Elasticsearch Java API简介

加入依赖

我本地的Elasticsearch的版本是2.1.0,因此加入相应的maven依赖

<dependency>
<groupId>org.elasticsearch</groupId>
<artifactId>elasticsearch</artifactId>
<version>2.1.0</version>
</dependency>

创建Client

Elasticsearch Client分为Node Client和TransportClient。

  • Node Client:节点本身也是Elasticsearch集群的节点,也进入Elasticsearch集群和别的Elasticsearch集群中的节点一样
  • TransportClient:轻量级的Client,使用Netty线程池,Socket连接到ES集群。本身不加入到集群,只作为请求的处理

一般我们使用TransportClient。创建Client的实例如下:

	private TransportClient client = null;

    @Before
public void createElaCLient() throws UnknownHostException {
//如果集群是默认名称的话可以不设置集群名称
Settings settings = Settings.settingsBuilder().put("cluster.name","elasticsearch").build();
client = TransportClient.builder().settings(settings).build().addTransportAddress(new InetSocketTransportAddress(InetAddress.getByName("master"),9300));
} /**
* 关闭ela客户端
*/
@After
public void closeElaClient(){
if(client != null){
client.close();
}
}

client.transport.sniff嗅探功能

你可以设置client.transport.sniff为true来使客户端去嗅探整个集群的状态,把集群中其它机器的ip地址加到客户端中,这样做的好处是一般你不用手动设置集群里所有集群的ip到连接客户端,它会自动帮你添加,并且自动发现新加入集群的机器。代码实例如下:

	private TransportClient client = null;

    @Before
public void createElaCLient() throws UnknownHostException {
//如果集群是默认名称的话可以不设置集群名称
Settings settings = Settings.settingsBuilder().put("cluster.name","elasticsearch").put("client.transport.sniff",true).build();
client = TransportClient.builder().settings(settings).build().addTransportAddress(new InetSocketTransportAddress(InetAddress.getByName("master"),9300));
}

注意:当ES服务器监听使用内网服务器IP而访问使用外网IP时,不要使用client.transport.sniff为true,在自动发现时会使用内网IP进行通信,导致无法连接到ES服务器,而直接使用addTransportAddress方法进行指定ES服务器

测试Client连接到Elasticsearch集群

代码如下:

@Test
public void testConnection(){
List<DiscoveryNode> discoveryList = client.connectedNodes();
for(DiscoveryNode node : discoveryList){
System.out.println(node.getName());
}
}

创建/删除Index和Type信息

	/**
* 创建索引
*/
@Test
public void createIndex(){
if(client != null){
client.admin().indices().create(new CreateIndexRequest("test_index")).actionGet();
}
} /**
* 清除索引
*/
@Test
public void clearIndex(){
IndicesExistsResponse indicesExistsResponse = client.admin().indices().exists(new IndicesExistsRequest("test_index")).actionGet();
if(indicesExistsResponse.isExists()){
client.admin().indices().delete(new DeleteIndexRequest("test_index")).actionGet();
}
} /**
* 定义索引的映射类型(mapping)
*/
@Test
public void defineIndexTypeMapping(){
try {
XContentBuilder builder = XContentFactory.jsonBuilder();
builder.startObject()
.startObject("test")
.startObject("properties")
.startObject("id").field("type","long").field("store","yes").endObject()
.startObject("name").field("type","string").field("store","yes").field("index","not_analyzed").endObject()
.endObject()
.endObject()
.endObject();
PutMappingRequest mappingRequest = Requests.putMappingRequest("test_index").type("test").source(builder);
client.admin().indices().putMapping(mappingRequest).actionGet();
} catch (IOException e) {
e.printStackTrace();
}
} /**
* 删除index下的某个type
*/
@Test
public void deleteType(){
if(client != null){
client.prepareDelete().setIndex("test_index").setType("test").execute().actionGet();
}
}

这里自定义了某个Type的索引映射(Mapping),默认ES会自动处理数据类型的映射:针对整型映射为long,浮点数为double,字符串映射为string,时间为date,true或false为boolean。

注意:针对字符串,ES默认会做“analyzed”处理,即先做分词、去掉stop words等处理再index。如果你需要把一个字符串做为整体被索引到,需要把这个字段这样设置:field(“index”, “not_analyzed”)。

索引数据

	/**
* 批量索引
*/
@Test
public void indexData(){
BulkRequestBuilder requestBuilder = client.prepareBulk();
for(Person person : personList){
String obj = getIndexDataFromHotspotData(person);
if(obj != null){
requestBuilder.add(client.prepareIndex("test_index","test",String.valueOf(person.getId())).setRefresh(true).setSource(obj));
}
}
BulkResponse bulkResponse = requestBuilder.execute().actionGet();
if(bulkResponse.hasFailures()){
Iterator<BulkItemResponse> it = bulkResponse.iterator();
while(it.hasNext()){
BulkItemResponse itemResponse = it.next();
if(itemResponse.isFailed()){
System.out.println(itemResponse.getFailureMessage());
}
}
}
} /**
* 单个索引数据
* @return
*/
@Test
public void indexHotspotData() {
String jsonSource = getIndexDataFromHotspotData(new Person(1004,"jim"));
if (jsonSource != null) {
IndexRequestBuilder requestBuilder = client.prepareIndex("test_index",
"test").setRefresh(true);
requestBuilder.setSource(jsonSource)
.execute().actionGet();
}
}
public String getIndexDataFromHotspotData(Person p){
String result = null;
if(p != null){
try {
XContentBuilder builder = XContentFactory.jsonBuilder();
builder.startObject().field("id",p.getId()).field("name",p.getName()).endObject();
result = builder.string();
} catch (IOException e) {
e.printStackTrace();
}
}
return result;
}

查询数据

ES支持分页查询获取数据,也可以一次性获取大量数据,需要使用Scroll Search,QueryBuilder是一个查询条件

	public List<Long> searchData(QueryBuilder builder){
List<Long> ids = new ArrayList<>();
SearchResponse response = client.prepareSearch("test_index").setTypes("test").setQuery(builder).setSize(10).execute().actionGet();
SearchHits hits = response.getHits();
for(SearchHit hit : hits){
Long id = (Long) hit.getSource().get("id");
ids.add(id);
}
return ids;
}
上一篇:java单例模式详解


下一篇:[Hibernate系列—] 3. 映射文件和使用SchemaExport制作自己主动Schema