Lyndon (or anyone who might have thoughts) - I like the pattern started for making better use of docker to solve TinkerPop testing issues. We need more smart things to happen with docker to keep our code easy to build, test and manage. Now that i'm digging into it, I have developed some thoughts/questions.
1. I see the docker-compose.yml file and that it references the Gremlin Server image of: tinkerpop/gremlin-server:3.5 but doesn't that just start a Gremlin Server on 8182 which would start it with an empty graph by default? where does the test data for all the different graphs required by Gherkin get loaded? 2. Is there a way to get docker-compose to use the local SNAPSHOT of Gremlin Server rather than pinning to tinkerpop/gremlin-server:3.5? 3. I think that this model gets rid of the complicated configuration of the .glv file which "turns on" the Maven <profile> for GLV building. We can simply require Maven+docker-compose to do a build and streamline all of that sort of stuff. Does that sound right?
