Neo4j只显示一个节点的1对多关系

Question

i'm trying to solve a problem of the 1: many relationship display in neo4j. My dataset is as below

child,desc,type,parent
1,PGD,Exchange,0
2,MSE 1,MSE,1
3,MSE 2,MSE,1
4,MSE 3,MSE,1
5,MSE 4,MSE,1
6,BRAS 1,BRAS,2
6,BRAS 1,BRAS,3
7,BRAS 2,BRAS,4
7,BRAS 2,BRAS,5
10,NPE 1,NPE,6
11,NPE 2,NPE,7
12,OLT,OLT,10
12,OLT,OLT,11
13,FDC,FDC,12
14,FDP,FDP,13
15,Cust 1,Customer,14
16,Cust 2,Customer,14
17,Cust 3,Customer,14

LOAD CSV WITH HEADERS FROM 'file:///FTTH_sample.csv' AS line 
CREATE(:ftthsample
{child_id:line.child,
desc:line.desc,
type:line.type,
parent_id:line.parent});

//Relations 
match (child:ftthsample),(parent:ftthsample)
where child.child_id=parent.parent_id 
create (child)-[:test]->(parent)

//Query:
MATCH (child)-[childrel:test*]-(elem)-[parentrel:test*]->(parent) 
WHERE elem.desc='FDP'
RETURN child,childrel,elem,parentrel

It returns a display as below.

I want the duplicate nodes to be displayed as one. Newbie with Neo4J. Can anyone of the experts help please?

Answer 1

This seems like an error in your graph creation query. You have a few lines in your query specifying the same node multiple times, but with multiple parents:

6,BRAS 1,BRAS,2
6,BRAS 1,BRAS,3

I'm guessing you actually want this to be a single node, with parent relationships to nodes with the given parent ids, instead of two separate nodes.

Let's adjust your import query. Instead of using a CREATE on each line, we'll use MERGE, and just on the child_id , which seems to be your primary key (maybe consider just using id instead, as a node can have an id on its own, without having to consider the context of whether it's a parent or child). We can use the ON CREATE clause after MERGE to add in the remaining properties only if the MERGE resulted in node creation (instead of matching to an existing node.

That will ensure we only have one node created per child_id .

Rather than having to rematch the child, we can use the child node we just created, match on the parent, and create the relationship.

LOAD CSV WITH HEADERS FROM 'file:///FTTH_sample.csv' AS line 
MERGE(child:ftthsample {child_id:line.child})
ON CREATE SET
child.desc = line.desc,
child.type = line.type

WITH child, line.parent as parentId
MATCH (parent:ftthsample)
WHERE parent.child_id = parentId
MERGE (child)-[:test]->(parent)

Note that we haven't added line.parent as a property. It's not needed, since we only use that to create relationships, and after the relationships are there, we won't need those again.