Jena with LDF server

Hello, I am relatively new to the field and was wondering how Jena codes can be invoked in the LDF server available at http://linkeddatafragments.org/. The above LDF server is written using server side java script and is based on node.js package. I want to write some code in Jena (in form of some

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

Hi Andy and Stephen, Thank you very much for your help and efforts. Hmm, well, to be honest, I totally agree with what you are discussing about :D, but what should I do now? Regards, Yang On 03/16/2015 09:17 PM, Andy Seaborne wrote: On 16/03/15 18:00, Stephen Allen wrote: On Mon, Mar 16, 2

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

On 16/03/15 18:00, Stephen Allen wrote: On Mon, Mar 16, 2015 at 12:21 PM, Andy Seaborne wrote: On 16/03/15 16:07, Stephen Allen wrote: This looks like a possible combination of two bugs. In TextDocProducerTriples, there is a ThreadLocal called inTransaction that should be subclassing ThreadL

Re: How to write RDF model to a Semantic Database

Have you tried looking at the official documentation? Here is an example with transactions: http://jena.apache.org/documentation/tdb/tdb_transactions.html#write-transactions On Mon, Mar 16, 2015 at 6:54 PM, Prasad Priyadarshana Fernando wrote: > Can anyone give me an example for writing the RDF

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

On Mon, Mar 16, 2015 at 12:21 PM, Andy Seaborne wrote: > On 16/03/15 16:07, Stephen Allen wrote: > >> This looks like a possible combination of two bugs. In >> TextDocProducerTriples, there is a ThreadLocal called inTransaction that >> should be subclassing ThreadLocal and overriding the initialV

Re: How to write RDF model to a Semantic Database

Can anyone give me an example for writing the RDF module to a TDB data set? Thanks *Prasad Priyadarshana Fernando * Mobile: +1 330 283 5827 On Sun, Mar 15, 2015 at 1:46 PM, Prasad Priyadarshana Fernando < bpp...@gmail.com> wrote: > Hi, > > Can I have s

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

Hi Andy, Thank you very much for your enthusiastic help :D Then the current config is changed to: ``` <#tdb_inf_ds> a ja:RDFDataset ; ja:defaultGraph <#tdb_graph> ; . <#tdb_graph> a tdb:GraphTDB ; tdb:dataset <#tdb_ds> . <#tdb_ds> a tdb:DatasetTDB; tdb:location "Data";

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

On 16/03/15 16:07, Stephen Allen wrote: This looks like a possible combination of two bugs. In TextDocProducerTriples, there is a ThreadLocal called inTransaction that should be subclassing ThreadLocal and overriding the initialValue() method to return false. It is not doing so. This would be O

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

Sorry - I was too hasty to asking the question. The text dataset has to be a dataset. There are two in your configuration: <#tdb_inf_ds> and <#tdb_ds>. You said <#tdb_ds> works. The other case is to still have a general purpose dataset with a single graph in it but skip the RDFS. Where it

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

This looks like a possible combination of two bugs. In TextDocProducerTriples, there is a ThreadLocal called inTransaction that should be subclassing ThreadLocal and overriding the initialValue() method to return false. It is not doing so. This would be OK if the start()/finish() methods were bei

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

Hi Andy, Thank you for the explanation for the updates. As for the current issue, only :text_dataset -> <#tdb_ds> works. :text_dataset -> <#tdb_graph> -> <#tdb_ds> will lead to such an exception: [2015-03-16 15:26:25] Server ERROR Exception in initialization: caught: com.hp.hpl.jena.rdf.m

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

Hi Yang, That's really useful. Was in both cases of :text_dataset -> <#tdb_graph> -> <#tdb_ds> and :text_dataset -> <#tdb_ds> or just the latter. 1.1.1 to 1.1.2. is a maintenance update to Fuseki1. Fuseki2 server is restructured to make the security work, add the admin operations so

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

Hi Andy, Ha, your suggestion is to the point precisely. It works after removing RDFS reasoner, in both 1.1.2 and 2! So it seems the problem is from RDFS side. From 1.1.1 to 2 is a huge leap, but is there also a big improvement from 1.1.1 to 1.1.2? They collaborated very well before. Regards,

