Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:7053
Return-Path: <cschneid@cschneid.com>
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Delivered-To: mailing list internals@lists.php.net
Received: (qmail 87576 invoked by uid 1010); 13 Jan 2004 16:31:12 -0000
Delivered-To: ezmlm-scan-internals@lists.php.net
Delivered-To: ezmlm-internals@lists.php.net
Received: (qmail 87531 invoked from network); 13 Jan 2004 16:31:11 -0000
Received: from unknown (HELO xaxa.search.ch) (195.141.85.117)
  by pb1.pair.com with SMTP; 13 Jan 2004 16:31:11 -0000
Received: from localhost (localhost [127.0.0.1])
	by xaxa.search.ch (Postfix) with ESMTP
	id 5AB206DB1F; Tue, 13 Jan 2004 17:31:11 +0100 (CET)
Received: by xaxa.search.ch (Postfix, from userid 65534)
	id 56B1D6DB24; Tue, 13 Jan 2004 17:31:10 +0100 (CET)
Received: from cschneid.com (ultrafilter-i [192.168.85.2])
	by xaxa.search.ch (Postfix) with ESMTP
	id 03D446DB1F; Tue, 13 Jan 2004 17:31:09 +0100 (CET)
Message-ID: <40041D4B.9060900@cschneid.com>
Date: Tue, 13 Jan 2004 17:31:07 +0100
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5) Gecko/20031009
X-Accept-Language: de-ch, en-us, en
MIME-Version: 1.0
To: Adam Trachtenberg <adam@trachtenberg.com>
Cc: Rob Richards <rrichards@ctindustries.net>,
	internals@lists.php.net
References: <Pine.LNX.4.58.0401130248410.21920@miranda.org> <00fe01c3d9d1$6e8618a0$f7dea8c0@cyberware.local> <4003FEFA.2090700@cschneid.com> <CAE69F12-45DE-11D8-B3DB-003065FB1208@trachtenberg.com>
In-Reply-To: <CAE69F12-45DE-11D8-B3DB-003065FB1208@trachtenberg.com>
X-Enigmail-Version: 0.76.7.0
X-Enigmail-Supports: pgp-inline, pgp-mime
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Spam-Checker-Version: SpamAssassin 2.61 (1.212.2.1-2003-12-09-exp) on 
	xaxa.search.ch
X-Spam-Level: 
X-Spam-Status: No, hits=0.1 required=5.0 tests=BAYES_44,HTML_MESSAGE 
	autolearn=ham version=2.61
X-Virus-Scanned: by AMaViS 0.3.12pre8
Subject: Re: [PHP-DEV] SimpleXML: Moving Forward
From: cschneid@cschneid.com (Christian Schneider)

Adam Trachtenberg wrote:
> Ugh. This is pretty much the limit of what I think is reasonable for 
> SimpleXML to handle. It think the API would be more consistent if the 

Agreed. I was just curious how it behaves if I push it to the limit :-)

 > However, I don't really see any way around this otherwise. Either it's
 >  general or not. It can't be both. (Unless there's some magical type
 > that's both an array and a scalar.) I'm willing to put up with this

Elements are already a magical type which is an object and an array.
Making foreach work on a both the scalar and array incantation of child 
elements seems very useful to me. And usefulness seems to be the goal of 
SimpleXML as far as I understand.

> maxOccurs attributes in the schema for an element. If maxOccurs > 1, 
> then the elements would be placed in an array even if there was only one 
> element in that particular instance.

I don't like the idea:  Different behaviour with or without schema. I 
write code without schema first. If I decide to add a schema later code 
has to be rewritten. Not good.

> Again, my opinion is that arbitrary XML documents are best parsed using 
> DOM and well-defined ones are best parsed using SimpleXML.

I agree but I think I'll have something like
<config>
   <option>1</option>
   <option>2</option>
...
</config>
with 0 - many options quite often and iterating should be a simple 
foreach IMHO.

> AFAIK, it's actually also impossible to find out the name of the 
> document element using SimpleXML, even using XPath.

Yup, right now the only way is probably to wrap it in a dummy tag before 
giving it to SimpleXML. Sounds like a good idea anyway to me, if I have 
a domain specific document with varying content I'd probably do
<domain>...</domain> anyway.

>> I would actually expect abcd but only once:
>> Node Value: abcd
>>
>> Concatenating all text parts _and_ returning them once for each part 
>> definitely seems wrong.
> 
> Aren't those two lines contradictory? :)

Why? Right now it returns abcd twice which is definitely wrong. 
Returning ab and cd or (preferably IMHO) abcd once seems right.

- Chris