Re: Agenda 2010-10-19

On 19/10/10 13:59, Axel Polleres wrote:
>
> On 19 Oct 2010, at 09:48, Andy Seaborne wrote:
>
>>>                • Basic Update Test cases, see http://www.w3.org/2009/sparql/docs/tests/basic-update/
>>
>>
>> On 18/10/10 17:01, Alexandre Passant wrote:
>>> Besides that, the 4 TC looks of for me.
>>> Haven't check the DELETE ones so far.
>>
>> basic-update:
>>
>> 1/ :insert-data-spo-named1 has mysterious ut:graph
>
> in what sense "mysterious"?

The manifest uses both ut:graph and ut:data with ut:graphData.

:insert-data-spo-named1

  mf:result [ ut:result ut:success ;
              ut:graphData  [ *ut:graph* <spo.ttl> ;
                            rdfs:label "http://example.org/g1" ]
               ] .

but:

:insert-data-spo-named2
     mf:result [ ut:result ut:success ;
                 ut:graphData  [ *ut:data* <spo2.ttl> ;
                                 rdfs:label "http://example.org/g1" ]
               ] .

> The usage of  "ut:graphData [ ut:graph "... is explained in the README.html here:

No, it's not!

There is no mention of ut:graph anywhere in the README. v1.19

The text you quote uses ut:data, something that is wrong in modeling 
terms.  Doubly so as it copies a mistake from the query manifest 9check 
the domain).

> "[...] the object of the ut:graphData property indicates the named graph components of the Graph Store. Named graphs are described either - analogous to the qt:graphData property from the query test vocabulary - by explicit URI reference (in which case the graph name is supposed to correspong to the respective URI reference), or the object of the ut:graphData property may be a resource further described in terms of a reference to the graph by the ut:data property and a reference to its name under which it is accessible in the graph store using the rdfs:label property with plain literal value. The ability to assign a "name" differnt from the URI reference name explicitly to a named graph in a graph store is needed to denote different graphs by the same name to describe the status of a named graph prior and after execution of an update."

[[
or the object of the ut:graphData property may be a resource further 
described in terms of a reference to the graph by the *ut:data* property
]]

I have been assuming that graphData is a bNode + properties or a URI, 
but not URI + properties.

>
> let's discuss unclear aspects of this in the call...

I hope today we can work on the process for developing tests.  How we 
contribute, how we ensure coverage, what criteria we are using for 
approval, preconditions for getting on the agenda to be approved,
do we have mixed approves/unapproved sets or approve sets, what the 
target is (all stores? stores that implement empty graphs, failure 
handling ...).

	Andy

Received on Tuesday, 19 October 2010 13:16:42 UTC