Re: Federated query

You'd need to embed SELECT DISTINCT inside the SERVICE statement SERVICE { Foo } Translates to SELECT * WHERE { Foo } on the remote endpoint, Foo can be a sub-query if you want to apply modifiers to the remote query Rob On 16/03/2015 13:05, "Mark Feblowitz" wrote: >That reminds me of a questi

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

Thanks - that suggest it is not Fuseki itself but other changes in this release. I don't understand why it's aborting at all, uploads aren't supposed to abort. One more question: What happens if there is no RDFS? Either using a dataset directly for :text_dataset or just removing the RDFS wr

Re: Federated query

That reminds me of a question I have: for a DISTINCT result from the service, would one embed a SELECT DISTINCT within the SERVICE statement? Or is that assumed to be performed in the enclosing context? Mark > On Mar 16, 2015, at 6:12 AM, Andy Seaborne wrote: > > On 16/03/15 00:37, Kamalraj

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

Hi Andy, Thanks for your reminder. We are using 1.1.1. But I just gave a try on 1.1.2. The result was almost the same: ``` 13:47:38 INFO [1] Upload: Graph: default (6 triple(s)) 13:47:38 WARN Exception during abort (operation attempts to continue): Can't abort a write lock-transaction 13:47

Re: Programmatic SPARQL 1.1

You can use SPIN API an operate the SPARQL query structure as RDF. We have developed a helper class for this purpose: https://github.com/Graphity/graphity-client/blob/master/src/main/java/org/graphity/processor/update/ModifyBuilder.java On Tue, Mar 10, 2015 at 9:32 AM, Maarten van Dessel wrote:

Re: Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

Hi Yang, One quick question - does your configuration work with Fuseki 1.1.2? It was released at the same time as Fuseki2 (last Friday). It would help in knowing if this is a text indexing issue of a Fuseki server issue. 2.0.0 and 1.1.2 use the same text indexing code. Andy On 16

Re: Programmatic SPARQL 1.1

For the sake of completeness I send my current solution to this problem... "How to create a programmatic SPARQL 1.1 Update query" ElementGroup queryPattern = ... ElementTriplesBlock deleteTriples = ... ElementTriplesBlock insertTriples = ... ParameterizedSparqlString qstring = new ParameterizedSp

Unable to enable full text search in Fuseki2 given a working config file previously used in Fuseki1

Hi there, We upgraded to Fuseki 2 as soon as it is released. However the configuration file we are using for Fuseki 1 is no longer working (or compatible?) with Fuseki 2. We spent quite a lot of time to tune it but we failed. There is no sufficient reference for this issue on your website, ei

Re: Changed model

That's a good way for a single model. For a whole dataset, there is DatasetGraphMonitor which combines a DatasetGraph and a notifier DatasetChanges. e.g. DatasetChangesCapture If you wrap your Dataset(Graph) with DatasetGraphMonitor you should get a call on every add/delete. There is suppo

Re: Federated query

On 16/03/15 00:37, Kamalraj Jairam wrote: Hello, if i create federated query like CONSTRUCT { ?s ?o .} WHERE { SERVICE { ?s ?o } SERV

Re: TDB Optimization

The biggest single thing you can and should do is to upgrade to Jena 2.13.0 as Andy already suggested 2.11.0 is from Sep 2013 I.e. 18 months ago, there have been 5 releases of Jena (and thus TDB) since then many of which have included substantial improvements to the query optimiser Rob On 15/03/

Re: Changed model

I found the solution after some more research: https://mail-archives.apache.org/mod_mbox/jena-users/201311.mbox/%3ccaoqrjk5dtbiqrjja__ru-jqmzpkcdsktgksms+fhzkge8qo...@mail.gmail.com%3e

Changed model

Hi everyone, Is there any way to discover if an UpdateRequest (eg. "INSERT { } WHERE { }") truly changed a model after execution? There are cases where a WHERE-clause can be to restrictive, which would result in the original model. Code example of how I see the solution: Model model

Re: TDB Optimization

On Sat, 14 Mar 2015 02:43:59 +0100, Abduladem Eljamel wrote: Hi All I loaded DBPedia RDF dump File to JenaTDB store. The triples were loaded successively including the indexingphase and “stats,opt” file was created; however, the SPARQL queries are notworking for complex queries. It tak