| More along the lines of a stored procedure than an extension to SPARQL. And more advanced than a simple stored query. More like a stored procedure backed by java code.
Functions look neat, but it doesn’t look like they get access to the underlying connection. Håvard On 13 Jan 2020, at 23:51, Jeen Broekstra <jeen.broekstra@xxxxxxxxx> wrote:
Cheers,
Jeen
Hi,
I got to thinking about a feature I wished had been available when working on my previous project utilizing Stardog.
In regular SQL databases you can use PL/SQL to add custom functions. One common function that I’ve always needed is support for extracting all data related to some resource based on rules that I’ve come up with. Usually some sort of recursive traversal that halts if it traverses a certain property or some other criteria.
This is usually not possible with just SPARQL since you can’t do ?a ?b* ?c.
When running the RDF4J server/workbench, I wonder if anyone has requested something like a class loader approach to side load functions?
A user could easily implement this on their client, but network latency and query parsing will add a decent amount of overhead.
Håvard
_______________________________________________
rdf4j-dev mailing list
rdf4j-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/rdf4j-dev
|
