Trying to understand recursive relationship patterns

[rmjantz]

Hi,I am trying to use recursive relationship patterns and feel I am either doing something wrong or else I don't understand them. I have a graph that has 4 nodes conncted by three relationships, as in (n1)-->(n2)-->(n3)-->(n4). I am trying a match like 'match( n:Node)-[r*1..3]->(m:Node)' - the match should find three hops, but never returns more than two. What am I missing? Thanks.

[andyfengHKU]

Interesting, it should return all path up to 3 hop. Do u mind share a reproducible example with us?

[rmjantz]

I think I figured out what is going on.
If I use this query with the above graph 'match( n:Node)-[r1..3]->(m:Node) return properties(nodes(r), 'name');' (i.e., passing the relationship variable to properties) it tells me that it traversed 3 hops (correct) but only shows me two nodes (n1 and n2)
If I change the query to 'match p = ( n:Node)-[r1..3]->(m:Node) return properties(nodes(p), 'name');' (i.e., assigning the que ry result to p and passing that to properties) it gives me all four nodes, which is what i expect.
Q: is the behavior of the first query correct? It seems like it should also show me four nodes.

[andyfengHKU]

It is expected behavior, in the case of

p = ( n:Node)-[r1..3]->(m:Node) 

you can think of p = n + r + m. If r contains the same number of nodes as p, then these are semantically equivalent expression and it will be redundant to semantically equivalent expressions in different ways.

[rmjantz]

I will work one up.

…

On Tue, Jun 10, 2025, 9:34 PM andyfeng ***@***.***> wrote: Interesting, it should return all path up to 3 hop. Do u mind share a reproducible example with us? — Reply to this email directly, view it on GitHub <#5520 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A3MD2POHPMDUI4E2OABH2SL3C6IUPAVCNFSM6AAAAAB7BDDAFKVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTGNBSG43DGNY> . You are receiving this because you authored the thread.Message ID: ***@***.***>