- From: Jacek Kopecky <jacek.kopecky@deri.org>
- Date: Fri, 04 May 2007 21:23:25 +0200
- To: Tammo van Lessen <tammo.van.lessen@iaas.uni-stuttgart.de>
- Cc: public-ws-semann@w3.org
Hi Tammo, the schema mapping annotations point directly to the mappings. The annotation doesn't give any information about the type of the mapping language, but in general, resolving the URI will give you enough information through metadata or the file itself. Schema mapping annotations are lists of URIs of alternative mappings; I guess if there are multiple, the processor can try the first, if it doesn't understand it, try the second etc. As an optimization, a processor can have working mappings pre-cached, so it may recognize a known URI among the alternatives and use that right away. For actual lifting and lowering implementations, you can check out a draft description of how to use XSLT for lifting and lowering RDF data at [1]. It contains a few examples and it provides an XSLT library called RDFXSLT (for now) which helps with creating the lowering - it provides a shield against the RDF/XML idiosyncrasies. [1] http://www.wsmo.org/TR/d24/d24.2/v0.1/20070412/rdfxslt.html The document may be linked from the SAWSDL WG page soon. Can you please write me about how you use SAWSDL? I'd very much appreciate that. 8-) Hope it helps, Jacek On Fri, 2007-05-04 at 15:13 +0200, Tammo van Lessen wrote: > Hi all, > > While thinking about how to implement a generic lifting and lowering > service using SA-WSDL description, I'm wondering how a mapping processor > can find out which mapping language is used in the document referenced > by the {lifting|lowering}SchemeMapping attribute? Or are these URIs just > virtual descriptors for a {mapping-language, mapping-document} tuple > that is stored somewhere else? > > And another question: Are there already implementations or plans to > implement such a L&L component? > > Best regards, > Tammo >
Received on Friday, 4 May 2007 19:23:32 UTC