RE: Grddl Squirrel 6 test cases

Le jeudi 09 mars 2006 à 16:41 +0000, McBride, Brian a écrit :
> The RDF syntax spec says
> 
> [[
> If the RDF/XML is a standalone XML document (identified by presentation
> as an application/rdf+xml RDF MIME type object, or by some other means)
> then the grammar may start with production doc or production
> nodeElement.
> ]]
> 
> The significance of this is that it opens the way for an RDF document to
> have a dataview:transformation attribute on its root element.

OK, I understand now.

>   The other
> question is if it is not served with an RDF mimetype should one apply
> transforms in the namespace doc itself.

Well, if the namespace document itself is RDF (namely, with an rdf:RDF
root element), I think the "stop when you hit RDF/XML" rule should
apply. 
 
> > Another option is to say that a GRDDL implementation should 
> > not attempt to incorporate non-Valid RDF statements at any 
> > stage in the process (in which case the interpretation is in 
> > fact different in the end).
> 
> We'd need to be clear about what "non-valid RDF statements" are.  Do you
> mean invalid RDF syntax?

Yes, that's what I meant. But from you said, the MIME-Type does actually
matter to determine whether a document is RDF-valid or not.

> > Also, at some point, RDF/XML could be integrated into a 
> > non-RDF root element; is that still the case? If so, we 
> > probably need test cases to deal with that possibility as well.
> 
> Hmm, I'm not clear about that.  Clearly one can include the RDF/XML in
> some other root element, but I'm not sure if one is then allowed to
> interpret it as RDF/XML - can the context modify the meaning?

I guess I was thinking to what you just pointed out: if a document is
served as application/rdf+xml, the root element doesn't need to be
rdf:RDF; but I guess we can apply the end-of-recursion rule again in
this case, by defining hitting an RDF/XML document as:
* an XML document whose root element is rdf:RDF
* a document served as application/rdf+xml

But indeed, all of this could use some formalization.

Dom
-- 
Dominique Hazaël-Massieux - http://www.w3.org/People/Dom/
W3C/ERCIM
mailto:dom@w3.org