W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > April to June 2004

Re: test case results expressed in ResultSet terms

From: Jos De_Roo <jos.deroo@agfa.com>
Date: Sun, 27 Jun 2004 13:04:56 +0200
To: eric@w3.org
Cc: Jos De Roo <jdroo@skynet.be>, public-rdf-dawg@w3.org, public-rdf-dawg-request@w3.org
Message-ID: <OF488C43C2.9A88ECF3-ONC1256EC0.00390400-C1256EC0.003CCCDC@agfa.com>

EricP wrote:
[...]
>> I was also trying to run your "nested optional arcs" test case
>> and had to make some corrections in our implementation so that
>> in fact only those 4 results are returned (also using a design
>> that allows result interpretation as it refers to the query).
>> 
>> So querying the facts given in
>> http://eulersharp.sourceforge.net/2004/04test/ericP.n3
>> with filter
>> http://eulersharp.sourceforge.net/2004/04test/ericF.n3
>> results in
>> http://eulersharp.sourceforge.net/2004/04test/ericE.n3
>
> Yeah, the trick (for me, at least) was to make sure I NULL'd
> all the otherwise unbound variables in any optional part that
> was not satisfied. I think this is pretty easy to define so
> I'm not to worried about this as a requirement from that
> perspective. However, it is worth noting that introducing
> optionals does mean we have an ordered set of constraints.

when we have found all solutions coming through the filter
we put them in a new engine and take out the unique triples
using the unification method, but indeed, you are right, we
have to build something (plan to do that later..) to do an
automatic clause and premis reordering to make it order
independent.


> I'd like to experiment with Euler(sharp?). I guessed a CVSROOT:
>   cvs -d:pserver:anonymous@cvs.sf.net:/cvsroot/eulersharp login
> but didn't guess a module name. I tried Euler, euler, EulerSharp,
> and eulersharp.

Oops.. will have to write some more documentation..
(ther's now some at 
http://eulersharp.sourceforge.net/2004/01swap/Euler.txt)

You could checkout module 2004/01swap
$ cvs -z3 -d:pserver:anonymous@cvs.sourceforge.net:/cvsroot/eulersharp co 
2004/01swap
and then run
$ java -cp 2004/01swap/classes euler.EulerRunner --nope --think \
  http://eulersharp.sourceforge.net/2004/04test/ericP \
  --filter http://eulersharp.sourceforge.net/2004/04test/ericF
for the java version, or
$ 2004/01swap/bin/Release/Euler --nope --think \
  http://eulersharp.sourceforge.net/2004/04test/ericP \
  --filter http://eulersharp.sourceforge.net/2004/04test/ericF
for the .NET version

(yesterday I only checked in the sources, and today I've
checked in also the classes and bin, so it could take a
while before can get them)

-- 
Jos De Roo, AGFA http://www.agfa.com/w3c/jdroo/
Received on Sunday, 27 June 2004 07:05:31 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:19 GMT