l", (char **)&haystack, &haystack_len, (char **)&needle, &needle_len, &loffset) == FAILURE) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "grapheme_strrpos: unable to parse input param", 0 TSRMLS_CC );
+			 
+		RETURN_FALSE;
+	}
+
+	if ( OUTSIDE_STRING(loffset, haystack_len) ) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_strpos: Offset not contained in string", 1 TSRMLS_CC );
+		
+		RETURN_FALSE;
+	}
+
+	/* we checked that it will fit: */
+	offset = (int32_t) loffset;
+	
+	/* the offset is 'grapheme count offset' so it still might be invalid - we'll check it later */
+
+	if (needle_len == 0) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_strpos: Empty delimiter", 1 TSRMLS_CC );
+		
+		RETURN_FALSE;
+	}
+
+	is_ascii = grapheme_ascii_check(haystack, haystack_len) >= 0;
+
+	if ( is_ascii ) {
+		unsigned char *needle_dup, *haystack_dup;
+
+		needle_dup = (unsigned char *)estrndup((char *)needle, needle_len);
+		php_strtolower((char *)needle_dup, needle_len);
+		haystack_dup = (unsigned char *)estrndup((char *)haystack, haystack_len);
+		php_strtolower((char *)haystack_dup, haystack_len);
+
+		ret_pos = grapheme_strrpos_ascii(haystack_dup, haystack_len, needle_dup, needle_len, offset);
+		
+		efree(haystack_dup);
+		efree(needle_dup);
+
+		if ( ret_pos >= 0 ) {
+			RETURN_LONG(ret_pos);
+		}
+
+		/* if the needle was ascii too, we are done */
+
+		if (  grapheme_ascii_check(needle, needle_len) >= 0 ) {
+			RETURN_FALSE;
+		}
+
+		/* else we need to continue via utf16 */
+	}
+
+	ret_pos = grapheme_strrpos_utf16(haystack, haystack_len, needle, needle_len, offset, 1 /* f_ignore_case */ TSRMLS_CC);
+
+	if ( ret_pos >= 0 ) {
+		RETURN_LONG(ret_pos);
+	} else {
+		RETURN_FALSE;
+	}
+	
+
+}
+/* }}} */
+
+/* {{{ proto string grapheme_substr(string str, int start [, int length])
+   Returns part of a string */
+PHP_FUNCTION(grapheme_substr)
+{
+	unsigned char *str, *sub_str;
+	UChar *ustr;
+	int str_len, sub_str_len, ustr_len;
+	long lstart = 0, length = 0;
+	int32_t start = 0;
+	int iter_val;
+	UErrorCode status;
+	unsigned char u_break_iterator_buffer[U_BRK_SAFECLONE_BUFFERSIZE];
+	UBreakIterator* bi = NULL;
+	int sub_str_start_pos, sub_str_end_pos;
+	int32_t (*iter_func)(UBreakIterator *);
+
+	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "sl|l", (char **)&str, &str_len, &lstart, &length) == FAILURE) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "grapheme_substr: unable to parse input param", 0 TSRMLS_CC );
+			 
+		RETURN_FALSE;
+	}
+
+	if ( OUTSIDE_STRING(lstart, str_len) ) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_substr: start not contained in string", 1 TSRMLS_CC );
+		
+		RETURN_FALSE;
+	}
+
+	/* we checked that it will fit: */
+	start = (int32_t) lstart;
+
+	/* the offset is 'grapheme count offset' so it still might be invalid - we'll check it later */
+
+	if ( grapheme_ascii_check(str, str_len) >= 0 ) {
+		grapheme_substr_ascii((char *)str, str_len, start, length, ZEND_NUM_ARGS(), (char **) &sub_str, &sub_str_len);
+
+		if ( NULL == sub_str ) {
+			RETURN_FALSE;
+		}
+
+		RETURN_STRINGL(((char *)sub_str), sub_str_len, 1);
+	}
+
+	ustr = NULL;
+	ustr_len = 0;
+	status = U_ZERO_ERROR;
+	intl_convert_utf8_to_utf16(&ustr, &ustr_len, (char *)str, str_len, &status);
+
+	if ( U_FAILURE( status ) ) {
+		/* Set global error code. */
+		intl_error_set_code( NULL, status TSRMLS_CC );
+
+		/* Set error messages. */
+		intl_error_set_custom_msg( NULL, "Error converting input string to UTF-16", 1 TSRMLS_CC );
+		efree( ustr );
+		RETURN_FALSE;
+	}
+
+	bi = grapheme_get_break_iterator((void*)u_break_iterator_buffer, &status TSRMLS_CC );
+
+	if( U_FAILURE(status) ) {
+		RETURN_FALSE;
+	}
+	
+	ubrk_setText(bi, ustr, ustr_len,	&status);
+
+	if ( start < 0 ) {
+		iter_func = ubrk_previous;
+		ubrk_last(bi);
+		iter_val = 1;
+	}
+	else {
+		iter_func = ubrk_next;
+		iter_val = -1;
+	}
+
+	sub_str_start_pos = 0;
+
+	while ( start ) {
+		sub_str_start_pos = iter_func(bi);
+
+		if ( UBRK_DONE == sub_str_start_pos ) {
+			break;
+		}
+
+		start += iter_val;
+	}
+
+	if ( 0 != start || sub_str_start_pos >= ustr_len ) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_substr: start not contained in string", 1 TSRMLS_CC );
+	
+		efree(ustr);
+		ubrk_close(bi);
+		RETURN_FALSE;
+	}
+
+	if (ZEND_NUM_ARGS() <= 2) {
+
+		/* no length supplied, return the rest of the string */
+
+		sub_str = NULL;
+		sub_str_len = 0;
+		status = U_ZERO_ERROR;
+		intl_convert_utf16_to_utf8((char **)&sub_str, &sub_str_len, ustr + sub_str_start_pos, ustr_len - sub_str_start_pos, &status);
+
+		efree( ustr );
+		ubrk_close( bi );
+
+		if ( U_FAILURE( status ) ) {
+			/* Set global error code. */
+			intl_error_set_code( NULL, status TSRMLS_CC );
+
+			/* Set error messages. */
+			intl_error_set_custom_msg( NULL, "Error converting output string to UTF-8", 1 TSRMLS_CC );
+
+			efree( sub_str );
+
+			RETURN_FALSE;
+		}
+
+		/* return the allocated string, not a duplicate */
+		RETURN_STRINGL(((char *)sub_str), sub_str_len, 0);
+	}
+
+	/* find the end point of the string to return */
+
+	if ( length < 0 ) {
+		iter_func = ubrk_previous;
+		ubrk_last(bi);
+		iter_val = 1;
+	}
+	else {
+		iter_func = ubrk_next;
+		iter_val = -1;
+	}
+
+	sub_str_end_pos = 0;
+
+	while ( length ) {
+		sub_str_end_pos = iter_func(bi);
+
+		if ( UBRK_DONE == sub_str_end_pos ) {
+			break;
+		}
+
+		length += iter_val;
+	}
+	
+	if ( UBRK_DONE == sub_str_end_pos ) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_substr: length not contained in string", 1 TSRMLS_CC );
+	
+		efree(ustr);
+		ubrk_close(bi);
+		RETURN_FALSE;
+	}
+
+	sub_str = NULL;
+	status = U_ZERO_ERROR;
+	intl_convert_utf16_to_utf8((char **)&sub_str, &sub_str_len, ustr + sub_str_start_pos, ( sub_str_end_pos - sub_str_start_pos ), &status);
+
+	efree( ustr );
+	ubrk_close( bi );
+
+	if ( U_FAILURE( status ) ) {
+		/* Set global error code. */
+		intl_error_set_code( NULL, status TSRMLS_CC );
+
+		/* Set error messages. */
+		intl_error_set_custom_msg( NULL, "Error converting output string to UTF-8", 1 TSRMLS_CC );
+
+		if ( NULL != sub_str )
+			efree( sub_str );
+
+		RETURN_FALSE;
+	}
+
+	 /* return the allocated string, not a duplicate */
+	RETURN_STRINGL(((char *)sub_str), sub_str_len, 0);
+
+}
+/* }}} */
+
+/* {{{	strstr_common_handler */
+static void strstr_common_handler(INTERNAL_FUNCTION_PARAMETERS, int f_ignore_case)
+{
+	unsigned char *haystack, *needle, *found;
+	int haystack_len, needle_len;
+	int ret_pos, uchar_pos;
+	zend_bool part = 0;
+
+	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "ss|b", (char **)&haystack, &haystack_len, (char **)&needle, &needle_len, &part) == FAILURE) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "grapheme_strstr: unable to parse input param", 0 TSRMLS_CC );
+			 
+		RETURN_FALSE;
+	}
+
+	if (needle_len == 0) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_strpos: Empty delimiter", 1 TSRMLS_CC );
+		
+		RETURN_FALSE;
+	}
+
+
+	if ( !f_ignore_case ) {
+
+		/* ASCII optimization: quick check to see if the string might be there
+		 * I realize that 'offset' is 'grapheme count offset' but will work in spite of that 
+		*/
+		found = (unsigned char *)php_memnstr((char *)haystack, (char *)needle, needle_len, (char *)haystack + haystack_len);
+
+		/* if it isn't there the we are done */
+		if ( !found ) {
+			RETURN_FALSE;
+		}
+
+		/* if it is there, and if the haystack is ascii, we are all done */
+		if ( grapheme_ascii_check(haystack, haystack_len) >= 0 ) {
+			size_t found_offset = found - haystack;
+
+			if (part) {
+				RETURN_STRINGL(((char *)haystack) , found_offset, 1);
+			} else {
+				RETURN_STRINGL(((char *)found), haystack_len - found_offset, 1);
+			}
+		}
+
+	}
+
+	/* need to work in utf16 */
+	ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, 0, &uchar_pos, f_ignore_case TSRMLS_CC );
+
+	if ( ret_pos < 0 ) {
+		RETURN_FALSE;
+	}
+
+	/* uchar_pos is the 'nth' Unicode character position of the needle */
+
+	ret_pos = 0;
+	U8_FWD_N(haystack, ret_pos, haystack_len, uchar_pos);
+
+	if (part) {
+		RETURN_STRINGL(((char *)haystack), ret_pos, 1);
+	} 
+	else {
+		RETURN_STRINGL(((char *)haystack) + ret_pos, haystack_len - ret_pos, 1);
+	}
+
+}
+/* }}} */
+
+/* {{{ proto string grapheme_strstr(string haystack, string needle[, bool part])
+   Finds first occurrence of a string within another */
+PHP_FUNCTION(grapheme_strstr)
+{
+	strstr_common_handler(INTERNAL_FUNCTION_PARAM_PASSTHRU, 0 /* f_ignore_case */);
+}
+/* }}} */
+
+/* {{{ proto string grapheme_stristr(string haystack, string needle[, bool part])
+   Finds first occurrence of a string within another */
+PHP_FUNCTION(grapheme_stristr)
+{
+	strstr_common_handler(INTERNAL_FUNCTION_PARAM_PASSTHRU, 1 /* f_ignore_case */);
+}
+/* }}} */
+
+/* {{{ grapheme_extract_charcount_iter - grapheme iterator for grapheme_extract MAXCHARS */
+inline int32_t
+grapheme_extract_charcount_iter(UBreakIterator *bi, int32_t csize, unsigned char *pstr, int32_t str_len)
+{
+	int pos = 0, prev_pos = 0;
+	int ret_pos = 0, prev_ret_pos = 0;
+
+	while ( 1 ) {
+		pos = ubrk_next(bi);
+
+		if ( UBRK_DONE == pos ) {
+			break;
+		}
+
+		/* if we are beyond our limit, then the loop is done */
+		if ( pos > csize ) {
+			break;
+		}
+
+		/* update our pointer in the original UTF-8 buffer by as many characters
+		   as ubrk_next iterated over */
+
+		prev_ret_pos = ret_pos;
+		U8_FWD_N(pstr, ret_pos, str_len, pos - prev_pos);
+
+		if ( prev_ret_pos == ret_pos ) {
+			/* something wrong - malformed utf8? */
+			break;
+		}
+
+		prev_pos = pos;
+	}
+
+	return ret_pos;
+}
+/* }}} */
+
+/* {{{ grapheme_extract_bytecount_iter - grapheme iterator for grapheme_extract MAXBYTES */
+inline int32_t
+grapheme_extract_bytecount_iter(UBreakIterator *bi, int32_t bsize, unsigned char *pstr, int32_t str_len)
+{
+	int pos = 0, prev_pos = 0;
+	int ret_pos = 0, prev_ret_pos = 0;
+
+	while ( 1 ) {
+		pos = ubrk_next(bi);
+
+		if ( UBRK_DONE == pos ) {
+			break;
+		}
+
+		prev_ret_pos = ret_pos;
+		U8_FWD_N(pstr, ret_pos, str_len, pos - prev_pos);
+
+		if ( ret_pos > bsize ) {
+			ret_pos = prev_ret_pos;
+			break;
+		}
+
+		if ( prev_ret_pos == ret_pos ) {
+			/* something wrong - malformed utf8? */
+			break;
+		}
+
+		prev_pos = pos;
+	}
+
+	return ret_pos;
+}
+/* }}} */
+
+/* {{{ grapheme_extract_count_iter - grapheme iterator for grapheme_extract COUNT */
+inline int32_t
+grapheme_extract_count_iter(UBreakIterator *bi, int32_t size, unsigned char *pstr, int32_t str_len)
+{
+	int pos = 0, next_pos = 0;
+	int ret_pos = 0;
+
+	while ( size ) {
+		next_pos = ubrk_next(bi);
+
+		if ( UBRK_DONE == next_pos ) {
+			break;
+		}
+		pos = next_pos;
+		size--;
+	}
+
+	/* pos is one past the last UChar - and represent the number of code units to 
+		advance in the utf-8 buffer
+	*/
+
+	U8_FWD_N(pstr, ret_pos, str_len, pos);
+
+	return ret_pos;
+}
+/* }}} */
+
+/* {{{ grapheme extract iter function pointer array */
+typedef int32_t (*grapheme_extract_iter)(UBreakIterator * /*bi*/, int32_t /*size*/, unsigned char * /*pstr*/, int32_t /*str_len*/);
+
+static grapheme_extract_iter grapheme_extract_iters[] = {
+	&grapheme_extract_count_iter,
+	&grapheme_extract_bytecount_iter,
+	&grapheme_extract_charcount_iter,
+};
+/* }}} */
+
+/* {{{ proto string grapheme_extract(string str, int size[, int extract_type[, int start[, int next]]])
+	Function to extract a sequence of default grapheme clusters */
+PHP_FUNCTION(grapheme_extract)
+{
+	unsigned char *str, *pstr;
+	UChar *ustr;
+	int str_len, ustr_len;
+	long size; /* maximum number of grapheme clusters, bytes, or characters (based on extract_type) to return */
+	long lstart = 0; /* starting position in str in bytes */
+	int32_t start = 0;
+	long extract_type = GRAPHEME_EXTRACT_TYPE_COUNT;
+	UErrorCode status;
+	unsigned char u_break_iterator_buffer[U_BRK_SAFECLONE_BUFFERSIZE];
+	UBreakIterator* bi = NULL;
+	int ret_pos;
+	zval *next = NULL; // return offset of next part of the string
+
+	if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "sl|llz", (char **)&str, &str_len, &size, &extract_type, &lstart, &next) == FAILURE) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "grapheme_extract: unable to parse input param", 0 TSRMLS_CC );
+			 
+		RETURN_FALSE;
+	}
+
+	if ( NULL != next ) {
+		if ( !PZVAL_IS_REF(next) ) {
+			intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+				 "grapheme_extract: 'next' was not passed by reference", 0 TSRMLS_CC );
+			 
+			RETURN_FALSE;
+		}
+		else {
+			/* initialize next */
+            ZVAL_LONG(next, start);
+		}
+	}
+
+	if ( extract_type < GRAPHEME_EXTRACT_TYPE_MIN || extract_type > GRAPHEME_EXTRACT_TYPE_MAX ) {
+
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			 "grapheme_extract: unknown extract type param", 0 TSRMLS_CC );
+			 
+		RETURN_FALSE;
+	}
+
+	if ( lstart > INT32_MAX || lstart < 0 || lstart >= str_len ) {
+
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_extract: start not contained in string", 1 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	/* we checked that it will fit: */
+	start = (int32_t) lstart;
+
+	pstr = str + start;
+
+	/* just in case pstr points in the middle of a character, move forward to the start of the next char */
+	if ( !UTF8_IS_SINGLE(*pstr) && !U8_IS_LEAD(*pstr) ) {
+		unsigned char *str_end = str + str_len;
+
+		while ( !UTF8_IS_SINGLE(*pstr) && !U8_IS_LEAD(*pstr) ) {
+			pstr++;
+			if ( pstr >= str_end ) {
+				intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+								"grapheme_extract: invalid input string", 0 TSRMLS_CC );
+			 
+				RETURN_FALSE;
+			}
+		}
+	}
+
+	str_len -= (pstr - str);
+
+	/* if the string is all ASCII up to size+1 - or str_len whichever is first - then we are done.
+		(size + 1 because the size-th character might be the beginning of a grapheme cluster)
+	 */
+	
+	if ( -1 != grapheme_ascii_check(pstr, size + 1 < str_len ? size + 1 : str_len ) ) {
+		if ( NULL != next ) {
+			ZVAL_LONG(next, start+size);
+		}
+		RETURN_STRINGL(((char *)pstr), size, 1);
+	}
+
+	/* convert the strings to UTF-16. */
+	ustr = NULL;
+	ustr_len = 0;
+	status = U_ZERO_ERROR;
+	intl_convert_utf8_to_utf16(&ustr, &ustr_len, (char *)pstr, str_len, &status );
+
+	if ( U_FAILURE( status ) ) {
+		/* Set global error code. */
+		intl_error_set_code( NULL, status TSRMLS_CC );
+
+		/* Set error messages. */
+		intl_error_set_custom_msg( NULL, "Error converting input string to UTF-16", 1 TSRMLS_CC );
+
+		if ( NULL != ustr )
+			efree( ustr );
+
+		RETURN_FALSE;
+	}
+
+	bi = NULL;
+	status = U_ZERO_ERROR;
+	bi = grapheme_get_break_iterator(u_break_iterator_buffer, &status TSRMLS_CC );
+
+	ubrk_setText(bi, ustr, ustr_len, &status);
+
+	/* if the caller put us in the middle of a grapheme, we can't detect it in all cases since we
+		can't back up. So, we will not do anything. */
+
+	/* now we need to find the end of the chunk the user wants us to return */
+
+	ret_pos = (*grapheme_extract_iters[extract_type])(bi, size, pstr, str_len);
+
+	efree(ustr);
+	ubrk_close(bi);
+
+	if ( NULL != next ) {
+		ZVAL_LONG(next, start+ret_pos);
+	}
+
+	RETURN_STRINGL(((char *)pstr), ret_pos, 1);
+}
+
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: fdm=marker
+ * vim: noet sw=4 ts=4
+ */
+

