Re: Namespace conformance & SAX2

[Chris Nokleberg <chris@sixlegs.com>]

Here's an expanded quote from the spec.
(http://www.w3.org/TR/1999/REC-xml-names-19990114/#Conformance)

  The effect of conformance is that in such a document: 

  * All element types and attribute names contain either zero or one colon. 
  * No entity names, PI targets, or notation names contain any colons.

  Strictly speaking, attribute values declared to be of types ID,
  IDREF(S), ENTITY(IES), and NOTATION are also Names, and thus should be
  colon-free. However, the declared type of attribute values is only
  available to processors which read markup declarations, for example
  validating processors. Thus, unless the use of a validating processor
  has been specified, there can be no assurance that the contents of
  attribute values have been checked for conformance to this
  specification.

To me, this last sentence implies that there *is* an "assurance" that
all processors will check the first two constraints, and that
validating processors should additionally check ID, IDREF, ENTITY, and
NOTATION.

Given this, it makes the most sense to me if the first two constraints
are treated as fatal errors, and the additional constraints as
nonfatal.

However, I'd be happy if you were right and parsers are not required
to check these constraints at all--it would make my life easier.

--Chris


On Fri, May 18, 2001 at 11:28:39AM -0700, David Brownell wrote:
> Actually, any parser is trivially conformant with the namespace
> spec if it completely ignores those constraints.  The conformance
> constraint is for documents only ... meaning, document creation
> facilities are the only ones constrained.
> 
> Some of my review comments for the namespace spec requested
> (repeatedly!) clarification on what XML processors need to do,
> but it wasn't forthcoming (over several spec iterations).  So it
> was clear to me that the relevant W3C folk didn't want to require
> XML processors to do anything.
> 
> Would you propose that such constraints be reported as fatal
> errors (WFness violations), nonfatal ones (like validity problems
> and a few other random errors), warnings, or something else?
> 
> - Dave
> 
> 
> ----- Original Message ----- 
> From: "Chris Nokleberg" <chris@sixlegs.com>
> To: <xml-dev@lists.xml.org>
> Sent: Friday, May 18, 2001 10:28 AM
> Subject: Namespace conformance & SAX2
> 
> 
> > The Namespace spec adds additional conformance constraints beyond XML 1.0:
> >   * All element types and attribute names contain either zero or one colon.
> >   * No entity names, PI targets, or notation names contain any colons.
> > 
> > Is a non-validating SAX2 parser required to check these constraints?
> > My suggestion is that it should be required to check iff the
> > http://xml.org/sax/features/namespaces feature is true.
> > 
> > Thanks,
> > Chris
> > 
> 
>