Useful notes on getting started including links to tutorials.

Beginner tutorials on using the gremlin console and basic query syntax:

http://tinkerpop.apache.org/docs/3.3.3/tutorials/getting-started/

Example of how to set up the gremlin console to point to a remote server:

https://docs.aws.amazon.com/neptune/latest/userguide/access-graph-gremlin-console.html

Getting started example of using the java library

https://docs.aws.amazon.com/neptune/latest/userguide/access-graph-gremlin-java.html

Some AWS specific Gremlin implementation notes:

https://docs.aws.amazon.com/neptune/latest/userguide/access-graph-gremlin-differences.html

Assuming a Neptune instance is running and an EC2 instance is all set up.

Curently the EC2 is not exposed publically so to connect to Neptune you need to open an ssh tunnel to forward requests to localhost:8182 to EC2 - you will need to keep this open.

ssh -i ~/.ssh/ons-web-development.pem -L 8182:<AWS_NEPTUNE_ENDPOINT>:8182 ec2-user@<AWS_EC2_PORT>

Gremlin console is a terminal tool which allows you query the database. Follow the AWS steps to set it up. (Including the instuctions on how to connect to a remote DB):

https://docs.aws.amazon.com/neptune/latest/userguide/access-graph-gremlin-console.html

From your home directory on the EC2 box run:

wget https://archive.apache.org/dist/tinkerpop/3.3.2/apache-tinkerpop-gremlin-server-3.3.2-bin.zip

unzip apache-tinkerpop-gremlin-server-3.3.2-bin.zip

cd apache-tinkerpop-gremlin-server-3.3.2

Start the server and point it at your AWS Neptune instance.

./bin/gremlin-server.sh conf/gremlin-server.yaml

From your machine if you curl localhost:8182 you should get a response similar to:

{
  "requestId": "38b2c520-d97a-f5ff-414e-7d9a5c6de398",
  "code": "MissingParameterException",
  "detailedMessage": "no gremlin script supplied"
}

You are successfull able to hit the Gremlin server running on the EC2 instance (whoop!)

The app is as basic as it comes and simply takes input from the console and uses the gremlin client to send queries to the gremlin server. Simply type in your query and hit enter (or enter exit to quit).

From the project root

Build it

mvn clean package -DskipTests=true 

Run it

java -jar target/dp-neptune-spike-1.0-SNAPSHOT.jar -query

Execute a query (see the gremlin tutorials for syntax)

grem-query> g.addV("person").property("name", "Dave").next()
grem-query> g.V().valueMap(true)
> {id=50b2c521-8a91-457f-4a82-eb4d29c6ef67, label=person, name=[Dave]}

The above example runs 2 queries, the first creates a new vertex labelled "person" with a name property "Dave". The second queries for all vertices and returns them with all their properties and labels - which in this case returns the vertex created in the previous query.