CODEWITHSUNDEEP
neo4jcypher
adam
adam

Reputation: 978

Any length relationship between two nodes that share the same label

I'm trying to find a query that will show me any length relationship that exists between two nodes that share the same index. Basically, if there is any overlap between for a specific label. My graph is Pretty simple and not particularly large: 

(m:`Campaign`), (n:`Politician`), (o:`Assistant`), (p:`Staff`), (q:`Aid`),  (s:`Contributor`)
(m)<-[:Campaigns_for]-(n)
(o)<-[:works_for]-(m)
(p)<-[:works_for]-(o)
(q)<-[:volunteers_for]-(p)
(m)<-[:contributes_to]-(s)

I want to find all the shared nodes and their relationships between Campaigns. so far i have: 

MATCH (n:`Campaign`)-[r*]-(m:`Campaign`)
RETURN n,count(r) as R,m
ORDER BY R DESC

but it's not returning everyhing I want, I want in addition to the counts, the labels of each relationship and the names of the nodes in between.

Upvotes: 1

Views: 150

Answers (1)

BtySgtMajor
BtySgtMajor

Reputation: 1462

Assuming that "names of the nodes" means "return the name property of the node" (you could always substitute in "labels(n)" if you're after labels), then something like this might work, but you have some aggregation going on here so you may need to parse a bit:

MATCH p =(a:Campaign)-[r*]-(b:Campaign)
RETURN a, length(relationships(p)) AS count, b, extract(x IN relationships(p)| type(x)), extract(x IN nodes(p)| x.name)
ORDER BY count DESC

I'm also assuming that when you say "not returning everything you want", you mean that in addition to what's currently returned in your result set, you want just those other items you listed.

Keep in mind it might also be possible to have a cycle in your graph (not knowing too much about your particular graph), so, you may want to check your beginning and ending nodes.

Upvotes: 2

Related Questions