Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:29822
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain gmail.com designates 66.249.92.169 as permitted sender)
DomainKey-Status: bad
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=beta;
        h=received:message-id:date:from:user-agent:mime-version:to:subject:content-type:content-transfer-encoding;
        b=CLleCHmRyTPbTcZB3+lhfXuQsD5VH8D0vRJqRdBNHjYUaELtZVL/lWCTHGzm34fzrBM6cVNtXOEznoAUnHTvAK4NLL2DS2Qvx994SYgOZ0ktyCcYf+WiOuNrDHpf+gjgR0lwvVrvBRmLCD16xsyUFWcO0wUf2UclAMmTPZEIFF8=
Message-ID: <4658DD8A.8050004@gmail.com>
Date: Sun, 27 May 2007 03:23:22 +0200
User-Agent: Thunderbird 2.0.0.0 (Windows/20070326)
MIME-Version: 1.0
To:  internals@lists.php.net
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Subject: RE: Fixing PECL + Core
From: giunta.gaetano@gmail.com (Gaetano Giunta)

Just to add my experience (even though the original poster explicitly 
asked for single extensions not to be mentioned) from a 
not-completely-unrelated problem: some extensions have an "internal" 
version number that is not always updated when the userland API of said 
extension is changed - see eg. Json and the recent change of behavior on 
decoding scalar values.

This makes it very hard for people maintaining php libraries to code 
defensively and test the version in use to enable workarounds:
- the extension version number is useless (if not managed correctly)
- the php version number is useless, since the extension might have been 
downloaded and compiled from pecl and not correspond to the version that 
is distributed with the core

The only option left in this situation is to test the functionality 
needed first, then use it, much as it is done in js - a very sad state 
of things... (of course I would not recommend disabling upgrade / 
backport of a php extension from pecl as a solution)

The fact that the extension lives in both pecl and core and the two 
might be slightly out of sync does not help at all when trying to find 
out the cause of regressions...

Bye
Gaetano