Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:61140
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain gmail.com designates 209.85.160.170 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <op.what7eicidpuyk@damnation>
References: <4FFD11C1.3090007@sugarcrm.com>
	<1342029955.13569.17.camel@guybrush>
	<4FFDC10F.6050106@sugarcrm.com>
	<CAEZPtU7Ntim8Oo1HCJ=3+KceB9t2GRaB=w9pgmcOLaLJAgjomw@mail.gmail.com>
	<4FFDDBAD.7070100@sugarcrm.com>
	<CAEZPtU6TqQ5Vez+8UPd3vfwMxVHhADhxhwjhLR5WzSo7O7SpAA@mail.gmail.com>
	<4FFDDF66.2070806@sugarcrm.com>
	<op.what7eicidpuyk@damnation>
Date: Wed, 11 Jul 2012 22:43:14 +0200
Message-ID: <CAEZPtU5ZeRQ_t6P9b+rJYBwP7PDzddck_CzZ5cWgzXEMm62Vvg@mail.gmail.com>
To: Gustavo Lopes <glopes@nebm.ist.utl.pt>
Cc: Stas Malyshev <smalyshev@sugarcrm.com>, =?ISO-8859-1?Q?Johannes_Schl=FCter?= <johannes@schlueters.de>, 
	PHP Internals <internals@lists.php.net>
Content-Type: text/plain; charset=ISO-8859-1
Subject: Re: [PHP-DEV] Re: bug 18556 - tolower & locales
From: pierre.php@gmail.com (Pierre Joye)

hi,

On Wed, Jul 11, 2012 at 10:36 PM, Gustavo Lopes <glopes@nebm.ist.utl.pt> wrote:
> Em Wed, 11 Jul 2012 22:17:42 +0200, Stas Malyshev <smalyshev@sugarcrm.com>
> escreveu:
>
>
>> And we regularly fix bugs in such functions in stable versions.
>>
>>> If it is not good enough for 5.3, it is not good enough for 5.4.
>>
>>
>> Sorry, I can't accept this argument. There were numerous things that are
>> in 5.4 but not in 5.3.
>
>
> I also think this is not appropriate for 5.4, but for different reasons.
>
> You're changing the Zend API. You introduce new public functions. This means
> an extension that uses the new symbols will be compatible with PHP 5.4.6,
> but not before. When this happens, it's very annoying (like the PHP_FE_END
> in 5.3, though that only affected source compatibility -- this also affects
> binary compatibility).
>

That alone is a no go for 5.4 and 5.3.


> You also change the behavior of zend_binary_strncasecmp(). Just like you had
> to change ext/standard/string.c, other extensions may also have to (not to
> mention the risk you missed something).

That adds way too much risks, for little to no gain (sorry for the few
people suffering from this bug).

cheers,
-- 
Pierre

@pierrejoye | http://blog.thepimp.net | http://www.libgd.org