Sparql test cases

[GoogleCodeExporter]

As briefly discussed with Andreas, I would like to create a whole SPARQL test 
suite that covers as much scenarios as possible.

To do that, we could use (I asked to author and is ok for him, I'm waiting from 
OReilly permission) the examples (both ttl and rq files in book "Leaning 
SPARQL" [1])

So we will create a test case with several test methods that use and assert the 
examples in the book.

In case OReilly doesn't allow such usage I'll use those examples in order to 
create our own set of datafiles.  

[1] http://www.learningsparql.com/

Original issue reported on code.google.com by a.gazzarini@gmail.com on 19 Feb 2014 at 2:42

[GoogleCodeExporter]

The author told me (via email) that is Ok. I asked for OReilly permissions but 
still no response. Anyway, reading their policies [1] I believe there's no 
problem at all. 

I will submit the first set of tests soon. 

[1] http://www.oreillynet.com/pub/a/oreilly/ask_tim/2001/codepolicy.html

Original comment by a.gazzarini@gmail.com on 21 Feb 2014 at 3:59

• Changed state: Started

[GoogleCodeExporter]

Hi Andrea,

I think we should do the SPARQL testing in a much more structured fashion. 
Maybe we should reuse the SPARQL test cases/data from W3C [1]. This way, we are 
sure that we have all SPARQL 1.1 features reflected/tested in our tests.

In general, we should revisit the overall tests and see that we can improve the 
test coverage. I think some the test classes (especially in the core module) 
should be revised.

What do you think?

Kind regards
Andreas

[1] http://www.w3.org/2009/sparql/docs/tests/

Original comment by andreas.josef.wagner on 29 Jun 2014 at 1:32

[GoogleCodeExporter]

Yes, nice idea.
The data provided in the book is good because it is already packed in 
different files (the exNN.* you see in the test/resources). There are a 
lot of other files in the source code associated with the book and I'm 
doing that gradually.

But (I think) we can do both, using also code and tests from W3C. The 
current test data is under a folder learningSPARQL and the test is 
called LearningSPARQLTest. We could have a similar structure for W3C.

In general, that's the reason why I left open this issue: it is a 
cross-cutting issue, it can be completed, improved and integrated with 
other things, relase by release.

Yes, I agree, also here we could have something like a cross-cutting 
issue because I think a good coverage is a long term goal. The 
dictionary implementation for example, should have a good coverage but 
tests took a (moreless) good 60% of the overall effort. So thinking in 
those terms for all current tests I think is not a trivial task.

Best,
Andrea

Original comment by a.gazzarini@gmail.com on 29 Jun 2014 at 5:42