> can you by chance share your dataset (or a generator that creates a similar > one)? Then I'd like to profile the query.
Thanks for the offer, Michael. The dataset is a direct copy from my production data, so it is not possible to share it. > > One other thing. From what I understand, those queries should return > > the same result, but they don't: > > > > start c=(typeIndex,node_type,"C"), e=(typeIndex,node_type,"E") > > match(c)-->(e) where (c.node_name = "name") return c,e > > > > start c=(typeIndex,node_type,"C") match(c)-->(e) where (c.node_name = > > "name") and (e.node_type = "E") return c,e > > What do you expect and what did they return ? And is the node_type > property indexed exactly as node_type in the index, and are _all_ nodes > connected to c just _e_ nodes with the appropriate type? The first query returns the expected results, the second query returns zero results. Node_type is correctly indexed because it is working with the first query. C nodes are connected with different nodes, but some of them are E nodes. This is basically a query like: "find all male actors (node_type:C) who played in SiFi movies (node_type:E) with lastname = name" BTW: I would assume that the following query is also equivalent, but haven't tested it yet: start c=(typeNameIndex,`node_type:"C" and node_name:"name"`), e=(typeIndex,node_type,"E") match(c)-->(e) return c,e _______________________________________________ Neo4j mailing list [email protected] https://lists.neo4j.org/mailman/listinfo/user
