Hi there, Don't think there is better documentation other than this: http://docs.janusgraph.org/latest/serializer.html I did try to create my own version of an Object serializer once but I just

Hi, I have been evaluating JanusGraph for the past few weeks and am still unsure of what files are crucial to run an instance of JanusGraph. Are there any example of a simple project that has the bare

I mean open new API to get the customized data or post formatted data and load to JanusGraph.

Hi Joe, To shed some more light on the running figures you presented, I ran some tests on my own cluster: 1. I loaded the default janusgraph-hbase table with the following simple script from the

I created a separate issue to track index tool improvement - https://github.com/JanusGraph/janusgraph/issues/461 Robert Dale

Feel free to add to this issue - https://github.com/JanusGraph/janusgraph/issues/460 Robert Dale

Thanks a lot for the clarification. However, here is what is happening: gremlin> mgmt.awaitGraphIndexStatus(graph, 'movieid').call() ==>GraphIndexStatusReport[success=false, indexName='movieid',

Actually, I still prefer what I suggested -- that the code example of Graph of the Gods be used to illustrate the complete example beginning to end. I already reviewed the page you referred to and

Is there any document about how the data is serialized in Cassandra? We'd like to convert those raw data back to original form.

Please note "Graph indexes built against newly defined property keys, i.e. property keys that are defined in the same management transaction as the index, are immediately available. Graph indexes

There are no ready to run tools for this. You will have to write a small script to parse the csv file, create vertexes and edges. You can search for 'titan bulk load csv' for some loading strategies.

You can do that. I create each index in separate transactions. I push the `awaitGraphIndexStatus` to a thread pool so subsequent indexes can be created. Then after all that, I loop through and ENABLE

Something like this? http://tinkerpop.apache.org/docs/current/reference/#vertex-steps Robert Dale

Hi Elis, Did you figure this out? I am in the same predicament. I have a simple graph of movies and persons who have acted or directed (edge) in those movies. I have downloaded a sample of the IMDB

Hi folks, So I created an index like so: gremlin> graph.tx().rollback() ==>null gremlin> mgmt = graph.openManagement() ==>org.janusgraph.graphdb.database.management.ManagementSystem@6843fdc4 gremlin>

I do have a follow-up question. Do the vertex ids have to be unique across all vertices? I am assuming they do. So, it is not like I can have vertex ids unique for within a label 'person' but not

Thanks a lot Robert! We should documents it somewhere as clearly as you articulated it! Rohit

I think the root of my problems is that I am trying to create two indexes on the same property key, but with different label constraints on the vertices. This seemed like a very natural and useful

There's a wealth of information in the Apache TinkerPop reference documentation. I would consider it prerequisite reading material because JanusGraph does not duplicate all of its information. As

Robert, Yes, I looked at those but those did not help clarify it as well as looking at the code that creates the Graph of the Gods schema did. I think it will be very useful to add the code from the