ext/intl/grapheme/grapheme_util.c

@@ -0,0 +1,619 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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.               |
+   +----------------------------------------------------------------------+
+   | Author: Ed Batutis <ed@batutis.com>                                  |
+   +----------------------------------------------------------------------+
+ */
+
+/* {{{ includes */
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <php.h>
+#include "grapheme.h"
+#include "grapheme_util.h"
+#include "intl_common.h"
+
+#include <unicode/utypes.h>
+#include <unicode/ucol.h>
+#include <unicode/ustring.h>
+#include <unicode/ubrk.h>
+
+#include "ext/standard/php_string.h"
+
+ZEND_EXTERN_MODULE_GLOBALS( intl )
+
+/* }}} */
+
+/* {{{ grapheme_close_global_iterator - clean up */
+void
+grapheme_close_global_iterator( TSRMLS_D )
+{
+	UBreakIterator *global_break_iterator = INTL_G( grapheme_iterator );
+
+	if ( NULL != global_break_iterator ) {
+		ubrk_close(global_break_iterator);
+	}
+}
+/* }}} */
+
+/* {{{ grapheme_intl_case_fold: convert string to lowercase */
+void
+grapheme_intl_case_fold(UChar** ptr_to_free, UChar **str, int32_t *str_len, UErrorCode *pstatus )
+{
+    UChar *dest;
+    int32_t dest_len, size_required;
+
+    /* allocate a destination string that is a bit larger than the src, hoping that is enough */
+    dest_len = (*str_len) + ( *str_len / 10 );
+    dest = (UChar*) eumalloc(dest_len);
+
+    *pstatus = U_ZERO_ERROR;
+    size_required = u_strFoldCase(dest, dest_len, *str, *str_len, U_FOLD_CASE_DEFAULT, pstatus);
+
+    dest_len = size_required;
+
+    if ( U_BUFFER_OVERFLOW_ERROR == *pstatus ) {
+
+        dest = (UChar*) eurealloc(dest, dest_len);
+
+        *pstatus = U_ZERO_ERROR;
+        size_required = u_strFoldCase(dest, dest_len, *str, *str_len, U_FOLD_CASE_DEFAULT, pstatus);
+    }
+
+    if ( U_FAILURE(*pstatus) ) {
+        return;
+    }
+
+    if ( NULL != ptr_to_free) {
+        efree(*ptr_to_free);
+        *ptr_to_free = dest;
+    }
+
+    *str = dest;
+    *str_len = dest_len;
+
+    return;
+}
+/* }}} */
+
+/* {{{ grapheme_substr_ascii f='from' - starting point, l='length' */
+void
+grapheme_substr_ascii(char *str, int str_len, int f, int l, int argc, char **sub_str, int *sub_str_len)
+{
+    *sub_str = NULL;
+
+    if (argc > 2) {
+        if ((l < 0 && -l > str_len)) {
+            return;
+        } else if (l > str_len) {
+            l = str_len;
+        }
+    } else {
+        l = str_len;
+    }
+
+    if (f > str_len || (f < 0 && -f > str_len)) {
+        return;
+    }
+
+    if (l < 0 && (l + str_len - f) < 0) {
+        return;
+    }
+
+    /* if "from" position is negative, count start position from the end
+     * of the string
+     */
+    if (f < 0) {
+        f = str_len + f;
+        if (f < 0) {
+            f = 0;
+        }
+    }
+
+
+    /* if "length" position is negative, set it to the length
+     * needed to stop that many chars from the end of the string
+     */
+    if (l < 0) {
+        l = (str_len - f) + l;
+        if (l < 0) {
+            l = 0;
+        }
+    }
+
+    if (f >= str_len) {
+        return;
+    }
+
+    if ((f + l) > str_len) {
+        l = str_len - f;
+    }
+
+    *sub_str = str + f;
+    *sub_str_len = l;
+
+    return;
+}
+/* }}} */
+
+/* {{{ grapheme_strrpos_utf16 - strrpos using utf16 */
+int
+grapheme_strrpos_utf16(unsigned char *haystack, int32_t haystack_len, unsigned char*needle, int32_t needle_len, int32_t offset, int f_ignore_case TSRMLS_DC)
+{
+    UChar *uhaystack, *puhaystack, *uhaystack_end, *uneedle;
+    int32_t uhaystack_len, uneedle_len;
+    UErrorCode status;
+    unsigned char u_break_iterator_buffer[U_BRK_SAFECLONE_BUFFERSIZE];
+    UBreakIterator* bi = NULL;
+    int ret_pos, pos;
+
+    /* convert the strings to UTF-16. */
+    uhaystack = NULL;
+    uhaystack_len = 0;
+    status = U_ZERO_ERROR;
+    intl_convert_utf8_to_utf16(&uhaystack, &uhaystack_len, (char *) haystack, haystack_len, &status );
+
+    if ( U_FAILURE( status ) ) {
+        /* Set global error code. */
+        intl_error_set_code( NULL, status TSRMLS_CC );
+
+        /* Set error messages. */
+        intl_error_set_custom_msg( NULL, "Error converting input string to UTF-16", 1 TSRMLS_CC );
+        efree( uhaystack );
+        return -1;
+    }
+
+    if ( f_ignore_case ) {
+        grapheme_intl_case_fold(&uhaystack, &uhaystack, &uhaystack_len, &status );
+    }
+
+    /* get a pointer to the haystack taking into account the offset */
+    bi = NULL;
+    status = U_ZERO_ERROR;
+    bi = grapheme_get_break_iterator(u_break_iterator_buffer, &status TSRMLS_CC );
+
+    puhaystack = grapheme_get_haystack_offset(bi, uhaystack, uhaystack_len, offset);
+
+    if ( NULL == puhaystack ) {
+        intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_strpos: Offset not contained in string", 1 TSRMLS_CC );
+        efree( uhaystack );
+        ubrk_close (bi);
+        return -1;
+    }
+
+    uneedle = NULL;
+    uneedle_len = 0;
+    status = U_ZERO_ERROR;
+    intl_convert_utf8_to_utf16(&uneedle, &uneedle_len, (char *) needle, needle_len, &status );
+
+    if ( U_FAILURE( status ) ) {
+        /* Set global error code. */
+        intl_error_set_code( NULL, status TSRMLS_CC );
+
+        /* Set error messages. */
+        intl_error_set_custom_msg( NULL, "Error converting input string to UTF-16", 1 TSRMLS_CC );
+        efree( uhaystack );
+        efree( uneedle );
+        ubrk_close (bi);
+        return -1;
+    }
+
+    if ( f_ignore_case ) {
+        grapheme_intl_case_fold(&uneedle, &uneedle, &uneedle_len, &status );
+    }
+
+    ret_pos = -1;   /* -1 represents 'not found' */
+
+    /* back up until there's needle_len characters to compare */
+
+    uhaystack_end = uhaystack + uhaystack_len;
+    pos = ubrk_last(bi);
+    puhaystack = uhaystack + pos;
+
+    while ( uhaystack_end - puhaystack < uneedle_len ) {
+
+        pos = ubrk_previous(bi);
+
+        if ( UBRK_DONE == pos ) {
+            break;
+        }
+
+        puhaystack = uhaystack + pos;
+    }
+
+    /* is there enough haystack left to hold the needle? */
+    if ( ( uhaystack_end - puhaystack ) < uneedle_len ) {
+        /* not enough, not found */
+        goto exit;
+    }
+
+    while ( UBRK_DONE != pos ) {
+
+        if (!u_memcmp(uneedle, puhaystack, uneedle_len)) {  /* needle_len - 1 in zend memnstr? */
+
+            /* does the grapheme in the haystack end at the same place as the last grapheme in the needle? */
+
+            if ( ubrk_isBoundary(bi, pos + uneedle_len) ) {
+
+                /* found it, get grapheme count offset */
+                ret_pos = grapheme_count_graphemes(bi, uhaystack, pos);
+                break;
+            }
+
+            /* set position back */
+            ubrk_isBoundary(bi, pos);
+        }
+
+        pos = ubrk_previous(bi);
+        puhaystack = uhaystack + pos;
+    }
+
+exit:
+    efree( uhaystack );
+    efree( uneedle );
+    ubrk_close (bi);
+
+    return ret_pos;
+}
+
+/* }}} */
+
+/* {{{ grapheme_strpos_utf16 - strrpos using utf16*/
+int
+grapheme_strpos_utf16(unsigned char *haystack, int32_t haystack_len, unsigned char*needle, int32_t needle_len, int32_t offset, int32_t *puchar_pos, int f_ignore_case TSRMLS_DC)
+{
+	UChar *uhaystack, *puhaystack, *uneedle;
+	int32_t uhaystack_len, uneedle_len;
+	int ret_pos;
+	unsigned char u_break_iterator_buffer[U_BRK_SAFECLONE_BUFFERSIZE];
+	UBreakIterator* bi;
+	UErrorCode status;
+
+	*puchar_pos = -1;
+
+	/* convert the strings to UTF-16. */
+
+	uhaystack = NULL;
+	uhaystack_len = 0;
+	status = U_ZERO_ERROR;
+	intl_convert_utf8_to_utf16(&uhaystack, &uhaystack_len, (char *) haystack, haystack_len, &status );
+
+	if ( U_FAILURE( status ) ) {
+		/* Set global error code. */
+		intl_error_set_code( NULL, status TSRMLS_CC );
+
+		/* Set error messages. */
+		intl_error_set_custom_msg( NULL, "Error converting input string to UTF-16", 1 TSRMLS_CC );
+		efree( uhaystack );
+		return -1;
+	}
+
+	/* get a pointer to the haystack taking into account the offset */
+	bi = NULL;
+	status = U_ZERO_ERROR;
+	bi = grapheme_get_break_iterator(u_break_iterator_buffer, &status TSRMLS_CC );
+	
+	puhaystack = grapheme_get_haystack_offset(bi, uhaystack, uhaystack_len, offset);
+	uhaystack_len = (uhaystack_len - ( puhaystack - uhaystack));
+
+	if ( NULL == puhaystack ) {
+	
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR, "grapheme_strpos: Offset not contained in string", 1 TSRMLS_CC );
+		
+		efree( uhaystack );
+		ubrk_close (bi);
+					
+		return -1;
+	}
+
+	if ( f_ignore_case ) {
+		grapheme_intl_case_fold(&uhaystack, &puhaystack, &uhaystack_len, &status );
+	}
+
+	uneedle = NULL;
+	uneedle_len = 0;
+	status = U_ZERO_ERROR;
+	intl_convert_utf8_to_utf16(&uneedle, &uneedle_len, (char *) needle, needle_len, &status );
+
+	if ( U_FAILURE( status ) ) {
+		/* Set global error code. */
+		intl_error_set_code( NULL, status TSRMLS_CC );
+
+		/* Set error messages. */
+		intl_error_set_custom_msg( NULL, "Error converting input string to UTF-16", 1 TSRMLS_CC );
+		efree( uhaystack );
+		efree( uneedle );
+		ubrk_close (bi);
+		
+		return -1;
+	}
+
+	if ( f_ignore_case ) {
+		grapheme_intl_case_fold(&uneedle, &uneedle, &uneedle_len, &status );
+	}
+
+	ret_pos = grapheme_memnstr_grapheme(bi, puhaystack, uneedle, uneedle_len, puhaystack + uhaystack_len );
+	
+	*puchar_pos = ubrk_current(bi);
+
+	efree( uhaystack );
+	efree( uneedle );
+	ubrk_close (bi);
+
+	return ret_pos;
+}
+
+/* }}} */
+
+/* {{{ grapheme_ascii_check: ASCII check */
+int grapheme_ascii_check(const unsigned char *day, int32_t len)
+{
+	int ret_len = len;
+	while ( len-- ) {
+	if ( *day++ > 0x7f )
+		return -1;
+	}
+
+	return ret_len;
+}
+
+/* }}} */
+
+/* {{{ grapheme_split_string: find and optionally return grapheme boundaries */
+int grapheme_split_string(const UChar *text, int32_t text_length, int boundary_array[], int boundary_array_len TSRMLS_DC )
+{
+	unsigned char u_break_iterator_buffer[U_BRK_SAFECLONE_BUFFERSIZE];
+	UErrorCode		status = U_ZERO_ERROR;
+	int ret_len, pos;
+	UBreakIterator* bi;
+
+	bi = grapheme_get_break_iterator((void*)u_break_iterator_buffer, &status TSRMLS_CC );
+
+	if( U_FAILURE(status) ) {
+		return -1;
+	}
+	
+	ubrk_setText(bi, text, text_length,	&status);
+
+	pos = 0;
+	
+	for ( ret_len = 0; pos != UBRK_DONE; ) {
+	
+		pos = ubrk_next(bi);
+		
+		if ( pos != UBRK_DONE ) {
+		
+			if ( NULL != boundary_array && ret_len < boundary_array_len ) {
+				boundary_array[ret_len] = pos;
+			}
+
+			ret_len++;
+		}
+	}
+	 		
+	ubrk_close(bi);
+	
+	return ret_len;
+}
+/* }}} */
+
+/* {{{ grapheme_count_graphemes */
+inline int32_t
+grapheme_count_graphemes(UBreakIterator *bi, UChar *string, int32_t string_len)
+{
+	int ret_len = 0;
+	int pos = 0;
+	UErrorCode		status = U_ZERO_ERROR;
+	
+	ubrk_setText(bi, string, string_len, &status);
+
+	do {
+	
+		pos = ubrk_next(bi);
+		
+		if ( UBRK_DONE != pos ) {
+			ret_len++;
+		}
+		
+	} while ( UBRK_DONE != pos );
+	
+	return ret_len;
+}
+/* }}} */
+
+/* {{{ grapheme_memnstr_grapheme: find needle in haystack using grapheme boundaries */
+inline int32_t
+grapheme_memnstr_grapheme(UBreakIterator *bi, UChar *haystack, UChar *needle, int32_t needle_len, UChar *end)
+{
+	UChar *p = haystack;
+	UChar ne = needle[needle_len-1];
+	UErrorCode status;
+	int32_t grapheme_offset;
+	
+	end -= needle_len;
+
+	while (p <= end) {
+
+		if ((p = u_memchr(p, *needle, (end-p+1))) && ne == p[needle_len-1]) {
+
+			if (!u_memcmp(needle, p, needle_len - 1)) {  /* needle_len - 1 works because if needle_len is 1, we've already tested the char */
+
+				/* does the grapheme end here? */
+
+				status = U_ZERO_ERROR;
+				ubrk_setText (bi, haystack, (end - haystack) + needle_len, &status);
+
+				if ( ubrk_isBoundary (bi, (p - haystack) + needle_len) ) {
+
+					/* found it, get grapheme count offset */
+					grapheme_offset = grapheme_count_graphemes(bi, haystack, (p - haystack));
+
+					return grapheme_offset;
+				}
+			}
+		}
+
+		if (p == NULL) {
+			return -1;
+		}
+
+		p++;
+	}
+
+	return -1;
+}
+
+/* }}} */
+
+/* {{{ grapheme_memrstr_grapheme: reverse find needle in haystack using grapheme boundaries */
+inline void *grapheme_memrchr_grapheme(const void *s, int c, int32_t n)
+{
+	register unsigned char *e;
+
+	if (n <= 0) {
+		return NULL;
+	}
+
+	for (e = (unsigned char *)s + n - 1; e >= (unsigned char *)s; e--) {
+		if (*e == (unsigned char)c) {
+			return (void *)e;
+		}
+	}
+
+	return NULL;
+}
+/* }}} */
+
+/* {{{ 	grapheme_get_haystack_offset - bump the haystack pointer based on the grapheme count offset */
+UChar *
+grapheme_get_haystack_offset(UBreakIterator* bi, UChar *uhaystack, int32_t uhaystack_len, int32_t offset)
+{
+	UErrorCode		status;
+	int32_t pos;
+	int32_t (*iter_op)(UBreakIterator* bi);
+	int iter_incr;
+
+	if ( NULL != bi ) {
+		status = U_ZERO_ERROR;
+		ubrk_setText (bi, uhaystack, uhaystack_len, &status);
+	}
+
+	if ( 0 == offset ) {
+		return uhaystack;
+	}
+	
+	if ( offset < 0 ) {
+		iter_op = ubrk_previous;
+		ubrk_last(bi); /* one past the end */
+		iter_incr = 1;
+	}
+	else {
+		iter_op = ubrk_next;
+		iter_incr = -1;
+	}
+	
+	pos = 0;
+	
+	while ( pos != UBRK_DONE && offset != 0 ) {
+	
+		pos = iter_op(bi);
+		
+		if ( UBRK_DONE != pos ) {
+			offset += iter_incr;
+		}
+	}
+
+	if ( offset != 0 ) {
+		return NULL;
+	}
+	
+	return uhaystack + pos;
+}
+/* }}} */
+
+/* {{{ grapheme_strrpos_ascii: borrowed from the php ext/standard/string.c */
+ int32_t
+grapheme_strrpos_ascii(unsigned char *haystack, int32_t haystack_len, unsigned char *needle, int32_t needle_len, int32_t offset)
+{
+	unsigned char *p, *e;
+
+	if (offset >= 0) {
+		p = haystack + offset;
+		e = haystack + haystack_len - needle_len;
+	} else {
+		p = haystack;
+		if (needle_len > -offset) {
+			e = haystack + haystack_len - needle_len;
+		} else {
+			e = haystack + haystack_len + offset;
+		}
+	}
+
+	if (needle_len == 1) {
+		/* Single character search can shortcut memcmps */
+		while (e >= p) {
+			if (*e == *needle) {
+				return (e - p + (offset > 0 ? offset : 0));
+			}
+			e--;
+		}
+		return -1;
+	}
+
+	while (e >= p) {
+		if (memcmp(e, needle, needle_len) == 0) {
+			return (e - p + (offset > 0 ? offset : 0));
+		}
+		e--;
+	}
+
+	return -1;
+}
+
+/* }}} */
+
+/* {{{ grapheme_get_break_iterator: get a clone of the global character break iterator */
+UBreakIterator* 
+grapheme_get_break_iterator(void *stack_buffer, UErrorCode *status TSRMLS_DC )
+{
+	int32_t buffer_size;
+
+	UBreakIterator *global_break_iterator = INTL_G( grapheme_iterator );
+
+	if ( NULL == global_break_iterator ) {
+
+		global_break_iterator = ubrk_open(UBRK_CHARACTER, 
+											NULL,	/* icu default locale - locale has no effect on this iterator */
+											NULL,	/* text not set in global iterator */
+											0,		/* text length = 0 */
+											status);
+
+		INTL_G(grapheme_iterator) = global_break_iterator;
+	}
+
+	buffer_size = U_BRK_SAFECLONE_BUFFERSIZE;
+
+	return ubrk_safeClone(global_break_iterator, stack_buffer, &buffer_size, status);
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: fdm=marker
+ * vim: noet sw=4 ts=4
+ */
+

ext/intl/grapheme/grapheme_util.h

@@ -0,0 +1,59 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Ed Batutis <ed@batutis.com>                                 |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef GRAPHEME_GRAPHEME_UTIL_H
+#define GRAPHEME_GRAPHEME_UTIL_H
+
+#include "php_intl.h"
+#include "intl_convert.h"
+
+/* get_break_interator: get a break iterator from the global structure */
+UBreakIterator* grapheme_get_break_iterator(void *stack_buffer, UErrorCode *status TSRMLS_DC );
+
+void
+grapheme_substr_ascii(char *str, int32_t str_len, int32_t f, int32_t l, int argc, char **sub_str, int *sub_str_len);
+
+int
+grapheme_strrpos_utf16(unsigned char *haystack, int32_t haystack_len, unsigned char*needle, int32_t needle_len, int32_t offset, int f_ignore_case TSRMLS_DC);
+
+int
+grapheme_strpos_utf16(unsigned char *haystack, int32_t haystack_len, unsigned char*needle, int32_t needle_len, int32_t offset, int *puchar_pos, int f_ignore_case TSRMLS_DC);
+
+int grapheme_ascii_check(const unsigned char *day, int32_t len);
+
+int grapheme_split_string(const UChar *text, int32_t text_length, int boundary_array[], int boundary_array_len TSRMLS_DC );
+
+inline int32_t
+grapheme_count_graphemes(UBreakIterator *bi, UChar *string, int32_t string_len);
+
+inline int32_t
+grapheme_memnstr_grapheme(UBreakIterator *bi, UChar *haystack, UChar *needle, int32_t needle_len, UChar *end);
+
+inline void *grapheme_memrchr_grapheme(const void *s, int c, int32_t n);
+
+UChar *
+grapheme_get_haystack_offset(UBreakIterator* bi, UChar *uhaystack, int32_t uhaystack_len, int32_t offset);
+
+int32_t
+grapheme_strrpos_ascii(unsigned char *haystack, int32_t haystack_len, unsigned char *needle, int32_t needle_len, int32_t offset);
+
+UBreakIterator* 
+grapheme_get_break_iterator(void *stack_buffer, UErrorCode *status TSRMLS_DC );
+
+/* OUTSIDE_STRING: check if (possibly negative) long offset is outside the string with int32_t length */
+#define OUTSIDE_STRING(offset, max_len) ( offset < INT32_MIN || offset > INT32_MAX || (offset < 0 ? -offset > (long) max_len : offset >= (long) max_len) )
+
+#endif // GRAPHEME_GRAPHEME_UTIL_H

ext/intl/intl_common.h

@@ -0,0 +1,44 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   |          Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef INTL_COMMON_H
+#define INTL_COMMON_H
+/* Auxiliary macros */
+
+#include <php.h>
+#include <unicode/utypes.h>
+
+#ifndef UBYTES
+# define UBYTES(len) ((len) * sizeof(UChar))
+#endif
+
+#ifndef eumalloc
+# define eumalloc(size)  (UChar*)safe_emalloc(size, sizeof(UChar), 0)
+#endif
+
+#ifndef eurealloc
+# define eurealloc(ptr, size)  (UChar*)erealloc((ptr), size * sizeof(UChar))
+#endif
+
+#define USIZE(data) sizeof((data))/sizeof(UChar)
+#define UCHARS(len) ((len) / sizeof(UChar))
+
+#define INTL_Z_STRVAL_P(str) (UChar*) Z_STRVAL_P(str)
+#define INTL_Z_STRLEN_P(str) UCHARS( Z_STRLEN_P(str) )
+
+#endif /* INTL_COMMON_H */

ext/intl/intl_convert.c

@@ -0,0 +1,154 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <php.h>
+
+#include "intl_common.h"
+#include "intl_convert.h"
+
+/* {{{ intl_convert_utf8_to_utf16
+ * Convert given string from UTF-8 to UTF-16 to *target buffer.
+ *
+ * It *target is NULL then we allocate a large enough buffer,
+ * store the converted string into it, and make target point to it.
+ *
+ * Otherwise, if *target is non-NULL, we assume that it points to a
+ * dynamically allocated buffer of *target_len bytes length.
+ * In this case the buffer will be used to store the converted string to,
+ * and may be resized (made larger) if needed.
+ *
+ * @param target      Where to place the result.
+ * @param target_len  Result length.
+ * @param source      String to convert.
+ * @param source_len  Length of the source string.
+ * @param status      Conversion status.
+ *
+ * @return void       This function does not return anything.
+ */
+void intl_convert_utf8_to_utf16(
+	UChar**     target, int* target_len,
+	const char* src,    int  src_len,
+	UErrorCode* status )
+{
+	UChar*      dst_buf = NULL;
+	int32_t     dst_len = 0;
+
+	// If *target is NULL determine required destination buffer size (pre-flighting).
+	// Otherwise, attempt to convert source string; if *target buffer is not large enough
+	// it will be resized appropriately.
+	*status = U_ZERO_ERROR;
+
+	u_strFromUTF8( *target, *target_len, &dst_len, src, src_len, status );
+
+	if( *status == U_ZERO_ERROR )
+	{
+		// String is converted successfuly
+		(*target)[dst_len] = 0;
+		*target_len = dst_len;
+		return;
+	}
+
+	// Bail out if an unexpected error occured.
+	// (U_BUFFER_OVERFLOW_ERROR means that *target buffer is not large enough).
+	// (U_STRING_NOT_TERMINATED_WARNING usually means that the input string is empty).
+	if( *status != U_BUFFER_OVERFLOW_ERROR && *status != U_STRING_NOT_TERMINATED_WARNING )
+		return;
+
+	// Allocate memory for the destination buffer (it will be zero-terminated).
+	dst_buf = eumalloc( dst_len + 1 );
+
+	// Convert source string from UTF-8 to UTF-16.
+	*status = U_ZERO_ERROR;
+	u_strFromUTF8( dst_buf, dst_len+1, NULL, src, src_len, status );
+	if( U_FAILURE( *status ) )
+	{
+		efree( dst_buf );
+		return;
+	}
+
+	dst_buf[dst_len] = 0;
+
+	if( *target )
+		efree( *target );
+
+	*target     = dst_buf;
+	*target_len = dst_len;
+}
+/* }}} */
+
+/* {{{ intl_convert_utf16_to_utf8
+ * Convert given string from UTF-16 to UTF-8.
+ *
+ * @param target      Where to place the result.
+ * @param target_len  Result length.
+ * @param source      String to convert.
+ * @param source_len  Length of the source string.
+ * @param status      Conversion status.
+ *
+ * @return void       This function does not return anything.
+ */
+void intl_convert_utf16_to_utf8(
+	char**       target, int* target_len,
+	const UChar* src,    int  src_len,
+	UErrorCode*  status )
+{
+	char*       dst_buf = NULL;
+	int32_t     dst_len;
+
+	// Determine required destination buffer size (pre-flighting).
+	*status = U_ZERO_ERROR;
+	u_strToUTF8( NULL, 0, &dst_len, src, src_len, status );
+
+	// Bail out if an unexpected error occured.
+	// (U_BUFFER_OVERFLOW_ERROR means that *target buffer is not large enough).
+	// (U_STRING_NOT_TERMINATED_WARNING usually means that the input string is empty).
+	if( *status != U_BUFFER_OVERFLOW_ERROR && *status != U_STRING_NOT_TERMINATED_WARNING )
+		return;
+
+	// Allocate memory for the destination buffer (it will be zero-terminated).
+	dst_buf = emalloc( dst_len+1 );
+
+	// Convert source string from UTF-8 to UTF-16.
+	*status = U_ZERO_ERROR;
+	u_strToUTF8( dst_buf, dst_len, NULL, src, src_len, status );
+	if( U_FAILURE( *status ) )
+	{
+		efree( dst_buf );
+		return;
+	}
+
+	// U_STRING_NOT_TERMINATED_WARNING is OK for us => reset 'status'.
+	*status = U_ZERO_ERROR;
+
+	dst_buf[dst_len] = 0;
+	*target     = dst_buf;
+	*target_len = dst_len;
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/intl_convert.h

