W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > April to June 2004

BRQL implementation

From: Seaborne, Andy <andy.seaborne@hp.com>
Date: Thu, 24 Jun 2004 10:37:18 +0100
Message-ID: <E864E95CB35C1C46B72FEA0626A2E80803615C65@0-mail-br1.hpl.hp.com>
To: public-rdf-dawg@w3.org

I have done a prototype implementation of some of BRQL by adding features to
Jena's RDQL implementation:


It also passes the RDQL JUnit test suite in Jena.  As the prototype took one
day, don't expect a reliable, solid implementation.  Indeed, it is no more
than proof of concept.  There are various limitations and you can write very
silly queries involving CONSTRUCT and OPTIONAL (they will fail at run time
but they should be parser errors).

Implementation (a whole Jena installation with BRQL)

Run the command "java -cp ... jena.rdfquery ....." to run BRQL.
Some trivial examples are run by the script "examples" in:

Most of the time went into the OPTIONALs parts - it is a streaming
evaluation of queries with (multiple) optional clauses.  Bindings in one
optional part are allowed to effect another optional part with evaluation
being in the order the clauses appear in the query.  This is untested and
unchecked - it wasn't an objective for this initial prototype.


DESCRIBE <uri> not done.
DESCRIBE is done by bNode graph closure and is fixed in this implementation.

Indeed, the Java interface to DESCRIBE results is going to be reworked.

Received on Thursday, 24 June 2004 05:37:35 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:00:27 UTC