五.全文检索ElasticSearch经典入门-ElasticSearch Java实战

Posted 墨家巨子@俏如来

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了五.全文检索ElasticSearch经典入门-ElasticSearch Java实战相关的知识,希望对你有一定的参考价值。

前言

今天开始我们学习如何使用Java来操作ES,这里会讲两种操作方式,一是使用ES提供的jar包来操作,二是使用SpringBootData来操作,第二种方式是企业里面使用的较多的方式了。

Java 操作ES

环境集成

创建一个普通的Java工程,首先我们需要导入ES的jar包

<dependency>
   <groupId>org.elasticsearch.client</groupId>
    <artifactId>transport</artifactId>
    <version>6.8.6</version>
</dependency>
<dependency>
    <groupId>junit</groupId>
    <artifactId>junit</artifactId>
    <version>4.12</version>
    <scope>compile</scope>
</dependency>

然后创建ES的客户端

public class ESClientUtil 

    public static TransportClient getClient()
        TransportClient client = null;
        Settings settings = Settings.builder().put("cluster.name", "elasticsearch").build();
        try 
            client = new PreBuiltTransportClient(settings).addTransportAddress(
            new TransportAddress(InetAddress.getByName("127.0.0.1"), 9300));
         catch (UnknownHostException e) 
            e.printStackTrace();
        
        return client;
    


添加文档

编写测试类 ,往指定的索引库中添加文档

@Test
public void testAdd() 
  //获取客户端对象
  TransportClient client = ESClientUtil.getClient();

  //创建索引
  IndexRequestBuilder indexRequestBuilder = client.prepareIndex("orders", "_doc", "1");
  Map<String,Object> data = new HashMap<>();
  data.put("id",1);
  data.put("title","买了一个鼠标");
  data.put("count",1);
  data.put("amount",100.00);
  data.put("status",1);
  //获取结果
  IndexResponse indexResponse = indexRequestBuilder.setSource(data).get();
  //打印给过
  System.out.println(indexResponse);
  client.close();

打印结果如下 :

IndexResponse[index=orders,type=_doc,id=1,version=18,result=updated,seqNo=19,primaryTerm=3,shards=“total”:2,“successful”:1,“failed”:0]

获取一个文档

@Test
public void testGet() 
	//获取客户端对象
	TransportClient client = ESClientUtil.getClient();
	//获取文档
	GetResponse response = client.prepareGet("orders", "_doc", "1").get();
	//打印结果
	System.out.println(response);

打印结果如下:amount=100.0, count=1, id=1, title=买了一个鼠标, status=1

更新一个文档

@Test

public void testUpdate()
    //获取客户端对象
    TransportClient client = ESClientUtil.getClient();

    //修改索引
    UpdateRequestBuilder updateRequestBuilder = client.prepareUpdate("orders", "_doc", "1");
    Map<String,Object> data = new HashMap<>();
    data.put("id",1);
    data.put("title","买了一个鼠标");
    data.put("count",2);
    data.put("amount",100.00);
    //data.put("status",1);
    //获取结果设置修改内容
    UpdateResponse updateResponse = updateRequestBuilder.setDoc(data).get();

    System.out.println(updateResponse);
    client.close();

打印结果如下:UpdateResponse[index=orders,type=_doc,id=1,version=19,seqNo=20,primaryTerm=3,result=updated,shards=ShardInfototal=2, successful=1, failures=[]]

注意:这种修改方式是不会影响到未修改的字段的,比如上面注释掉status字段,status字段不会被修改。也就是说这种方式是局部修改。

删除一个文档

 @Test
 public void testDelete() 
      //获取客户端对象
      TransportClient client = ESClientUtil.getClient();
      //执行删除
      DeleteResponse response = client.prepareDelete("orders", "_doc", "1").get();
      //打印结果
      System.out.println(response);
  

打印结果:DeleteResponse[index=orders,type=_doc,id=1,version=20,result=deleted,shards=ShardInfototal=2, successful=1, failures=[]]

批量操作

同时提交多个操作

@Test
public void testBuilkAdd()
   //获取客户端对象
   TransportClient client = ESClientUtil.getClient();

   BulkRequestBuilder bulkRequestBuilder = client.prepareBulk();