@@ -0,0 +1,33 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef INTL_CONVERT_H
+#define INTL_CONVERT_H
+
+#include <unicode/ustring.h>
+
+void intl_convert_utf8_to_utf16(
+	UChar**     target, int* target_len,
+	const char* src,    int  src_len,
+	UErrorCode* status );
+
+void intl_convert_utf16_to_utf8(
+	char**       target, int* target_len,
+	const UChar* src,    int  src_len,
+	UErrorCode*  status );
+
+#endif // INTL_CONVERT_H

ext/intl/intl_data.h

@@ -0,0 +1,87 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   |          Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef INTL_DATA_H
+#define INTL_DATA_H
+
+#include <unicode/utypes.h>
+
+#include "intl_error.h"
+
+/* Mock object to generalize error handling in sub-modules.
+   Sub-module data structures should always have error as first element 
+   for this to work! 
+*/
+typedef struct _intl_data {
+	zend_object		zo;
+	intl_error		error;
+} intl_object;
+
+#define INTL_METHOD_INIT_VARS(oclass, obj)		\
+	zval*             object  = NULL;			\
+	oclass##_object*  obj     = NULL;			\
+	intl_error_reset( NULL TSRMLS_CC );			
+
+#define INTL_DATA_ERROR(obj)				(((intl_object *)(obj))->error)
+#define INTL_DATA_ERROR_P(obj)				(&(INTL_DATA_ERROR((obj))))
+#define INTL_DATA_ERROR_CODE(obj)			INTL_ERROR_CODE(INTL_DATA_ERROR((obj)))
+
+#define INTL_METHOD_FETCH_OBJECT(oclass, obj)									\
+	obj = (oclass##_object *) zend_object_store_get_object( object TSRMLS_CC );	\
+    intl_error_reset( INTL_DATA_ERROR_P(obj) TSRMLS_CC );						\
+
+#define INTL_METHOD_CHECK_STATUS(obj, msg)											\
+    intl_error_set_code( NULL, INTL_DATA_ERROR_CODE((obj)) TSRMLS_CC );				\
+    if( U_FAILURE( INTL_DATA_ERROR_CODE((obj)) ) )									\
+    {																				\
+        intl_errors_set_custom_msg( INTL_DATA_ERROR_P((obj)), msg, 0 TSRMLS_CC );	\
+        RETURN_FALSE;										\
+    }
+
+#define INTL_METHOD_RETVAL_UTF8(obj, ustring, ulen, free_it)									\
+{																								\
+	char *u8value;																				\
+	int u8len;																					\
+	intl_convert_utf16_to_utf8(&u8value, &u8len, ustring, ulen, &INTL_DATA_ERROR_CODE((obj)));	\
+	if((free_it)) {																				\
+		efree(ustring);																			\
+	}																							\
+	INTL_METHOD_CHECK_STATUS((obj), "Error converting value to UTF-8");							\
+	RETVAL_STRINGL(u8value, u8len, 0);															\
+}
+
+#define INTL_MAX_LOCALE_LEN 64
+
+#define INTL_CHECK_LOCALE_LEN(locale_len)												\
+	if((locale_len) > INTL_MAX_LOCALE_LEN) {											\
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,									\
+	"Locale string too long, should be no longer than 64 characters", 0 TSRMLS_CC );	\
+		RETURN_NULL();																	\
+	}
+
+#define INTL_CHECK_LOCALE_LEN_OBJ(locale_len, object)									\
+	if((locale_len) > INTL_MAX_LOCALE_LEN) {											\
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,									\
+	"Locale string too long, should be no longer than 64 characters", 0 TSRMLS_CC );	\
+		zval_dtor(object);																\
+		ZVAL_NULL(object);																\
+		RETURN_NULL();																	\
+	}
+
+
+#endif // INTL_DATA_H

ext/intl/intl_error.c

@@ -0,0 +1,215 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   |          Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <php.h>
+
+#include "php_intl.h"
+#include "intl_error.h"
+
+ZEND_EXTERN_MODULE_GLOBALS( intl )
+
+/* {{{ intl_error* intl_g_error_get()
+ * Return global error structure.
+ */
+static intl_error* intl_g_error_get( TSRMLS_D )
+{
+	return &INTL_G( g_error );
+}
+/* }}} */
+
+/* {{{ void intl_free_custom_error_msg( intl_error* err )
+ * Free mem.
+ */
+static void intl_free_custom_error_msg( intl_error* err TSRMLS_DC )
+{
+	if( !err && !( err = intl_g_error_get( TSRMLS_C ) ) )
+		return;
+
+	if( !err->free_custom_error_message )
+		return;
+
+	efree( err->custom_error_message );
+
+	err->custom_error_message      = NULL;
+	err->free_custom_error_message = 0;
+}
+/* }}} */
+
+/* {{{ intl_error* intl_error_create()
+ * Create and initialize  internals of 'intl_error'.
+ */
+intl_error* intl_error_create( TSRMLS_D )
+{
+	intl_error* err = ecalloc( 1, sizeof( intl_error ) );
+
+	intl_error_init( err TSRMLS_CC );
+
+	return err;
+}
+/* }}} */
+
+/* {{{ void intl_error_init( intl_error* coll_error )
+ * Initialize internals of 'intl_error'.
+ */
+void intl_error_init( intl_error* err TSRMLS_DC )
+{
+	if( !err && !( err = intl_g_error_get( TSRMLS_C ) ) )
+		return;
+
+	err->code                      = U_ZERO_ERROR;
+	err->custom_error_message      = NULL;
+	err->free_custom_error_message = 0;
+}
+/* }}} */
+
+/* {{{ void intl_error_reset( intl_error* err )
+ * Set last error code to 0 and unset last error message
+ */
+void intl_error_reset( intl_error* err TSRMLS_DC )
+{
+	if( !err && !( err = intl_g_error_get( TSRMLS_C ) ) )
+		return;
+
+	err->code = U_ZERO_ERROR;
+
+	intl_free_custom_error_msg( err TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ void intl_error_set_custom_msg( intl_error* err, char* msg, int copyMsg )
+ * Set last error message to msg copying it if needed.
+ */
+void intl_error_set_custom_msg( intl_error* err, char* msg, int copyMsg TSRMLS_DC )
+{
+	if( !msg )
+		return;
+
+	if( !err && !( err = intl_g_error_get( TSRMLS_C ) ) )
+		return;
+
+	// Free previous message if any
+	intl_free_custom_error_msg( err TSRMLS_CC );
+
+	// Mark message copied if any
+	err->free_custom_error_message = copyMsg;
+
+	// Set user's error text message
+	err->custom_error_message = copyMsg ? estrdup( msg ) : msg;
+}
+/* }}} */
+
+/* {{{ const char* intl_error_get_message( intl_error* err )
+ * Create output message in format "<intl_error_text>: <extra_user_error_text>".
+ */
+char* intl_error_get_message( intl_error* err TSRMLS_DC )
+{
+	const char* uErrorName = NULL;
+	char*       errMessage = 0;
+
+	if( !err && !( err = intl_g_error_get( TSRMLS_C ) ) )
+		return estrdup( "" );
+
+	uErrorName = u_errorName( err->code );
+
+	// Format output string
+	if( err->custom_error_message )
+	{
+		spprintf( &errMessage, 0, "%s: %s", err->custom_error_message, uErrorName );
+	}
+	else
+	{
+		spprintf( &errMessage, 0, "%s", uErrorName );
+	}
+
+	return errMessage;
+}
+/* }}} */
+
+/* {{{ void intl_error_set_code( intl_error* err, UErrorCode err_code )
+ * Set last error code.
+ */
+void intl_error_set_code( intl_error* err, UErrorCode err_code TSRMLS_DC )
+{
+	if( !err && !( err = intl_g_error_get( TSRMLS_C ) ) )
+		return;
+
+	err->code = err_code;
+}
+/* }}} */
+
+/* {{{ void intl_error_get_code( intl_error* err )
+ * Return last error code.
+ */
+UErrorCode intl_error_get_code( intl_error* err TSRMLS_DC )
+{
+	if( !err && !( err = intl_g_error_get( TSRMLS_C ) ) )
+		return U_ZERO_ERROR;
+
+	return err->code;
+}
+/* }}} */
+
+/* {{{ void intl_error_set( intl_error* err, UErrorCode code, char* msg, int copyMsg )
+ * Set error code and message.
+ */
+void intl_error_set( intl_error* err, UErrorCode code, char* msg, int copyMsg TSRMLS_DC )
+{
+	intl_error_set_code( err, code TSRMLS_CC );
+	intl_error_set_custom_msg( err, msg, copyMsg TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ void intl_errors_reset( intl_error* err )
+ */
+void intl_errors_reset( intl_error* err TSRMLS_DC )
+{
+	intl_error_reset( err TSRMLS_CC );
+	intl_error_reset( NULL TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ void intl_errors_set_custom_msg( intl_error* err, char* msg, int copyMsg )
+ */
+void intl_errors_set_custom_msg( intl_error* err, char* msg, int copyMsg TSRMLS_DC )
+{
+	intl_error_set_custom_msg( err, msg, copyMsg TSRMLS_CC );
+	intl_error_set_custom_msg( NULL, msg, copyMsg TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ intl_errors_set_code( intl_error* err, UErrorCode err_code )
+ */
+void intl_errors_set_code( intl_error* err, UErrorCode err_code TSRMLS_DC )
+{
+	intl_error_set_code( err, err_code TSRMLS_CC );
+	intl_error_set_code( NULL, err_code TSRMLS_CC );
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/intl_error.h

@@ -0,0 +1,46 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   |          Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef INTL_ERROR_H
+#define INTL_ERROR_H
+
+#include <unicode/utypes.h>
+
+#define INTL_ERROR_CODE(e) (e).code
+
+typedef struct _intl_error {
+	UErrorCode      code;
+	char*           custom_error_message;
+	int             free_custom_error_message;
+} intl_error;
+
+intl_error* intl_error_create( TSRMLS_D );
+void        intl_error_init( intl_error* err TSRMLS_DC );
+void        intl_error_reset( intl_error* err TSRMLS_DC );
+void        intl_error_set_code( intl_error* err, UErrorCode err_code TSRMLS_DC );
+void        intl_error_set_custom_msg( intl_error* err, char* msg, int copyMsg TSRMLS_DC );
+void        intl_error_set( intl_error* err, UErrorCode code, char* msg, int copyMsg TSRMLS_DC );
+UErrorCode  intl_error_get_code( intl_error* err TSRMLS_DC );
+char*       intl_error_get_message( intl_error* err TSRMLS_DC );
+
+// Wrappers to synchonize object's and global error structures.
+void        intl_errors_reset( intl_error* err TSRMLS_DC );
+void        intl_errors_set_custom_msg( intl_error* err, char* msg, int copyMsg TSRMLS_DC );
+void        intl_errors_set_code( intl_error* err, UErrorCode err_code TSRMLS_DC );
+
+#endif // INTL_ERROR_H

ext/intl/locale/locale.c

@@ -0,0 +1,70 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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.               |
+   +----------------------------------------------------------------------+
+   | Author: Kirti Velankar <kirtig@yahoo-inc.com>                  	  |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "locale_class.h"
+#include "locale.h"
+
+#include <unicode/utypes.h>
+#include <unicode/uloc.h>
+#include <unicode/ustring.h>
+
+/* {{{ locale_register_constants
+ * Register constants common for the both (OO and procedural)
+ * APIs.
+ */
+void locale_register_constants( INIT_FUNC_ARGS )
+{
+	if( !Locale_ce_ptr )
+	{
+		zend_error( E_ERROR, "Locale class not defined" );
+		return;
+	}
+
+	#define LOCALE_EXPOSE_CONST(x) REGISTER_LONG_CONSTANT(#x, x, CONST_CS)
+	#define LOCALE_EXPOSE_CLASS_CONST(x) zend_declare_class_constant_long( Locale_ce_ptr, ZEND_STRS( #x ) - 1, ULOC_##x TSRMLS_CC );
+	#define LOCALE_EXPOSE_CUSTOM_CLASS_CONST_STR(name, value) zend_declare_class_constant_string( Locale_ce_ptr, ZEND_STRS( name ) - 1, value TSRMLS_CC );
+
+	LOCALE_EXPOSE_CLASS_CONST( ACTUAL_LOCALE );
+	LOCALE_EXPOSE_CLASS_CONST( VALID_LOCALE );
+
+	zend_declare_class_constant_null(Locale_ce_ptr, ZEND_STRS("DEFAULT_LOCALE") - 1 TSRMLS_CC);
+
+	LOCALE_EXPOSE_CUSTOM_CLASS_CONST_STR( "LANG_TAG", LOC_LANG_TAG);
+	LOCALE_EXPOSE_CUSTOM_CLASS_CONST_STR( "EXTLANG_TAG", LOC_EXTLANG_TAG);
+	LOCALE_EXPOSE_CUSTOM_CLASS_CONST_STR( "SCRIPT_TAG", LOC_SCRIPT_TAG);
+	LOCALE_EXPOSE_CUSTOM_CLASS_CONST_STR( "REGION_TAG", LOC_REGION_TAG);
+	LOCALE_EXPOSE_CUSTOM_CLASS_CONST_STR( "VARIANT_TAG",LOC_VARIANT_TAG);
+	LOCALE_EXPOSE_CUSTOM_CLASS_CONST_STR( "GRANDFATHERED_LANG_TAG",LOC_GRANDFATHERED_LANG_TAG);
+	LOCALE_EXPOSE_CUSTOM_CLASS_CONST_STR( "PRIVATE_TAG",LOC_PRIVATE_TAG);
+
+	#undef LOCALE_EXPOSE_CUSTOM_CLASS_CONST_STR
+	#undef LOCALE_EXPOSE_CLASS_CONST
+	#undef LOCALE_EXPOSE_CONST
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/locale/locale.h

@@ -0,0 +1,36 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Kirti Velankar <kirtig@yahoo-inc.com>                 	  |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef LOCALE_LOCALE_H
+#define LOCALE_LOCALE_H
+
+#include <php.h>
+
+void locale_register_constants( INIT_FUNC_ARGS );
+ 
+#define OPTION_DEFAULT NULL
+#define LOC_LANG_TAG "language"
+#define LOC_SCRIPT_TAG "script"
+#define LOC_REGION_TAG "region"
+#define LOC_VARIANT_TAG "variant"
+#define LOC_EXTLANG_TAG "extlang"
+#define LOC_GRANDFATHERED_LANG_TAG "grandfathered"
+#define LOC_PRIVATE_TAG "private"
+#define LOC_CANONICALIZE_TAG "canonicalize"
+
+#define LOCALE_INI_NAME "intl.default_locale"
+
+#endif // LOCALE_LOCALE_H

ext/intl/locale/locale_class.c

@@ -0,0 +1,126 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Kirti Velankar <kirtig@yahoo-inc.com>                  	  |
+   +----------------------------------------------------------------------+
+ */
+
+#include <unicode/uloc.h>
+
+#include "php_intl.h"
+
+#include "intl_error.h"
+#include "locale_class.h"
+#include "locale_methods.h"
+#include "locale.h"
+
+zend_class_entry *Locale_ce_ptr = NULL;
+
+
+/////////////////////////////////////////////////////////////////////////////
+// 'Locale' class registration structures & functions
+/////////////////////////////////////////////////////////////////////////////
+
+/* {{{ Locale methods arguments info */
+// NOTE: modifying 'locale_XX_args' do not forget to
+//       modify approptiate 'locale_XX_args' for
+//       the procedural API.
+
+static
+ZEND_BEGIN_ARG_INFO_EX( locale_0_args, 0, 0, 0 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( locale_1_arg, 0, 0, 1 )
+	ZEND_ARG_INFO( 0, arg1 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( locale_2_args, 0, 0, 2 )
+	ZEND_ARG_INFO( 0, arg1 )
+	ZEND_ARG_INFO( 0, arg2 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( locale_3_args, 0, 0, 3 )
+        ZEND_ARG_INFO( 0, arg1 )
+        ZEND_ARG_INFO( 0, arg2 )
+        ZEND_ARG_INFO( 0, arg3 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( locale_4_args, 0, 0, 4 )
+        ZEND_ARG_INFO( 0, arg1 )
+        ZEND_ARG_INFO( 0, arg2 )
+        ZEND_ARG_INFO( 0, arg3 )
+        ZEND_ARG_INFO( 0, arg4 )
+ZEND_END_ARG_INFO()
+
+/* }}} */
+
+/* {{{ Locale_class_functions
+ * Every 'Locale' class method has an entry in this table
+ */
+
+function_entry Locale_class_functions[] = {
+	ZEND_FENTRY( getDefault, zif_locale_get_default , locale_0_args , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC  )
+	ZEND_FENTRY( setDefault, zif_locale_set_default , locale_1_arg , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( getPrimaryLanguage, ZEND_FN( locale_get_primary_language ), locale_1_arg , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( getScript, ZEND_FN( locale_get_script ), locale_1_arg , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( getRegion, ZEND_FN( locale_get_region ), locale_1_arg , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( getKeywords, ZEND_FN( locale_get_keywords ), locale_1_arg , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( getDisplayScript, ZEND_FN( locale_get_display_script ), locale_2_args , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( getDisplayRegion, ZEND_FN( locale_get_display_region ), locale_2_args , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( getDisplayName, ZEND_FN( locale_get_display_name ), locale_2_args , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( getDisplayLanguage, ZEND_FN( locale_get_display_language ), locale_2_args , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( getDisplayVariant, ZEND_FN( locale_get_display_variant ), locale_2_args , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( composeLocale, ZEND_FN( locale_compose ), locale_1_arg , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( parseLocale, ZEND_FN( locale_parse ), locale_1_arg , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( getAllVariants, ZEND_FN( locale_get_all_variants ), locale_1_arg , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( filterMatches, ZEND_FN( locale_filter_matches ), locale_3_args, ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( lookup, ZEND_FN( locale_lookup ), locale_4_args, ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( canonicalize, ZEND_FN( locale_canonicalize ), locale_1_arg , ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	{ NULL, NULL, NULL }
+};
+/* }}} */
+
+/* {{{ locale_register_Locale_class
+ * Initialize 'Locale' class
+ */
+void locale_register_Locale_class( TSRMLS_D )
+{
+	zend_class_entry ce;
+
+	// Create and register 'Locale' class.
+	INIT_CLASS_ENTRY( ce, "Locale", Locale_class_functions );
+	ce.create_object = NULL;
+	Locale_ce_ptr = zend_register_internal_class( &ce TSRMLS_CC );
+
+	// Declare 'Locale' class properties.
+	if( !Locale_ce_ptr )
+	{
+		zend_error( E_ERROR,
+			"Locale: Failed to register Locale class.");
+		return;
+	}
+}
+/* }}} */
+
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/locale/locale_class.h

@@ -0,0 +1,46 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Kirti Velankar <kirtig@yahoo-inc.com>                  	  |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef LOCALE_CLASS_H
+#define LOCALE_CLASS_H
+
+#include <php.h>
+
+#include "intl_common.h"
+#include "intl_error.h"
+
+#include <unicode/uloc.h>
+
+typedef struct {
+	zend_object     zo;
+
+	// ICU locale
+	char*      		uloc1;
+
+} Locale_object;
+
+
+void locale_register_Locale_class( TSRMLS_D );
+
+extern zend_class_entry *Locale_ce_ptr;
+
+/* Auxiliary macros */
+
+#define LOCALE_METHOD_INIT_VARS       \
+    zval*             	object  = NULL;   \
+    intl_error_reset( NULL TSRMLS_CC ); \
+
+#endif // #ifndef LOCALE_CLASS_H

ext/intl/locale/locale_methods.h

@@ -0,0 +1,47 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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.               |
+   +----------------------------------------------------------------------+
+   | Author: Kirti Velankar <kirtig@yahoo-inc.com>                        |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef LOCALE_METHODS_H
+#define LOCALE_METHODS_H
+
+#include <php.h>
+
+
+PHP_FUNCTION( locale_get_primary_language );
+PHP_FUNCTION( locale_get_script );
+PHP_FUNCTION( locale_get_region );
+PHP_FUNCTION( locale_get_all_variants);
+
+PHP_NAMED_FUNCTION( zif_locale_get_default );
+PHP_NAMED_FUNCTION( zif_locale_set_default );
+
+PHP_FUNCTION( locale_get_display_name );
+PHP_FUNCTION( locale_get_display_language );
+PHP_FUNCTION( locale_get_display_script );
+PHP_FUNCTION( locale_get_display_region );
+PHP_FUNCTION( locale_get_display_variant );
+
+PHP_FUNCTION( locale_get_keywords );
+PHP_FUNCTION( locale_canonicalize);
+
+PHP_FUNCTION( locale_compose);
+PHP_FUNCTION( locale_parse);
+
+PHP_FUNCTION( locale_filter_matches);
+PHP_FUNCTION( locale_lookup);
+PHP_FUNCTION( locale_canonicalize);
+
+#endif // LOCALE_METHODS_H

