Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:39579
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: unknown (pb1.pair.com: domain php.net does not designate 85.214.94.56 as permitted sender)
Date: Sun, 3 Aug 2008 23:25:01 +0200
Reply-To: Marcus Boerger <helly@php.net>
Message-ID: <1815659426.20080803232501@marcus-boerger.de>
To: "Robin Fernandes" <robinf@php.net>
CC: "Marcus Boerger" <helly@php.net>, internals <internals@lists.php.net>
In-Reply-To: <5a8807d10808031324k2f83789ak233d75ec704a6c31@mail.gmail.com>
References: <5a8807d10808030407j14f51cb6ga7d917d51c6206ac@mail.gmail.com> <93357385.20080803141751@marcus-boerger.de> <5a8807d10808031324k2f83789ak233d75ec704a6c31@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Subject: Re: [PHP-DEV] Intended behavior of ArrayObject::exchangeArray() ?
From: helly@php.net (Marcus Boerger)

Hello Robin,

Sunday, August 3, 2008, 10:24:17 PM, you wrote:

> Hi Marcus!

>>> 1. What should happen when the argument is an object?
>>
>> Seems like an error message is missing there. It allows to take an instance
>> of another ArrayObject/Iterator and use the array from that.
>>
>> In case any other Object is passed it is ignored. What do you feel?

> Sounds alright to me, but it's inconsistent with __construct(), which
> in addition to using the storage array from ArrayObject/Iterator
> instances, can use the properties from any other kind of object. Also,
> just to be clear, I should point out that it is not the current
> behaviour. Passing any object to exchangeArray(), even an instance of
> ArrayObject/Iterator, eradicates the container:

> <?php
> $ao1 = new ArrayObject(array('one'));
> $ao2 = new ArrayObject(array('two'));
> $ao1->exchangeArray($ao2);
> var_dump($ao1);
?>>

> Output on PHP 5.3.0-dev (cli) (built: Jul 28 2008 00:21:59):
> object(ArrayObject)#1 (0) {
> }


>>> Here's a suggested patch against 5_3 that implements this (and
>>> includes some tests): http://pastebin.ca/1091668
>>> Note that I'm not an internals or SPL expert, so let me know if I
>>> missed something.
>>
>> Seems about right. Can you attach the patch as a text file (I cannot open
>> the link).

> Attached. I just naively moved some logic from __construct() into a
> separate function and called it from both __construct() and
> exchangeArray().

Well, naively or not, this seems about perfectly correct. Why don't you
simply apply this to HEAD and 5.3?

Best regards,
 Marcus