Re: Fix for #53727 (is_subclass_of resolution with interfaces)

Hey Paul, The only thing we are attempting to do here is make is_subclass_of() more consistent when using class names (strings), something instanceof cannot handle. Given the following: interface A {} class B implements A {} class C extends B {} // currently false, and correct as B is not a subclass of B var_dump(is_subclass_of('B', 'B')); // currently false, but SHOULD be true var_dump(is_subclass_of('B', 'A')); // true since C's parent class is B var_dump(is_subclass_of('C', 'B')); // currently true, but inconsistent with is_subclass_of('B', 'A') // why should having a parent matter? var_dump(is_subclass_of('C', 'A')); Currently, this (pseudo-code) is the workaround in ZF2 we are finding ourselves go with: function someFunc($class, $type) { return (array_key_exists(

    $type,
    (class_parents($class, true) + class_implements($class, true)

)); } -ralph On 6/29/11 3:20 PM, Paul Dragoonis wrote:

On Wed, Jun 29, 2011 at 8:49 PM, Ralph Schindler<ralph@smashlabs.com> wrote:

Correct. I was hasty in that example, the first was copied& tested (and is reflected in the test, as is that variation of what I wrote up.) Either way, test and patch work in 5_3.

Doesn't this functionality confuse matters? If this patch is added, is there now no difference between instanceof and is_subclass_of(). If this is the case my question is then why do we have two methods to do the same thing? I thought instanceof was for parent classes + interfaces.. and is_subclass_of() was just for parent classes. Regards, Paul Dragoonis.

Thanks, -ralph On 6/29/11 2:42 PM, David Zülke wrote:

On 29.06.2011, at 21:39, Ralph Schindler wrote:

interface A {} class B implements A {} class C extends B {} var_dump(is_subclass_of('B', 'A')); // true

Typo there; that should be 'C', not 'B'. David

-- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php