- From: Jeremy Carroll <jjc@hpl.hp.com>
- Date: Wed, 28 Mar 2007 15:25:05 +0100
- To: GRDDL Working Group <public-grddl-wg@w3.org>
Summary: concerning http://www.w3.org/2001/sw/grddl-wg/td/pendinglist#loopx I think this is significantly harder than what the WG intends, and may merit a substantive change to the spec. ******************** I wrote the following test, which the Jena GRDDL reader fails. http://www.w3.org/2001/sw/grddl-wg/td/pendinglist#loopx An XML document is its own namespace document, with grddl transformation, specifying a namespaceTransformation, which specifies a further namespaceTransformation. i.e. http://www.w3.org/2001/sw/grddl-wg/td/loopx has a GRDDL result being: <> g:namespaceTransformation <loopx.xsl> . from the embeddedRDF transform. By looking at its namespace doc, which is itself, it has a grddl result being <> g:namespaceTransformation <loopy.xsl> . Hence, by looking (one again?) at its namespace doc, it has a GRDDL result <> g:namespaceTransformation <one.xsl> . Hence, by looking (yet again?) at its namespace doc, it has a GRDDL result _:a rdf:value "one" . This gives http://www.w3.org/2001/sw/grddl-wg/td/loopx-output The jena grddl reader, doesn't keep looking again, and so only finds the first two triples. I think this is significantly harder than what the WG intends, and may merit a substantive change to the spec. Jeremy -- Hewlett-Packard Limited registered Office: Cain Road, Bracknell, Berks RG12 1HN Registered No: 690597 England
Received on Wednesday, 28 March 2007 14:25:34 UTC