Re: LC Status and implementation reports

Hi Eric
On Feb 2, 2012, at 1:30 AM, Eric Prud'hommeaux wrote:

> On Jan 31, 2012 10:17 AM, "Boris Villazón Terrazas" <bvillazon@fi.upm.es> wrote:
> >
> > Hi Eric
> >
> >
> > On 31/01/2012 15:37, Eric Prud'hommeaux wrote:
> >>
> >> * ashok malhotra<ashok.malhotra@oracle.com>  [2012-01-26 06:35-0800]
> >>>
> >>> On 1/26/2012 6:25 AM, Eric Prud'hommeaux wrote:
> >>>>
> >>>> Where are we in terms of being able to tell the world how to submit tests results?
> >>>
> >>> We are running late on this.  First, are we happy with the test cases?  We have not reviewed them in some time.
> >>> Second, did we decide a format for submitting test case results?  We discussed this but I don't remember a resolution.
> >>
> >> I recall that we were happy to emulate the SPARQL test report harness.
> >
> > Thanks for the reminder! ....
> > Sorry but I'm not sure of a few things, and I want to be sure that the options are
> >
> > A.
> > 1. A particular R2RML implementation download the Test Suite, run it and generate a set of nq files (as output of the TCs).
> > 2. The owner of the implementation download the TestHarness software, run against the generated nq files, and the software generates test report
> > 3. The owner of the implementation upload the test report
> 
> This ishow SPARQL did it, and my understanding of how we've decided to do it.
> 
Ok, thanks!
So, this is the explanation that [1] should include. It's clear now.
I'm going to update [1] tomorrow.

Thanks again

Boris


[1] http://www.w3.org/2001/sw/rdb2rdf/wiki/Submitting_Test_Results


> > B.
> > 1. A particular R2RML implementation download the Test Suite, run it and generate a set of nq files (as output of the TCs).
> > 2. The owner of the implementation upload the results (the set of nq files).
> > 3. We run the TestHarness software against the results (uploaded previously), and the software generates the test report
> >
> > In both cases the format of submitting the R2RML test case results is nq, right?
> >
> > For the DM we'll have ttl files (instead of nq), right?
> >
> > Boris
> >
> > P.S. Sorry again if my questions are trivial
> 
> Easily answered, but important.
> 
> >> Here's an example:
> >>
> >> -------------------------------------------------------------------------------
> >> # Example RDB2RDF test report for:
> >> #<myProject>  : a doap:project with a doap:name .
> >> #<myTestHarness>  : a earl:Software
> >>
> >> @prefix rdbf:<http://www.w3.org/2001/sw/rdb2rdf/test-cases/#>
> >> @prefix doap:<http://usefulinc.com/ns/doap#>.
> >> @prefix earl:<http://www.w3.org/ns/earl#>.
> >> @prefix rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#>.
> >> @prefix rdfs:<http://www.w3.org/2000/01/rdf-schema#>.
> >> @prefix xml:<http://www.w3.org/XML/1998/namespace>.
> >> @prefix dct:<http://purl.org/dc/terms/>.
> >>
> >> <myProject>
> >>        a doap:Project ;
> >>        doap:name "My Project Name" ;
> >>        doap:homepage<http://myproject.example/home/>  .
> >>
> >> <myTestHarness>
> >>        a earl:Software ;
> >>        dct:title "My Implementation RDB2RDF test harness" .
> >>
> >>
> >> [] a earl:Assertion;
> >>        earl:assertedBy<myTestHarness>  ; # my harness asserted that
> >>        earl:subject<myProject>  ;        # my project
> >>        earl:result [ a earl:TestResult ;
> >>                earl:outcome earl:pass    # passed (vs. earl:fail)
> >>        ] ;
> >>        earl:test rdbf:D000-1table0rows . # the RDB2RDF test.
> >> …
> >> -------------------------------------------------------------------------------
> >>
> >>
> >>
> >>> Third, should we create a separate mailing list for test case reports?
> >>>
> >>> Ashok
> >>>
> >