ext/intl/msgformat/msgformat.c

@@ -0,0 +1,256 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <unicode/ustring.h>
+#include <unicode/umsg.h>
+
+#include "php_intl.h"
+#include "msgformat_class.h"
+#include "intl_convert.h"
+
+/* {{{ proto MessageFormatter MesssageFormatter::create( string $locale, string $pattern )
+ * Create formatter. }}} */
+/* {{{ proto MessageFormatter msgfmt_create( string $locale, string $pattern )
+ * Create formatter.
+ */
+PHP_FUNCTION( msgfmt_create )
+{
+	char*       locale;
+	char*       pattern;
+	int         locale_len = 0, pattern_len = 0;
+	UChar*      spattern     = NULL;
+	int         spattern_len = 0;
+	zval*       object;
+	MessageFormatter_object* mfo;
+
+	intl_error_reset( NULL TSRMLS_CC );
+
+	// Parse parameters.
+	if( zend_parse_parameters( ZEND_NUM_ARGS() TSRMLS_CC, "ss",
+		&locale, &locale_len, &pattern, &pattern_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"msgfmt_create: unable to parse input parameters", 0 TSRMLS_CC );
+
+		RETURN_NULL();
+	}
+
+	INTL_CHECK_LOCALE_LEN(locale_len);
+	// Create a MessageFormatter object and save the ICU formatter into it.
+	if( ( object = getThis() ) == NULL )
+		object = return_value;
+
+	if( Z_TYPE_P( object ) != IS_OBJECT )
+		object_init_ex( object, MessageFormatter_ce_ptr );
+
+	MSG_FORMAT_METHOD_FETCH_OBJECT;
+
+	// Convert pattern (if specified) to UTF-16.
+	if(pattern && pattern_len) {
+		intl_convert_utf8_to_utf16(&spattern, &spattern_len, pattern, pattern_len, &INTL_DATA_ERROR_CODE(mfo));
+		if( U_FAILURE( INTL_DATA_ERROR_CODE((mfo)) ) )
+		{
+			intl_error_set( NULL, INTL_DATA_ERROR_CODE( mfo ),
+				"msgfmt_create: error converting pattern to UTF-16", 0 TSRMLS_CC );
+			zval_dtor(return_value);
+			RETURN_NULL();
+		}
+	} else {
+		spattern_len = 0;
+		spattern = NULL;
+	}
+
+	if(locale_len == 0) {
+		locale = INTL_G(default_locale);
+	}
+
+	if(msfgotmat_fix_quotes(&spattern, &spattern_len, &INTL_DATA_ERROR_CODE(mfo)) != SUCCESS) {
+		intl_error_set( NULL, U_INVALID_FORMAT_ERROR,
+			"msgfmt_create: error converting pattern to quote-friendly format", 0 TSRMLS_CC );
+		zval_dtor(return_value);
+		RETURN_NULL();
+	}
+
+	(mfo)->mf_data.orig_format = estrndup(pattern, pattern_len);
+	(mfo)->mf_data.orig_format_len = pattern_len;
+	
+	// Create an ICU message formatter.
+	MSG_FORMAT_OBJECT(mfo) = umsg_open(spattern, spattern_len, locale, NULL, &INTL_DATA_ERROR_CODE(mfo));
+
+	if(spattern) {
+		efree(spattern);
+	}
+
+	if( U_FAILURE( INTL_DATA_ERROR_CODE((mfo)) ) )
+	{
+		intl_error_set( NULL, INTL_DATA_ERROR_CODE( mfo ),
+			"msgfmt_create: message formatter creation failed", 0 TSRMLS_CC );
+		zval_dtor(return_value);
+		RETURN_NULL();
+	}
+}
+/* }}} */
+
+/* {{{ proto void MessageFormatter::__construct( string $locale, string $pattern )
+ * MessageFormatter object constructor.
+ */
+PHP_METHOD( MessageFormatter, __construct )
+{
+	char*       locale;
+	char*       pattern;
+	int         locale_len, pattern_len = 0;
+	UChar*      spattern     = NULL;
+	int         spattern_len = 0;
+	zval*       object;
+	MessageFormatter_object* mfo;
+
+	intl_error_reset( NULL TSRMLS_CC );
+
+	object = getThis();
+
+	// Parse parameters.
+	if( zend_parse_parameters( ZEND_NUM_ARGS() TSRMLS_CC, "ss",
+		&locale, &locale_len, &pattern, &pattern_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"__construct: unable to parse input params", 0 TSRMLS_CC );
+		zval_dtor(object);
+		ZVAL_NULL(object);
+		RETURN_NULL();
+	}
+
+	INTL_CHECK_LOCALE_LEN_OBJ(locale_len, object);
+	mfo = (MessageFormatter_object *) zend_object_store_get_object( object TSRMLS_CC );
+
+	intl_error_reset( &mfo->mf_data.error TSRMLS_CC );
+
+	// Convert pattern to UTF-16.
+	if(pattern && pattern_len) {
+		intl_convert_utf8_to_utf16(&spattern, &spattern_len, pattern, pattern_len, &INTL_DATA_ERROR_CODE(mfo));
+		if( U_FAILURE( INTL_DATA_ERROR_CODE((mfo)) ) )
+		{
+			intl_error_set( NULL, INTL_DATA_ERROR_CODE( mfo ),
+				"__construct: Error converting pattern to UTF-16", 0 TSRMLS_CC );
+			zval_dtor(object);
+			ZVAL_NULL(object);
+			RETURN_NULL();
+		}
+	} else {
+		spattern_len = 0;
+		spattern = NULL;
+	}
+
+	if(locale_len == 0) {
+		locale = INTL_G(default_locale);
+	}
+
+	if(msfgotmat_fix_quotes(&spattern, &spattern_len, &INTL_DATA_ERROR_CODE(mfo)) != SUCCESS) {
+		intl_error_set( NULL, U_INVALID_FORMAT_ERROR,
+			"__construct: error converting pattern to quote-friendly format", 0 TSRMLS_CC );
+		zval_dtor(object);
+		ZVAL_NULL(object);
+		RETURN_NULL();
+	}
+
+	(mfo)->mf_data.orig_format = estrndup(pattern, pattern_len);
+	(mfo)->mf_data.orig_format_len = pattern_len;
+	
+	// Create an ICU message formatter.
+	MSG_FORMAT_OBJECT(mfo) = umsg_open(spattern, spattern_len, locale, NULL, &INTL_DATA_ERROR_CODE(mfo));
+
+	if(spattern && spattern_len) {
+		efree(spattern);
+	}
+
+	if( U_FAILURE( INTL_DATA_ERROR_CODE((mfo)) ) )
+	{
+		intl_error_set( NULL, INTL_DATA_ERROR_CODE(mfo),
+			"__construct: message formatter creation failed", 0 TSRMLS_CC );
+		zval_dtor(object);
+		ZVAL_NULL(object);
+		RETURN_NULL();
+	}
+}
+/* }}} */
+
+/* {{{ proto int MessageFormatter::getErrorCode()
+ * Get formatter's last error code. }}} */
+/* {{{ proto int msgfmt_get_error_code( MessageFormatter $nf )
+ * Get formatter's last error code.
+ */
+PHP_FUNCTION( msgfmt_get_error_code )
+{
+	zval*                    object  = NULL;
+	MessageFormatter_object*  mfo     = NULL;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O",
+		&object, MessageFormatter_ce_ptr ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"msgfmt_get_error_code: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	mfo = (MessageFormatter_object *) zend_object_store_get_object( object TSRMLS_CC );
+
+	// Return formatter's last error code.
+	RETURN_LONG( INTL_DATA_ERROR_CODE(mfo) );
+}
+/* }}} */
+
+/* {{{ proto string MessageFormatter::getErrorMessage( )
+ * Get text description for formatter's last error code. }}} */
+/* {{{ proto string msgfmt_get_error_message( MessageFormatter $coll )
+ * Get text description for formatter's last error code.
+ */
+PHP_FUNCTION( msgfmt_get_error_message )
+{
+	char*                    message = NULL;
+	zval*                    object  = NULL;
+	MessageFormatter_object*  mfo     = NULL;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O",
+		&object, MessageFormatter_ce_ptr ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"msgfmt_get_error_message: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	mfo = (MessageFormatter_object *) zend_object_store_get_object( object TSRMLS_CC );
+
+	// Return last error message.
+	message = intl_error_get_message( &mfo->mf_data.error TSRMLS_CC );
+	RETURN_STRING( message, 0);
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/msgformat/msgformat.h

@@ -0,0 +1,27 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef MSG_FORMATTER_H
+#define MSG_FORMATTER_H
+
+#include <php.h>
+
+PHP_FUNCTION( msgfmt_create );
+PHP_FUNCTION( msgfmt_get_error_code );
+PHP_FUNCTION( msgfmt_get_error_message );
+PHP_METHOD( MessageFormatter, __construct );
+
+#endif // MSG_FORMAT_H

ext/intl/msgformat/msgformat_attr.c

@@ -0,0 +1,144 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "msgformat_class.h"
+#include "msgformat_attr.h"
+#include "intl_convert.h"
+
+#include <unicode/ustring.h>
+
+
+/* {{{ proto string MessageFormatter::getPattern( )
+ * Get formatter pattern. }}} */
+/* {{{ proto string msgfmt_get_pattern( MessageFormatter $mf )
+ * Get formatter pattern.
+ */
+PHP_FUNCTION( msgfmt_get_pattern )
+{
+	UChar  value_buf[64];
+	int    length = USIZE( value_buf );
+	UChar* value  = value_buf;
+	MSG_FORMAT_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O", &object, MessageFormatter_ce_ptr ) == FAILURE )
+	{
+		intl_error_set( INTL_DATA_ERROR_P(mfo), U_ILLEGAL_ARGUMENT_ERROR,	
+			"msgfmt_get_pattern: unable to parse input params", 0 TSRMLS_CC );
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	MSG_FORMAT_METHOD_FETCH_OBJECT;
+
+	if(mfo->mf_data.orig_format) {
+		RETURN_STRINGL(mfo->mf_data.orig_format, mfo->mf_data.orig_format_len, 1);
+	}
+
+	RETURN_FALSE;
+}
+/* }}} */
+
+/* {{{ proto bool MessageFormatter::setPattern( string $pattern )
+ * Set formatter pattern. }}} */
+/* {{{ proto bool msgfmt_set_pattern( MessageFormatter $mf, string $pattern )
+ * Set formatter pattern.
+ */
+PHP_FUNCTION( msgfmt_set_pattern )
+{
+	char*       value = NULL;
+	int         value_len = 0;
+	int         spattern_len = 0;
+	UChar*	    spattern  = NULL;
+	MSG_FORMAT_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Os",
+		&object, MessageFormatter_ce_ptr, &value, &value_len ) == FAILURE )
+	{
+		intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,	
+			"msgfmt_set_pattern: unable to parse input params", 0 TSRMLS_CC);
+		RETURN_FALSE;
+	}
+
+	MSG_FORMAT_METHOD_FETCH_OBJECT;
+
+	// Convert given pattern to UTF-16.
+	intl_convert_utf8_to_utf16(&spattern, &spattern_len, value, value_len, &INTL_DATA_ERROR_CODE(mfo));
+	INTL_METHOD_CHECK_STATUS(mfo, "Error converting pattern to UTF-16" );
+
+	if(msfgotmat_fix_quotes(&spattern, &spattern_len, &INTL_DATA_ERROR_CODE(mfo)) != SUCCESS) {
+		intl_error_set( NULL, U_INVALID_FORMAT_ERROR,
+			"msgfmt_set_pattern: error converting pattern to quote-friendly format", 0 TSRMLS_CC );
+		RETURN_FALSE;
+	}
+
+	// TODO: add parse error information
+	umsg_applyPattern(MSG_FORMAT_OBJECT(mfo), spattern, spattern_len, NULL, &INTL_DATA_ERROR_CODE(mfo));
+	efree(spattern);
+	INTL_METHOD_CHECK_STATUS(mfo, "Error setting symbol value");
+
+	if(mfo->mf_data.orig_format) {
+		efree(mfo->mf_data.orig_format);
+	}
+	mfo->mf_data.orig_format = estrndup(value, value_len);
+	mfo->mf_data.orig_format_len = value_len;
+
+	RETURN_TRUE;
+}
+/* }}} */
+
+/* {{{ proto string MessageFormatter::getLocale()
+ * Get formatter locale. }}} */
+/* {{{ proto string msgfmt_get_locale(MessageFormatter $mf)
+ * Get formatter locale.
+ */
+PHP_FUNCTION( msgfmt_get_locale )
+{
+	char *loc;
+	MSG_FORMAT_METHOD_INIT_VARS;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "O",
+		&object, MessageFormatter_ce_ptr ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"msgfmt_get_locale: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	MSG_FORMAT_METHOD_FETCH_OBJECT;
+
+	loc = (char *)umsg_getLocale(MSG_FORMAT_OBJECT(mfo));
+	RETURN_STRING(loc, 1);
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/msgformat/msgformat_attr.h

@@ -0,0 +1,26 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef MSG_FORMAT_ATTR_H
+#define MSG_FORMAT_ATTR_H
+
+#include <php.h>
+
+PHP_FUNCTION( msgfmt_set_pattern );
+PHP_FUNCTION( msgfmt_get_pattern );
+PHP_FUNCTION( msgfmt_get_locale );
+
+#endif // MSG_FORMAT_ATTR_H

ext/intl/msgformat/msgformat_class.c

@@ -0,0 +1,127 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#include <unicode/unum.h>
+
+#include "msgformat_class.h"
+#include "php_intl.h"
+#include "msgformat_data.h"
+#include "msgformat_format.h"
+#include "msgformat_parse.h"
+#include "msgformat.h"
+#include "msgformat_attr.h"
+
+zend_class_entry *MessageFormatter_ce_ptr = NULL;
+
+/////////////////////////////////////////////////////////////////////////////
+// Auxiliary functions needed by objects of 'MessageFormatter' class
+/////////////////////////////////////////////////////////////////////////////
+
+/* {{{ MessageFormatter_objects_dtor */
+static void MessageFormatter_object_dtor(void *object, zend_object_handle handle TSRMLS_DC )
+{
+	zend_objects_destroy_object( object, handle TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ MessageFormatter_objects_free */
+void MessageFormatter_object_free( zend_object *object TSRMLS_DC )
+{
+	MessageFormatter_object* mfo = (MessageFormatter_object*)object;
+
+	zend_object_std_dtor( &mfo->zo TSRMLS_CC );
+
+	msgformat_data_free( &mfo->mf_data TSRMLS_CC );
+
+	efree( mfo );
+}
+/* }}} */
+
+/* {{{ MessageFormatter_object_create */
+zend_object_value MessageFormatter_object_create(zend_class_entry *ce TSRMLS_DC)
+{
+	zend_object_value    retval;
+	MessageFormatter_object*     intern;
+
+	intern = ecalloc( 1, sizeof(MessageFormatter_object) );
+	msgformat_data_init( &intern->mf_data TSRMLS_CC );
+	zend_object_std_init( &intern->zo, ce TSRMLS_CC );
+
+	retval.handle = zend_objects_store_put(
+		intern,
+		MessageFormatter_object_dtor,
+		(zend_objects_free_object_storage_t)MessageFormatter_object_free,
+		NULL TSRMLS_CC );
+
+	retval.handlers = zend_get_std_object_handlers();
+
+	return retval;
+}
+/* }}} */
+
+/////////////////////////////////////////////////////////////////////////////
+// 'MessageFormatter' class registration structures & functions
+/////////////////////////////////////////////////////////////////////////////
+
+/* {{{ MessageFormatter_class_functions
+ * Every 'MessageFormatter' class method has an entry in this table
+ */
+
+static function_entry MessageFormatter_class_functions[] = {
+	PHP_ME( MessageFormatter, __construct, NULL, ZEND_ACC_PUBLIC|ZEND_ACC_CTOR )
+	ZEND_FENTRY(  create, ZEND_FN( msgfmt_create ), NULL, ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	PHP_NAMED_FE( format, ZEND_FN( msgfmt_format ), NULL )
+	ZEND_FENTRY(  formatMessage, ZEND_FN( msgfmt_format_message ), NULL, ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	PHP_NAMED_FE( parse, ZEND_FN( msgfmt_parse ), NULL )
+	ZEND_FENTRY(  parseMessage, ZEND_FN( msgfmt_parse_message ), NULL, ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	PHP_NAMED_FE( setPattern, ZEND_FN( msgfmt_set_pattern ), NULL )
+	PHP_NAMED_FE( getPattern, ZEND_FN( msgfmt_get_pattern ), NULL )
+	PHP_NAMED_FE( getLocale, ZEND_FN( msgfmt_get_locale ), NULL )
+	PHP_NAMED_FE( getErrorCode, ZEND_FN( msgfmt_get_error_code ), NULL )
+	PHP_NAMED_FE( getErrorMessage, ZEND_FN( msgfmt_get_error_message ), NULL )
+	{ NULL, NULL, NULL }
+};
+/* }}} */
+
+/* {{{ msgformat_register_class
+ * Initialize 'MessageFormatter' class
+ */
+void msgformat_register_class( TSRMLS_D )
+{
+	zend_class_entry ce;
+
+	// Create and register 'MessageFormatter' class.
+	INIT_CLASS_ENTRY( ce, "MessageFormatter", MessageFormatter_class_functions );
+	ce.create_object = MessageFormatter_object_create;
+	MessageFormatter_ce_ptr = zend_register_internal_class( &ce TSRMLS_CC );
+
+	// Declare 'MessageFormatter' class properties.
+	if( !MessageFormatter_ce_ptr )
+	{
+		zend_error(E_ERROR, "Failed to register MessageFormatter class");
+		return;
+	}
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/msgformat/msgformat_class.h

@@ -0,0 +1,41 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef MSG_FORMAT_CLASS_H
+#define MSG_FORMAT_CLASS_H
+
+#include <php.h>
+
+#include "intl_common.h"
+#include "intl_error.h"
+#include "intl_data.h"
+#include "msgformat_data.h"
+
+typedef struct {
+	zend_object     zo;
+	msgformat_data  mf_data;
+} MessageFormatter_object;
+
+void msgformat_register_class( TSRMLS_D );
+extern zend_class_entry *MessageFormatter_ce_ptr;
+
+/* Auxiliary macros */
+
+#define MSG_FORMAT_METHOD_INIT_VARS		INTL_METHOD_INIT_VARS(MessageFormatter, mfo)
+#define MSG_FORMAT_METHOD_FETCH_OBJECT	INTL_METHOD_FETCH_OBJECT(MessageFormatter, mfo)
+#define MSG_FORMAT_OBJECT(mfo)			(mfo)->mf_data.umsgf
+
+#endif // #ifndef MSG_FORMAT_CLASS_H

ext/intl/msgformat/msgformat_data.c

@@ -0,0 +1,98 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <unicode/ustring.h>
+#include "msgformat_data.h"
+
+/* {{{ void msgformat_data_init( msgformat_data* mf_data )
+ * Initialize internals of msgformat_data.
+ */
+void msgformat_data_init( msgformat_data* mf_data TSRMLS_DC )
+{
+	if( !mf_data )
+		return;
+
+	mf_data->umsgf = NULL;
+	mf_data->orig_format = NULL;
+	intl_error_reset( &mf_data->error TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ void msgformat_data_free( msgformat_data* mf_data )
+ * Clean up memory allocated for msgformat_data
+ */
+void msgformat_data_free( msgformat_data* mf_data TSRMLS_DC )
+{
+	if( !mf_data )
+		return;
+
+	if( mf_data->umsgf )
+		umsg_close( mf_data->umsgf );
+
+	if(mf_data->orig_format) {
+		efree(mf_data->orig_format);
+		mf_data->orig_format = NULL;
+	}
+
+	mf_data->umsgf = NULL;
+	intl_error_reset( &mf_data->error TSRMLS_CC );
+}
+/* }}} */
+
+/* {{{ msgformat_data* msgformat_data_create()
+ * Allocate memory for msgformat_data and initialize it with default values.
+ */
+msgformat_data* msgformat_data_create( TSRMLS_D )
+{
+	msgformat_data* mf_data = ecalloc( 1, sizeof(msgformat_data) );
+
+	msgformat_data_init( mf_data TSRMLS_CC );
+
+	return mf_data;
+}
+/* }}} */
+
+int msfgotmat_fix_quotes(UChar **spattern, uint32_t *spattern_len, UErrorCode *ec) 
+{
+	if(*spattern && *spattern_len && u_strchr(*spattern, (UChar)'\'')) {
+		UChar *npattern = emalloc(sizeof(UChar)*(2*(*spattern_len)+1));
+		uint32_t npattern_len;
+		npattern_len = umsg_autoQuoteApostrophe(*spattern, *spattern_len, npattern, 2*(*spattern_len)+1, ec);
+		efree(*spattern);
+		if( U_FAILURE(*ec) )
+		{
+			return FAILURE;
+		}
+		npattern = erealloc(npattern, sizeof(UChar)*(npattern_len+1));
+		*spattern = npattern;
+		*spattern_len = npattern_len;
+	}
+	return SUCCESS;
+}
+
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/msgformat/msgformat_data.h

@@ -0,0 +1,41 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef MSG_FORMAT_DATA_H
+#define MSG_FORMAT_DATA_H
+
+#include <php.h>
+
+#include <unicode/umsg.h>
+
+#include "intl_error.h"
+
+typedef struct {
+	// error hangling
+	intl_error      error;
+
+	// formatter handling
+	UMessageFormat* umsgf;
+	char*			orig_format;
+	ulong			orig_format_len;
+} msgformat_data;
+
+msgformat_data* msgformat_data_create( TSRMLS_D );
+void msgformat_data_init( msgformat_data* mf_data TSRMLS_DC );
+void msgformat_data_free( msgformat_data* mf_data TSRMLS_DC );
+int msfgotmat_fix_quotes(UChar **spattern, uint32_t *spattern_len, UErrorCode *ec);
+
+#endif // MSG_FORMAT_DATA_H

ext/intl/msgformat/msgformat_format.c

@@ -0,0 +1,184 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <unicode/ustring.h>
+
+#include "php_intl.h"
+#include "msgformat_class.h"
+#include "msgformat_format.h"
+#include "msgformat_data.h"
+#include "msgformat_helpers.h"
+#include "intl_convert.h"
+
+#ifndef Z_ADDREF_P
+#define Z_ADDREF_P(z) ((z)->refcount++)
+#endif
+
+/* {{{ */
+static void msgfmt_do_format(MessageFormatter_object *mfo, zval *args, zval *return_value TSRMLS_DC) 
+{
+	zval **fargs;
+	int count;
+	UChar* formatted = NULL;
+	int formatted_len = 0;
+	HashPosition pos;
+	int i;
+
+	count = zend_hash_num_elements(Z_ARRVAL_P(args));
+
+	if(count < umsg_format_arg_count(MSG_FORMAT_OBJECT(mfo))) {
+		// Not enough aguments for format!
+		intl_error_set( INTL_DATA_ERROR_P(mfo), U_ILLEGAL_ARGUMENT_ERROR,
+			"msgfmt_format: not enough parameters", 0 TSRMLS_CC );
+		RETVAL_FALSE;
+		return;
+	}
+
+	fargs = safe_emalloc(count, sizeof(zval *), 0);
+
+	zend_hash_internal_pointer_reset_ex(Z_ARRVAL_P(args), &pos);
+	for(i=0;i<count;i++) {
+		zval **val;
+		zend_hash_get_current_data_ex(Z_ARRVAL_P(args), (void **)&val, &pos);
+		fargs[i] = *val;
+		Z_ADDREF_P(fargs[i]);
+		/* TODO: needs refcount increase here? */
+		zend_hash_move_forward_ex(Z_ARRVAL_P(args), &pos);
+	}
+
+	umsg_format_helper(MSG_FORMAT_OBJECT(mfo), count, fargs, &formatted, &formatted_len, &INTL_DATA_ERROR_CODE(mfo) TSRMLS_CC);
+
+	for(i=0;i<count;i++) {
+		zval_ptr_dtor(&fargs[i]);
+	}
+
+	efree(fargs);
+
+	if (formatted && U_FAILURE( INTL_DATA_ERROR_CODE(mfo) ) ) {
+			efree(formatted);
+	}
+
+	INTL_METHOD_CHECK_STATUS( mfo, "Number formatting failed" );
+	INTL_METHOD_RETVAL_UTF8( mfo, formatted, formatted_len, 1 );
+}
+/* }}} */
+
+/* {{{ proto mixed MessageFormatter::format( array $args )
+ * Format a message. }}} */
+/* {{{ proto mixed msgfmt_format( MessageFormatter $nf, array $args )
+ * Format a message.
+ */
+PHP_FUNCTION( msgfmt_format )
+{
+	zval *args;
+	MSG_FORMAT_METHOD_INIT_VARS;
+
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Oa",
+		&object, MessageFormatter_ce_ptr,  &args ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"msgfmt_format: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	MSG_FORMAT_METHOD_FETCH_OBJECT;
+
+	msgfmt_do_format(mfo, args, return_value TSRMLS_CC);
+}
+/* }}} */
+
+/* {{{ proto mixed MessageFormatter::formatMessage( string $locale, string $pattern, array $args )
+ * Format a message. }}} */
+/* {{{ proto mixed msgfmt_format_message( string $locale, string $pattern, array $args )
+ * Format a message.
+ */
+PHP_FUNCTION( msgfmt_format_message )
+{
+	zval       *args;
+	UChar      *spattern = NULL;
+	int         spattern_len = 0;
+	char       *pattern = NULL;
+	int         pattern_len = 0;
+	char       *slocale = NULL;
+	int         slocale_len = 0;
+	MessageFormatter_object mf = {0};
+	MessageFormatter_object *mfo = &mf;
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "ssa",
+		  &slocale, &slocale_len, &pattern, &pattern_len, &args ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"msgfmt_format_message: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	msgformat_data_init(&mfo->mf_data TSRMLS_CC);
+
+	if(pattern && pattern_len) {
+		intl_convert_utf8_to_utf16(&spattern, &spattern_len, pattern, pattern_len, &INTL_DATA_ERROR_CODE(mfo));
+		if( U_FAILURE(INTL_DATA_ERROR_CODE((mfo))) )
+		{
+			intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+				"msgfmt_format_message: error converting pattern to UTF-16", 0 TSRMLS_CC );
+			RETURN_FALSE;
+		}
+	} else {
+		spattern_len = 0;
+		spattern = NULL;
+	}
+
+	if(slocale_len == 0) {
+		slocale = INTL_G(default_locale);
+	}
+
+	if(msfgotmat_fix_quotes(&spattern, &spattern_len, &INTL_DATA_ERROR_CODE(mfo)) != SUCCESS) {
+		intl_error_set( NULL, U_INVALID_FORMAT_ERROR,
+			"msgfmt_format_message: error converting pattern to quote-friendly format", 0 TSRMLS_CC );
+		RETURN_FALSE;
+	}
+
+	// Create an ICU message formatter.
+	MSG_FORMAT_OBJECT(mfo) = umsg_open(spattern, spattern_len, slocale, NULL, &INTL_DATA_ERROR_CODE(mfo));
+	if(spattern && spattern_len) {
+		efree(spattern);
+	}
+	INTL_METHOD_CHECK_STATUS(mfo, "Creating message formatter failed");
+
+	msgfmt_do_format(mfo, args, return_value TSRMLS_CC);
+
+	// drop the temporary formatter
+	msgformat_data_free(&mfo->mf_data TSRMLS_CC);
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/msgformat/msgformat_format.h

