- From: Dominique Hazael-Massieux <dom@w3.org>
- Date: Wed, 30 May 2007 14:32:20 +0200
- To: Harry Halpin <hhalpin@ibiblio.org>
- Cc: public-grddl-wg@w3.org
- Message-Id: <1180528340.6202.71.camel@altostratustier>
Hi, Le mardi 29 mai 2007 à 14:25 -0400, Harry Halpin a écrit : > While I know you are very busy, the GRDDL Spec is getting ready > for CR and it would make a lot of sense if the W3C GRDDL Service [1] was > compliant with the test cases [2]. Could you check this and get back to > the WG? I've started running the tests; on testlist1.rdf [4], I get one failure on the atom test, since it outputs turtle that is not supported by the XSLT service. On testlist2.rdf [5], I get a failure on multipleRepresentations [6], but this is probably due to the fact that http://www.w3.org/2001/sw/grddl-wg/td/multipleRepresentations.rdf doesn't exist. I also fail on the httpHeaders test since I haven't tried to implement it - I'll try to get a stab at implementing it. I failed on multiProfile due to the use of relative URIs in the profile attribute; I'm not sure this is very kasher - using relative URIs when a <base href> may get defined later - but the spec doesn't seem to forbid it, so I made it handle it. testft.py still claims that I fail the test, but I'm not sure that's true - for some (unidentified yet) reasons, I get duplicate in the results file (attached), but I'm not sure it really is against the spec or not. I fail most tests in testlist3.rdf [7], but since they are "robustness" tests and this service is mainly experimental, my motivation to fix those is not very high... (the EARL outputs of the testing for each of the tests lists is are attached to this message) > One easy way might be to use one of the GRDDL implementations on the > backend that is already compliant with the spec and test cases [3] > like > Jena if there are problems with the service being compliant that woudl > be too > timeconsuming to correct by hand. I think it probably makes sense that W3C support a real GRDDL online service, but I'm not sure the XSLT-based solution is the right base for this; I agree that starting from a more sound implementation would be the right thing to do, but I fear I don't have the cycles to set up such a system anytime soon... Dom > [1]http://www.w3.org/2003/11/rdf-in-xhtml-demo > and > http://www.w3.org/2004/01/rdxh/grddl-xml-demo > [2]http://www.w3.org/TR/grddl-tests/ > [3]http://www.w3.org/2001/sw/grddl-wg/td/test_results 4. http://www.w3.org/2001/sw/grddl-wg/td/testlist1.rdf 5. http://www.w3.org/2001/sw/grddl-wg/td/testlist2.rdf 6. http://www.w3.org/2001/sw/grddl-wg/td/multipleRepresentations 7. http://www.w3.org/2001/sw/grddl-wg/td/testlist3.rdf
Attachments
- text/rdf attachment: earl_out.rdf
- text/rdf attachment: earl_out2.rdf
- application/xml attachment: multiprofile-output.rdf.result
- text/rdf attachment: earl_out3.rdf
Received on Wednesday, 30 May 2007 12:33:08 UTC