Details
-
Type:
Improvement
-
Status: Closed
-
Priority:
Major
-
Resolution: Implemented
-
Affects Version/s: 3.2.0-incubating
-
Fix Version/s: 3.2.0-incubating
-
Component/s: hadoop
-
Labels:None
Description
so that the user can simply do graph.compute()... without having to specify the graph computer.
As of 3.0, we HadoopGraph does NOT have a default graph computer. This was done to not favor one technology over the other. We could allow the configuration to have a gremlin.hadoop.defaultGraphComputer=org.tinkerpop.SparkGRaphcomputer
That configuration would be carried through the various result sets so that the user would only have to choose the graph computer initially (or it is pre-selected by the graph implementation) and from that point on would be defaulted.
Activity
- All
- Comments
- Work Log
- History
- Activity
- Transitions
This is the logic. This will go into an upcoming PR that has a bunch of other Hadoop configuration related work.
@Override public GraphComputer compute() { if (this.configuration.containsKey(Constants.GREMLIN_HADOOP_DEFAULT_GRAPH_COMPUTER)) { try { return this.compute((Class<? extends GraphComputer>) Class.forName(this.configuration.getString(Constants.GREMLIN_HADOOP_DEFAULT_GRAPH_COMPUTER))); } catch (final Exception e) { throw new IllegalStateException(e.getMessage(), e); } } else throw new IllegalArgumentException("There is no default GraphComputer for HadoopGraph. Use HadoopGraph.compute(class) or gremlin.hadoop.defaultGraphComputer to specify the GraphComputer to use."); }