使用Neo4j load csv命令的困难

Question

我有一个csv文件，有3800000,0000行，每一行都有两个名字，意思是两个人互相认识：

name,friend
a,b
a,c
a,d
f,a
e,a
d,g

我需要计算所有的人之间的关系。

首先，我使用密码：

CREATE (n:Patent {name:'a'})
CREATE (n:Patent {name:'b'})
CREATE (n:Patent {name:'c'})
CREATE (n:Patent {name:'d'})
CREATE (n:Patent {name:'e'})
CREATE (n:Patent {name:'f'})
CREATE (n:Patent {name:'g'})



MATCH (a:Patent {name:'a'}), 
      (b:Patent {name:'b'}) 
MERGE (a)-[:CITED]->(b)


MATCH (a:Patent {name:'a'}), 
      (b:Patent {name:'c'}) 
MERGE (a)-[:CITED]->(b)


MATCH (a:Patent {name:'a'}), 
      (b:Patent {name:'d'}) 
MERGE (a)-[:CITED]->(b)


MATCH (a:Patent {name:'f'}), 
      (b:Patent {name:'a'}) 
MERGE (a)-[:CITED]->(b)


MATCH (a:Patent {name:'e'}), 
      (b:Patent {name:'a'}) 
MERGE (a)-[:CITED]->(b)



MATCH (a:Patent {name:'d'}), 
      (b:Patent {name:'g'}) 
MERGE (a)-[:CITED]->(b)

然后，我得到了结果：

​

​

但是行数太大，我尝试了load csv批处理导入命令：

:auto USING PERIODIC COMMIT 1000
LOAD CSV WITH HEADERS FROM "file:///friend_relation.csv" AS line
MATCH (a:Friend {name: line.name}), 
      (b:Friend {name: line.friend}) MERGE (a)-[:KNOWS]->(b)

它失败了：

​

​

我把大文件分成多个样本文件，并使用csv load来实现我的目标，我的问题是什么？

Answer 1

如果不返回任何内容，这是因为MATCH子句没有找到任何内容：

试着使用

:auto USING PERIODIC COMMIT 1000
LOAD CSV WITH HEADERS FROM "file:///friend_relation.csv" AS line
MERGE (a:Friend {name: line.name})
MERGE (b:Friend {name: line.friend}) 
MERGE (a)-[:KNOWS]->(b)