Re: Update

> On Oct 11, 2015, at 2:00 PM, Karima Rafes <karima.rafes@gmail.com> wrote:
> 
> Hello
> 
> Before to change the SPARQL test suite, there is a more urgent
> problem. I noted a lot of problems in the implementation of SPARQL
> protocol in the triplestores (the federated query with SPARQL is of
> the science fiction between editors of triplestores). Each editor has
> his own definition of SPARQL protocol.
> 
> I propose to implement if you want in PHP a very simple library for
> testing all the protocol query and update (via Get and Post with same
> parameters). In parallel, I can test automatically their online
> triplestores for helping the developers to converge.

I’m not sure why a proprietary service is necessary to describe and run such tests; most developers will want to be able to run such tests locally, IMO. Test manifests include “accept” and “response” values, which can provide much of what is necessary. Other specs (e.g. JSON-LD and CSV on the Web) invent new properties to add to an individual test entry for specifying HTTP header values, redirections, and so-forth. Shouldn’t it just be a matter of creating a new test entry to describe the attributes of an HTTP request, and the expected HTTP response (value and other metadata)? Perhaps we can operate from a set of requirements for tests, and then come up with additions to <http://www.w3.org/2001/sw/DataAccess/tests/test-manifest#> and/or <http://www.w3.org/2001/sw/DataAccess/tests/test-query#> for describing such envelope information.

> If I develop these new tests without participation of editors, no
> triplestores would pass these new tests. So I'll let you decide if
> it's useful to develop new tests for the protocol of SPARQL. If some
> editors(minimum 2) are OK to converge, I can propose implement quickly
> these tests via TFT [1].

The best way to manage that is to create one or more issues at https://github.com/w3c/rdf-tests/issues to clarify specific test requirements and get community buy-in before adding PRs for tests implementing these test requirements. Then, different implementations can run their implementations against these tests before broader consensus is achieved to make them official.

> For info, the service http://sparqlscore.com/ uses the W3C's
> repository in github. If you update the test suite SPARQL in your
> repository, you will see the results after 24 hours on my service.

This looks like a great service.

Gregg

> Thanks for your help
> Bye
> Karima
> 
> [1] https://github.com/BorderCloud/TFT
> 
> 2015-10-10 19:04 GMT+02:00 Gregg Kellogg <gregg@greggkellogg.net>:
>> Welcome to new people and organizations who have joined the CG; we now have 17 participants.
>> 
>> Things are a bit slow right now, but we have some issues for updating tests pending [1]:
>> 
>> * rename negative syntax tests to .err (Turtle)
>> * missing prefixes in two negative syntax tests (Turtle)
>> * invalid character in localName_…
>> * Curate CWM/SWAP tests
>> * Add Turtle, Trig, RDF/XML and SPARQL tests to better validate relative IRI resolution
>> 
>> I would also be good for someone to propose changes to the SPARQL test suite for RDF 1.1 alignment, and other issues that have been noted.
>> 
>> Anyone can propose issues, Pull Requests gladly accepted. Reminder: to merge such changes into the main branch requires consensus. See [2] for more information.
>> 
>> Gregg Kellogg
>> gregg@greggkellogg.net
>> 
>> [1] https://github.com/w3c/rdf-tests/issues
>> [2] http://w3c.github.io/rdf-tests/
>> 
>>

Received on Monday, 12 October 2015 16:54:46 UTC