News for hackers. Stuff that matters.

A comment on Crux SQL.

Re:Crux SQL by liminal on Tuesday August 04, 2020 @02:14PM (#24052841)
This looks really interesting - especially the temporal aspects. Does anyone have any insight into using it with RDF/SPARQL data/queries?
[ Parent | on HN ]
Re:Crux SQL by lgessler on Tuesday August 04, 2020 @02:21PM (#24052922)
From their docs[0]:

> The REST API also provides an experimental endpoint for SPARQL 1.1 Protocol queries under /sparql/, rewriting the query into the Crux Datalog dialect. Only a small subset of SPARQL is supported and no other RDF features are available.

[0]: https://opencrux.com/docs

[ Parent | on HN ]
Re:Crux SQL by refset on Tuesday August 04, 2020 @04:21PM (#24054100)
The context here is that we benchmark ourselves against the likes of Neo4j and RDF4J using SPARQL test suites, in particular WatDiv [0] and LUBM [1] which are specifically designed for stress testing various subgraph matching queries [2]. Therefore we had to translate the data sets and queries from RDF/SPARQL to edn/Datalog etc.

There is an open issue in regards more general RDF support with details on the kinds of things Crux would need to add: https://github.com/juxt/crux/issues/317

[0] https://dsg.uwaterloo.ca/watdiv/

[1] http://swat.cse.lehigh.edu/projects/lubm/

[2] https://en.wikipedia.org/wiki/Subgraph_isomorphism_problem

[ Parent | on HN ]