@@ -0,0 +1,25 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef MSG_FORMAT_FORMAT_H
+#define MSG_FORMAT_FORMAT_H
+
+#include <php.h>
+
+PHP_FUNCTION( msgfmt_format );
+PHP_FUNCTION( msgfmt_format_message );
+
+#endif // MSG_FORMAT_FORMAT_H

ext/intl/msgformat/msgformat_helpers.cpp

@@ -0,0 +1,211 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <math.h>
+#include <unicode/msgfmt.h>
+
+extern "C" {
+#include "php_intl.h"
+#include "msgformat_class.h"
+#include "msgformat_format.h"
+#include "msgformat_helpers.h"
+#include "intl_convert.h"
+}
+
+U_NAMESPACE_BEGIN
+/**
+ * This class isolates our access to private internal methods of
+ * MessageFormat.  It is never instantiated; it exists only for C++
+ * access management.
+ */
+class MessageFormatAdapter {
+public:
+    static const Formattable::Type* getArgTypeList(const MessageFormat& m,
+                                                   int32_t& count);
+};
+const Formattable::Type*
+MessageFormatAdapter::getArgTypeList(const MessageFormat& m,
+                                     int32_t& count) {
+    return m.getArgTypeList(count);
+}
+U_NAMESPACE_END
+
+U_CFUNC int32_t umsg_format_arg_count(UMessageFormat *fmt) 
+{
+	int32_t fmt_count = 0;
+	MessageFormatAdapter::getArgTypeList(*(const MessageFormat*)fmt, fmt_count);
+	return fmt_count;
+}
+
+U_CFUNC void umsg_format_helper(UMessageFormat *fmt, int arg_count, zval **args, UChar **formatted, int *formatted_len, UErrorCode *status TSRMLS_DC)
+{
+	int fmt_count = 0;
+    const Formattable::Type* argTypes =
+		MessageFormatAdapter::getArgTypeList(*(const MessageFormat*)fmt, fmt_count);
+	Formattable* fargs = new Formattable[fmt_count ? fmt_count : 1];
+
+	for(int32_t i = 0; i < fmt_count; ++i) {
+        UChar  *stringVal = NULL;
+		int     stringLen = 0;
+		int64_t tInt64 = 0;
+
+		switch(argTypes[i]) {
+			case Formattable::kDate:
+				convert_to_long_ex(&args[i]);
+				fargs[i].setDate(U_MILLIS_PER_SECOND * (double)Z_LVAL_P(args[i]));
+				break;
+
+			case Formattable::kDouble:
+				convert_to_double_ex(&args[i]);
+			    fargs[i].setDouble(Z_DVAL_P(args[i]));
+				break;
+            
+	        case Formattable::kLong:
+				convert_to_long_ex(&args[i]);
+			    fargs[i].setLong(Z_LVAL_P(args[i]));
+				break;
+
+	        case Formattable::kInt64:
+				if(Z_TYPE_P(args[i]) == IS_DOUBLE) {
+					tInt64 = (int64_t)Z_DVAL_P(args[i]);
+				} else if(Z_TYPE_P(args[i]) == IS_LONG) {
+					tInt64 = (int64_t)Z_LVAL_P(args[i]);
+				} else {
+					SEPARATE_ZVAL_IF_NOT_REF(&args[i]);
+					convert_scalar_to_number( args[i] TSRMLS_CC );
+					tInt64 = (Z_TYPE_P(args[i]) == IS_DOUBLE)?(int64_t)Z_DVAL_P(args[i]):Z_LVAL_P(args[i]);
+				}
+			    fargs[i].setInt64(tInt64);
+				break;
+            
+	        case Formattable::kString:
+		        convert_to_string_ex(&args[i]);
+				intl_convert_utf8_to_utf16(&stringVal, &stringLen, Z_STRVAL_P(args[i]), Z_STRLEN_P(args[i]), status);
+				if(U_FAILURE(*status)){
+					delete[] fargs;
+					return;
+				}
+				fargs[i].setString(stringVal);
+				efree(stringVal);
+			    break;
+            
+			case Formattable::kArray:
+			case Formattable::kObject:
+				*status = U_UNSUPPORTED_ERROR;
+				delete[] fargs;
+				return;
+        }		
+	}
+
+    UnicodeString resultStr;
+    FieldPosition fieldPosition(0);
+    
+    /* format the message */
+    ((const MessageFormat*)fmt)->format(fargs, fmt_count, resultStr, fieldPosition, *status);
+
+    delete[] fargs;
+
+    if(U_FAILURE(*status)){
+        return;
+    }
+
+	*formatted_len = resultStr.length();
+	*formatted = eumalloc(*formatted_len+1);
+	resultStr.extract(*formatted, *formatted_len+1, *status);
+}
+
+#define cleanup_zvals() for(int j=i;j>=0;j--) { zval_ptr_dtor((*args)+i); }
+
+U_CFUNC void umsg_parse_helper(UMessageFormat *fmt, int *count, zval ***args, UChar *source, int source_len, UErrorCode *status)
+{
+    UnicodeString srcString(source, source_len);
+    Formattable *fargs = ((const MessageFormat*)fmt)->parse(srcString, *count, *status);
+
+	if(U_FAILURE(*status)) {
+		return;
+	}
+
+	*args = (zval **)safe_emalloc(*count, sizeof(zval *), 0);
+
+    // assign formattables to varargs
+    for(int32_t i = 0; i < *count; i++) {
+	    int64_t aInt64;
+		double aDate;
+		UnicodeString temp;
+		char *stmp;
+		int stmp_len;
+
+		ALLOC_INIT_ZVAL((*args)[i]);
+		
+		switch(fargs[i].getType()) {
+        case Formattable::kDate:
+			aDate = ((double)fargs[i].getDate())/U_MILLIS_PER_SECOND;
+			if(aDate > LONG_MAX || aDate < -LONG_MAX) {
+				ZVAL_DOUBLE((*args)[i], aDate<0?ceil(aDate):floor(aDate));
+			} else {
+				ZVAL_LONG((*args)[i], (long)aDate);
+			}
+            break;
+
+        case Formattable::kDouble:
+			ZVAL_DOUBLE((*args)[i], (double)fargs[i].getDouble());
+            break;
+
+        case Formattable::kLong:
+			ZVAL_LONG((*args)[i], fargs[i].getLong());
+            break;
+
+        case Formattable::kInt64:
+            aInt64 = fargs[i].getInt64();
+			if(aInt64 > LONG_MAX || aInt64 < -LONG_MAX) {
+				ZVAL_DOUBLE((*args)[i], (double)aInt64);
+			} else {
+				ZVAL_LONG((*args)[i], (long)aInt64);
+			}
+            break;
+
+        case Formattable::kString:
+            fargs[i].getString(temp);
+			intl_convert_utf16_to_utf8(&stmp, &stmp_len, temp.getBuffer(), temp.length(), status);
+			if(U_FAILURE(*status)) {
+				cleanup_zvals();
+				return;
+			}
+			ZVAL_STRINGL((*args)[i], stmp, stmp_len, 0);
+            break;
+
+        case Formattable::kObject:
+        case Formattable::kArray:
+            *status = U_ILLEGAL_ARGUMENT_ERROR;
+			cleanup_zvals();
+            break;
+        }
+    }
+	delete[] fargs;
+}
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/msgformat/msgformat_helpers.h

@@ -0,0 +1,25 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef MSG_FORMAT_HELPERS_H
+#define MSG_FORMAT_HELPERS_H
+
+int32_t umsg_format_arg_count(UMessageFormat *fmt); 
+void umsg_format_helper(UMessageFormat *fmt, int arg_count, zval **args,
+						UChar **formatted, int *formatted_len, UErrorCode *status TSRMLS_DC);
+void umsg_parse_helper(UMessageFormat *fmt, int *count, zval ***args,
+					   UChar *source, int source_len, UErrorCode *status);
+#endif // MSG_FORMAT_HELPERS_H

ext/intl/msgformat/msgformat_parse.c

@@ -0,0 +1,157 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <unicode/ustring.h>
+
+#include "php_intl.h"
+#include "msgformat_class.h"
+#include "msgformat_parse.h"
+#include "msgformat_data.h"
+#include "msgformat_helpers.h"
+#include "intl_convert.h"
+
+/* {{{ */
+static void msgfmt_do_parse(MessageFormatter_object *mfo, char *source, int src_len, zval *return_value TSRMLS_DC) 
+{
+	zval **fargs;
+	int count = 0;
+	int i;
+	UChar *usource = NULL;
+	int usrc_len = 0;
+
+	intl_convert_utf8_to_utf16(&usource, &usrc_len, source, src_len, &INTL_DATA_ERROR_CODE(mfo));
+	INTL_METHOD_CHECK_STATUS(mfo, "Converting parse string failed");
+
+	umsg_parse_helper(MSG_FORMAT_OBJECT(mfo), &count, &fargs, usource, usrc_len, &INTL_DATA_ERROR_CODE(mfo));
+	efree(usource);
+	INTL_METHOD_CHECK_STATUS(mfo, "Parsing failed");
+
+	array_init(return_value);
+	for(i=0;i<count;i++) {
+		add_next_index_zval(return_value, fargs[i]);
+	}
+	efree(fargs);
+}
+/* }}} */
+
+/* {{{ proto array MessageFormatter::parse( string $source )
+ * Parse a message }}} */
+/* {{{ proto array msgfmt_parse( MessageFormatter $nf, string $source )
+ * Parse a message.
+ */
+PHP_FUNCTION( msgfmt_parse )
+{
+	char *source;
+	int source_len;
+	MSG_FORMAT_METHOD_INIT_VARS;
+
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "Os",
+		&object, MessageFormatter_ce_ptr,  &source, &source_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"msgfmt_parse: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	// Fetch the object.
+	MSG_FORMAT_METHOD_FETCH_OBJECT;
+
+	msgfmt_do_parse(mfo, source, source_len, return_value TSRMLS_CC);
+}
+/* }}} */
+
+/* {{{ proto array MessageFormatter::formatMessage( string $locale, string $pattern, string $source )
+ * Parse a message. }}} */
+/* {{{ proto array numfmt_parse_message( string $locale, string $pattern, string $source )
+ * Parse a message.
+ */
+PHP_FUNCTION( msgfmt_parse_message )
+{
+	UChar      *spattern = NULL;
+	int         spattern_len = 0;
+	char       *pattern = NULL;
+	int         pattern_len = 0;
+	char       *slocale = NULL;
+	int         slocale_len = 0;
+	char       *source = NULL;
+	int         src_len = 0;
+	MessageFormatter_object mf = {0};
+	MessageFormatter_object *mfo = &mf;
+
+	// Parse parameters.
+	if( zend_parse_parameters( ZEND_NUM_ARGS() TSRMLS_CC, "sss",
+		  &slocale, &slocale_len, &pattern, &pattern_len, &source, &src_len ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+			"msgfmt_parse_message: unable to parse input params", 0 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	msgformat_data_init(&mfo->mf_data TSRMLS_CC);
+
+	if(pattern && pattern_len) {
+		intl_convert_utf8_to_utf16(&spattern, &spattern_len, pattern, pattern_len, &INTL_DATA_ERROR_CODE(mfo));
+		if( U_FAILURE(INTL_DATA_ERROR_CODE((mfo))) )
+		{
+			intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+				"msgfmt_parse_message: error converting pattern to UTF-16", 0 TSRMLS_CC );
+			RETURN_FALSE;
+		}
+	} else {
+		spattern_len = 0;
+		spattern = NULL;
+	}
+
+	if(slocale_len == 0) {
+		slocale = INTL_G(default_locale);
+	}
+
+	if(msfgotmat_fix_quotes(&spattern, &spattern_len, &INTL_DATA_ERROR_CODE(mfo)) != SUCCESS) {
+		intl_error_set( NULL, U_INVALID_FORMAT_ERROR,
+			"msgfmt_parse_message: error converting pattern to quote-friendly format", 0 TSRMLS_CC );
+		RETURN_FALSE;
+	}
+
+	// Create an ICU message formatter.
+	MSG_FORMAT_OBJECT(mfo) = umsg_open(spattern, spattern_len, slocale, NULL, &INTL_DATA_ERROR_CODE(mfo));
+	if(spattern && spattern_len) {
+		efree(spattern);
+	}
+	INTL_METHOD_CHECK_STATUS(mfo, "Creating message formatter failed");
+
+	msgfmt_do_parse(mfo, source, src_len, return_value TSRMLS_CC);
+
+	// drop the temporary formatter
+	msgformat_data_free(&mfo->mf_data TSRMLS_CC);
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/msgformat/msgformat_parse.h

@@ -0,0 +1,25 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Stanislav Malyshev <stas@zend.com>                          |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef MSG_FORMAT_PARSE_H
+#define MSG_FORMAT_PARSE_H
+
+#include <php.h>
+
+PHP_FUNCTION( msgfmt_parse );
+PHP_FUNCTION( msgfmt_parse_message );
+
+#endif // MSG_FORMAT_PARSE_H

ext/intl/normalizer/normalizer.c

@@ -0,0 +1,68 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Ed Batutis <ed@batutis.com>                                 |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "normalizer_class.h"
+#include "normalizer.h"
+
+#include <unicode/utypes.h>
+#include <unicode/unorm.h>
+#include <unicode/ustring.h>
+
+/* {{{ normalizer_register_constants
+ * Register constants common for the both (OO and procedural)
+ * APIs.
+ */
+void normalizer_register_constants( INIT_FUNC_ARGS )
+{
+	if( !Normalizer_ce_ptr )
+	{
+		zend_error( E_ERROR, "Normalizer class not defined" );
+		return;
+	}
+
+	#define NORMALIZER_EXPOSE_CONST(x) REGISTER_LONG_CONSTANT(#x, x, CONST_CS)
+	#define NORMALIZER_EXPOSE_CLASS_CONST(x) zend_declare_class_constant_long( Normalizer_ce_ptr, ZEND_STRS( #x ) - 1, NORMALIZER_##x TSRMLS_CC );
+	#define NORMALIZER_EXPOSE_CUSTOM_CLASS_CONST(name, value) zend_declare_class_constant_long( Normalizer_ce_ptr, ZEND_STRS( name ) - 1, value TSRMLS_CC );
+
+	// Normalization form constants
+	NORMALIZER_EXPOSE_CLASS_CONST( NONE );
+	NORMALIZER_EXPOSE_CLASS_CONST( FORM_D );
+	NORMALIZER_EXPOSE_CLASS_CONST( NFD );
+	NORMALIZER_EXPOSE_CLASS_CONST( FORM_KD );
+	NORMALIZER_EXPOSE_CLASS_CONST( NFKD );
+	NORMALIZER_EXPOSE_CLASS_CONST( FORM_C );
+	NORMALIZER_EXPOSE_CLASS_CONST( NFC );
+	NORMALIZER_EXPOSE_CLASS_CONST( FORM_KC );
+	NORMALIZER_EXPOSE_CLASS_CONST( NFKC );
+
+	#undef NORMALIZER_EXPOSE_CUSTOM_CLASS_CONST
+	#undef NORMALIZER_EXPOSE_CLASS_CONST
+	#undef NORMALIZER_EXPOSE_CONST
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/normalizer/normalizer.h

@@ -0,0 +1,37 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Ed Batutis <ed@batutis.com>                                 |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef NORMALIZER_NORMALIZER_H
+#define NORMALIZER_NORMALIZER_H
+
+#include <php.h>
+#include <unicode/utypes.h>
+#include <unicode/unorm.h>
+
+#define NORMALIZER_NONE UNORM_NONE
+#define NORMALIZER_FORM_D UNORM_NFD
+#define NORMALIZER_NFD UNORM_NFD
+#define NORMALIZER_FORM_KD UNORM_NFKD
+#define NORMALIZER_NFKD UNORM_NFKD
+#define NORMALIZER_FORM_C UNORM_NFC
+#define NORMALIZER_NFC UNORM_NFC
+#define NORMALIZER_FORM_KC UNORM_NFKC
+#define NORMALIZER_NFKC UNORM_NFKC
+#define NORMALIZER_DEFAULT UNORM_DEFAULT
+
+void normalizer_register_constants( INIT_FUNC_ARGS );
+
+#endif // NORMALIZER_NORMALIZER_H

ext/intl/normalizer/normalizer_class.c

@@ -0,0 +1,82 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Ed Batutis <ed@batutis.com>                                 |
+   +----------------------------------------------------------------------+
+ */
+
+#include "normalizer_class.h"
+#include "php_intl.h"
+#include "normalizer_normalize.h"
+#include "intl_error.h"
+
+#include <unicode/unorm.h>
+
+zend_class_entry *Normalizer_ce_ptr = NULL;
+
+/////////////////////////////////////////////////////////////////////////////
+// 'Normalizer' class registration structures & functions
+/////////////////////////////////////////////////////////////////////////////
+
+/* {{{ Normalizer methods arguments info */
+
+static
+ZEND_BEGIN_ARG_INFO_EX( normalizer_3_args, 0, 0, 3 )
+	ZEND_ARG_INFO( 0, arg1 )
+	ZEND_ARG_INFO( 0, arg2 )
+	ZEND_ARG_INFO( 0, arg3 )
+ZEND_END_ARG_INFO()
+
+/* }}} */
+
+/* {{{ Normalizer_class_functions
+ * Every 'Normalizer' class method has an entry in this table
+ */
+
+function_entry Normalizer_class_functions[] = {
+	ZEND_FENTRY( normalize, ZEND_FN( normalizer_normalize ), normalizer_3_args, ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	ZEND_FENTRY( isNormalized, ZEND_FN( normalizer_is_normalized ), normalizer_3_args, ZEND_ACC_PUBLIC|ZEND_ACC_STATIC )
+	{ NULL, NULL, NULL }
+};
+/* }}} */
+
+/* {{{ normalizer_register_Normalizer_class
+ * Initialize 'Normalizer' class
+ */
+void normalizer_register_Normalizer_class( TSRMLS_D )
+{
+	zend_class_entry ce;
+
+	// Create and register 'Normalizer' class.
+	INIT_CLASS_ENTRY( ce, "Normalizer", Normalizer_class_functions );
+	ce.create_object = NULL;
+	Normalizer_ce_ptr = zend_register_internal_class( &ce TSRMLS_CC );
+
+	// Declare 'Normalizer' class properties.
+	if( !Normalizer_ce_ptr )
+	{
+		zend_error( E_ERROR,
+			"Normalizer: attempt to create properties "
+			"on a non-registered class." );
+		return;
+	}
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/normalizer/normalizer_class.h

