Subject: Re: DTD element ordering From: Horst Szillat Date: Wed, 29 Mar 1995 14:01:44 -0800 (PST)
How the Web Was Won
Subject: Re: DTD element ordering From: Horst Szillat Date: Wed, 29 Mar 1995 14:01:44 -0800 (PST)
Organization: Unlimited Surprise Systems, Berlin
Lines: 25
Message-ID: 
References: <1995Mar28.225425.29439@rsg1.er.usgs.gov>
NNTP-Posting-Host: berlin.snafu.de
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Sender: szillat@berlin
In-Reply-To: <1995Mar28.225425.29439@rsg1.er.usgs.gov> 

On Tue, 28 Mar 1995, Peter N. Schweitzer wrote:

> I have a DTD with an element like this:
> 
> 
> 
> This means that NAME must contain A and C, and may contain any number
> of B's.  Further, it means that the elements A, B, and C can occur
> in any order (because I used '&' rather than ',' to separate them).
> However, I observe (using sgmls 1.1.91) that if B is repeated, all
> occurrences of B must be together, i.e. I get errors from
That is right. Your content model says:
The parts (A), (B*) and (C) have to appear once and in an arbitrary order.
Try:

      (B*,((A,B*,C)|(C,B*,A)),B*)

Explanation needed?
Anyone who desagrees or has a shorter/easyer content model?

Horst Szillat

email: szillat@berlin.snafu.de
url  : http://www.snafu.de/~szillat/



From rballard@cnj.digex.net Sun Mar 30 10:28:18 1995
Status: RO
X-Status: 
Newsgroups: comp.text.sgml
Path: news3.digex.net!news2.digex.net!howland.reston.ans.net!Germany.EU.net!EU.net!sun4nl!hsahgl!jta_swets