Neo4j创建自动索引

一、创建Neo4j的Legacy indexing

　　1.为节点创建索引

　　官方API的创建示例为：

将一节点添加至索引：

public static void AddNodeIndex(String nid)
    {
        String txUri=SERVER_ROOT_URI+"index/node/favorites";
        WebResource resource = Client.create().resource(txUri);
        String entity="{"value" : "n204","uri" : "http://192.168.209.128:7474/db/data/node/"+nid+"","key" : "n201"}";
        ClientResponse response = resource.accept(MediaType.APPLICATION_JSON)
                .type(MediaType.APPLICATION_JSON).entity(entity)
                .post(ClientResponse.class);

        System.out.println(response.getStatus());
        System.out.println(response.getEntity(String.class));
        response.close();
    }

ps:nid是要添加索引的节点ID  value是索引值  key是索引名称

2.通过属性查找节点

public static void GetNodeByIndex()
    {
        String txUri=SERVER_ROOT_URI+"index/node/favorites/n201/n201";
        WebResource resource = Client.create().resource(txUri);
        
        ClientResponse response = resource.accept(MediaType.APPLICATION_JSON)
                .type(MediaType.APPLICATION_JSON)
                .get(ClientResponse.class);

        System.out.println(response.getStatus());
        System.out.println(response.getEntity(String.class));
        response.close();
    }

txUri路径中：favorites为刚创建的索引名称，第一个n201是节点索引key，第二个n201是节点索引值

二、自动创建索引（Legacy Automatic Indexes）

What default configuration means depends on how you have configured your database. If you haven’t
changed any indexing configuration, it means the indexes will be using a Lucene-based backend.

数据库配置之后，就可以自动创建索引

1.配置文件配置

Auto-indexing must be enabled through configuration before we can create or configure them. Firstly
ensure that you’ve added some config like this into your server’s conf/neo4j.properties file:

打开conf/neo4j.properties文件如图

配置下面的节点

# Enable auto-indexing for nodes, default is false.
node_auto_indexing=true

# The node property keys to be auto-indexed, if enabled.
node_keys_indexable=name,ki

# Enable auto-indexing for relationships, default is false.
relationship_auto_indexing=true

# The relationship property keys to be auto-indexed, if enabled.
relationship_keys_indexable=name,ki

Node_keys_indexable、relationship_keys_indexable对应节点、关系的属性

配置完成之后重启服务

重启三个节点的集群

2。测试索引

插入一个节点和关系

    // 创建节点
    @Test
    public void test2() {
        URI uri = CreateSimpleGraph.createNode();
        CreateSimpleGraph.addProperty(uri, "name", "张三");
        URI uri1 = CreateSimpleGraph.createNode();
        CreateSimpleGraph.addProperty(uri1, "name", "李四");
    }
// 为节点设置关系
    @Test
    public void test6() {
        for (int i = 1; i < 2; i++) {
            try {
                URI suri = new URI("http://192.168.209.128:7474/db/data/node/171391");
                String uri1="http://192.168.209.128:7474/db/data/node/";
                URI euri = new URI("http://192.168.209.128:7474/db/data/node/171392");
                URI reluri= CreateSimpleGraph.addRelationship(suri, euri, "家人","{"ki" : "1234567890", "name" : "无" }");
                System.out.println(reluri);
            } catch (URISyntaxException e) {
                // 异常信息输出该内容
                e.printStackTrace();
            }
        }
    }

3.通过属性查找节点

public static void GetNodeByAutoIndex(String ki)
    {
//        String txUri=SERVER_ROOT_URI+"index/node/node_auto_index/name/"+ki;
        String txUri=SERVER_ROOT_URI+"index/auto/node/ki/"+ki;
        WebResource resource = Client.create().resource(txUri);
        
        ClientResponse response = resource.accept(MediaType.APPLICATION_JSON)
                .type(MediaType.APPLICATION_JSON)
                .get(ClientResponse.class);

        System.out.println(response.getStatus());
        System.out.println(response.getEntity(String.class));
        response.close();
    }
    
    public static void GetRelationshipByAutoIndex(String ki)
    {
//        String txUri=SERVER_ROOT_URI+"index/node/node_auto_index/name/"+ki;
        String txUri=SERVER_ROOT_URI+"index/auto/relationship/ki/"+ki;
        WebResource resource = Client.create().resource(txUri);
        
