为单个Neo4j节点使用多变量的Cypher查询问题

问题原因 #

你尝试的查询无结果，核心是混淆了Neo4j内部节点ID和自定义id属性：

• 原查询直接复用变量（如nodeTo1）是直接匹配同一节点，而你改用A{id: nodeTo0.id}时，若节点没有自定义id属性，或者nodeTo0.id并非指向该属性，就会匹配不到节点。
• 即使存在自定义id属性，这种写法也不如直接变量关联高效，且容易出错。

正确实现方式 #

方式1：通过WHERE子句绑定同一节点 #

在每个后续MATCH中，先用新变量匹配节点，再通过WHERE指定它与前一个节点是同一个：

MATCH (nodeFrom0:Z{}) -[nodeRelation0:FLOWS_TO]-> (nodeTo0:A{})
MATCH (nodeFrom1:A{}) -[nodeRelation1:FLOWS_TO]-> (nodeTo1:A{})
WHERE nodeFrom1 = nodeTo0
MATCH (nodeFrom2:A{}) -[nodeRelation2:FLOWS_TO]-> (nodeTo2:A{})
WHERE nodeFrom2 = nodeTo1
MATCH (nodeFrom3:A{}) -[nodeRelation3:FLOWS_TO]-> (nodeTo3:B{})
WHERE nodeFrom3 = nodeTo2
MATCH (nodeFrom4:B{}) -[nodeRelation4:FLOWS_TO]-> (nodeTo4:C{})
WHERE nodeFrom4 = nodeTo3
MATCH (nodeFrom5:C{}) -[nodeRelation5:FLOWS_TO]-> (nodeTo5:B{})
WHERE nodeFrom5 = nodeTo4
MATCH (nodeFrom6:B{}) -[nodeRelation6:FLOWS_TO]-> (nodeTo6:D{})
WHERE nodeFrom6 = nodeTo5
MATCH (nodeFrom7:D{}) -[nodeRelation7:FLOWS_TO]-> (nodeTo7:B{})
WHERE nodeFrom7 = nodeTo6
MATCH (nodeFrom8:B{}) -[nodeRelation8:FLOWS_TO]-> (nodeTo8:E{})
WHERE nodeFrom8 = nodeTo7

RETURN nodeFrom0,nodeRelation0,nodeTo0,nodeFrom1,nodeRelation1,nodeTo1,nodeFrom2,nodeRelation2,nodeTo2,nodeFrom3,nodeRelation3,nodeTo3,nodeFrom4,nodeRelation4,nodeTo4,nodeFrom5,nodeRelation5,nodeTo5,nodeFrom6,nodeRelation6,nodeTo6,nodeFrom7,nodeRelation7,nodeTo7,nodeFrom8,nodeRelation8,nodeTo8

方式2：直接在MATCH中链式关联（更简洁） #

如果只是需要用新变量指代同一节点，也可以在MATCH语句中直接将前一个节点的变量作为当前MATCH的起点，同时给它起新别名：

MATCH (nodeFrom0:Z{}) -[nodeRelation0:FLOWS_TO]-> (nodeTo0:A{})
MATCH (nodeFrom1:A{}) = nodeTo0 -[nodeRelation1:FLOWS_TO]-> (nodeTo1:A{})
MATCH (nodeFrom2:A{}) = nodeTo1 -[nodeRelation2:FLOWS_TO]-> (nodeTo2:A{})
MATCH (nodeFrom3:A{}) = nodeTo2 -[nodeRelation3:FLOWS_TO]-> (nodeTo3:B{})
MATCH (nodeFrom4:B{}) = nodeTo3 -[nodeRelation4:FLOWS_TO]-> (nodeTo4:C{})
MATCH (nodeFrom5:C{}) = nodeTo4 -[nodeRelation5:FLOWS_TO]-> (nodeTo5:B{})
MATCH (nodeFrom6:B{}) = nodeTo5 -[nodeRelation6:FLOWS_TO]-> (nodeTo6:D{})
MATCH (nodeFrom7:D{}) = nodeTo6 -[nodeRelation7:FLOWS_TO]-> (nodeTo7:B{})
MATCH (nodeFrom8:B{}) = nodeTo7 -[nodeRelation8:FLOWS_TO]-> (nodeTo8:E{})

RETURN nodeFrom0,nodeRelation0,nodeTo0,nodeFrom1,nodeRelation1,nodeTo1,nodeFrom2,nodeRelation2,nodeTo2,nodeFrom3,nodeRelation3,nodeTo3,nodeFrom4,nodeRelation4,nodeTo4,nodeFrom5,nodeRelation5,nodeTo5,nodeFrom6,nodeRelation6,nodeTo6,nodeFrom7,nodeRelation7,nodeTo7,nodeFrom8,nodeRelation8,nodeTo8

适配JSON数组生成场景的模板 #

针对你从[{node1:xxxx, relation:xxxx, node2:xxxx},...]生成查询的需求，可遵循以下规则：

1. 第一条JSON直接生成基础MATCH语句：MATCH (nodeFrom0:<node1标签>{id: '<node1 ID>'}) -[nodeRelation0:<relation类型>]-> (nodeTo0:<node2标签>{id: '<node2 ID>'})
2. 后续每条JSON，若node1 ID等于前一条的node2 ID，则生成：

   MATCH (nodeFromN:<node1标签>{id: '<node1 ID>'}) -[nodeRelationN:<relation类型>]-> (nodeToN:<node2标签>{id: '<node2 ID>'})
   WHERE nodeFromN = nodeTo{N-1}
   

   或者更简洁的：

   MATCH (nodeFromN:<node1标签>{id: '<node1 ID>'}) = nodeTo{N-1} -[nodeRelationN:<relation类型>]-> (nodeToN:<node2标签>{id: '<node2 ID>'})
   

这样既能保证用新变量指代同一节点，又能正确匹配到数据。

内容的提问来源于stack exchange，提问作者spd