Neo4j 路径查询/排序路径中的属性

Question

电影数据集上的非常简单的最短路径查询：

MATCH path=shortestPath((p1:Person)-[*0..15]-(p2:Person)) WHERE toLower(p1.name) = toLower('Halle Berry') AND toLower(p2.name) = toLower('Tom Hanks') 
RETURN relationships(path) AS relationships, nodes(path) AS nodes

如何对角色数组进行排序？ IE。在这部特别的电影中，他们扮演了很多角色。我可以在应用程序端对其进行排序，但最好在 Neo 端进行排序。说到节点，我一直在用 UNWIND 做这件事，但这似乎对路径属性不起作用。

Answer 1

您可以访问路径的元素并一次获取一个属性。这很乏味，但它完成了工作。

elements[index] 为起始节点； elements[index+1] 是关系，elements[index+2] 是结束节点。 ID(node)是节点id，type(node)是关系类型

MATCH path=shortestPath((p1:Person)-[*0..15]-(p2:Person)) 
WHERE toLower(p1.name) = toLower('Halle Berry') AND toLower(p2.name) = toLower('Tom Hanks') 
WITH apoc.path.elements(path) AS elements, nodes(path) AS nodes
UNWIND range(0, size(elements)-2) AS index
WITH nodes, elements, index WHERE index %2 = 0
WITH nodes, ID(elements[index]) AS start, ID(elements[index+1]) AS identity, ID(elements[index+2]) AS end, type(elements[index+1]) as type, elements[index+1] as roles
UNWIND roles as role
WITH nodes, start, end, identity, type, apoc.coll.sort(role.roles) as roles
RETURN collect({start: start, end: end, identity: identity, type: type, properties: {roles: roles}}) as relationships, nodes

RESULT:
╒══════════════════════════════════════════════════════════════════════╤══════════════════════════════════════════════════════════════════════╕
│"relationships"                                                       │"nodes"                                                               │
╞══════════════════════════════════════════════════════════════════════╪══════════════════════════════════════════════════════════════════════╡
│[{"identity":139,"start":106,"end":105,"type":"ACTED_IN","properties":│[{"name":"Halle Berry","born":1966},{"title":"Cloud Atlas","tagline":"│
│{"roles":["Jocasta Ayrs","Luisa Rey","Meronym","Ovid"]}},{"identity":1│Everything is connected","released":2012},{"name":"Tom Hanks","born":1│
│37,"start":105,"end":71,"type":"ACTED_IN","properties":{"roles":["Derm│956}]                                                                 │
│ot Hoggins","Dr. Henry Goose","Isaac Sachs","Zachry"]}}]              │                                                                      │
└──────────────────────────────────────────────────────────────────────┴──────────────────────────────────────────────────────────────────────┘

====已编辑=== 您可以 UNWIND 角色然后对其进行排序。最后，再次收藏为字典列表。

MATCH path=shortestPath((p1:Person)-[*0..15]-(p2:Person)) 
WHERE toLower(p1.name) = toLower('Halle Berry') AND toLower(p2.name) = toLower('Tom Hanks') 
WITH relationships(path) AS relationships, nodes(path) AS nodes
UNWIND relationships as role
WITH apoc.coll.sort(role.roles) as roles, nodes 
RETURN collect({roles: roles}) as relationships, nodes

RESULT:
╒══════════════════════════════════════════════════════════════════════╤══════════════════════════════════════════════════════════════════════╕
│"relationships"                                                       │"nodes"                                                               │
╞══════════════════════════════════════════════════════════════════════╪══════════════════════════════════════════════════════════════════════╡
│[{"roles":["Jocasta Ayrs","Luisa Rey","Meronym","Ovid"]},{"roles":["De│[{"name":"Halle Berry","born":1966},{"title":"Cloud Atlas","tagline":"│
│rmot Hoggins","Dr. Henry Goose","Isaac Sachs","Zachry"]}]             │Everything is connected","released":2012},{"name":"Tom Hanks","born":1│
│                                                                      │956}]                                                                 │
└──────────────────────────────────────────────────────────────────────┴──────────────────────────────────────────────────────────────────────┘

Neo4j 路径查询/排序路径中的属性

Neo4j path query / sorting properties in path

neo4j

cypher