Re: Validing parser required?

[James Aylett <james@xxxxxxxxxxxx>]

On Fri, Jul 02, 2004 at 11:02:33AM -0400, Sam Ruby wrote:

> I would like to see Atom "cleanly and thoroughly specified". 
> People should not be required to use a validating parser simply 
> because the spec is silent on whether this is allowed or not, and 
> therefore must be allowed.
> 
> One way to handle this is the way SOAP[5] does :
> 
> >A SOAP message MUST NOT contain a Document Type Declaration.
> 
> That seems pretty clear, and certainly is very effective.

I have two thoughts on this:

 (1) a DocType declaration could be of use in /creating/ an Atom
     document, because it allows a producer to validate immediately on
     production. However if we're stating that an Atom document
     must be able to be understood entirely by a non-validating
     parser, there is no other use of a DocType declaration (at least,
     that I can see) - except for named entity declarations, which
     are only really of use if you're writing this stuff by hand.

 (2) the language to express that will be difficult to make clear. The
     best I've come up with is: 

       An Atom document MUST be expressed such that its meaning is
       identical whether a validating or non-validating XML parser is
       used.

     And I'm not terribly happy with that phrasing in all sorts of
     ways.
 
And of course we can suggest using a schema, not a DTD, for validity
checks. So I'd probably be happy with outlawing DocType declarations
in Atom.

James

-- 
/--------------------------------------------------------------------------\
  James Aylett                                                  xapian.org
  james@xxxxxxxxxxxx                               uncertaintydivision.org