Looks like this is because you are indexing your property as text. See https://docs.janusgraph.org/index-backend/text-search/#full-text-search_1 I believe the query has(“country", P.neq(null)) would

I've tried with the graph of the gods (steps shown below), and the indexes seems to be persisted and working fine (don't see any warning when querying by name or age). So I guess I have some issue

$ ./bin/gremlin.sh gremlin> graph = JanusGraphFactory.open("/graph/graph.properties") gremlin> g = graph.traversal() gremlin> g.V().has("country",

Can you show the profile output? Just add .profile() to the end of your query. 「Laura Morales <lauretas@...>」在 2021年7月23日 週五，下午4:13 寫道：

I was able to open my lucene folder with a tool called "luke" for peeking at the data, and all the "country" values seems to be in there. I assume by looking at this that the index was created and

Unfortunately this does not seem to remove the warning. WARN org.janusgraph.graphdb.transaction.StandardJanusGraphTx - Query requires iterating over all vertices [(country <> null)]. For better

Try rewriting your query as: ` g.V().has("country", P.neq(null)).values("country").dedup().order().limit(10)` FYI starting from the upcoming 0.6.0 release, has("country") will leverage the mixed

My ".properties" file: gremlin.graph=org.janusgraph.core.JanusGraphFactory storage.backend=berkeleyje storage.directory=/graph/db index.search.backend=lucene

Hi, it means your query is not leveraging your indexes. Can you provide the query and also the output of `mgmt.printIndexes()` (or how you created the indices)?

I was able to create a "Composite Index", but I cannot seem to be able to create a "Mixed Index" with berkeleyjs and lucene. What I've done: - I've modified the ".properties" file

Hi All, I am trying to run SparkGraphComputer using tinkerpop 3.4.1 on Spark2.3 having Hadoop3 but getting issues, not able to connect with Spark. As per

Thank you a lot for the help, I could not figure out what the problem was! Is it normal that the requirement is so high? 5GB of space for an empty database? And can it be changed somehow with some

Hi Laura, Down the stacktrace you can note: After some googling you can conclude that your JanusGraph disk is short of 4.7 GB of space for running JanusGraph with BerkeleyjeDB. Maybe check the logs

Hi Laura, A similar question was posed recently: https://lists.lfaidata.foundation/g/janusgraph-users/message/5986 So, 1. Only for the CompositeIndex 2. In your specific example, you could use the

What am I doing wrong? I cannot get Janus to start. $ java -version openjdk version "1.8.0_252" OpenJDK Runtime Environment (build 1.8.0_252-8u252-b09-1~deb9u1-b09) OpenJDK 64-Bit Server VM (build

What are the practical implications of using Object.class as a property type, instead of the other native types (eg. String.class, Integer.class, etc.)? IIUC Object.class means that a property can

If I setup an empty graph with persistent storage, for example berkeley (thus not an in-memory graph), can I load a graphml/graphson file and have it all added to the

Hi, You can first try to write a custom VertexProgram for belief propagation with Apache TinkerPop. A custom VertexProgram supports the massive message parsing needed for belief propagation. If it

Hi Laura, JanusGraph support for loading data does not go further than the Traversal API (used in https://tinkerpop.apache.org/docs/current/reference/#addvertex-step ) and the JanusGraphManagement API

I'm moving from AWS Neptune architecture to JanusGraph in CentOS. Anyone has tried implementing Belief Propagation with JanusGraph?