Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:77482
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain gmail.com designates 209.85.216.176 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <BLU436-SMTP25081263DCD1A7897FB9134E2B30@phx.gbl>
References: <BLU436-SMTP25081263DCD1A7897FB9134E2B30@phx.gbl>
Date: Mon, 22 Sep 2014 21:50:45 +0200
Message-ID: <CAEZPtU6hmGg4AzER7ySn_v1DeTHFmiwHodc2Zmbq78d4Kohb7w@mail.gmail.com>
To: Bob Weinand <bobwei9@hotmail.com>
Cc: PHP internals <internals@lists.php.net>
Content-Type: text/plain; charset=UTF-8
Subject: Re: [PHP-DEV] The ABI and bug 68057
From: pierre.php@gmail.com (Pierre Joye)

On Mon, Sep 22, 2014 at 8:48 PM, Bob Weinand <bobwei9@hotmail.com> wrote:
> https://bugs.php.net/bug.php?id=68057
>
> is a regression introduced in PHP 5.6.
> So code which still worked with PHP 5.5 breaks in 5.6.
>
> The big issue is here, that the fix for this bug requires ABI breakage. (changing an unsigned short into a zend_uint)
>
> I know that ABI compatibility is a pretty serious issue here, but it also means that some perfectly valid PHP 5.5 code which is supposed to still work on 5.6 will silently break.
> So, is it worth to break ABI or does it have to wait for 5.7 or the next major?


Is the actual code used by the parser exposed? If yes it will be
somehow bad as ext maintainers relying on this will need two different
builds. I do not think there are many affected exts out there tho'. If
not, then it is easy :)

-- 
Pierre

@pierrejoye | http://www.libgd.org