@@ -0,0 +1,52 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Ed Batutis <ed@batutis.com>                                 |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef NORMALIZER_CLASS_H
+#define NORMALIZER_CLASS_H
+
+#include <php.h>
+
+#include "intl_common.h"
+#include "intl_error.h"
+
+#include <unicode/unorm.h>
+
+typedef struct {
+	zend_object     zo;
+
+	// error value not used currently
+	intl_error  err;
+
+} Normalizer_object;
+
+#define NORMALIZER_ERROR(co) (co)->err
+#define NORMALIZER_ERROR_P(co) &(NORMALIZER_ERROR(co))
+
+#define NORMALIZER_ERROR_CODE(co)   INTL_ERROR_CODE(NORMALIZER_ERROR(co))
+#define NORMALIZER_ERROR_CODE_P(co) &(INTL_ERROR_CODE(NORMALIZER_ERROR(co)))
+
+void normalizer_register_Normalizer_class( TSRMLS_D );
+void normalizer_object_init( Normalizer_object* co TSRMLS_DC );
+void normalizer_object_destroy( Normalizer_object* co TSRMLS_DC );
+
+extern zend_class_entry *Normalizer_ce_ptr;
+
+/* Auxiliary macros */
+
+#define NORMALIZER_METHOD_INIT_VARS     \
+    intl_error_reset( NULL TSRMLS_CC ); \
+
+#endif // #ifndef NORMALIZER_CLASS_H

ext/intl/normalizer/normalizer_normalize.c

@@ -0,0 +1,256 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5														  |
+   +----------------------------------------------------------------------+
+   | 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:			  |
+   | http://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: Ed Batutis <ed@batutis.com>								  |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php_intl.h"
+#include "unicode/unorm.h"
+#include "normalizer.h"
+#include "normalizer_class.h"
+#include "normalizer_normalize.h"
+#include "intl_convert.h"
+
+/* {{{ proto string Normalizer::normalize( string $input [, string $form = FORM_C] )
+ * Normalize a string. }}} */
+/* {{{ proto string normalizer_normalize( string $input [, string $form = FORM_C] )
+ * Normalize a string.
+ */
+PHP_FUNCTION( normalizer_normalize )
+{
+	char*			input = NULL;
+	// form is optional, defaults to FORM_C
+	long			form = NORMALIZER_DEFAULT;
+	int			input_len = 0;
+		
+	UChar*			uinput = NULL;
+	int			uinput_len = 0;
+	int			expansion_factor = 1;
+	UErrorCode		status = U_ZERO_ERROR;
+		
+	UChar*			uret_buf = NULL;
+	int			uret_len = 0;
+		
+	char*			ret_buf = NULL;
+	int32_t			ret_len = 0;
+
+	int32_t			size_needed;
+		
+	NORMALIZER_METHOD_INIT_VARS
+
+	intl_error_reset( NULL TSRMLS_CC );
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "s|l",
+				&input, &input_len, &form ) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+						 "normalizer_normalize: unable to parse input params", 1 TSRMLS_CC );
+
+		RETURN_NULL();
+	}
+
+	expansion_factor = 1;
+
+	switch(form) {
+		case NORMALIZER_NONE:
+			break;
+		case NORMALIZER_FORM_D:
+			expansion_factor = 3;
+			break;
+		case NORMALIZER_FORM_KD:
+			expansion_factor = 3;
+			break;
+		case NORMALIZER_FORM_C:
+		case NORMALIZER_FORM_KC:
+			break;
+		default:
+			intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+						"normalizer_normalize: illegal normalization form", 1 TSRMLS_CC );
+			RETURN_NULL();
+	}
+
+	/*
+	 * Normalize string (converting it to UTF-16 first).
+	 */
+
+	// First convert the string to UTF-16.
+	intl_convert_utf8_to_utf16(&uinput, &uinput_len, input, input_len, &status );
+
+	if( U_FAILURE( status ) )
+	{
+		// Set global error code.
+		intl_error_set_code( NULL, status TSRMLS_CC );
+
+		// Set error messages.
+		intl_error_set_custom_msg( NULL, "Error converting input string to UTF-16", 1 TSRMLS_CC );
+		efree( uinput );
+		RETURN_NULL();
+	}
+
+
+	// Allocate memory for the destination buffer for normalization
+	uret_len = uinput_len * expansion_factor;
+	uret_buf = eumalloc( uret_len + 1 );
+
+	// normalize
+	size_needed = unorm_normalize( uinput, uinput_len, form, (int32_t) 0 /* options */, uret_buf, uret_len, &status);
+	
+	// Bail out if an unexpected error occured.
+	// (U_BUFFER_OVERFLOW_ERROR means that *target buffer is not large enough).
+	// (U_STRING_NOT_TERMINATED_WARNING usually means that the input string is empty).
+	if( U_FAILURE(status) && status != U_BUFFER_OVERFLOW_ERROR && status != U_STRING_NOT_TERMINATED_WARNING ) {
+		efree( uret_buf );
+		efree( uinput );
+		RETURN_NULL();
+	}
+
+	if ( size_needed > uret_len ) {
+		// realloc does not seem to work properly - memory is corrupted
+		// uret_buf =  eurealloc(uret_buf, size_needed + 1);
+		efree( uret_buf );
+		uret_buf = eumalloc( size_needed + 1 );
+		uret_len = size_needed;
+
+		status = U_ZERO_ERROR;
+
+		// try normalize again
+		size_needed = unorm_normalize( uinput, uinput_len, form, (int32_t) 0 /* options */, uret_buf, uret_len, &status);
+
+		// Bail out if an unexpected error occured.
+		if( U_FAILURE(status)  ) {
+			// Set error messages.
+			intl_error_set_custom_msg( NULL,"Error normalizing string", 1 TSRMLS_CC );
+			efree( uret_buf );
+			efree( uinput );
+			RETURN_NULL();
+		}
+	}
+
+	efree( uinput );
+
+	// the buffer we actually used
+	uret_len = size_needed;
+
+	// Convert normalized string from UTF-16 to UTF-8.
+	intl_convert_utf16_to_utf8( &ret_buf, &ret_len, uret_buf, uret_len, &status );
+	efree( uret_buf );
+	if( U_FAILURE( status ) )
+	{
+		intl_error_set( NULL, status,
+				"normalizer_normalize: error converting normalized text UTF-8", 1 TSRMLS_CC );
+		RETURN_NULL();
+	}
+
+	// Return it.
+	RETVAL_STRINGL( ret_buf, ret_len, FALSE );
+}
+/* }}} */
+
+/* {{{ proto bool Normalizer::isNormalized( string $input [, string $form = FORM_C] )
+ * Test if a string is in a given normalization form. }}} */
+/* {{{ proto bool normalizer_is_normalize( string $input [, string $form = FORM_C] )
+ * Test if a string is in a given normalization form.
+ */
+PHP_FUNCTION( normalizer_is_normalized )
+{
+	char*	 	input = NULL;
+	// form is optional, defaults to FORM_C
+	long		form = NORMALIZER_DEFAULT;
+	int		input_len = 0;
+
+	UChar*	 	uinput = NULL;
+	int		uinput_len = 0;
+	UErrorCode	status = U_ZERO_ERROR;
+		
+	UBool		uret = FALSE;
+		
+	NORMALIZER_METHOD_INIT_VARS
+
+	intl_error_reset( NULL TSRMLS_CC );
+
+	// Parse parameters.
+	if( zend_parse_method_parameters( ZEND_NUM_ARGS() TSRMLS_CC, getThis(), "s|l",
+				&input, &input_len, &form) == FAILURE )
+	{
+		intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+				"normalizer_is_normalized: unable to parse input params", 1 TSRMLS_CC );
+
+		RETURN_FALSE;
+	}
+
+	switch(form) {
+		// case NORMALIZER_NONE: not allowed - doesn't make sense
+
+		case NORMALIZER_FORM_D:
+		case NORMALIZER_FORM_KD:
+		case NORMALIZER_FORM_C:
+		case NORMALIZER_FORM_KC:
+			break;
+		default:
+			intl_error_set( NULL, U_ILLEGAL_ARGUMENT_ERROR,
+						"normalizer_normalize: illegal normalization form", 1 TSRMLS_CC );
+			RETURN_NULL();
+	}
+
+
+	/*
+	 * Test normalization of string (converting it to UTF-16 first).
+	 */
+
+	// First convert the string to UTF-16.
+	intl_convert_utf8_to_utf16(&uinput, &uinput_len, input, input_len, &status );
+
+	if( U_FAILURE( status ) )
+	{
+		// Set global error code.
+		intl_error_set_code( NULL, status TSRMLS_CC );
+
+		// Set error messages.
+		intl_error_set_custom_msg( NULL, "Error converting string to UTF-16.", 1 TSRMLS_CC );
+		efree( uinput );
+		RETURN_FALSE;
+	}
+
+
+	// test string
+	uret = unorm_isNormalizedWithOptions( uinput, uinput_len, form, (int32_t) 0 /* options */, &status);
+	
+	efree( uinput );
+
+	// Bail out if an unexpected error occured.
+	if( U_FAILURE(status)  ) {
+		// Set error messages.
+		intl_error_set_custom_msg( NULL,"Error testing if string is the given normalization form.", 1 TSRMLS_CC );
+		RETURN_FALSE;
+	}
+
+	if ( uret )
+		RETURN_TRUE;
+				
+	RETURN_FALSE;
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/normalizer/normalizer_normalize.h

@@ -0,0 +1,25 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Ed Batutis <ed@batutis.com>                                 |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef NORMALIZER_NORMALIZE_H
+#define NORMALIZER_NORMALIZE_H
+
+#include <php.h>
+
+PHP_FUNCTION( normalizer_normalize );
+PHP_FUNCTION( normalizer_is_normalized );
+
+#endif // NORMALIZER_NORMALIZE_H

ext/intl/php_intl.c

@@ -0,0 +1,490 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   |          Stanislav Malyshev <stas@zend.com>                          |
+   |          Kirti Velankar <kirtig@yahoo-inc.com>   			  |
+   +----------------------------------------------------------------------+
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+
+#include "php_intl.h"
+#include "intl_error.h"
+#include "collator/collator_class.h"
+#include "collator/collator.h"
+#include "collator/collator_attr.h"
+#include "collator/collator_compare.h"
+#include "collator/collator_sort.h"
+#include "collator/collator_convert.h"
+#include "collator/collator_locale.h"
+#include "collator/collator_create.h"
+#include "collator/collator_error.h"
+
+#include "formatter/formatter.h"
+#include "formatter/formatter_class.h"
+#include "formatter/formatter_attr.h"
+#include "formatter/formatter_format.h"
+#include "formatter/formatter_main.h"
+#include "formatter/formatter_parse.h"
+
+#include "msgformat/msgformat.h"
+#include "msgformat/msgformat_class.h"
+#include "msgformat/msgformat_attr.h"
+#include "msgformat/msgformat_format.h"
+#include "msgformat/msgformat_parse.h"
+
+#include "normalizer/normalizer.h"
+#include "normalizer/normalizer_class.h"
+#include "normalizer/normalizer_normalize.h"
+
+#include "locale/locale.h"
+#include "locale/locale_class.h"
+#include "locale/locale_methods.h"
+
+#include "dateformat/dateformat.h"
+#include "dateformat/dateformat_class.h"
+#include "dateformat/dateformat_attr.h"
+#include "dateformat/dateformat_format.h"
+#include "dateformat/dateformat_parse.h"
+#include "dateformat/dateformat_data.h"
+
+#include "msgformat/msgformat.h"
+#include "common/common_error.h"
+
+#include <unicode/uloc.h>
+#include <ext/standard/info.h>
+
+#include "php_ini.h"
+#define INTL_MODULE_VERSION PHP_INTL_VERSION
+
+/*
+ * locale_get_default has a conflict since ICU also has 
+ * a function with the same  name
+ * in fact ICU appends the version no. to it also
+ * Hence the following undef for ICU version
+ * Same true for the locale_set_default function 
+*/
+#undef locale_get_default
+#undef locale_set_default
+
+ZEND_DECLARE_MODULE_GLOBALS( intl )
+
+/* {{{ Arguments info */
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_static_0_args, 0, 0, 0 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_static_1_arg, 0, 0, 1 )
+	ZEND_ARG_INFO( 0, arg1 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_static_2_args, 0, 0, 2 )
+	ZEND_ARG_INFO( 0, arg1 )
+	ZEND_ARG_INFO( 0, arg2 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_0_args, 0, 0, 1 )
+	ZEND_ARG_OBJ_INFO( 0, object, Collator, 0 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_1_arg, 0, 0, 2 )
+	ZEND_ARG_OBJ_INFO( 0, object, Collator, 0 )
+	ZEND_ARG_INFO( 0, arg1 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_2_args, 0, 0, 3 )
+	ZEND_ARG_OBJ_INFO( 0, object, Collator, 0 )
+	ZEND_ARG_INFO( 0, arg1 )
+	ZEND_ARG_INFO( 0, arg2 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( collator_sort_args, 0, 0, 2 )
+	ZEND_ARG_OBJ_INFO( 0, object, Collator, 0 )
+	ZEND_ARG_ARRAY_INFO( 1, arr, 0 )
+	ZEND_ARG_INFO( 0, sort_flags )
+ZEND_END_ARG_INFO()
+
+ZEND_BEGIN_ARG_INFO_EX( numfmt_parse_arginfo, 0, 0, 2 )
+	ZEND_ARG_INFO( 0, formatter )
+	ZEND_ARG_INFO( 0, string )
+	ZEND_ARG_INFO( 0, type )
+	ZEND_ARG_INFO( 1, position )
+ZEND_END_ARG_INFO()
+
+ZEND_BEGIN_ARG_INFO_EX( numfmt_parse_currency_arginfo, 0, 0, 3 )
+	ZEND_ARG_INFO( 0, formatter )
+	ZEND_ARG_INFO( 0, string )
+	ZEND_ARG_INFO( 1, currency )
+	ZEND_ARG_INFO( 1, position )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( locale_0_args, 0, 0, 0 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( locale_1_arg, 0, 0, 1 )
+        ZEND_ARG_INFO( 0, arg1 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( locale_2_args, 0, 0, 2 )
+        ZEND_ARG_INFO( 0, arg1 )
+        ZEND_ARG_INFO( 0, arg2 )
+ZEND_END_ARG_INFO()
+static
+ZEND_BEGIN_ARG_INFO_EX( locale_3_args, 0, 0, 3 )
+        ZEND_ARG_INFO( 0, arg1 )
+        ZEND_ARG_INFO( 0, arg2 )
+        ZEND_ARG_INFO( 0, arg3 )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( locale_4_args, 0, 0, 4 )
+        ZEND_ARG_INFO( 0, arg1 )
+        ZEND_ARG_INFO( 0, arg2 )
+        ZEND_ARG_INFO( 0, arg3 )
+        ZEND_ARG_INFO( 0, arg4 )
+ZEND_END_ARG_INFO()
+
+#define intl_0_args collator_static_0_args
+#define intl_1_arg collator_static_1_arg
+
+static
+ZEND_BEGIN_ARG_INFO_EX( normalizer_args, 0, 0, 1 )
+	ZEND_ARG_INFO( 0, input )
+	ZEND_ARG_INFO( 0, form )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( grapheme_1_arg, 0, 0, 1 )
+        ZEND_ARG_INFO( 0, string )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( grapheme_search_args, 0, 0, 2 )
+        ZEND_ARG_INFO( 0, haystack )
+        ZEND_ARG_INFO( 0, needle )
+        ZEND_ARG_INFO( 0, offset )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( grapheme_substr_args, 0, 0, 2 )
+        ZEND_ARG_INFO( 0, string )
+        ZEND_ARG_INFO( 0, start )
+        ZEND_ARG_INFO( 0, length )
+ZEND_END_ARG_INFO()
+
+static
+ZEND_BEGIN_ARG_INFO_EX( grapheme_strstr_args, 0, 0, 2 )
+        ZEND_ARG_INFO( 0, haystack )
+        ZEND_ARG_INFO( 0, needle )
+        ZEND_ARG_INFO( 0, before_needle )
+ZEND_END_ARG_INFO()
+
+ZEND_BEGIN_ARG_INFO_EX( grapheme_extract_args, 0, 0, 2 )
+		ZEND_ARG_INFO( 0, arg1 )
+		ZEND_ARG_INFO( 0, arg2 )
+		ZEND_ARG_INFO( 0, arg3 )
+		ZEND_ARG_INFO( 0, arg4 )
+		ZEND_ARG_INFO( 1, arg5 )  /* 1 = pass by reference */
+ZEND_END_ARG_INFO()
+
+/* }}} */
+
+/* {{{ intl_functions
+ *
+ * Every user visible function must have an entry in intl_functions[].
+ */
+zend_function_entry intl_functions[] = {
+
+	// collator functions
+	PHP_FE( collator_create, collator_static_1_arg )
+	PHP_FE( collator_compare, collator_2_args )
+	PHP_FE( collator_get_attribute, collator_1_arg )
+	PHP_FE( collator_set_attribute, collator_2_args )
+	PHP_FE( collator_get_strength, collator_0_args )
+	PHP_FE( collator_set_strength, collator_1_arg )
+	PHP_FE( collator_sort, collator_sort_args )
+	PHP_FE( collator_sort_with_sort_keys, collator_sort_args )
+	PHP_FE( collator_asort, collator_sort_args )
+	PHP_FE( collator_get_locale, collator_1_arg )
+	PHP_FE( collator_get_error_code, collator_0_args )
+	PHP_FE( collator_get_error_message, collator_0_args )
+
+	// formatter functions
+	PHP_FE( numfmt_create, NULL )
+	PHP_FE( numfmt_format, NULL )
+	PHP_FE( numfmt_parse, numfmt_parse_arginfo )
+	PHP_FE( numfmt_format_currency, NULL )
+	PHP_FE( numfmt_parse_currency, numfmt_parse_currency_arginfo )
+	PHP_FE( numfmt_set_attribute, NULL )
+	PHP_FE( numfmt_get_attribute, NULL )
+	PHP_FE( numfmt_set_text_attribute, NULL )
+	PHP_FE( numfmt_get_text_attribute, NULL )
+	PHP_FE( numfmt_set_symbol, NULL )
+	PHP_FE( numfmt_get_symbol, NULL )
+	PHP_FE( numfmt_set_pattern, NULL )
+	PHP_FE( numfmt_get_pattern, NULL )
+	PHP_FE( numfmt_get_locale, NULL )
+	PHP_FE( numfmt_get_error_code, NULL )
+	PHP_FE( numfmt_get_error_message, NULL )
+
+	// normalizer functions
+	PHP_FE( normalizer_normalize, normalizer_args )
+	PHP_FE( normalizer_is_normalized, normalizer_args )
+
+	//Locale functions
+	PHP_NAMED_FE( locale_get_default, zif_locale_get_default, locale_0_args )
+	PHP_NAMED_FE( locale_set_default, zif_locale_set_default, locale_1_arg )
+	PHP_FE( locale_get_primary_language, locale_1_arg )
+	PHP_FE( locale_get_script, locale_1_arg )
+	PHP_FE( locale_get_region, locale_1_arg )
+	PHP_FE( locale_get_keywords, locale_1_arg )
+	PHP_FE( locale_get_display_script, locale_2_args )
+	PHP_FE( locale_get_display_region, locale_2_args )
+	PHP_FE( locale_get_display_name, locale_2_args )
+	PHP_FE( locale_get_display_language, locale_2_args)
+	PHP_FE( locale_get_display_variant, locale_2_args )
+	PHP_FE( locale_compose, locale_1_arg )
+	PHP_FE( locale_parse, locale_1_arg )
+	PHP_FE( locale_get_all_variants, locale_1_arg )
+	PHP_FE( locale_filter_matches, locale_3_args )
+	PHP_FE( locale_canonicalize, locale_1_arg )
+	PHP_FE( locale_lookup, locale_4_args )
+
+	// MessageFormatter functions
+	PHP_FE( msgfmt_create, NULL )
+	PHP_FE( msgfmt_format, NULL )
+	PHP_FE( msgfmt_format_message, NULL )
+	PHP_FE( msgfmt_parse, NULL )
+	PHP_FE( msgfmt_parse_message, NULL )
+	PHP_FE( msgfmt_set_pattern, NULL )
+	PHP_FE( msgfmt_get_pattern, NULL )
+	PHP_FE( msgfmt_get_locale, NULL )
+	PHP_FE( msgfmt_get_error_code, NULL )
+	PHP_FE( msgfmt_get_error_message, NULL )
+
+	// IntlDateFormatter functions
+	PHP_FE( datefmt_create, NULL )
+	PHP_FE( datefmt_get_datetype, NULL )
+	PHP_FE( datefmt_get_timetype, NULL )
+	PHP_FE( datefmt_get_calendar, NULL )
+	PHP_FE( datefmt_set_calendar, NULL )
+	PHP_FE( datefmt_get_locale, NULL )
+	PHP_FE( datefmt_get_timezone_id, NULL )
+	PHP_FE( datefmt_set_timezone_id, NULL )
+	PHP_FE( datefmt_get_pattern, NULL )
+	PHP_FE( datefmt_set_pattern, NULL )
+	PHP_FE( datefmt_is_lenient, NULL )
+	PHP_FE( datefmt_set_lenient, NULL )
+	PHP_FE( datefmt_format, NULL )
+	PHP_FE( datefmt_parse, NULL )
+        PHP_FE( datefmt_localtime , NULL )
+	PHP_FE( datefmt_get_error_code, NULL )
+	PHP_FE( datefmt_get_error_message, NULL )
+
+	// grapheme functions
+	PHP_FE( grapheme_strlen, grapheme_1_arg )
+	PHP_FE( grapheme_strpos, grapheme_search_args )
+	PHP_FE( grapheme_stripos, grapheme_search_args )
+	PHP_FE( grapheme_strrpos, grapheme_search_args )
+	PHP_FE( grapheme_strripos, grapheme_search_args )
+	PHP_FE( grapheme_substr, grapheme_substr_args )
+	PHP_FE( grapheme_strstr, grapheme_strstr_args )
+	PHP_FE( grapheme_stristr, grapheme_strstr_args )
+	PHP_FE( grapheme_extract, grapheme_extract_args )
+
+	// common functions
+	PHP_FE( intl_get_error_code, intl_0_args )
+	PHP_FE( intl_get_error_message, intl_0_args )
+	PHP_FE( intl_is_failure, intl_1_arg )
+	PHP_FE( intl_error_name, intl_1_arg )
+
+	{ NULL, NULL, NULL }
+};
+/* }}} */
+
+
+/* {{{ INI Settings */
+PHP_INI_BEGIN()
+    STD_PHP_INI_ENTRY(LOCALE_INI_NAME, NULL, PHP_INI_ALL, OnUpdateStringUnempty, default_locale, zend_intl_globals, intl_globals)
+
+PHP_INI_END()
+/* }}} */
+
+
+static PHP_GINIT_FUNCTION(intl);
+
+/* {{{ intl_module_entry */
+zend_module_entry intl_module_entry = {
+#if ZEND_MODULE_API_NO >= 20010901
+	STANDARD_MODULE_HEADER,
+#endif
+	"intl",
+	intl_functions,
+	PHP_MINIT( intl ),
+	PHP_MSHUTDOWN( intl ),
+	PHP_RINIT( intl ),
+	PHP_RSHUTDOWN( intl ),
+	PHP_MINFO( intl ),
+	INTL_MODULE_VERSION,
+	PHP_MODULE_GLOBALS(intl),   /* globals descriptor */
+	PHP_GINIT(intl),            /* globals ctor */
+	NULL,                       /* globals dtor */
+	NULL,                       /* post deactivate */
+	STANDARD_MODULE_PROPERTIES_EX
+};
+/* }}} */
+
+#ifdef COMPILE_DL_INTL
+ZEND_GET_MODULE( intl )
+#endif
+
+/* {{{ intl_init_globals */
+static PHP_GINIT_FUNCTION(intl)
+{
+	memset( intl_globals, 0, sizeof(zend_intl_globals) );
+}
+/* }}} */
+
+/* {{{ PHP_MINIT_FUNCTION
+ */
+PHP_MINIT_FUNCTION( intl )
+{
+	//For the default locale php.ini setting
+	REGISTER_INI_ENTRIES();
+
+	REGISTER_LONG_CONSTANT("INTL_MAX_LOCALE_LEN", INTL_MAX_LOCALE_LEN, CONST_CS);
+
+	// Register 'Collator' PHP class
+	collator_register_Collator_class( TSRMLS_C );
+
+	// Expose Collator constants to PHP scripts
+	collator_register_constants( INIT_FUNC_ARGS_PASSTHRU );
+
+	// Register 'NumberFormatter' PHP class
+	formatter_register_class( TSRMLS_C );
+
+	// Expose NumberFormatter constants to PHP scripts
+	formatter_register_constants( INIT_FUNC_ARGS_PASSTHRU );
+
+	// Register 'Normalizer' PHP class
+	normalizer_register_Normalizer_class( TSRMLS_C );
+
+	// Expose Normalizer constants to PHP scripts
+	normalizer_register_constants( INIT_FUNC_ARGS_PASSTHRU );
+	
+	// Register 'Locale' PHP class
+	locale_register_Locale_class( TSRMLS_C );
+
+	// Expose Locale constants to PHP scripts
+	locale_register_constants( INIT_FUNC_ARGS_PASSTHRU );
+
+	msgformat_register_class(TSRMLS_C);
+
+	grapheme_register_constants( INIT_FUNC_ARGS_PASSTHRU );
+
+	// Register 'DateFormat' PHP class
+	dateformat_register_IntlDateFormatter_class( TSRMLS_C );
+
+	// Expose DateFormat constants to PHP scripts
+	dateformat_register_constants( INIT_FUNC_ARGS_PASSTHRU );
+
+	// Expose ICU error codes to PHP scripts.
+	intl_expose_icu_error_codes( INIT_FUNC_ARGS_PASSTHRU );
+
+	// Global error handling.
+	intl_error_init( NULL TSRMLS_CC );
+
+	//Set the default_locale value
+    	if( INTL_G(default_locale) == NULL ) {
+        	INTL_G(default_locale) = pestrdup(uloc_getDefault(), 1) ;
+	}
+
+	return SUCCESS;
+}
+/* }}} */
+
+/* {{{ PHP_MSHUTDOWN_FUNCTION
+ */
+PHP_MSHUTDOWN_FUNCTION( intl )
+{
+    //For the default locale php.ini setting
+    UNREGISTER_INI_ENTRIES();
+
+    return SUCCESS;
+}
+/* }}} */
+
+/* {{{ PHP_RINIT_FUNCTION
+ */
+PHP_RINIT_FUNCTION( intl )
+{
+	//Set the default_locale value
+    if( INTL_G(default_locale) == NULL ) {
+        INTL_G(default_locale) = pestrdup(uloc_getDefault(), 1) ;
+    }
+	return SUCCESS;
+}
+/* }}} */
+
+/* {{{ PHP_RSHUTDOWN_FUNCTION
+ */
+PHP_RSHUTDOWN_FUNCTION( intl )
+{
+	if(INTL_G(current_collator)) {
+		INTL_G(current_collator) = NULL;
+	}
+	if (INTL_G(grapheme_iterator)) {
+		grapheme_close_global_iterator( TSRMLS_C );
+		INTL_G(grapheme_iterator) = NULL;
+	}
+
+	intl_error_reset( NULL TSRMLS_CC);
+	return SUCCESS;
+}
+/* }}} */
+
+/* {{{ PHP_MINFO_FUNCTION
+ */
+PHP_MINFO_FUNCTION( intl )
+{
+	php_info_print_table_start();
+	php_info_print_table_header( 2, "Internationalization support", "enabled" );
+	php_info_print_table_row( 2, "version", INTL_MODULE_VERSION );
+	php_info_print_table_end();
+
+    //For the default locale php.ini setting
+    DISPLAY_INI_ENTRIES() ;
+}
+/* }}} */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/php_intl.h

