玩 GraphX 的时候知道了 GraphFrames ，玩 GraphFrames 的时候知道了 Cypher ，于是入坑 Neo4j。
Neo4j的安装和开发环境的搭建
https://neo4j.com/download/other-releases/，从上述链接选择合适的版本下载即可。
上述链接仅供参考。由于不同版本之间存在差异，细节上会有些许不同。（the devil is in the details.）
笔者的安装环境：
Windows7 64
jdk-8u144-windows-x64.exe
neo4j-community-3.2.6-windows.zip
可以参考 这里 搭建开发环境。当然，也可以在 IDEA 的 maven 工程中添加如下依赖，就这么简单。
<!-- Neo4j dependency -->
<dependency>
<groupId>org.neo4j</groupId>
<artifactId>neo4j</artifactId>
<version>3.2.6</version>
</dependency>
Neo4j 提供 JAVA API 以编程方式执行所有数据库操作。它支持两种类型的API：
1、Neo4j 原生的 Java API
2、Neo4j Cypher Java API
Neo4j 原生 Java API 是一种低级别的纯 JAVA API，用于执行数据库操作。Neo4j Cypher Java API 是简单而强大的 JAVA API，用于执行所有CQL命令以执行数据库操作。
Neo4j 原生的 Java API
import org.neo4j.graphdb.*;
import org.neo4j.graphdb.factory.GraphDatabaseFactory;
import java.io.File;
public class Neo4jNativeJavaAPI {
private static void registerShutdownHook(final GraphDatabaseService graphDB) {
// Registers a shutdown hook for the Neo4j instance so that it shuts down nicely
// when the VM exits (even if you "Ctrl-C" the running example before it's completed)
/*为了确保neo4j数据库的正确关闭，我们可以添加一个关闭钩子方法 registerShutdownHook。 *这个方法的意思就是在jvm中增加一个关闭的钩子， *当jvm关闭的时候，会执行系统中已经设置的所有通过方法addShutdownHook添加的钩子， *当系统执行完这些钩子后，jvm才会关闭。 *所以这些钩子可以在jvm关闭的时候进行内存清理、对象销毁等操作。*/
Runtime.getRuntime().addShutdownHook(
new Thread() {
public void run() {
//Shutdown the Database
System.out.println("Server is shutting down");
graphDB.shutdown();
}
}
);
}
public static void main(String[] args) {
//指定 Neo4j 存储路径
File file = new File("D:\\neo4j-community-3.2.6\\data\\databases\\graph.db");
//Create a new Object of Graph Database
GraphDatabaseService graphDB = new GraphDatabaseFactory().newEmbeddedDatabase(file);
System.out.println("Server is up and Running");
try(Transaction tx = graphDB.beginTx()){
/** * 新增User节点 * 添加Lable以区分节点类型 * 每个节点设置name属性 */
Node user1 = graphDB.createNode(MyLabels.USERS);
user1.setProperty("name", "John Johnson");
Node user2 = graphDB.createNode(MyLabels.USERS);
user2.setProperty("name", "Kate Smith");
Node user3 = graphDB.createNode(MyLabels.USERS);
user3.setProperty("name", "Jack Jeffries");
/** * 为user1添加Friend关系 * 注：Neo4j的关系是有向的箭头，正常来讲Friend关系应该是双向的， * 此处为了简单起见，将关系定义成单向的，不会影响后期的查询 */
user1.createRelationshipTo(user2,MyRelationshipTypes.IS_FRIEND_OF);
user1.createRelationshipTo(user3,MyRelationshipTypes.IS_FRIEND_OF);
/** * 新增Movie节点 * 添加Lable以区分节点类型 * 每个节点设置name属性 */
Node movie1 = graphDB.createNode(MyLabels.MOVIES);
movie1.setProperty("name", "Fargo");
Node movie2 = graphDB.createNode(MyLabels.MOVIES);
movie2.setProperty("name", "Alien");
Node movie3 = graphDB.createNode(MyLabels.MOVIES);
movie3.setProperty("name", "Heat");
/** * 为User节点和Movie节点之间添加HAS_SEEN关系, HAS_SEEN关系设置stars属性 */
Relationship relationship1 = user1.createRelationshipTo(movie1, MyRelationshipTypes.HAS_SEEN);
relationship1.setProperty("stars", 5);
Relationship relationship2 = user2.createRelationshipTo(movie3, MyRelationshipTypes.HAS_SEEN);
relationship2.setProperty("stars", 3);
Relationship relationship6 = user2.createRelationshipTo(movie2, MyRelationshipTypes.HAS_SEEN);
relationship6.setProperty("stars", 6);
Relationship relationship3 = user3.createRelationshipTo(movie1, MyRelationshipTypes.HAS_SEEN);
relationship3.setProperty("stars", 4);
Relationship relationship4 = user3.createRelationshipTo(movie2, MyRelationshipTypes.HAS_SEEN);
relationship4.setProperty("stars", 5);
tx.success();
System.out.println("Done successfully");
} catch (Exception e) {
e.printStackTrace();
} finally {
graphDB.shutdown(); //关闭数据库
}
//Register a Shutdown Hook
registerShutdownHook(graphDB);
}
}
/** * Label类型枚举类 */
enum MyLabels implements Label {
MOVIES, USERS
}
/** * 关系类型枚举类 */
enum MyRelationshipTypes implements RelationshipType {
IS_FRIEND_OF, HAS_SEEN
}运行完毕后，在 neo4j-community-3.2.6 目录下执行
./bin/neo4j.bat console输出如下
2017-11-01 03:24:32.585+0000 INFO ======== Neo4j 3.2.6 ========
2017-11-01 03:24:32.601+0000 INFO Starting...
2017-11-01 03:24:33.552+0000 INFO Bolt enabled on 127.0.0.1:7687.
2017-11-01 03:24:36.714+0000 INFO Started.
2017-11-01 03:24:37.571+0000 INFO Remote interface available at http://localhost:7474/此时，在 http://localhost:7474/ 能够看到如下图
Neo4j Cypher Java API
在 Java 程序中用 Cypher 对上图进行查询
import org.neo4j.graphdb.*;
import org.neo4j.graphdb.factory.GraphDatabaseFactory;
import java.io.File;
public class Neo4jCypherJavaAPI {
public static void main(String[] args) {
//指定 Neo4j 存储路径
File file = new File("D:\\neo4j-community-3.2.6\\data\\databases\\graph.db");
//Create a new Object of Graph Database
GraphDatabaseService graphDB = new GraphDatabaseFactory().newEmbeddedDatabase(file);
System.out.println("Server is up and Running");
try(Transaction tx = graphDB.beginTx()){
//通过Cypher查询获得结果
StringBuilder sb = new StringBuilder();
sb.append("MATCH (john)-[:IS_FRIEND_OF]->(USER)-[:HAS_SEEN]->(movie) ");
sb.append("RETURN movie");
Result result = graphDB.execute(sb.toString());
//遍历结果
while(result.hasNext()){
//get("movie")和查询语句的return movie相匹配
Node movie = (Node) result.next().get("movie");
System.out.println(movie.getId() + " : " + movie.getProperty("name"));
}
tx.success();
System.out.println("Done successfully");
} catch (Exception e) {
e.printStackTrace();
} finally {
graphDB.shutdown(); //关闭数据库
}
}
}输出如下：
Server is up and Running
8 : Alien
9 : Heat
8 : Alien
7 : Fargo
Done successfully