- From: Dan Connolly <connolly@w3.org>
- Date: Wed, 09 Jun 2004 07:40:05 -0500
- To: danny666@virgilio.it
- Cc: "Sean B. Palmer" <sean+rixt@infomesh.net>, RDF in XHTML task force <public-rdf-in-xhtml-tf@w3.org>
On Wed, 2004-06-09 at 05:44, Danny Ayers wrote:
> Dan, Sean,
>
> Please correct me if I'm wrong, I haven't had time to pore over the source:
> Both glean.py and garner.py discover and apply a series of XSLTs to the
> source XHTML, reading the resultant RDF/XML into a triplestore/model.
> The merged model is then serialised out as RDF/XML.
>
> Can't the same result be achieved without the Redland/Pyrple dependency?
Almost...
> Isn't a purely syntactical merge possible - just top & tail ('deroot')
> the individual RDF/XML results, then glue them back into a common root.
Yes, that's how the first version of glean.py worked.
But more recent versions support "following your nose" thru
profile/namespace documents, GRDDLing those, and looking
for statements of the form
profile dataview:profileTransformation ?XFORM.
so there's some non-trivial RDF-level processing going on. I thought
about doing that with XSLT, but then thought better of it.
> I would guess XSLT could do this whilst retaining all the required
> namespace declarations.
>
> Feasible?
>
> Cheers,
> Danny.
--
Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Wednesday, 9 June 2004 08:39:38 UTC