[jira] Created: (CLEREZZA-467) Add support for SPARQL queries against multiple named graph on the Jena TDB store

["Olivier Grisel (JIRA)" <ji...@apache.org>]

Add support for SPARQL queries against multiple named graph on the Jena TDB store
---------------------------------------------------------------------------------

                 Key: CLEREZZA-467
                 URL: https://issues.apache.org/jira/browse/CLEREZZA-467
             Project: Clerezza
          Issue Type: Improvement
            Reporter: Olivier Grisel


Jena TDB supports queries involving mutliple named graph (along with a default graph).

It would be very useful if those named graphs could be mapped to the Clerezza graph ids used by TcProvider.

Implementing this will likely require a change in the current filesystem directory structure used by the TdbTcProvider#getMGraph implementation: graphs are currently stored in isolated datasets under separate folders on the hard drive. Instead one should use a common dataset folder and make TdbTcProvider#getMGraph calls:

		Model model = dataset.getNamedModel(graphUri);

instead of the current use of the default model of each isolated dataset:

		Model model = dataset.getDefaultModel();

A migration path from the old storage structure to the new one should be provided as part of the release notes.

Relevant parts of the Jena TDB documentation:

http://www.openjena.org/wiki/TDB/Datasets
http://www.openjena.org/wiki/TDB/DynamicDatasets



--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira