Re: HTML 3.2 Content Models

Dianne Gorman (dkgsoft@oznet02.ozemail.com.au)
Thu, 9 May 1996 08:08:33 +1000


Message-Id: <199605082207.IAA16348@oznet02.ozemail.com.au>
From: "Dianne Gorman" <dkgsoft@oznet02.ozemail.com.au>
To: murray@spyglass.com (Murray Altheim)
Date: Thu, 9 May 1996 08:08:33 +1000
Subject: Re: HTML 3.2 Content Models
Cc: www-html@w3.org

On  8 May 96 at 15:47, you wrote:

> Dianne Gorman <dkgsoft@oznet02.ozemail.com.au> writes:
> >I have constructed hierarchical trees of the content models in the
> >html 3.2 dtd.
> >http://www.ozemail.com.au/~dkgsoft/html/  if you are interested.
> >
> >These are done by hand, so be gentle if you find errors.
> 
> Dianne,
> 
> There are tools to do this without errors. If you're interested, I maintain

Yes - thanks, I know about Earl Hood's excellent tools.  My tree is 
very much more compact, and I learn a lot more about a DTD by 
carefully following through content models than I would by running a 
script.  The self-inflicted pain is purely a learning exercise - and 
I find the results very useful.  

> >PLAINTEXT appears to be an orphan - it is in the preformatted text
> >section of the DTD, but it doesn't seem to be in the content models
> >anywhere
> 
> PLAINTEXT is an anomaly, given that its intended usage is not
> SGML-conformant. Basically, everything after <PLAINTEXT> was not to be
> parsed as SGML content. This invalidates any document, given that there
> cannot then be any closing BODY or HTML tags. I think we deal with it
> internally in our DTDs by keeping its allowed usage outside of BODY, but
> I'd have to check. If there are levels of deprecation, this one rates about
> a ten.

In the HTML 3.2 DTD, PLAINTEXT is outside of HTML, not just outside 
of BODY.

Regards
Dianne
*********************************************
Dianne Gorman - Sydney, Australia
email: dkgsoft@ozemail.com.au
www:   http://www.ozemail.com.au/~dkgsoft/index.html
*********************************************