W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > April to June 2011

Re: ACTION-423: Include a test on nodes in path of length zero come from specified named graph

From: Matthew Perry <matthew.perry@oracle.com>
Date: Tue, 05 Apr 2011 09:09:05 -0400
Message-ID: <4D9B1471.4000009@oracle.com>
To: SPARQL Working Group <public-rdf-dawg@w3.org>
Hi,

I agree with the new result that Andy posted. It looks like there is a mistake in the existing test.

- Matt

On 4/5/2011 4:15 AM, Andy Seaborne wrote:
>
> On 05/04/11 02:16, Lee Feigenbaum wrote:
>> On 4/4/2011 9:02 PM, Gregory Williams wrote:
>>> On Apr 4, 2011, at 8:28 PM, Lee Feigenbaum wrote:
>>>
>>>> Thanks, Matt!
>>>>
>>>> It would be great if people can try these out (Greg, Andy, Olivier,
>>>> ...?) so that we can approve these tomorrow if OK.
> ARQ passes pp34 and pp35 (path-ng-01.rq and path-ng-02.rq)
>
>>> I pass the new tests, but as Andy and I have discussed on irc, I think
>>> we approved some other pp tests last week that are actually wrong, so
>>> we should revisit that at some point.
>> Could you let us know which ones those are?
> pp29
>
> Got: 5
> # i.e. what we now think is right:
> ------
> | s  |
> ======
> | :a |
> | :a |
> | :c |
> | :a |
> | :c |
> ------
> Expected: 4
> # i.e. what the test has
> ------
> | s  |
> ======
> | :c |
> | :a |
> | :a |
> | :a |
> ------
>
> ARQ had a bug in this evaluator in that it was expanding {2,} backwards
> (the "2" at the tail of the path not the head).  This test has a
> grounded tail and variable head which triggered the bug.
>
> 	Andy
>
>
>
>> thanks,
>> Lee
>>
>>> thanks,
>>> .greg
>>>
>>>
Received on Tuesday, 5 April 2011 13:09:32 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:46 GMT