On 10 July 2012 22:23, Jerven Bolleman <[email protected]> wrote:
> Hi Holger, Peter,
>
> I was investigating the possibility to upgrade SPIN-API 1.2.0 to depend on
> Jena 2.7.2. And wanted to share my results. Holger, are you interested in
> working more publicly on the spin opensource part? (Copyright assignments
> are possibility etc...) and if so how would you like to proceed?
>
> First thing is that there are very few changes between Jena 2.6.4 and Jena
> 2.7. I took the feature/jena27upgrade branch from Peter on github as a
> starting point. I had a few issues.
I have also made some other backwards incompatible changes to Visitors
internally, and added parameters to many methods (keeping previous
method signatures as wrappers for the new methods where possible) to
avoid relying on singletons deep down in the SPIN hierarchy, so there
may be other sideeffects that are not picked up by the very small test
suite. Previously Holger had said that he would be hesitant to take
the full patch as it was so large, but if you can isolate the
TemplateGroup changes from my other changes it will be easier to
maintain the patch and easier for Holger/etc., to review and accept
it.
> Template Group
>
> The first is the Template group that peter talked about. Which breaks one
> test case in my project. One where I validate that the SKOS constraints
> really work.
>
>
> * or {*} property paths
>
> The second is the changing of com.hp.hpl.jena.sparql.path.P_OneOrMore
> into com.hp.hpl.jena.sparql.path.P_OneOrMore1 or
> com.hp.hpl.jena.sparql.path.P_OneOrMoreN.
>
> For now I changed these into com.hp.hpl.jena.sparql.path.P_OneOrMore1. This
> allows all spin test cases to pass. And also allows 358 of my codes unit
> test to pass.Jena's serious lack of java doc in these areas means I am not
> sure what the difference between these options are. Except that the
> PathWriter translates P_OneOrMore1 as * and P_OneOrMoreN as {*}. The {*}
> does not seem to be allowed by the 05 January 2012 Sparql 1.1 draft. So the
> version I picked should be correct.
>
>
> VALUES from LC 2
>
> The thirds is the addition of an ElementData i.e. the current proposed
> VALUES keyword. But I have no instance of that in my queries as this is a
> last call 2 feauture. The spinrdf serialization format should be extended
> for this case. It would simplefy some of my constraint queries.
>
>
> Non SPIN problems
>
> The main change I had to do in my queries was simple.
> I had used the following.
>
> SELECT * WHERE {?this a :something} GROUP BY ?this
>
> These where are longer allowed but are easily fixed by changing them.
>
> SELECT ?this WHERE {?this a :something} GROUP BY ?this
>
>
> The repo having this code is on
> https://github.com/JervenBolleman/spin/tree/feature/jena27upgrade
>
> I will try to flesh out the ElementData values code. But Holger/Topbraid
> devs could you discuss how you would like to collaborate.
>
> Regards,
> Jerven
>
> On 01/26/2012 12:13 AM, Holger Knublauch wrote:
>>
>> Hi Peter,
>>
>> sorry but you are way ahead of me here. We will look into Jena 2.7
>> migration after 3.6 (which is currently our focus to get out of the door)
>>
>> Regards,
>> Holger
Now that 3.6 is out of the door, and Jena has had some time to
stabilise at Apache, it may be a good time to look into this. The
TemplateGroup changes appear to be the only backwards incompatible
changes since 2.6.4 (although they did occur in 2.6.5 as I pointed out
previously).
In addition, having a SPIN-API reference implementation that works
with a modern version of Jena will help the uptake of SPIN overall. I
understand that it will require changes to the rest of the TopBraid
suite of products as they may use TemplateGroup themselves, or another
class that has been moved or renamed, so it may not be a trivial
upgrade, but if TemplateGroup and TemplateVisitor turn out to be the
only changes necessary it shouldn't take too long for someone familiar
with the code (and having access to a comprehensive test suite) to
perform the necessary changes.
>>
>> On Jan 24, 2012, at 3:19 PM, Peter Ansell wrote:
>>
>>> Hi,
>>>
>>> I am wanting to upgrade my spin-api fork to use jena-2.7.0-incubating
>>> that was released recently so that I can start using jena-2.7.0 in
>>> other linked projects, but I am having a few issues.
>>>
>>> SPIN-API currently relies on
>>> com.hp.hpl.jena.sparql.syntax.TemplateGroup and
>>> com.hp.hpl.jena.sparql.syntax.TemplateVisitor that were removed in
>>> jena-2.6.5 and hence they are not available in 2.7.0.
>>>
>>> In order to fix the TemplateVisitor issue I removed
>>> AbstractTemplateVisitor. This removal may have caused a sideeffect in
>>> SPINConstraints.convertAskToConstruct where I had to remove the
>>> reference to TemplateGroup and replace it with something. Naively I
>>> replaced it with a call to getConstructTemplate but that appears to be
>>> null in at least some cases, so two of the tests are now failing. I
>>> don't have a comprehensive test suite for SPIN so it is difficult to
>>> say what the solution might be. In addition, looking at the test using
>>> a debugger doesn't appear to show anything interesting.
>>>
>>> The Git branch that I have been working on is:
>>>
>>> https://github.com/ansell/spin/tree/feature/jena27upgrade
>>>
>>> Thanks,
>>>
>>> Peter
>>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Group "TopBraid Suite Users", the topics of which include TopBraid
>>> Composer,
>>> TopBraid Live, TopBraid Ensemble, SPARQLMotion and SPIN.
>>> To post to this group, send email to
>>> [email protected]
>>> To unsubscribe from this group, send email to
>>> [email protected]
>>> For more options, visit this group at
>>> http://groups.google.com/group/topbraid-users?hl=en
>>
>>
>
>
> --
> -------------------------------------------------------------------
> Jerven Bolleman [email protected]
> SIB Swiss Institute of Bioinformatics Tel: +41 (0)22 379 58 85
> CMU, rue Michel Servet 1 Fax: +41 (0)22 379 58 58
> 1211 Geneve 4,
> Switzerland www.isb-sib.ch - www.uniprot.org
> Follow us at https://twitter.com/#!/uniprot
> -------------------------------------------------------------------
>
>
> -- You received this message because you are subscribed to the Google
> Group "TopBraid Suite Users", the topics of which include Enterprise
> Vocabulary Network (EVN), TopBraid Composer, TopBraid Live,
> TopBraid Ensemble, SPARQLMotion, SPARQL Web Pages and SPIN.
> To post to this group, send email to
> [email protected]
> To unsubscribe from this group, send email to
> [email protected]
> For more options, visit this group at
> http://groups.google.com/group/topbraid-users?hl=en
-- You received this message because you are subscribed to the Google
Group "TopBraid Suite Users", the topics of which include Enterprise Vocabulary
Network (EVN), TopBraid Composer, TopBraid Live,
TopBraid Ensemble, SPARQLMotion, SPARQL Web Pages and SPIN.
To post to this group, send email to
[email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/topbraid-users?hl=en