   Map<String,Object> data1 = new HashMap<>();
   data1.put("id",1);
   data1.put("title","买了一个鼠标");
   data1.put("count",2);
   data1.put("amount",100.00);

   //添加操作
   bulkRequestBuilder.add(client.prepareIndex("orders", "_doc", "1").setSource(data1));

   Map<String,Object> data2 = new HashMap<>();
   data2.put("id",2);
   data2.put("title","买了一个鼠标");
   data2.put("count",2);
   data2.put("amount",100.00);
   //添加操作
   bulkRequestBuilder.add(client.prepareIndex("orders", "_doc", "2").setSource(data2));

   Map<String,Object> data3 = new HashMap<>();
   data3.put("id",2);
   data3.put("title","买了一个鼠标");
   data3.put("count",2);
   data3.put("amount",200.00);

   //修改操作
   bulkRequestBuilder.add(client.prepareUpdate("orders", "_doc", "2").setDoc(data3));


   BulkResponse bulkItemResponses = bulkRequestBuilder.get();
   Iterator<BulkItemResponse> iterator = bulkItemResponses.iterator();
   while(iterator.hasNext())
       BulkItemResponse next = iterator.next();
       System.out.println(next.getResponse());
   
   client.close();
 

打印结果
IndexResponse[index=orders,type=_doc,id=1,version=1,result=created,seqNo=22,primaryTerm=3,shards=“total”:2,“successful”:1,“failed”:0]
IndexResponse[index=orders,type=_doc,id=2,version=4,result=updated,seqNo=3,primaryTerm=3,shards=“total”:2,“successful”:1,“failed”:0]
UpdateResponse[index=orders,type=_doc,id=2,version=5,seqNo=4,primaryTerm=3,result=updated,shards=ShardInfototal=2, successful=1, failures=[]]

DSL查询

案例:查询订单标题中包含:鼠标,价格在10-2000之间,状态为1,按照价格正排序

@Test
    public void testSearch()
        //获取客户端对象
        TransportClient client = ESClientUtil.getClient();

        SearchRequestBuilder searchRequestBuilder = client.prepareSearch("orders");
        //设置分页
        searchRequestBuilder.setFrom(0);
        searchRequestBuilder.setSize(10);
        searchRequestBuilder.addSort("amount", SortOrder.ASC);

        //查询条件
        BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
        //dsl查询
        List<QueryBuilder> must = boolQueryBuilder.must();
        must.add(QueryBuilders.matchQuery("title" , "鼠标"));

        //dsl过滤
        List<QueryBuilder> filter = boolQueryBuilder.filter();
        filter.add(QueryBuilders.rangeQuery("amount").lte(2000).gte(10));
        filter.add(QueryBuilders.termQuery("status",1));

        searchRequestBuilder.setQuery(boolQueryBuilder);

        SearchResponse searchResponse = searchRequestBuilder.get();

        SearchHits hits = searchResponse.getHits();

        System.out.println("条数:"+hits.getTotalHits());
        for (SearchHit hit : hits.getHits()) 
            System.out.println(hit.getSourceAsMap());

        

        client.close();
    

SpringBootData操作ES

环境集成

第一步:导入SpringBoot提整合ES的依赖 spring-boot-starter-data-elasticsearch

<!--SpringBoot-->
<parent>
   <groupId> org.springframework.boot</groupId>
   <artifactId>spring-boot-starter-parent</artifactId>
   <version>2.2.5.RELEASE</version>
</parent>

<dependencies>
	<dependency>
	    <groupId>org.springframework.boot</groupId>
	    <artifactId>spring-boot-starter-data-elasticsearch</artifactId>
	</dependency>
	<dependency>
	    <groupId>org.springframework.boot</groupId>
	     <artifactId>spring-boot-starter-test</artifactId>
	 </dependency>
	  <dependency>
            <groupId>org.projectlombok</groupId>
            <artifactId>lombok</artifactId>
        </dependency>
</dependencies>

第二步:然后需要在yml中对ES进行配置 , 如果是集群配置增加uri即可,单个配置如下:

spring:
  elasticsearch:
    rest:
      uris:
        - http://localhost:9200

第三步:编写启动类

@SpringBootApplication
public class SearchStart 

    public static void main(String[] args) 
        SpringApplication.run(SearchStart.class);
    


