I would like to be able to specify the neo4j (2.0+) database folder when I open a connection to a graph using bulbs. The goal is to be able to open different or multiple neo4j databases (folders in data/) on the same machine without starting the neo4j server. This works with the Gremlin console. I suspect the route to doing this with bulbs would be to use the bulbs Rexster client (http://bulbflow.com/docs/api/bulbs/rexster/client/) and somehow automatically start a Rexster server pointing to the folder. Has anybody done this or can anyone contribute advice toward such goals?
many thanks
Bulbs was designed to make it easy to work with multiple graph databases.
Configure your rexster.xml
for each Neo4j database you want to run (each will have a different name and thus a different URL path), and then create a separate Bulbs Config
and Graph
object for each database:
>>> from bulbs.rexster import Graph, Config
>>>
>>> config = Config('http://localhost:8182/graphs/somegraph')
>>> g1 = Graph(config)
>>>
>>> config = Config('http://localhost:8182/graphs/anothergraph')
>>> g2 = Graph(config)
Here are the relevant docs...
Rexster Configuration (rexster.xml)
https://github.com/tinkerpop/rexster/wiki/Rexster-Configuration
Rexster/Neo4j2 Configuration
https://github.com/tinkerpop/rexster/wiki/Specific-Graph-Configurations#neo4j2
Bulbs Config