Working with rdflib 4.2 from Python, Sesame, and trying SERVICE (for the similar use case as Can I combine local and remote dataset within SPARQL query?). However, it seems like rdflib does not support the SERVICE:
Exception: ServiceGraphPattern not implemented
As I still want to combine local graph, which contains client program's local state and other context, which is supposed to be used together with a remote RDF storage via SPARQL.
What would be a possible workaround, except for storing local data on the remote first (it is probably ok to send local data to remote without persisting)? Possible complication is that remote endpoint (in Sesame) performs OWL reasoning, and I would like for the local data to benefit from that too. Loading graph from remote to local is not an option, as the dataset could be too big without local restrictions.
What options do I have if I still want to use rdflib with Sesame over SPARQL / SPARQLUpdate? How they compare in terms of computational efficiency?
EDIT: found one possible workaround described here: SPARQL query on the remote remote endpoint RDFLib / Redland but it is too low level.
