Thanks for the information Andy.
Currently we are using jena-2.8.8

Can you also please share some documentation on how can we use Fuseki with TDB 
and interface RDFConnection?

Regards,
Ashwani


On 06/01/20, 4:29 PM, "Andy Seaborne" <[email protected]> wrote:

    
    
    
    
    On 06/01/2020 05:50, Ashwani Rathi wrote:
    
    > Hi All,
    
    > We have the following method where we are using Jena 2
    
    
    
    Which version of jena 2 exactly?
    
    
    
    , where we are trying to fetch/ create Model object from ModelRDB 
    
    utility method using com.hp.hpl.jena.db.IDBConnection
    
    >      private Model getModelFromRDFStore(String modelName, IDBConnection 
conn)
    
    >      {
    
    >          Model model = null;
    
    >          if (!conn.containsModel(modelName)) {
    
    >              model = ModelRDB.createModel(conn, modelName);
    
    >          }
    
    >          else {
    
    >              model = ModelRDB.open(conn, modelName);
    
    >          }
    
    > 
    
    >          return model;
    
    >      }
    
    > 
    
    > com.hp.hpl.jena.db.IDBConnection basically encapsulates a JDBC connection.
    
    > 
    
    > Now support for com.hp.hpl.jena.db.IDBConnection and corresponding 
implementing classes has been removed from Jena 3.
    
    > This is the functionality which we are trying to replicate in Jena 3.
    
    
    
    See prvious reply - it depends on what the application is trying to do, 
    
    and what sort fo reliability of data it requires.
    
    
    
    For just a local data base, use TDB.
    
    Note: all storage in jena3 is based on "Datasets", not "Models". Use the 
    
    default model of a dataset for a single RDF graph in a database.
    
    
    
    For a database on a remote server shared between several instances of 
    
    your application, use Fuseki with TDB and interface RDFConnection.
    
    
    
    
    
    > 
    
    > So can you please guide how can we achieve this functionality in Jena 3 
using TDB or something else?
    
    
    
    Dataset dataset = TDB2Factory.connectDataset("some directory");
    
    dataset.getDefaultModel();
    
    
    
    but also all access to the database needs to be in a dataset transaction 
    
    See Txn and the documentation
    
    
    
    
https://urldefense.proofpoint.com/v2/url?u=https-3A__jena.apache.org_documentation_txn_&d=DwIDaQ&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=ld8gXdwUlVansMT0flGoGNN0AOkhQba_hAJM5PewJCA&m=CThVIE5cplymnNcjUUNuq_4CFBNj8BUjRg3PWigMKAE&s=lEE_JneU-vmNLseoh-8pF4OwY1uoaD2EZetxHezQnSs&e=
 
    
    
    
         Andy
    
    
    
    > 
    
    > Regards,
    
    > Ashwani
    
    > 
    
    > On 06/01/20, 3:55 AM, "Andy Seaborne" <[email protected]> wrote:
    
    > 
    
    >      
    
    >      
    
    >      On 05/01/2020 09:15, Ashwani Rathi wrote:
    
    >      > Hi,
    
    >      >
    
    >      > We are upgrading from jena 2 to jena 3 in our project.
    
    >      >
    
    >      > While jena used to have the following two classes in jena.jar:
    
    >      > com.hp.hpl.jena.db.DBConnection
    
    >      > com.hp.hpl.jena.db.IDBConnection
    
    >      
    
    >      That is the very old RDB layer; I am no sure it was ever released as
    
    >      Apache Jena release. Which version of Jena do you have? 2.6.4 or 
earlier?
    
    >      
    
    >      > there are no such classes in jena 3
    
    >      
    
    >      There have been a lot of changes, including going from RDF 1.0 to 
RDF 1.1.
    
    >      
    
    >      > We are using these classes in our project.
    
    >      >
    
    >      > Can someone guide me or provide a prototype on what claaes in jena 
3 can be used to provide functionality for these missing classes?
    
    >      
    
    >      The current persistent storage options are TDB (TDB1, and TDB2). 
These
    
    >      are custom RDF storage systems.  They scale better, load faster, run
    
    >      faster and are more robust than RDB. There is a SQL-backed storage
    
    >      system as well but we don't recommend it for any new work - and it is
    
    >      not compatible in any way with RDB.
    
    >      
    
    >      But if, as I suspect, this is a big version jump, maybe the place to
    
    >      start is describing what use of RDF is made by your project?  Is it
    
    >      using the Model API? Some early SPARQL?
    
    >      
    
    >           Andy
    
    >      
    
    >      > Regards,
    
    >      >
    
    >      > Ashwani
    
    >      >
    
    >      >
    
    >      
    
    > 
    
    > 
    
    


Reply via email to