@@ -0,0 +1,76 @@
+/*
+   +----------------------------------------------------------------------+
+   | PHP Version 5                                                        |
+   +----------------------------------------------------------------------+
+   | 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:           |
+   | http://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: Vadim Savchuk <vsavchuk@productengine.com>                  |
+   |          Dmitry Lakhtyuk <dlakhtyuk@productengine.com>               |
+   |          Stanislav Malyshev <stas@zend.com>                          |
+   |          Kirti Velankar <kirtig@yahoo-inc.com>                       |
+   +----------------------------------------------------------------------+
+ */
+
+#ifndef PHP_INTL_H
+#define PHP_INTL_H
+
+#include <php.h>
+
+#include "collator/collator_sort.h"
+#include "grapheme/grapheme.h"
+#include "intl_error.h"
+
+extern zend_module_entry intl_module_entry;
+#define phpext_intl_ptr &intl_module_entry
+
+#ifdef PHP_WIN32
+#define PHP_INTL_API __declspec(dllexport)
+#else
+#define PHP_INTL_API
+#endif
+
+#ifdef ZTS
+#include "TSRM.h"
+#endif
+
+ZEND_BEGIN_MODULE_GLOBALS(intl)
+	zval* current_collator;
+	char* default_locale;
+	collator_compare_func_t compare_func;
+	UBreakIterator* grapheme_iterator;
+	intl_error g_error;
+ZEND_END_MODULE_GLOBALS(intl)
+
+/* Macro to access request-wide global variables. */
+#ifdef ZTS
+#define INTL_G(v) TSRMG(intl_globals_id, zend_intl_globals *, v)
+#else
+#define INTL_G(v) (intl_globals.v)
+#endif
+
+ZEND_EXTERN_MODULE_GLOBALS(intl)
+
+PHP_MINIT_FUNCTION(intl);
+PHP_MSHUTDOWN_FUNCTION(intl);
+PHP_RINIT_FUNCTION(intl);
+PHP_RSHUTDOWN_FUNCTION(intl);
+PHP_MINFO_FUNCTION(intl);
+
+#define PHP_INTL_VERSION "1.0.0"
+
+#endif  /* PHP_INTL_H */
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */

ext/intl/tests/bug12887.phpt

@@ -0,0 +1,29 @@
+--TEST--
+locale_get_keywords() bug #12887
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+function ut_main()
+{
+	$res_str = '';
+	$keywords_arr = ut_loc_get_keywords( 'de_DE@currency=EUR;collation=PHONEBOOK;sort=PHONEBOOK' );
+	if ($keywords_arr) {
+		foreach( $keywords_arr as $key => $value){
+			$res_str .= "$key = $value\n";
+		}
+	}
+    $res_str .= "\n";
+    return $res_str;
+
+}
+
+include_once( 'ut_common.inc' );
+ut_run();
+
+?>
+--EXPECT--
+collation = PHONEBOOK
+currency = EUR
+sort = PHONEBOOK

ext/intl/tests/collation_customization.phpt

@@ -0,0 +1,70 @@
+--TEST--
+Collation customization
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Check effects of changing misc collattion options.
+ */
+
+
+function cmp_array( &$coll, $a )
+{
+    $res = '';
+    $prev = null;
+    foreach( $a as $i )
+    {
+        if( is_null( $prev ) )
+            $res .= "$i";
+        else
+        {
+            $eqrc = ut_coll_compare( $coll, $prev, $i );
+            $eq = $eqrc < 0 ? "<" : ( $eqrc > 0 ? ">" : "=" );
+            $res .= " $eq $i";
+        }
+
+        $prev = $i;
+    }
+    $res .= "\n";
+
+    return $res;
+}
+
+function check_alternate_handling( &$coll )
+{
+    $res = '';
+
+    ut_coll_set_strength( $coll, Collator::TERTIARY );
+    ut_coll_set_attribute( $coll, Collator::ALTERNATE_HANDLING, Collator::NON_IGNORABLE );
+
+    $res .= cmp_array( $coll, array( 'di Silva', 'Di Silva', 'diSilva', 'U.S.A.', 'USA' ) );
+
+    ut_coll_set_attribute( $coll, Collator::ALTERNATE_HANDLING, Collator::SHIFTED );
+
+    $res .= cmp_array( $coll, array( 'di Silva', 'diSilva', 'Di Silva', 'U.S.A.', 'USA' ) );
+
+    ut_coll_set_strength( $coll, Collator::QUATERNARY );
+
+    $res .= cmp_array( $coll, array( 'di Silva', 'diSilva', 'Di Silva', 'U.S.A.', 'USA' ) );
+    $res .= "\n";
+
+    return $res;
+}
+
+function ut_main()
+{
+    $coll = ut_coll_create( 'en_US' );
+
+    return
+        check_alternate_handling( $coll );
+}
+
+include_once( 'ut_common.inc' );
+ut_run();
+?>
+--EXPECT--
+di Silva < Di Silva < diSilva < U.S.A. < USA
+di Silva = diSilva < Di Silva < U.S.A. = USA
+di Silva < diSilva < Di Silva < U.S.A. < USA

ext/intl/tests/collator_asort.phpt

@@ -0,0 +1,242 @@
+--TEST--
+asort()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Sort associative arrays using various locales.
+ */
+
+
+$test_num = 1;
+
+/*
+ * Sort various arrays in specified locale.
+ */
+function sort_arrays( $locale, $test_arrays, $sort_flag = Collator::SORT_REGULAR )
+{
+    $res_str = '';
+
+    $coll = ut_coll_create( $locale );
+
+    foreach( $test_arrays as $test_array )
+    {
+        // Try to sort test data.
+        $res_val = ut_coll_asort( $coll, $test_array, $sort_flag );
+
+        // Return output data.
+        $res_dump = "\n" . dump( $test_array ) .
+                    "\n Result: " . dump( $res_val );
+
+		// Preppend test signature to output string
+        $md5 = md5( $res_dump );
+
+        global $test_num;
+        
+        $res_str .= "\n\n".
+                    "Test $test_num.$md5:" .
+                    $res_dump;
+        ++$test_num;
+    }
+
+    return $res_str;
+}
+
+/*
+ * Test main function.
+ */
+function ut_main()
+{
+    global $test_num;
+    $test_num = 1;
+    $res_str = '';
+
+    // Sort an array in SORT_REGULAR mode using en_US locale.
+    $test_params = array(
+        array( 'd' => 'y'  ,
+               'c' => 'i'  ,
+               'a' => 'k'  ),
+
+        array( 'a' => 'a'  ,
+               'b' => 'aaa',
+               'c' => 'aa' ),
+
+        array( 'a'  => 'a' ,
+               'aaa'=> 'a' ,
+               'aa' => 'a' ),
+
+        array( '1' => 'abc',
+               '5' => '!'  ,
+               '2' => null ,
+               '7' => ''   ),
+
+        array( '1' => '100',
+               '2' => '25' ,
+               '3' => '36' ),
+
+        array( '1' => 5    ,
+               '2' => '30' ,
+               '3' => 2    )
+    );
+
+    $res_str .= sort_arrays( 'en_US', $test_params );
+
+    // Sort an array in SORT_STRING mode using en_US locale.
+    $test_params = array(
+        array( '1' => '100',
+               '2' => '25' ,
+               '3' => '36' ),
+
+        array( '1' => 5    ,
+               '2' => '30' ,
+               '3' => 2    ),
+
+        array( '1' => 'd'  ,
+               '2' => ''   ,
+               '3' => ' a' ),
+
+        array( '1' => 'y'  ,
+               '2' => 'k'  ,
+               '3' => 'i'  )
+    );
+
+    $res_str .= sort_arrays( 'en_US', $test_params, Collator::SORT_STRING );
+
+    // Sort a non-ASCII array using ru_RU locale.
+    $test_params = array(
+        array( 'п' => 'у',
+               'б' => 'в',
+               'е' => 'а' ),
+
+        array( '1' => 'п',
+               '4' => '',
+               '7' => 'd',
+               '2' => 'пп' )
+    );
+
+    $res_str .= sort_arrays( 'ru_RU', $test_params );
+
+
+    // Sort an array using Lithuanian locale.
+    $test_params = array(
+        array( 'd' => 'y',
+               'c' => 'i',
+               'a' => 'k' )
+    );
+
+    $res_str .= sort_arrays( 'lt_LT', $test_params );
+
+    return $res_str . "\n";
+}
+
+include_once( 'ut_common.inc' );
+ut_run();
+?>
+--EXPECT--
+Test 1.162b81ac12878b817fc39063097e45b5:
+array (
+  'c' => 'i',
+  'a' => 'k',
+  'd' => 'y',
+)
+ Result: true
+
+Test 2.93d96e22f692d8a281b0a389f01f8d1e:
+array (
+  'a' => 'a',
+  'c' => 'aa',
+  'b' => 'aaa',
+)
+ Result: true
+
+Test 3.9f25de4482bc7b58de508e278113317c:
+array (
+  'aa' => 'a',
+  'aaa' => 'a',
+  'a' => 'a',
+)
+ Result: true
+
+Test 4.a85a41ea78e45b651080cfd98c0b431d:
+array (
+  7 => '',
+  2 => NULL,
+  5 => '!',
+  1 => 'abc',
+)
+ Result: true
+
+Test 5.99dc71f405b286e03d489061b36e6900:
+array (
+  2 => '25',
+  3 => '36',
+  1 => '100',
+)
+ Result: true
+
+Test 6.bf5bba243307c9d12934e756ad4be190:
+array (
+  3 => 2,
+  1 => 5,
+  2 => '30',
+)
+ Result: true
+
+Test 7.e4ee7024c61476e9e7a6c28b5e47df6f:
+array (
+  1 => '100',
+  2 => '25',
+  3 => '36',
+)
+ Result: true
+
+Test 8.5fa7033dd43784be0db1474eb48b83c8:
+array (
+  3 => 2,
+  2 => '30',
+  1 => 5,
+)
+ Result: true
+
+Test 9.588cdf4692bc09aa92ffe7e48f9e4579:
+array (
+  2 => '',
+  3 => ' a',
+  1 => 'd',
+)
+ Result: true
+
+Test 10.be02641a47ebcccd23e4183ca3a415f7:
+array (
+  3 => 'i',
+  2 => 'k',
+  1 => 'y',
+)
+ Result: true
+
+Test 11.153d9b11d1e5936afc917a94a4e11f34:
+array (
+  'е' => 'а',
+  'б' => 'в',
+  'п' => 'у',
+)
+ Result: true
+
+Test 12.6eaea3fa21b3b7d006ca7dfa27d4e282:
+array (
+  4 => '',
+  7 => 'd',
+  1 => 'п',
+  2 => 'пп',
+)
+ Result: true
+
+Test 13.8800d48abb960a59002eef77f1d73ae0:
+array (
+  'c' => 'i',
+  'd' => 'y',
+  'a' => 'k',
+)
+ Result: true

ext/intl/tests/collator_compare.phpt

@@ -0,0 +1,134 @@
+--TEST--
+compare()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Compare various string pairs using various locales.
+ */
+
+
+/*
+ * Converts comparison result to a character.
+ */
+function cmp_to_char( $comp_res )
+{
+    switch( $comp_res )
+    {
+    case 0:            // UCOL_EQUAL
+        return '=';
+    case 1:            // UCOL_GREATER
+        return '>';
+    case -1:           // UCOL_LESS
+        return '<';
+    default:
+        return '?';
+    }
+}
+
+/*
+ * Compare string pairs in the given array
+ * using specified locale.
+ */
+function compare_pairs( $locale, $test_array )
+{
+    $res_str = '';
+
+    $coll = ut_coll_create( $locale );
+
+    foreach( $test_array as $test_strings )
+    {
+        list( $str1, $str2 ) = $test_strings;
+
+        // Compare strings.
+        $res_val = cmp_to_char( ut_coll_compare( $coll, $str1, $str2 ) );
+
+        // Concatenate result strings.
+        $res_str .= dump( $str1 ) .
+                    ' ' . $res_val . ' ' .
+                    dump( $str2 ) . "\n";
+    }
+
+    return $res_str;
+
+}
+
+function ut_main()
+{
+    $res_str = '';
+
+    // Compare strings using en_US locale.
+    $test_params = array(
+        array( 'abc', 'abc' ),
+        array( 'Abc', 'abc' ),
+        array( 'a'  , 'abc' ),
+        array( 'a'  , ''    ),
+        array( ''  , ''     ),
+        array( 'a'  , 'b'   ),
+        array( 'ab'  , 'b'  ),
+        array( 'ab'  , 'a'  ),
+        array( 123  , 'abc' ),
+        array( 'ac' , null  ),
+        array( '.'  , '.'   ),
+        // Try to compare long strings.
+        array( 'abcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcde',
+               'abcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdea'),
+        array( null , null  )
+    );
+
+    $res_str .= compare_pairs( 'en_US', $test_params );
+
+
+    // Compare strings using ru_RU locale.
+    $test_params = array(
+        array( 'а',   'б' ),
+        array( 'а',   'аа' ),
+        array( 'аб', 'ба' ),
+        array( 'а',   ',' ),
+        array( 'а',   'b' ),
+        array( 'а',   'bb' ),
+        array( 'а',   'ab' ),
+        array( 'а',   null )
+    );
+
+    $res_str .= compare_pairs( 'ru_RU', $test_params );
+
+
+    // Compare strings using lt_LT locale.
+    $test_params = array(
+        array( 'y', 'k' )
+    );
+
+    $res_str .= compare_pairs( 'lt_LT', $test_params );
+
+    return $res_str;
+}
+
+include_once( 'ut_common.inc' );
+ut_run();
+?>
+--EXPECT--
+'abc' = 'abc'
+'Abc' > 'abc'
+'a' < 'abc'
+'a' > ''
+'' = ''
+'a' < 'b'
+'ab' < 'b'
+'ab' > 'a'
+123 < 'abc'
+'ac' > NULL
+'.' = '.'
+'abcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcde' < 'abcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdefabcdea'
+NULL = NULL
+'а' < 'б'
+'а' < 'аа'
+'аб' < 'ба'
+'а' > ','
+'а' > 'b'
+'а' > 'bb'
+'а' > 'ab'
+'а' > NULL
+'y' < 'k'

ext/intl/tests/collator_create.phpt

@@ -0,0 +1,81 @@
+--TEST--
+create()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Try creating collator with different locales
+ * with Procedural and Object methods.
+ */
+
+function ut_main()
+{
+    $res_str = '';
+
+    $locales = array(
+        'EN-US-ODESSA',
+        'UK_UA_ODESSA',
+        'uk-ua_CALIFORNIA@currency=;currency=GRN',
+        '',
+        'root',
+        'uk@currency=EURO',
+	'1234567891113151719212325272931333537394143454749515357596163656769717375777981838587899193959799'
+    );
+
+    foreach( $locales as $locale )
+    {
+        // Create Collator with the current locale.
+        $coll = ut_coll_create( $locale );
+        if( !is_object($coll) )
+        {
+            $res_str .= "Error creating collator with '$locale' locale: " .
+                 intl_get_error_message() . "\n";
+            continue;
+        }
+
+        // Get the requested, valid and actual locales.
+        $vloc = ut_coll_get_locale( $coll, Locale::VALID_LOCALE );
+        $aloc = ut_coll_get_locale( $coll, Locale::ACTUAL_LOCALE );
+
+        // Show them.
+        $res_str .= "Locale: '$locale'\n" .
+            "  ULOC_REQUESTED_LOCALE = '$locale'\n" .
+            "  ULOC_VALID_LOCALE     = '$vloc'\n" .
+            "  ULOC_ACTUAL_LOCALE    = '$aloc'\n";
+    }
+
+    return $res_str;
+}
+
+include_once( 'ut_common.inc' );
+ut_run();
+
+?>
+--EXPECTF--
+Locale: 'EN-US-ODESSA'
+  ULOC_REQUESTED_LOCALE = 'EN-US-ODESSA'
+  ULOC_VALID_LOCALE     = 'en_US'
+  ULOC_ACTUAL_LOCALE    = 'en'
+Locale: 'UK_UA_ODESSA'
+  ULOC_REQUESTED_LOCALE = 'UK_UA_ODESSA'
+  ULOC_VALID_LOCALE     = 'uk_UA'
+  ULOC_ACTUAL_LOCALE    = 'uk'
+Locale: 'uk-ua_CALIFORNIA@currency=;currency=GRN'
+  ULOC_REQUESTED_LOCALE = 'uk-ua_CALIFORNIA@currency=;currency=GRN'
+  ULOC_VALID_LOCALE     = 'uk_UA'
+  ULOC_ACTUAL_LOCALE    = 'uk'
+Locale: ''
+  ULOC_REQUESTED_LOCALE = ''
+  ULOC_VALID_LOCALE     = '%s'
+  ULOC_ACTUAL_LOCALE    = '%s'
+Locale: 'root'
+  ULOC_REQUESTED_LOCALE = 'root'
+  ULOC_VALID_LOCALE     = 'root'
+  ULOC_ACTUAL_LOCALE    = 'root'
+Locale: 'uk@currency=EURO'
+  ULOC_REQUESTED_LOCALE = 'uk@currency=EURO'
+  ULOC_VALID_LOCALE     = 'uk'
+  ULOC_ACTUAL_LOCALE    = 'uk'
+Error creating collator with '1234567891113151719212325272931333537394143454749515357596163656769717375777981838587899193959799' locale: Locale string too long, should be no longer than 64 characters: U_ILLEGAL_ARGUMENT_ERROR

ext/intl/tests/collator_get_error_code.phpt

@@ -0,0 +1,47 @@
+--TEST--
+get_error_code()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Retreive error code.
+ */
+
+
+/*
+ * Check if error code equals to expected one.
+ */
+function check_rc( $rc, $expected )
+{
+    return ( $rc === $expected ? "ok" : "failed" ) . "\n";
+} 
+
+function ut_main()
+{
+    $res = '';
+    $coll = ut_coll_create( 'ru_RU' );
+
+    // Try specifying a correct attribute.
+    ut_coll_get_attribute( $coll, Collator::NORMALIZATION_MODE );
+    $status = ut_coll_get_error_code( $coll );
+    $res .= check_rc( $status, U_ZERO_ERROR );
+
+    // Try specifying an incorrect attribute.
+    ut_coll_get_attribute( $coll, 12345 );
+    $status = ut_coll_get_error_code( $coll );
+    $res .= check_rc( $status, U_ILLEGAL_ARGUMENT_ERROR );
+
+    return $res;
+}
+
+# Suppress warning messages.
+error_reporting( E_ERROR );
+
+include_once( 'ut_common.inc' );
+ut_run();
+?>
+--EXPECT--
+ok
+ok

ext/intl/tests/collator_get_error_message.phpt

@@ -0,0 +1,36 @@
+--TEST--
+get_error_message()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Retreive error message.
+ */
+
+
+function ut_main()
+{
+    $res = '';
+    $coll = ut_coll_create( 'ru_RU' );
+
+    // Try specifying a correct attribute.
+    ut_coll_get_attribute( $coll, Collator::NORMALIZATION_MODE );
+    $status = ut_coll_get_error_message( $coll );
+    $res .= $status . "\n";
+
+    // Try specifying an incorrect attribute.
+    ut_coll_get_attribute( $coll, 12345 );
+    $status = ut_coll_get_error_message( $coll );
+    $res .= $status . "\n";
+
+    return $res;
+}
+
+include_once( 'ut_common.inc' );
+ut_run();
+?>
+--EXPECT--
+U_ZERO_ERROR
+Error getting attribute value: U_ILLEGAL_ARGUMENT_ERROR

ext/intl/tests/collator_get_locale.phpt

@@ -0,0 +1,52 @@
+--TEST--
+get_locale()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Try to specify valid and invalid locale types when getting locale.
+ */
+
+function ut_main()
+{
+    $locales = array(
+        Locale::VALID_LOCALE,
+        Locale::ACTUAL_LOCALE,
+        100,
+        -100,
+        -9999999999999,
+        9999999999999,
+        1.2,
+        9999999999999999999999999999999999999999999999
+    );
+
+    $coll = ut_coll_create( 'en_US' );
+    $res_str = '';
+
+    foreach( $locales as $locale )
+    {
+        $rc = ut_coll_get_locale( $coll, $locale );
+
+        $res_str .= sprintf(
+            "Locale of type %s is %s\n",
+            dump( $locale ),
+            dump( $rc ) );
+    }
+
+    return $res_str . "\n";
+}
+
+include_once( 'ut_common.inc' );
+ut_run();
+?>
+--EXPECT--
+Locale of type 1 is 'en_US'
+Locale of type 0 is 'en'
+Locale of type 100 is false
+Locale of type -100 is false
+Locale of type -9999999999999 is false
+Locale of type 9999999999999 is false
+Locale of type 1.2 is 'en_US'
+Locale of type 1.0E+46 is false

