CODEWITHSUNDEEP
tinkerpopgremlin-server
learnAndImprove
learnAndImprove

Reputation: 1359

Storing graph to gremlin server from in memory graph

I'm new to Graphs in general. I'm attempting to store a TinkerPopGraph that I've created dynamically to gremlin server to be able to issue gremlin queries against it.

Consider the following code:

    Graph inMemoryGraph;
    inMemoryGraph = TinkerGraph.open();
    inMemoryGraph.io(IoCore.graphml()).readGraph("test.graphml");
    GraphTraversalSource g = inMemoryGraph.traversal();

    List<Result> results = 
      client.submit("g.V().valueMap()").all().get();

I need some glue code. The gremlin query here is issued against the modern graph that is a default binding for the g variable. I would like to somehow store my inMemoryGraph so that when I run a gremlin query, its ran against my graph.

Upvotes: 0

Views: 1374

Answers (1)

stephen mallette
stephen mallette

Reputation: 46226

All graph configurations in Gremlin Server must occur through its YAML configuration file. Since you say you're connected to the modern graph I'll assume that you're using the default "modern" configuration file that ships with the standard distribution of Gremlin Server. If that is the case, then you should look at conf/gremlin-server-modern.yaml. You'll notice that this:

graphs: {
  graph: conf/tinkergraph-empty.properties}

That creates a Graph reference in Gremlin Server called "graph" which you can reference from scripts. Next, note this second configuration:

org.apache.tinkerpop.gremlin.jsr223.ScriptFileGremlinPlugin: {files: [scripts/generate-modern.groovy]}}}

Specifically, pay attention to scripts/generate-modern.groovy which is a Gremlin Server initialization script. Opening that up you will see this:

// an init script that returns a Map allows explicit setting of global bindings.
def globals = [:]

// Generates the modern graph into an "empty" TinkerGraph via LifeCycleHook.
// Note that the name of the key in the "global" map is unimportant.
globals << [hook : [
  onStartUp: { ctx ->
    ctx.logger.info("Loading 'modern' graph data.")
      org.apache.tinkerpop.gremlin.tinkergraph.structure.TinkerFactory.generateModern(graph)
  }
] as LifeCycleHook]

// define the default TraversalSource to bind queries to - this one will be named "g".
globals << [g : graph.traversal()]

The comments should do most of the explaining. The connection here is that you need to inject your graph initialization code into this script and assign your inMemoryGraph.traversal() to g or whatever variable name you wish to use to identify it on the server. All of this is described in the Reference Documentation.

There is a way to make this work in a more dynamic fashion, but it involves extending Gremlin Server through its interfaces. You would have to build a custom GraphManager - the interface can be found here. Then you would set the graphManager key in the server configuration file with the fully qualified name of your instance.

Upvotes: 2

Related Questions