        ClientResponse response = resource.accept(MediaType.APPLICATION_JSON)
                .type(MediaType.APPLICATION_JSON)
                .get(ClientResponse.class);

        System.out.println(response.getStatus());
        System.out.println(response.getEntity(String.class));
        response.close();
    }

关系的输出结果为：

[ {
  "extensions" : { },
  "metadata" : {
    "id" : 337,
    "type" : "家人"
  },
  "data" : {
    "name" : "无",
    "ki" : "1234567890"
  },
  "property" : "http://192.168.209.128:7474/db/data/relationship/337/properties/{key}",
  "start" : "http://192.168.209.128:7474/db/data/node/171391",
  "self" : "http://192.168.209.128:7474/db/data/relationship/337",
  "end" : "http://192.168.209.128:7474/db/data/node/171392",
  "type" : "家人",
  "properties" : "http://192.168.209.128:7474/db/data/relationship/337/properties"
} ]

这里说明一下，传值为中文的时候，查询不出来，可能需要编码，因为工作暂时没有用到，就没有再研究了

三、NEO4J批量插入和用户密码问题

1.批量操作

批量操作的官方文档地址：http://neo4j.com/docs/milestone/rest-api-batch-ops.html

public static int BatchInserterNode(String body) {
        // String
        // body="[{"method":"POST","to":"/node","body":{"name":"aaa","lead":"aaa1"},"id":0},{"method":"POST","to":"/node","body":{"name":"bbb","lead":"bbb1"},"id":1}]";
        // POST {} to the node entry point URI
        ClientResponse response = GetResourceInstance("batch")
                .accept(MediaType.APPLICATION_JSON)
                .type(MediaType.APPLICATION_JSON).entity(body)
                .post(ClientResponse.class);

        // System.out.println(String.format(
        // "POST to [%s], status code [%d], location header [%s]",
        // nodeEntryPointUri, response.getStatus(), location.toString()));
        String entity = response.getEntity(String.class);
        response.close();

        int status = response.getStatus();

        return status;
    }

//调用 
//批量插入速度测试
    @Test
    public void test10()
    {
    
        StringBuilder sbody=new StringBuilder();
        sbody.append("[");
        for (int i = 161337; i < 171337; i++) {
            if(i>161337)
            {
                sbody.append(",");
            }
            sbody.append("{");
            sbody.append(""method":"POST","to":"/node","body":{"name":"n"+i+"","lead":"a"+i+""},"id":"+i+"");
            sbody.append("}");
        }
        
        
        sbody.append("]");
        long st = System.currentTimeMillis();
        int status= CreateSimpleGraph.BatchInserterNode(sbody.toString());
        System.out.println("插入状态:"+status);
        long et = System.currentTimeMillis();
        System.out.println("插入10000条数据总共耗时：" + (et - st) + "毫秒");        
    }

批量插入的速度10000条大概在6秒左右吧

2.NEO4J密码操作

本次用的版本是：neo4j-enterprise-2.2.0-unix，连接数据库的时候需要输入用户名称密码。上个版本好像是不用的。

使用密码时：

public void checkDatabaseIsRunning() {
        // START SNIPPET: checkServer
        WebResource resource = Client.create().resource(SERVER_ROOT_URI);
        ClientResponse response = resource.get(ClientResponse.class);
        resource.addFilter(new HTTPBasicAuthFilter("neo4j", "123456"));
        if (response.getStatus() == 200) {
            System.out.println("连接成功!");
        } else {
            System.out.println("连接失败!");
        }
        // System.out.println(String.format("GET on [%s], status code [%d]",
        // SERVER_ROOT_URI, response.getStatus()));
        response.close();
        // END SNIPPET: checkServer
    }

如不需要使用密码，可以在配置文件进行配置

vi /conf/neo4j-server.properties这个文件

这里改成false即可，默认是True

参考文档

http://neo4j.com/docs/milestone/rest-api-batch-ops.html

http://neo4j.com/docs/milestone/rest-api-auto-indexes.html

http://neo4j.com/docs/milestone/rest-api-security.html