Open Tpt opened 4 years ago
What's the planned approach here, doing it via SPARQL in the background? I've tested several SHACL implementations and it looks like most of them do some in-memory validation, which might or might not work for large datasets.
From what I heard Stardog is translating shapes to SPARQL and validates them like this. This has the benefit that it can use indexes that are there anyway already.
I have no real plan at the moment. My current goal is pushing SPARQL update and implementing full text search, RDF* and query optimization. ShEX/SHACL is more something for the distant future. Using SPARQL (or at least a part of the SPARQL evaluator) is indeed probably the best way to do validation.
Ok I think that makes sense. To attract more people I think it's important to have a usable store that people can work with in production and then slowly start improving query planer/performance.
This will hopefully attract more users, contributors & money so other features like SHACL can be added.
Shex would be so cool! As an aside, do you know if there are any shex rust libraries? (I didn't find any after a quick look)
As an aside, do you know if there are any shex rust libraries?
Not at my knowledge, sadly.
Interested in SHACL support. Particularly SHACL Rules - https://www.w3.org/TR/shacl-af/#rules.
sophia is a Rust project that may also be interested in SHACL - https://github.com/pchampin/sophia_rs/issues/108
As an aside, do you know if there are any shex rust libraries?
Not at my knowledge, sadly.
I got to know someone working on a Rust implementation of ShEx, you can find the Github repository here.
It would be nice to have built-in support for shape validations in Oxigraph. It might be done by implementing ShEx or SHACL, or maybe even both.