W3C home > Mailing lists > Public > public-owl-wg@w3.org > April 2009

Re: Report on current test case coverage

From: Alan Ruttenberg <alanruttenberg@gmail.com>
Date: Tue, 28 Apr 2009 08:22:11 -0400
Message-ID: <29af5e2d0904280522q7e7f7dfbic0acf71c6d7cae03@mail.gmail.com>
To: mak@aifb.uni-karlsruhe.de
Cc: public-owl-wg@w3.org
Hi Markus,

I have started generating test cases for the functional syntax to RDF
Mapping, initializing the results using an implementation of the
mapping I've done and eyeballing the results. I haven't yet uploaded
them to the wiki but in their raw form they are at
http://svn.mumble.net:8080/svn/lsw/trunk/owl/owl2/tests/fs2rdf/

I'm suggesting that they be checked in two ways:
1) Using Jena's isIsomorphicWith (or equivalent alternative) to see
that the triples generated by a tool are the same as the ones in the
test, module bnodes.
2) Using backwards and forwards positive entailment tests in tools
that can read both functional syntax and RDF, with the premise and
conclusion being each in one of the two syntaxes.

I won't be at the TC this week, due to a conflict but will be working
on further tests before our following TC.

-Alan

On Tue, Apr 28, 2009 at 5:41 AM, Markus Krötzsch
<mak@aifb.uni-karlsruhe.de> wrote:
> Many of the new OWL 2 features are still lacking test cases to support tool
> developers. The online test case repository now provides overview pages to
> give you an impression of our test case coverage per feature:
>
> * http://km.aifb.uni-karlsruhe.de/projects/owltests/index.php/RDF/XML_coverage
> * http://km.aifb.uni-karlsruhe.de/projects/owltests/index.php/FS_coverage
>
> The "features" are merely syntactic here, hence the separation between RDF/XML
> syntax and FS syntax. Tests are counted on both pages as soon as we have added
> (manual) syntactic translations -- the RDF/XML page thus eventually will be
> more complete (but less precise since RDF/XML overloads various syntactic
> elements).
>
> Contributions are welcome -- recall that each use of OWL 2 that is supported
> by a test case is more likely to be supported properly by future tools.
>
> Markus
>
> P.S. The online interface for contributing tests has recently been simplified
> (even ;-) further.
>
> --
> Markus Krötzsch
> Institut AIFB, Universität Karlsruhe (TH), 76128 Karlsruhe
> phone +49 (0)721 608 7362          fax +49 (0)721 608 5998
> mak@aifb.uni-karlsruhe.de          www  http://korrekt.org
>
>
Received on Tuesday, 28 April 2009 12:23:12 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 28 April 2009 12:23:15 GMT