- From: Alberto Reggiori <alberto@asemantics.com>
- Date: Tue, 26 Oct 2004 15:46:49 +0200
- To: RDF Data Access Working Group <public-rdf-dawg@w3.org>
Steve, Andy, after having spent some cycles on doing simple regression tests using the DAWG proposed test-manifest.n3 http://www.w3.org/2001/sw/DataAccess/tests/test-manifest.n3 and the associated test-query.n3 http://www.w3.org/2001/sw/DataAccess/tests/test-query.n3 and wrote a simple test program to read a manifest and query it to run regression tests as described http://www.rdfstore.org/dawg-testcases/runtests.txt generating results for some SOURCE test cases http://www.rdfstore.org/dawg-testcases/results.html and comparing the current above DAWG query testing vocabularies with previous work done http://www.w3.org/2003/03/rdfqr-tests/summary.html and especially http://www.w3.org/2003/03/rdfqr-tests/summary.html#manifest and http://www.w3.org/2003/03/rdfqr-tests/rdf-query-testcases.html I found the following points relevant to discussion and worth to consider, to update/modify the manifest and query vocabularies: - all refs to inputs, outputs and queries must be relative (partially/already done) - or use some sensible xml:base if RDF/XML used - all manifest, query and tests should be possibly expressed in standard RDF/XML or N-Triples syntax (rather then non-standard N3/Turtle) - to allow a complete automation of the DAWG regression-tests machinery each input and output should be qualified / typed accordingly -> the documents have a syntax specified in the manifest: N-Triples, N3, Turtle, TriX or RDF/XML -> the following typed nodes can be used: -> qt:NT-Document, qt:N3-Document, qt:Turtle-Document, qt:TriX-Document and qt:RDF-XML-Document -> for DESCRIBE, CONSTRUCT and misc output formats eventually: -> qt:RDF-for-XML-Document [1] qt:XML-Document [2] qt:SPAQL-XML-Document [3] -> then each client software can associate the right MIME-types to document, without doing nasty regexps on file extensions (e.g. file.n3) to fire different parsers for different syntaxes -> a would assumed qt:query to have a default domain to qt:SPARQL-Document - add a qt:metadata to associate qt:data items to any addition SOURCE information -> see http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/ 0176.html - qt:numberRows might be added to each mf:ManifestEntry possibly ( me not having CVS access Eric/Andy what happened to http://www.w3.org/2003/03/rdfqr-tests/query.rdfs# and http://www.w3.org/2003/03/rdfqr-tests/manifest.rdfs# URLs ?) In particular in test-manifest.n3 - the second definition of mf:result should be named mf:status I think - add perhaps to rdfs:seeAlso to each mf:ManifestEntry to point to further infos if necessary I am looking forward to discuss further about the DAWG tests machinery in the upcoming meetings Yours Alberto [1] http://lists.w3.org/Archives/Public/www-archive/2004Sep/att-0013/_RDF- XML.html [2] http://lists.w3.org/Archives/Public/public-rdf-dawg/2004JulSep/ 0565.html [3] http://lists.w3.org/Archives/Public/public-rdf-dawg/2004JulSep/ 0577.html - Alberto Reggiori, Senior Partner, R&D @Semantics S.R.L. alberto@asemantics.com www.asemantics.com Milan Office, milano@asemantics.com, +39 0332 667092
Received on Tuesday, 26 October 2004 13:46:53 UTC