Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:44171
Return-Path: <rasmus@lerdorf.com>
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Delivered-To: mailing list internals@lists.php.net
Received: (qmail 71878 invoked from network); 5 Jun 2009 05:47:30 -0000
Received: from unknown (HELO lists.php.net) (127.0.0.1)
  by localhost with SMTP; 5 Jun 2009 05:47:30 -0000
Authentication-Results: pb1.pair.com smtp.mail=rasmus@lerdorf.com; spf=permerror; sender-id=unknown
Authentication-Results: pb1.pair.com header.from=rasmus@lerdorf.com; sender-id=unknown
Received-SPF: error (pb1.pair.com: domain lerdorf.com from 209.85.216.192 cause and error)
X-PHP-List-Original-Sender: rasmus@lerdorf.com
X-Host-Fingerprint: 209.85.216.192 mail-px0-f192.google.com  
Received: from [209.85.216.192] ([209.85.216.192:64353] helo=mail-px0-f192.google.com)
	by pb1.pair.com (ecelerity 2.1.1.9-wez r(12769M)) with ESMTP
	id 85/A0-00788-F61B82A4 for <internals@lists.php.net>; Fri, 05 Jun 2009 01:47:29 -0400
Received: by pxi30 with SMTP id 30so270522pxi.29
        for <internals@lists.php.net>; Thu, 04 Jun 2009 22:47:25 -0700 (PDT)
Received: by 10.114.177.9 with SMTP id z9mr4921196wae.103.1244180845304;
        Thu, 04 Jun 2009 22:47:25 -0700 (PDT)
Received: from ?216.145.54.158? (socks1.corp.yahoo.com [216.145.54.158])
        by mx.google.com with ESMTPS id n40sm12283199wag.30.2009.06.04.22.47.23
        (version=TLSv1/SSLv3 cipher=RC4-MD5);
        Thu, 04 Jun 2009 22:47:24 -0700 (PDT)
Message-ID: <4A28B16A.2040101@lerdorf.com>
Date: Thu, 04 Jun 2009 22:47:22 -0700
User-Agent: Thunderbird 2.0.0.18 (Macintosh/20081105)
MIME-Version: 1.0
To: Roman I <isroman@km.ru>
CC: internals@lists.php.net
References: <94.6F.05315.44EF11A4@pb1.pair.com> <77E6891A-AF77-47B5-94DB-FA69D4F7F198@roshambo.org> <DF.60.63575.1ED812A4@pb1.pair.com> <A52B781A-28BD-4609-811A-9C014E58F9F9@roshambo.org> <B5.6A.44672.BD4982A4@pb1.pair.com>
In-Reply-To: <B5.6A.44672.BD4982A4@pb1.pair.com>
X-Enigmail-Version: 0.95.7
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: Re: [PHP-DEV] Is there a technical reason <?= considered a "short
 tag"?
From: rasmus@lerdorf.com (Rasmus Lerdorf)

Roman I wrote:
> Philip Olson wrote:
>> On May 30, 2009, at 12:49 PM, Roman I wrote:
>>
>>> Philip Olson wrote:
>>>> On May 18, 2009, at 5:33 PM, Roman I wrote:
>>> This neither answers nor invalidates the original question, though.
>> Unfortunately I do not remember everything about this but do know that:
>>
>>    - Decoupling <?= from short_open_tag was discussed, and did not
>> happen
>>    - Allowing it to be enabled at runtime was discussed, and did not
>> happen
>>    - The issues with <?= and <? are similar... an invalid PI
>>
>> I've not read every thread or discussion on this (there are many) but
>> CVS speaks for itself. Here's one thought comparing the two viewpoints:
>>
>>    - http://marc.info/?l=php-internals&m=120614525214419&w=2
>>
>> The following RFC mentions this decoupling, yet was declined:
>>
>>    - http://wiki.php.net/rfc/shortags
>>
>> And lastly, the general topic of short tags came up recently while
>> several people discussed PHP 6:
>>
>>    - http://wiki.php.net/summits/pdmnotesmay09
>>
>> In summary: It's a topic people don't want to talk about. :)
>>
>> Regards,
>> Philip
> 
> Thanks for the references. I've seen the RFC, but I haven't seen that thread
> you've linked to. It does contain almost all things I could say here.
> However, I haven't found very strong argument against decoupling in that
> thread either. :)
> 
> Even without short tags, PHP will not be a well-formed XML processing
> instruction in many cases. For example,
> <?php $x = '<?>'; ?>
> or, more commonly:
> <root test="<?php echo $value; ?>" />
> 
> Plain <? has an additional problem of affecting the ability to include XML
> files from PHP source code. However, it seems to me that using the syntax
> above will have exactly the same consequences for XML well-formedness as
> using <?=. Since the syntax above is allowed with short_open_tags off,
> disallowing  <?= with that setting off looks like an overkill. That's my
> reasoning behind starting this thread, anyway.
> 
> Admittedly, I haven't said anything new here, but I've rephrased an old
> argument in a nicer way. :)Maybe it will do some good.

Right, I think that is actually the conclusion we came to at one point.
 I don't remember the discussion that caused the RFC to be declined.  I
would be in favour of reversing that decision wherever it came from.

Nobody is going to pass a PHP script through an XML parser, and PHP
itself will never be well-formed, so that is a lost cause.  We'd have to
start tossing CDATA blocks and write code like:

  if( 2 &lt; 3 ) echo "2 is less than 3";

And that just isn't going to happen.  The only valid argument against
short_tags itself is that it clashes with named PI tags and <?= does not
have that same problem.

-Rasmus