创建Document对象

第三步:编写Document对象 ,该对象是对存储到ES中的数据的封装,同时文档映射也是通过它来实现


//标记该对象是ES的文档对象
//indexName 索引库
//type 类型
@Document(indexName = "orders",type = "_doc")
@Data
@AllArgsConstructor
@NoArgsConstructor
public class OrderDoc 

    //标记为文档ID,该ID的值会作为document的id值
    @Id
    private Long id;
    /**
     * 标题需要分词,指定为text;并使用IK分词器
     * 一般需要作为关键字搜索的字段都要指定为text,因为需要分词且创建索引
     */
    @Field(type = FieldType.Text,analyzer = "ik_max_word",searchAnalyzer = "ik_max_word")
    //@Field(type = FieldType.Keyword)
    private String title;

    /**
     * 指定为integer类型
     */
    @Field(type = FieldType.Integer)
    private int count;
    /**
     * 状态指定为 integer类型
     */
    @Field(type = FieldType.Integer)
    private int status;

    /**
     * 金额
     */
    @Field(type = FieldType.Double)
    private BigDecimal amount;


创建Repository

第四步:SpringBootData提供了ElasticsearchRepository 来操作ES,该接口中包含了针对ES的CRUD方法,我们编写接口集成它即可使用

@Repository
public interface OrderRepository extends ElasticsearchRepository<OrderDoc,Long> 

注意:这里的泛型是当前Repository所要管理的实体类,也就是OrderDoc,Long是实体类ID的类型

创建索引和映射

编写一个基于SpringBoot的测试类如下:

@RunWith(SpringRunner.class)
@SpringBootTest(classes = SearchStart.class)
public class OrderESTest 
	//操作ES的template模板
    @Autowired
    private ElasticsearchRestTemplate template;
    
    @Autowired
    private OrderRepository orderRepository;

    @Test
    public void test()
   		//创建索引
        template.createIndex(CourseDoc.class);
        //创建映射
        template.putMapping(CourseDoc.class);
    

这里我注入了一个 ElasticsearchRestTemplate 工具类,通过它的 createIndex 可以创建索引库,SpringBoot指定解析 CourseDoc上的@Document注解中的index来确定索引库的名字。创建映射也是同样的道理。

基础CRUD

在SpringBoot中对ES的Crud显得极其简单,直接调用repository内置方法即可完成

@Test
    public void testAdd()
        orderRepository.save(new OrderDoc(1L,"买了一个表",1,1,new BigDecimal("200")));
    

    @Test
    public void testGet()
        Optional<OrderDoc> optional = orderRepository.findById(1l);
        System.out.println(optional.get());
    

    @Test
    public void testDelete()
        orderRepository.deleteById(1L);
    

注意:这里并没有update方法,因为save对象的时候如果ID已经存在,就会执行update操作。repository中还有很多的方法可以使用,你可以自己去调试。

DSL查询

在SpringBoot中我们通过 NativeSearchQueryBuilder 来构建查询条件,调用repository.search来执行查询。
案例:查询标题中包含鼠标 ,状态为1,金额在10-2000之间,按照金额倒排,查询第2页数据,每页10条

 @Test
public void testSearch()

    //查询构建器
    NativeSearchQueryBuilder builder = new NativeSearchQueryBuilder();

    //设置分页: 第2页 (0开始), 每页10数
    builder.withPageable(PageRequest.of(1,10));
    //设置排序 : 金额倒排
    builder.withSort(SortBuilders.fieldSort("amount").order(SortOrder.DESC));

    //构建组合查询
    BoolQueryBuilder boolQuery = QueryBuilders.boolQuery();

    //标题包含鼠标
    boolQuery.must(QueryBuilders.matchQuery("title","鼠标"))
            //状态值查询
            .filter(QueryBuilders.九.全文检索ElasticSearch经典入门-ElasticSearch映射修改

一.全文检索ElasticSearch经典入门-ElasticSearch介绍与安装

四.全文检索ElasticSearch经典入门-ElasticSearch DSL查询

九.全文检索ElasticSearch经典入门-ElasticSearch映射修改

十一.全文检索ElasticSearch经典入门-SpringBoot整合ElasticSearch7

十一.全文检索ElasticSearch经典入门-SpringBoot整合ElasticSearch7