Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:55802
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain akbkhome.com designates 202.81.246.113 as permitted sender)
Message-ID: <4E979AA4.6060001@akbkhome.com>
Date: Fri, 14 Oct 2011 10:12:52 +0800
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.21) Gecko/20110831 Icedove/3.1.13
MIME-Version: 1.0
To: Rasmus Lerdorf <rasmus@lerdorf.com>
CC: internals@lists.php.net
References: <4E969596.4090704@akbkhome.com> <4E970257.2010906@sugarcrm.com> <4E977A4B.4020609@akbkhome.com> <4E977D07.4010503@lerdorf.com>
In-Reply-To: <4E977D07.4010503@lerdorf.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [PHP-DEV] is_a fix for 5.4 and HEAD
From: alan@akbkhome.com (Alan Knowles)

On Friday, October 14, 2011 08:06 AM, Rasmus Lerdorf wrote:
> On 10/13/2011 04:54 PM, Alan Knowles wrote:
>> On Thursday, October 13, 2011 11:23 PM, Stas Malyshev wrote:
>>> On 10/13/11 12:39 AM, Alan Knowles wrote:
>>>> Can someone apply this to HEAD and PHP_5_4, or let me have karma.
>>>>
>>>> https://bugs.php.net/patch-display.php?bug_id=55475&patch=final_patch_for_5_4_and_HEAD&revision=latest
>>>>
>>>>
>>>> Thanks
>>>> Alan
>>>>
>>>>
>>> +/* {{{ proto bool is_subclass_of(mixed object_or_string, string
>>> class_name [, bool allow_string=true])
>>> +/* {{{ proto bool is_a(mixed object_or_string, string class_name [,
>>> bool allow_string=false])
>>>
>>> This is quite strange - two almost identical functions with completely
>>> different default semantics.
>> Yes, and if you can think of a way to depreciate them now so that it
>> does not break BC and could be changed later, please feel free, but I've
>> looked at it quite a bit and could not find any other way.
> Argh! Pet-peeve, it's deprecate. Depreciate means something completely
> different.
Agh to much accounting recently ;)
> I agree that it is slightly messy, but we have painted ourselves into a
> bit of a corner with the 5.3 mess. Stas, the whole point here is that
> changing the is_a() default in 5.3 caused huge problems, including
> security ones, so setting allow_string to false by default fixes that BC
> break in 5.3. Considering the huge amount of code hit by this I think we
> should keep that default in 5.4.
>
> Alan, does is_subclass_of() actually need the allow_string flag at all?
> It has always been documented take a mixed and hit the autoloader, so
> why would we need an option to disable the string?
I don't think I've used is_subclass_of for years, the flag is a knock on 
effect from sharing the same code, and may be useful in some edge cases 
(if somebody wanted to use subclass_of with user input.. ), since that 
bit does not affect me, I do not mind either way.

To disable for subclass of it is quite trivial, I can get a full patch 
ready (including updated tests) if we go that way.

if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, only_subclass ? 
"zs" : "zs|b", &obj, &class_name, &class_name_len, &allow_string) == 
FAILURE) {

Regards
Alan



> -Rasmus
>