Re: FORM content in DTD

newtonjs@vnet.net (Stan Newton)
Date: Thu, 13 Oct 94 21:54:27 EDT
Message-id: <199410140153.AA14333@char.vnet.net>
Reply-To: newtonjs@vnet.net
Originator: html-wg@oclc.org
Sender: html-wg@oclc.org
Precedence: bulk
From: newtonjs@vnet.net (Stan Newton)
To: Multiple recipients of list <html-wg@oclc.org>
Subject: Re: FORM content in DTD
X-Listprocessor-Version: 6.0c -- ListProcessor by Anastasios Kotsikonas
X-Comment: HTML Working Group (Private)
>To: Dave Raggett <dsr@hplb.hpl.hp.com>
>From: newtonjs@char.vnet.net (Stan Newton)
>Subject: Re: FORM content in DTD
>Cc: 
>Bcc: 
>X-Attachments: 
>
>>Stan said:
>>
>>> I don't believe this works when the Recommended switch is ON.
>>> In that case, the %body.content definition becomes
>>
>>> <!ENTITY % body.content "(%heading|%block|HR|ADDRESS)*"
>>
>>> which excludes %text.
>
>>Dave replied:
>>This is untrue. %block includes P which is defined as:
>>
>>        <!ELEMENT P - O (%text;)+>
>>
>>This is designed to ensure that when the Recommended switch is ON, all
>>text must be enclosed in an appropriate container.                ^^^^^
>^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>Dear Dave,
>
>I sorry but I wasn't very clear. This notion of containment is what is 
puzzling me. If I turn Recommended ON (which from an authoring viewpoint I 
like very much, by the way), I understand that this means text goes into 
containers like P. BUT..
>
>I don't see how it can work that way INSIDE the FORM tag. The FORM tag 
itself is the required container for text and furthermore the special FORM 
tags like SELECT and INPUT are not valid inside ordinary containers like P. 
So, it would seem to me, that it is therefore required AT ALL TIMES that a 
FORM tag must be able to contain text directly!! How else do I get INPUT, 
for example, into FORM without putting it in P where it is not allowed??
>
>Am I missing something here? If so, please someone set me straight. This 
present DTD looks to me like it is in error with respect to FORM content. 
>That's why I'm raising the question.
>
>Instead of my suggestion to edit %body.content, perhaps we should just 
define a %form.content entity in terms of the basic components instead of 
making multiple adjustments to %body.content. Such a change would make this 
important distinction clearer.
>