support SPARQL endpoints in addition to loading from turtle files?
KonradHoeffner opened this issue · comments
@Aklakan requests SPARQL endpoint support. As this would be a major addition and contrary to the initial direction, investigate how much effort this would take and what the performance would be.
btw, very nice project, and thanks for chosing Sophia to implement it :-)
You might want to look at https://github.com/pchampin/sophia_sparql_client .
I hacked it quickly as a poof of concept, but this should do the trick, and I'm willing to help improve it as needed.
PS: ultimately, I would like to include a generic SPARQL engine in Sophia, so that any Dataset (even in-memory ones) could be wrapped into a SparqlDataset. But that's not a small task...
@pchampin: Thanks! Sophia fits very well for this project because triples_with_s and triples_with_o are key for an RDF browser and sophia is optimized for that. So I'm not sure if I would change that to a SPARQL query and risk a performance overhead. I think I would rather have it as an alternative code path for people who do want to use RickView with a SPARQL endpoint but optimize RickView further for in-memory graphs using HDT, which also supports fast triples_with_s and triples_with_o queries but no SPARQL.
Not planning to implement this right now, may reopen it later.