Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:2719
Return-Path: <sniper@iki.fi>
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Delivered-To: mailing list internals@lists.php.net
Received: (qmail 5254 invoked from network); 24 Jun 2003 15:35:35 -0000
Received: from unknown (HELO netphobia.fi) (213.243.181.8)
  by pb1.pair.com with SMTP; 24 Jun 2003 15:35:35 -0000
Received: from localhost (jani@localhost)
	by netphobia.fi (8.11.6/8.11.6) with ESMTP id h5OFZU028341;
	Tue, 24 Jun 2003 18:35:31 +0300
X-Authentication-Warning: netphobia.fi: jani owned process doing -bs
Date: Tue, 24 Jun 2003 18:35:30 +0300 (EEST)
Sender: jani@netphobia.fi
Reply-To: Jani Taskinen <sniper@iki.fi>
To: Lars Torben Wilson <torben@php.net>
cc: internals@lists.php.net
In-Reply-To: <1056223169.965.68.camel@ali>
Message-ID: <Pine.LNX.4.44.0306241832220.27459-100000@netphobia.fi>
References: <1056223169.965.68.camel@ali>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: Re: [PHP-DEV] Array behaviour change intentional?
From: sniper@iki.fi (Jani Taskinen)

On Sat, 21 Jun 2003, Lars Torben Wilson wrote:

>Hi there,
>
>It seems that array appending has changed wrt negative indices. I had
>just documented the new behaviour but now I'm not entirely sure whether 
>I've just documented an unintentional change. Is this new behaviour what
>is desired (i.e. skipping straight from some negative key to 0 when
>appending with $arr[] = <expr>).
>
>
>
>Thanks for any insight,
>
>
>Torben
>
>
>Test script:
><?php
>$arr[-5] = "minus five";
>$arr[] = "minus four";
>var_dump($arr);
>?>
>
>
>Output PHP 4.2.1:
>array(2) {
>  [-5]=>
>  string(10) "minus five"
>  [-4]=>
>  string(10) "minus four"
>}

Output PHP 4.2.3:
array(2) {
  [-5]=>
  string(10) "minus five"
  [-4]=>
  string(10) "minus four"
}

Output PHP 4.3.1:
array(2) {
  [-5]=>
  string(10) "minus five"
  [0]=>
  string(10) "minus four"
}

>
>Output PHP 4.3.2:
>array(2) {
>  [-5]=>
>  string(10) "minus five"
>  [0]=>
>  string(10) "minus four"
>}

    Seems to have been changed in between 4.2.3 and 4.3.0.
    There is/was some bug reports about this too, iirc.
    
    Not very critical, I don't think there are that many scripts
    using this?
    
    --Jani