Neo4j Desktop导入csv文件

我的Neo4j时最新的V4.1.3，界面跟官网的教程可能不太一样。

本文参考Neo4j英文开发者文档链接如下：
https://neo4j.com/developer/desktop-csv-import/

数据来源：Youtube视频-Neo4j Batch Insertion from CSV

// neo4j-flight-csv-raw.csv
flight,airline,depart,arrive,capacity
23,Delta,ATL,DTW,160
24,Delta,DTW,ATL,160
27,Delta,DTW,BOS,160
28,Delta,BOS,DTW,160
35,Delta,DTW,PIT,128
36,Delta,PIT,DTW,128
37,Delta,ATL,BOS,160
38,Delta,BOS,ATL,160
43,Delta,ATL,PIT,160
44,Delta,PIT,ATL,160
45,Delta,BOS,PIT,160
46,Delta,PIT,BOS,160
101,Southwest,DTW,BOS,136
102,Southwest,BOS,DTW,136
103,Southwest,DTW,PIT,136
104,Southwest,PIT,DTW,136
1231,American,DTW,BOS,160
1232,American,DTW,PIT,160
1257,American,DTW,ATL,128
1258,American,BOS,ATL,160
1277,American,BOS,DTW,160
1278,American,PIT,DTW,160
1292,American,ATL,DTW,128
1292,American,ATL,BOS,160

// neo4j-airport-csv-raw.csv
label,city,state
DTW,Detroit,Michigan
ATL,Altanta,Georgia
PIT,Pittsburgh,Pennsylvania
BOS,Bostan,Massachussetts

声明：以下代码，即使你没装java的jdk也能跑起来，新手不要在环境上浪费过多时间。neo4j只认识utf-8编码的csv，不要直接把excel改成csv来用！

step1：
打开你的Neo4j桌面端，新建一个项目，并为此项目添加一个数据库实例（记住自己设置的用户名、密码），然后开启（start）本项目，之后会有登录操作，用户名默认是neo4j。

step2：
这时点数据库-设置，会出现本地访问的端口号界面，点击open folder-import打开该项目对应的本地import文件夹，把上述两个csv文件拖至此处。（只有执行了此操作，引入文件时才可以用file:///字样）

step3：
打开localhost:7474，会自动跳转页面至Neo4j网页版，然后就可以开始敲下面的代码了，依次敲完就好。

load csv with headers from "file:///neo4j-airport-csv-raw.csv" as airports return airports

load csv with headers from "file:///neo4j-flight-csv-raw.csv" as flights return flights

这两行代码，会读取文件得到对应的结构化信息。

---

create constraint on (f:Flight) assert f.flight is unique

create constraint on (a:Airport) assert a.label is unique

这两步是为Flight对象和Airport对象的label加上约束——唯一性。这样使得访问变得更快。

---

load csv with headers from "file:///neo4j-airport-csv-raw.csv" as airports
merge (a:Airport{label:airports.label})
	set a.city = airports.city, a.state = airports.state
return count(a)

生成各个机场节点，返回节点数目

match (a:Airport) return (a)

显示机场节点

---

LOAD CSV WITH HEADERS FROM 'file:///neo4j-flight-csv-raw.csv' AS flights
MERGE (f:Flight {number: flights.flight})
  SET f.airline = flights.airline, f.capacity = flights.capacity
RETURN count(f)

MATCH (f:Flight) RETURN f

同上

---

LOAD CSV WITH HEADERS FROM 'file:///neo4j-flight-csv-raw.csv' AS flights
MATCH (f:Flight {number:flights.flight})
MATCH (a:Airport {label:flights.arrive})
MERGE (f)-[r:arrives]->(a)

LOAD CSV WITH HEADERS FROM 'file:///neo4j-flight-csv-raw.csv' AS flights
MATCH (f:Flight {number:flights.flight})
MATCH (a:Airport {label:flights.depart})
MERGE (f)-[r:Depart]->(a)

以上两步是连接节点，形成关系

最后，敲入match (n) return (n)可以显示最终形成的图谱，如下图，航线、航班的对应关系一目了然。