TinkerPop3 Graph Structure Implementation for OrientDB
Java Scala
Latest commit d0e14e1 Nov 21, 2016 @mpollmeier mpollmeier committed on GitHub Merge pull request #112 from tshaug/master
makeActiveDb(); added to the methods getIndexManager() and getSchema()



Build Status Coverage Status Maven Central

Apache TinkerPop 3 Graph Structure Implementation for OrientDB. This started off as just a proof of concept, but thanks to a lot of help it's now in a really good shape.

Warning: While this is (as of now) the only Apache TinkerPop3 graph structure implementation for OrientDB, it's not the official one - it's not supported by the Orient team. The contributors focused on the functionality needed for their use cases, and it doesn't claim to be complete. UPDATE: since Nov 2016 OrientDB team started a process to officially support this module. A stable and supported release will be announced in the near future.

The main area that need some more work is index lookups - currently it does find the right index for a simple case, e.g. g.V.hasLabel("myLabel").has("someKey", "someValue"). However if there are multiple indexes on the same property, or if there the traversal should better use a composite index, that's not handled well yet. If you feel inclined you can add these cases to the OrientGraphIndexTest.java. The function that looks up indexes is OrientGraphStep.findIndex.


  • you can run the standard Apache TinkerPop test suite with mvn install -P release
  • there are some additional tests that you can run independently with mvn test
  • additionally there is a separate suite of tests in the tests-scala directory which you can run using sbt test
  • to automatically format the code (travis CI enforces a format check), just run mvn clean install


Have a look at the tests-scala which demonstrates the usage. There's also an orientdb example project in gremlin-scala-examples.

Labels and classes

Vertices and Edges are stored as classes based on their label. In order to allow vertices and edges to use the same label, the implementation prepends V_ or E_ in the class name:

  • vertex with label user -> classname V_user
  • edge with label user -> classname E_user


You might want to use orientdb-migrations to create a schema with indexes etc.


  • upgrade version: remove SNAPSHOT (driver/pom.xml and tests-scala/build.sbt)
  • commit on branch, push, create PR on github
  • await green light from travis
  • merge PR on github
  • mvn pull
  • mvn clean deploy -Prelease
  • git tag VERSION
  • bump versions to next SNAPSHOT (pom.xml, build.sbt)
  • git push
  • git push --tags