Does dgraph only have a default graph?
Dgraph’s model doesn’t have the “Named graphs” concept. But you can use that concept creating your structure like that.
The RDF format used in Dgraph is a simple N-Quads. Dgraph don’t holds the data as RDF. We use only in the mutation process, but you can use JSON too.
If you need to use the “Named graphs” concept. Just create a node for each name and use its UIDs for each data as a reference. Like in Get started with Dgraph
I’m not sure about terminology here. But I guess Dgraph is a “default graph”. But you’re free to elaborate mutiple graphs (in the same DB) and perform the “Named graphs” concept. But our RDF standard and Query language doesn’t predict the “Named graphs” concept.
Cheers.
Thanks. Just curious, given what you’ve said, why not use N-Triples as the input format?
N-Triples are not compatible with Dgraph’s model ‘a priori’.
In fact you can use N-Triples at any time with Dgraph (No support for symbols). But we use N-Quads because it’s a superset of N-Triples and fits perfectly with the Dgraph model.
-
Subject, Predicate, Object, Label
, aka -
Entity, Attribute, Other Entity / Value, Label
source: https://docs.dgraph.io/design-concepts/#edges
In Dgraph you have the possibility to enter four fields of information.
E.g: Entity + Attribute/Predicate + Value/Object + Facets.