ext/intl/tests/collator_get_set_attribute.phpt

@@ -0,0 +1,41 @@
+--TEST--
+get/set_attribute()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Try to set/get a collation attribute.
+ */
+
+
+/*
+ * Return current normalication mode.
+ */
+function check_val( $coll )
+{
+    $val = ut_coll_get_attribute( $coll, Collator::NORMALIZATION_MODE );
+    return sprintf( "%s\n", ( $val == Collator::OFF ? "off" : "on" ) );
+}
+
+function ut_main()
+{
+    $res = '';
+    $coll = ut_coll_create( 'en_US' );
+
+    ut_coll_set_attribute( $coll, Collator::NORMALIZATION_MODE, Collator::OFF );
+    $res .= check_val( $coll );
+
+    ut_coll_set_attribute( $coll, Collator::NORMALIZATION_MODE, Collator::ON );
+    $res .= check_val( $coll );
+
+    return $res;
+}
+
+include( 'ut_common.inc' );
+ut_run();
+?>
+--EXPECT--
+off
+on

ext/intl/tests/collator_get_set_strength.phpt

@@ -0,0 +1,41 @@
+--TEST--
+get/set_strength()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Try to set/get collation strength.
+ */
+
+/*
+ * Set given collation strength, then get it back
+ * and check if it's the same.
+ */
+function check_set_strength( $coll, $val )
+{
+    ut_coll_set_strength( $coll, $val );
+    $new_val = ut_coll_get_strength( $coll );
+    return ( $new_val == $val ? "ok" : "failed" ) . "\n";
+}
+
+function ut_main()
+{
+    $res = '';
+    $coll = ut_coll_create( 'en_US' );
+
+    $res .= check_set_strength( $coll, Collator::PRIMARY );
+    $res .= check_set_strength( $coll, Collator::SECONDARY );
+    $res .= check_set_strength( $coll, Collator::TERTIARY );
+
+    return $res;
+}
+
+include_once( 'ut_common.inc' );
+ut_run();
+?>
+--EXPECT--
+ok
+ok
+ok

ext/intl/tests/collator_sort.phpt

@@ -0,0 +1,247 @@
+--TEST--
+sort()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Sort arrays using various locales.
+ */
+
+
+$test_num = 1;
+
+/*
+ * Sort arrays in the given list using specified locale.
+ */
+function sort_arrays( $locale, $arrays, $sort_flag = Collator::SORT_REGULAR )
+{
+    $res_str = '';
+
+    $coll = ut_coll_create( $locale );
+
+    foreach( $arrays as $array )
+    {
+        // Sort array values
+        $res_val = ut_coll_sort( $coll, $array, $sort_flag );
+
+        // Concatenate the sorted array and function result
+        // with output string.
+        $res_dump = "\n" . dump( $array ) .
+                    "\n Result: " . dump( $res_val );
+
+		// Preppend test signature to output string
+        $md5 = md5( $res_dump );
+
+        global $test_num;
+        
+        $res_str .= "\n\n".
+                    "Test $test_num.$md5:" .
+                    $res_dump;
+        ++$test_num;
+    }
+
+    return $res_str;
+}
+
+function ut_main()
+{
+    global $test_num;
+    $test_num = 1;
+    $res_str = '';
+
+    // Sort an array in SORT_REGULAR mode using en_US locale.
+    $test_params = array(
+        array( 'abc', 'abd', 'aaa' ),
+        array( 'm'  , '1'  , '_'   ),
+        array( 'a'  , 'aaa', 'aa'  ),
+        array( 'ba' , 'b'  , 'ab'  ),
+        array( 'e'  , 'c'  , 'a'   ),
+        array( '100', '25' , '36'  ),
+        array( 5    , '30' , 2     ),
+        array( 'd'  , ''   , ' a'  ),
+        array( 'd ' , 'f ' , ' a'  ),
+        array( 'a'  , null , '3'   ),
+        array( 'y'  , 'k'  , 'i' )
+    );
+
+    $res_str .= sort_arrays( 'en_US', $test_params );
+
+    $test_params = array(
+        array( '100', '25' , '36'  ),
+        array( 5    , '30' , 2     ),
+        array( 'd'  , ''   , ' a'  ),
+        array( 'y'  , 'k'  , 'i' )
+    );
+
+    // Sort in en_US locale with SORT_STRING flag
+    $res_str .= sort_arrays( 'en_US', $test_params, Collator::SORT_STRING );
+
+
+    // Sort a non-ASCII array using ru_RU locale.
+    $test_params = array(
+        array( 'абг', 'абв', 'ааа', 'abc' ),
+        array( 'аа', 'ааа' , 'а' )
+    );
+
+    $res_str .= sort_arrays( 'ru_RU', $test_params );
+
+    // Sort an array using Lithuanian locale.
+    $test_params = array(
+        array( 'y'  , 'k'  , 'i' )
+    );
+
+    $res_str .= sort_arrays( 'lt_LT', $test_params );
+
+    return $res_str;
+}
+
+include_once( 'ut_common.inc' );
+ut_run();
+?>
+--EXPECT--
+Test 1.e8f1cd28133d79ecd660002f1c660d0e:
+array (
+  0 => 'aaa',
+  1 => 'abc',
+  2 => 'abd',
+)
+ Result: true
+
+Test 2.c2ded12173dd2996927378cae37eb275:
+array (
+  0 => '_',
+  1 => '1',
+  2 => 'm',
+)
+ Result: true
+
+Test 3.54071c968d71cb98c5d379145f8d7d38:
+array (
+  0 => 'a',
+  1 => 'aa',
+  2 => 'aaa',
+)
+ Result: true
+
+Test 4.19abe63d6f6dfef65b0e3c9ab4826b07:
+array (
+  0 => 'ab',
+  1 => 'b',
+  2 => 'ba',
+)
+ Result: true
+
+Test 5.9a8dc0a9bc771368c2f1fc3d02754610:
+array (
+  0 => 'a',
+  1 => 'c',
+  2 => 'e',
+)
+ Result: true
+
+Test 6.ab530b060e5e54a65bfb8b9f8fc61870:
+array (
+  0 => '25',
+  1 => '36',
+  2 => '100',
+)
+ Result: true
+
+Test 7.0718dd838509017bded2ed307a6e785f:
+array (
+  0 => 2,
+  1 => 5,
+  2 => '30',
+)
+ Result: true
+
+Test 8.923d65739c5219c634616ffd100a50e4:
+array (
+  0 => '',
+  1 => ' a',
+  2 => 'd',
+)
+ Result: true
+
+Test 9.289bc2f28e87d3201ec9d7e8477ae1b0:
+array (
+  0 => ' a',
+  1 => 'd ',
+  2 => 'f ',
+)
+ Result: true
+
+Test 10.de0fd958484f2377a645835d7fbcf124:
+array (
+  0 => NULL,
+  1 => '3',
+  2 => 'a',
+)
+ Result: true
+
+Test 11.dd2b8f0adb37c45d528cad1a0cc0f361:
+array (
+  0 => 'i',
+  1 => 'k',
+  2 => 'y',
+)
+ Result: true
+
+Test 12.1e6b4d6f7df9d4580317634ea46d8208:
+array (
+  0 => '100',
+  1 => '25',
+  2 => '36',
+)
+ Result: true
+
+Test 13.cec115dc9850b98dfbdf102efa09e61b:
+array (
+  0 => 2,
+  1 => '30',
+  2 => 5,
+)
+ Result: true
+
+Test 14.923d65739c5219c634616ffd100a50e4:
+array (
+  0 => '',
+  1 => ' a',
+  2 => 'd',
+)
+ Result: true
+
+Test 15.dd2b8f0adb37c45d528cad1a0cc0f361:
+array (
+  0 => 'i',
+  1 => 'k',
+  2 => 'y',
+)
+ Result: true
+
+Test 16.ca0e38a2e3147dd97070f2128f140934:
+array (
+  0 => 'abc',
+  1 => 'ааа',
+  2 => 'абв',
+  3 => 'абг',
+)
+ Result: true
+
+Test 17.91480b10473a0c96a4cd6d88c23c577a:
+array (
+  0 => 'а',
+  1 => 'аа',
+  2 => 'ааа',
+)
+ Result: true
+
+Test 18.fdd3fe3981476039164aa000bf9177f2:
+array (
+  0 => 'i',
+  1 => 'y',
+  2 => 'k',
+)
+ Result: true

ext/intl/tests/collator_sort_with_sort_keys.phpt

@@ -0,0 +1,189 @@
+--TEST--
+sort_with_sort_keys()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Sort arrays using various locales.
+ */
+
+
+$test_num = 1;
+
+/*
+ * Sort arrays in the given list using specified locale.
+ */
+function sort_arrays( $locale, $arrays )
+{
+    $res_str = '';
+
+    $coll = ut_coll_create( $locale );
+
+    foreach( $arrays as $array )
+    {
+        // Sort array values
+        $res_val = ut_coll_sort_with_sort_keys( $coll, $array );
+
+        // Concatenate the sorted array and function result
+        // with output string.
+        $res_dump = "\n" . dump( $array ) .
+                    "\n Result: " . dump( $res_val );
+        
+        
+        // Preppend test signature to output string
+        $md5 = md5( $res_dump );
+
+        global $test_num;
+
+        $res_str .= "\n\n".
+                    "Test $test_num.$md5:" .
+                    $res_dump;
+        ++$test_num;
+    }
+
+    return $res_str;
+}
+
+
+function ut_main()
+{
+    global $test_num;
+    $test_num = 1;
+    $res_str = '';
+
+    // Sort an array in SORT_REGULAR mode using en_US locale.
+    $test_params = array(
+        array( 'abc', 'abd', 'aaa' ),
+        array( 'm'  , '1'  , '_'   ),
+        array( 'a'  , 'aaa', 'aa'  ),
+        array( 'ba' , 'b'  , 'ab'  ),
+        array( 'e'  , 'c'  , 'a'   ),
+        array( 'd'  , ''   , ' a'  ),
+        array( 'd ' , 'f ' , ' a'  ),
+        array( 'a'  , null , '3'   ),
+        array( 'y'  , 'i'  , 'k'   )
+    );
+
+    $res_str .= sort_arrays( 'en_US', $test_params );
+
+    // Sort a non-ASCII array using ru_RU locale.
+    $test_params = array(
+        array( 'абг', 'абв', 'ааа', 'abc' ),
+        array( 'аа', 'ааа', 'а' )
+    );
+
+    $res_str .= sort_arrays( 'ru_RU', $test_params );
+
+    // Array with data for sorting.
+    $test_params = array(
+        array( 'y'  , 'i'  , 'k'   )
+    );
+
+    // Sort an array using Lithuanian locale.
+    $res_str .= sort_arrays( 'lt_LT', $test_params );
+
+    return $res_str . "\n";
+}
+
+include_once( 'ut_common.inc' );
+ut_run();
+?>
+--EXPECT--
+Test 1.e8f1cd28133d79ecd660002f1c660d0e:
+array (
+  0 => 'aaa',
+  1 => 'abc',
+  2 => 'abd',
+)
+ Result: true
+
+Test 2.c2ded12173dd2996927378cae37eb275:
+array (
+  0 => '_',
+  1 => '1',
+  2 => 'm',
+)
+ Result: true
+
+Test 3.54071c968d71cb98c5d379145f8d7d38:
+array (
+  0 => 'a',
+  1 => 'aa',
+  2 => 'aaa',
+)
+ Result: true
+
+Test 4.19abe63d6f6dfef65b0e3c9ab4826b07:
+array (
+  0 => 'ab',
+  1 => 'b',
+  2 => 'ba',
+)
+ Result: true
+
+Test 5.9a8dc0a9bc771368c2f1fc3d02754610:
+array (
+  0 => 'a',
+  1 => 'c',
+  2 => 'e',
+)
+ Result: true
+
+Test 6.923d65739c5219c634616ffd100a50e4:
+array (
+  0 => '',
+  1 => ' a',
+  2 => 'd',
+)
+ Result: true
+
+Test 7.289bc2f28e87d3201ec9d7e8477ae1b0:
+array (
+  0 => ' a',
+  1 => 'd ',
+  2 => 'f ',
+)
+ Result: true
+
+Test 8.de0fd958484f2377a645835d7fbcf124:
+array (
+  0 => NULL,
+  1 => '3',
+  2 => 'a',
+)
+ Result: true
+
+Test 9.dd2b8f0adb37c45d528cad1a0cc0f361:
+array (
+  0 => 'i',
+  1 => 'k',
+  2 => 'y',
+)
+ Result: true
+
+Test 10.ca0e38a2e3147dd97070f2128f140934:
+array (
+  0 => 'abc',
+  1 => 'ааа',
+  2 => 'абв',
+  3 => 'абг',
+)
+ Result: true
+
+Test 11.91480b10473a0c96a4cd6d88c23c577a:
+array (
+  0 => 'а',
+  1 => 'аа',
+  2 => 'ааа',
+)
+ Result: true
+
+Test 12.fdd3fe3981476039164aa000bf9177f2:
+array (
+  0 => 'i',
+  1 => 'y',
+  2 => 'k',
+)
+ Result: true

ext/intl/tests/dateformat_format.phpt

@@ -0,0 +1,297 @@
+--TEST--
+datefmt_format_code()
+--SKIPIF--
+<?php if( !extension_loaded( 'intl' ) ) print 'skip'; ?>
+--FILE--
+<?php
+
+/*
+ * Test for the datefmt_format  function
+ */
+
+
+function ut_main()
+{
+	$locale_arr = array (
+		'en_US'
+	);
+	
+	$datetype_arr = array (
+                IntlDateFormatter::FULL,
+                IntlDateFormatter::LONG,
+                IntlDateFormatter::MEDIUM,
+                IntlDateFormatter::SHORT,
+                IntlDateFormatter::NONE
+        );
+
+        $res_str = '';
+
+
+	$time_arr = array (
+		0,
+		-1200000,
+		1200000,
+		2200000000,
+		-2200000000,
+		90099999,
+		3600,
+		-3600
+	);
+
+	$localtime_arr1 = array (
+	    'tm_sec' => 24 ,
+	    'tm_min' => 3,
+	    'tm_hour' => 19,
+	    'tm_mday' => 3,
+	    'tm_mon' => 3,
+	    'tm_year' => 105,
+	    'tm_wday' => 0,
+	    'tm_yday' => 93,
+	    'tm_isdst' => 1
+	);
+	$localtime_arr2 = array (
+	    'tm_sec' => 24 ,
+	    'tm_min' => 3,
+	    'tm_hour' => 3,
+	    'tm_mday' => 3,
+	    'tm_mon' => 3,
+	    'tm_year' => 205,
+	    'tm_wday' => 5,
+	    'tm_yday' => 93,
+	    'tm_isdst' => 1
+	);
+	$localtime_arr3 = array (
+            'tm_sec' => 24 ,
+            'tm_min' => 3,
+            'tm_hour' => 3,
+            'tm_mday' => 3,
+            'tm_mon' => 3,
+            'tm_year' => -5,
+            'tm_wday' => 3,
+            'tm_yday' => 93,
+            'tm_isdst' => 1
+        );
+
+	$localtime_arr = array (
+		$localtime_arr1,
+		$localtime_arr2,
+		$localtime_arr3
+	);
+
+	//Test format with input as a timestamp : integer
+	foreach( $time_arr as $timestamp_entry){
+		$res_str .= "\n------------\n";
+		$res_str .= "\nInput timestamp is : $timestamp_entry";
+		$res_str .= "\n------------\n";
+		foreach( $locale_arr as $locale_entry ){
+			foreach( $datetype_arr as $datetype_entry )
+	{
+		$res_str .= "\nIntlDateFormatter locale= $locale_entry ,datetype = $datetype_entry ,timetype =$datetype_entry ";
+		//$fmt = ut_datefmt_create( $locale_entry , $datetype_entry ,$datetype_entry,'America/Los_Angeles', IntlDateFormatter::GREGORIAN  ,"dd/mm/yyyy");
+		$fmt = ut_datefmt_create( $locale_entry , $datetype_entry ,$datetype_entry,'America/Los_Angeles', IntlDateFormatter::GREGORIAN  );
+		//$fmt = ut_datefmt_create( $locale_entry , $datetype_entry ,$datetype_entry);
+		$formatted = ut_datefmt_format( $fmt , $timestamp_entry);
+		$res_str .= "\nFormatted timestamp is : $formatted";
+	}
+	}
+	}
+
+	//Test format with input as a localtime :array
+	foreach( $localtime_arr as $localtime_entry){
+		$res_str .= "\n------------\n";
+		$res_str .= "\nInput localtime is : ";
+		foreach( $localtime_entry as $key => $value){
+                    $res_str .= "$key : '$value' , ";
+		}
+
+		$res_str .= "\n------------\n";
+		foreach( $locale_arr as $locale_entry ){
+			foreach( $datetype_arr as $datetype_entry )
+	{
+		$res_str .= "\nIntlDateFormatter locale= $locale_entry ,datetype = $datetype_entry ,timetype =$datetype_entry ";
+		$fmt = ut_datefmt_create( $locale_entry , $datetype_entry ,$datetype_entry);
+		$formatted1 = ut_datefmt_format( $fmt , $localtime_entry);
+		if( intl_get_error_code() == U_ZERO_ERROR){
+			$res_str .= "\nFormatted localtime_array is : $formatted1";
+		}else{
+			$res_str .= "\nError while formatting as: '".intl_get_error_message()."'";
+		}
+	}
+	}
+	}
+
+	return $res_str;
+
+}
+
+include_once( 'ut_common.inc' );
+
+// Run the test
+ut_run();
+?>
+--EXPECT--
+------------
+
+Input timestamp is : 0
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted timestamp is : Wednesday, December 31, 1969 4:00:00 PM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted timestamp is : December 31, 1969 4:00:00 PM PST
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted timestamp is : Dec 31, 1969 4:00:00 PM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted timestamp is : 12/31/69 4:00 PM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted timestamp is : 19691231 04:00 PM
+------------
+
+Input timestamp is : -1200000
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted timestamp is : Wednesday, December 17, 1969 6:40:00 PM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted timestamp is : December 17, 1969 6:40:00 PM PST
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted timestamp is : Dec 17, 1969 6:40:00 PM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted timestamp is : 12/17/69 6:40 PM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted timestamp is : 19691217 06:40 PM
+------------
+
+Input timestamp is : 1200000
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted timestamp is : Wednesday, January 14, 1970 1:20:00 PM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted timestamp is : January 14, 1970 1:20:00 PM PST
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted timestamp is : Jan 14, 1970 1:20:00 PM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted timestamp is : 1/14/70 1:20 PM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted timestamp is : 19700114 01:20 PM
+------------
+
+Input timestamp is : 2200000000
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted timestamp is : Sunday, September 18, 2039 4:06:40 PM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted timestamp is : September 18, 2039 4:06:40 PM PDT
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted timestamp is : Sep 18, 2039 4:06:40 PM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted timestamp is : 9/18/39 4:06 PM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted timestamp is : 20390918 04:06 PM
+------------
+
+Input timestamp is : -2200000000
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted timestamp is : Saturday, April 14, 1900 5:53:20 PM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted timestamp is : April 14, 1900 5:53:20 PM PDT
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted timestamp is : Apr 14, 1900 5:53:20 PM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted timestamp is : 4/14/00 5:53 PM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted timestamp is : 19000414 05:53 PM
+------------
+
+Input timestamp is : 90099999
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted timestamp is : Wednesday, November 8, 1972 11:46:39 AM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted timestamp is : November 8, 1972 11:46:39 AM PST
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted timestamp is : Nov 8, 1972 11:46:39 AM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted timestamp is : 11/8/72 11:46 AM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted timestamp is : 19721108 11:46 AM
+------------
+
+Input timestamp is : 3600
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted timestamp is : Wednesday, December 31, 1969 5:00:00 PM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted timestamp is : December 31, 1969 5:00:00 PM PST
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted timestamp is : Dec 31, 1969 5:00:00 PM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted timestamp is : 12/31/69 5:00 PM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted timestamp is : 19691231 05:00 PM
+------------
+
+Input timestamp is : -3600
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted timestamp is : Wednesday, December 31, 1969 3:00:00 PM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted timestamp is : December 31, 1969 3:00:00 PM PST
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted timestamp is : Dec 31, 1969 3:00:00 PM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted timestamp is : 12/31/69 3:00 PM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted timestamp is : 19691231 03:00 PM
+------------
+
+Input localtime is : tm_sec : '24' , tm_min : '3' , tm_hour : '19' , tm_mday : '3' , tm_mon : '3' , tm_year : '105' , tm_wday : '0' , tm_yday : '93' , tm_isdst : '1' , 
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted localtime_array is : Sunday, April 3, 2005 7:03:24 PM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted localtime_array is : April 3, 2005 7:03:24 PM PDT
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted localtime_array is : Apr 3, 2005 7:03:24 PM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted localtime_array is : 4/3/05 7:03 PM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted localtime_array is : 20050403 07:03 PM
+------------
+
+Input localtime is : tm_sec : '24' , tm_min : '3' , tm_hour : '3' , tm_mday : '3' , tm_mon : '3' , tm_year : '205' , tm_wday : '5' , tm_yday : '93' , tm_isdst : '1' , 
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted localtime_array is : Friday, April 3, 2105 3:03:24 AM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted localtime_array is : April 3, 2105 3:03:24 AM PDT
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted localtime_array is : Apr 3, 2105 3:03:24 AM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted localtime_array is : 4/3/05 3:03 AM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted localtime_array is : 21050403 03:03 AM
+------------
+
+Input localtime is : tm_sec : '24' , tm_min : '3' , tm_hour : '3' , tm_mday : '3' , tm_mon : '3' , tm_year : '-5' , tm_wday : '3' , tm_yday : '93' , tm_isdst : '1' , 
+------------
+
+IntlDateFormatter locale= en_US ,datetype = 0 ,timetype =0 
+Formatted localtime_array is : Wednesday, April 3, 1895 3:03:24 AM PT
+IntlDateFormatter locale= en_US ,datetype = 1 ,timetype =1 
+Formatted localtime_array is : April 3, 1895 3:03:24 AM PDT
+IntlDateFormatter locale= en_US ,datetype = 2 ,timetype =2 
+Formatted localtime_array is : Apr 3, 1895 3:03:24 AM
+IntlDateFormatter locale= en_US ,datetype = 3 ,timetype =3 
+Formatted localtime_array is : 4/3/95 3:03 AM
+IntlDateFormatter locale= en_US ,datetype = -1 ,timetype =-1 
+Formatted localtime_array is : 18950403 03:03 AM