cancel
Showing results for 
Search instead for 
Did you mean: 

Head's Up! These forums are read-only. All users and content have migrated. Please join us at community.neo4j.com.

Match Variable length paths with different relationship types

Raj725
Node Clone

Hi,
I am trying to write a Cypher query to return the start and end nodes of all the patterns(subgraphs/paths) that are in the following regular expression:

A.(B|C.D)*.E

Start node: A
End node: E
Relationship Types: B,C,D
The path will have different combinations of B or C and D (C followed by D).

I know we can use "*" for variable length paths (like B*1..5) and "|" for OR between the relationships (like, :C|:D for either C or D).

How can I match C and D both sequentially?
(I tried :C|:D and :C|D both gives a result for C OR D)

Is it possible to write Cypher query for patterns (Regular Expression) as I mentioned above?

Thanks

1 ACCEPTED SOLUTION

I'm not aware of any way to do this at this time.

APOC path expander procedures have support for sequences of relationships, so if you just needed alternating :C and 😄 relationships it could support that, but we can't mix in :B as well.

One thing you could do is MATCH to the :C followed by 😄 pattern and create a new relationship for this:

MATCH (start)-[:C]-()-[:D]-(end)
CREATE (start)-[:CD]->(end)

That would allow you to use a path expander procedure from APOC and supply both the undirected :B relationship as well as the directed :CD relationship in the relationship filter:

MATCH (start:Node {name:'A'}), (end:Node {name:'E'})
CALL apoc.path.expandConfig(start, {terminatorNodes:[end], relationshipFilter:'B | CD>'}) YIELD path
RETURN path

View solution in original post

2 REPLIES 2

I'm not aware of any way to do this at this time.

APOC path expander procedures have support for sequences of relationships, so if you just needed alternating :C and 😄 relationships it could support that, but we can't mix in :B as well.

One thing you could do is MATCH to the :C followed by 😄 pattern and create a new relationship for this:

MATCH (start)-[:C]-()-[:D]-(end)
CREATE (start)-[:CD]->(end)

That would allow you to use a path expander procedure from APOC and supply both the undirected :B relationship as well as the directed :CD relationship in the relationship filter:

MATCH (start:Node {name:'A'}), (end:Node {name:'E'})
CALL apoc.path.expandConfig(start, {terminatorNodes:[end], relationshipFilter:'B | CD>'}) YIELD path
RETURN path

Thank you, Andrew.
It helped me!