# Right now i see this as the best option but we might shuffle code around # later if someone comes up with a better solution.
add new function MFB will follow soon
. zend_exception_get_default() -> zend_exception_get_default(TSRMLS_D) . zend_get_error_exception() -> zend_get_error_exception(TSRMLS_D) . added E_RECOVERABLE_ERROR . added ZEND_TOSTRING_FUNC_NAME . added __tostring function cache to zend_class_entry . added ZEND_NAMED_ME . modified ZEND_ME_MAPPING to support method flags . added ZEND_MN . method entries now use prefix "zim_" instead of "zif_" . drop EG(ze1_compatibility_mode) . changed cast handler, now without (int should_free): typedef int (*zend_object_cast_t)(zval *readobj, zval *retval, int type TSRMLS_DC); . changed get_iterator, now receives whether value is by ref: zend_object_iterator *(*get_iterator)(zend_class_entry *ce, zval *object, int by_ref TSRMLS_DC); . added zend_objects_store_add_ref_by_handle . added zend_objects_store_del_ref_by_handle . convert_to_explicit_type(pzv, type)
. Allow array($this, 'parent::method') for function 'pointers' . Spit out E_STRICT in case of erroneous use of function 'pointers'
…es on 64-bit platforms)
…lable and zend_make_callable on it. This functions allows to check if a php variable is a callable function and returns its function pointer as well as object if possible. # Commit this now so we can use it in 5.1.* series as discussed with Andi.
… to complement the property get/set methods)
…essages when working with non-PHP objects. # Using Z_OBJCE(object)->name is usually bad idea unless you know it's # a pure PHP object
…ndencies and conflicts.
functions for bool, double and binary safe string data
…LSE() used to return "" and not bool(false). It's not worth keeping it because STR_FREE() and zval_dtor() always have to check for it and it slows down the general case. In addition, it seems that empty_string has been abused quite a lot, and was used not only for setting zval's but generally in PHP code instead of "", which wasn't the intention. Last but not least, nuking empty_string should improve stability as I doubt every place correctly checked if they are not mistakenly erealloc()'ing it or calling efree() on it. NOTE: Some code is probably broken. Each extension maintainer should check and see that my changes are OK. Also, I haven't had time to touch PECL yet. Will try and do it tomorrow.