Titan 0.3.0 introduced the inmemory backend. I think there would be real value in being able to spin up an in memory graph and stuff a subset of the main graph inside of there to do what you'd like with it without changing the main graph, whether it be queries, modifications, or whatever. Depending on how Fulgora ends up shipping, this could be very useful.
Implementing this would be as simple as providing a with-in-memory method in graph.clj. It would take in a collection of identifiers of nodes and edges, suck all those objects in from the main graph, and then rebind all the methods inside of the with-in-memory to use the inmemory graph. Soliciting feedback from the Fulgora Folks to see what they will end up shipping.
Titan 0.3.0 introduced the
inmemory
backend. I think there would be real value in being able to spin up an in memory graph and stuff a subset of the main graph inside of there to do what you'd like with it without changing the main graph, whether it be queries, modifications, or whatever. Depending on how Fulgora ends up shipping, this could be very useful.Implementing this would be as simple as providing a
with-in-memory
method ingraph.clj
. It would take in a collection of identifiers of nodes and edges, suck all those objects in from the main graph, and then rebind all the methods inside of thewith-in-memory
to use the inmemory graph. Soliciting feedback from the Fulgora Folks to see what they will end up shipping.