How can I get all the hops in a path of unknown length with neo4j-python?

Question

MATCH (u:User {name: $user}), (target:Group {name: $group}), p=shortestPath((u)-[*]->(target)) RETURN p

When I run the above query in the Neo4j web UI, a fell graph of the resulting paths is displayed.

However, when I run the same query with the neo4j-python driver, only a Path objects with limited information are returned

<Path start=479557 end=404582 size=1>

How can I use Cypher and python to get complete path details, including all nodes and the relationships that connect them?

Answer 1

Depends on how you want to return data, but you can try something like this

MATCH (u:User {name: $user}), (target:Group {name: $group}),
p=shortestPath((u)-[*]->(target)) RETURN nodes(p),relationships(p)

Answer 2

Thanks for the help everyone! For reference, here's my complete example that converts paths to human-readable strings for console or email output.

def find_paths_from_standard_user_to_domain_admins(standard_user, domain_admins_group):
    """Returns a list of paths that a standard user could take to get domain admin credentials"""
    results = []
    query = "MATCH (u:User {name:{user}}), (target:Group {name: {group}})," \
            "p=allShortestPaths((u)-[*]->(target)) RETURN p"
    with driver.session() as session:
        with session.begin_transaction() as tx:
            for record in tx.run(query, user=standard_user, group=domain_admins_group):
                relationships = record["p"].relationships
                nodes = record["p"].nodes
                path = ""
                for i in (range(len(relationships))):
                    path += "{0}-[{1}]->".format(nodes[i]["name"], relationships[i].type)
                path += nodes[-1]["name"]
                results.append(path)
    return results

This is a query for a graph generated by the Bloodhound project, which builds graphs of Active Directory structures. It's extremely useful for domain admins, system architects, network defenders, and pentesters.