Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:19896
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Message-ID: <4369B150.2000205@fischer.name>
Date: Thu, 03 Nov 2005 07:42:24 +0100
User-Agent: Mozilla Thunderbird 1.0 (Windows/20041206)
MIME-Version: 1.0
To: Matteo Beccati <php@beccati.com>
CC: internals@lists.php.net
References: <4369513B.1020502@beccati.com>
In-Reply-To: <4369513B.1020502@beccati.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [PHP-DEV] What about a PHP 4.4.1pl1?
From: markus@fischer.name (Markus Fischer)

Matteo Beccati wrote:
> I newly subscribed to this ML after some time I didn't follow it, so I 
> beg pardon if I'm posting a question that was already discussed.
> 
> As you know, a few days ago PHP 4.4.1 was released. As current admin of 
> a widely used PHP project called phpAdsNew, I discovered just today that 
> there are compatibility problems with the latest PHP4 release. One of 
> our users preceded me and opened a bug report, which led to a prompt fix 
> by Ilia:
> 
> http://bugs.php.net/bug.php?id=35067
> 
> ( and its duplicate: http://bugs.php.net/bug.php?id=35070 )

The problem is also that the ChangeLog has no mention of this change of 
behaviour because it seems this change was done without having a bug to 
fix (at least not a bug in the database).

This change was first made in HEAD ( 
http://cvs.php.net/diff.php/php-src/ext/standard/basic_functions.c?r1=1.732&r2=1.733&ty=u 
) and then merged to all branches. But it doesn't reference a bug and 
doesn't describe what it tries to fix (" Make key() and current() take 
argument by value." doesn't explain much).

I also see that the initial change happened in HEAD, 5.1, 5.0 and 4.4 
but the revert only happened in HEAD, 5.1 and 4.4. Was 5.0 intentionally 
left out?

The interesting question is: why was this commited to a branch which 
should take only bug fixes when:
- there was no bug reference
- no real description what buggy behaviour it tries to fix
?

What I found on the list was a discussion started by Dimitry and the 
explaination was that "... makes it possible to pass return values of 
functions that return arrays.". "Makes is possible" doesn't sound like 
fixing a real bug to me.

- Markus