一、简介

Elasticsearch是一个分布式、RESTful风格的搜索和数据分析引擎,适用于各种数据类型,数字、文本、地理位置、结构化数据、非结构化数据;

在实际的工作中,历经过Elasticsearch从6.0到7.0的版本升级,而这次SpringBoot3和ES8.0的集成,虽然脚本的语法变化很小,但是Java客户端的API语法变化很大;

二、环境搭建

1、下载安装包

需要注意的是,这些安装包的版本要选择对应的,不然容易出问题;

软件包:elasticsearch-8.8.2-darwin-x86_64.tar.gz
分词器工具:elasticsearch-analysis-ik-8.8.2.zip
可视化工具:kibana-8.8.2-darwin-x86_64.tar.gz

2、服务启动

不论是ES还是Kibana,在首次启动后,会初始化很多配置文件,可以根据自己的需要做相关的配置调整,比如常见的端口调整,资源占用,安全校验等;

1、启动ES
elasticsearch-8.8.2/bin/elasticsearch

本地访问:localhost:9200

2、启动Kibana
kibana-8.8.2/bin/kibana

本地访问:http://localhost:5601

# 3、查看安装的插件
http://localhost:9200/_cat/plugins  ->  analysis-ik 8.8.2

三、工程搭建

1、工程结构

2、依赖管理

在starter-elasticsearch组件中,实际上依赖的是elasticsearch-java组件的8.7.1版本;

<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-data-elasticsearch</artifactId>
    <version>${spring-boot.version}</version>
</dependency>

3、配置文件

在上面环境搭建的过程中,已经禁用了用户和密码的登录验证,配置ES服务地址即可;

spring:
  # ElasticSearch配置
  elasticsearch:
    uris: localhost:9200

四、基础用法

1、实体类

通过Document和Field注解描述ES索引结构的实体类,注意这里JsonIgnoreProperties注解,解决索引中字段和实体类非一一对应的而引起的JSON解析问题;

@JsonIgnoreProperties(ignoreUnknown = true)
@Document(indexName = "contents_index", createIndex = false)
public class ContentsIndex implements Serializable {

    private static final long serialVersionUID=1L;

    @Field(type= FieldType.Integer)
    private Integer id;

    @Field(type= FieldType.Keyword)
    private String title;

    @Field(type= FieldType.Keyword)
    private String intro;

    @Field(type= FieldType.Text)
    private String content;

    @Field(type= FieldType.Integer)
    private Integer createId;

    @Field(type= FieldType.Keyword)
    private String createName;

    @Field(type= FieldType.Date,format = DateFormat.date_hour_minute_second)
    private Date createTime;
}

2、初始化索引

基于ElasticsearchTemplate类和上述实体类,实现索引结构的初始化,并且将tb_contents表中的数据同步到索引中,最后通过ID查询一条测试数据;

@Service
public class ContentsIndexService {
    private static final Logger log = LoggerFactory.getLogger(ContentsIndexService.class);

    @Resource
    private ContentsService contentsService ;
    @Resource
    private ElasticsearchTemplate template ;

    /**
     * 初始化索引结构和数据
     */
    public void initIndex (){
        // 处理索引结构
        IndexOperations indexOps = template.indexOps(ContentsIndex.class);
        if (indexOps.exists()){
            boolean delFlag = indexOps.delete();
            log.info("contents_index exists,delete:{}",delFlag);
            indexOps.createMapping(ContentsIndex.class);
        } else {
            log.info("contents_index not exists");
            indexOps.createMapping(ContentsIndex.class);
        }
        // 同步数据库表记录
        List<Contents> contentsList = contentsService.queryAll();
        if (contentsList.size() > 0){
            List<ContentsIndex> contentsIndexList = new ArrayList<>() ;
            contentsList.forEach(contents -> {
                ContentsIndex contentsIndex = new ContentsIndex() ;
                BeanUtils.copyProperties(contents,contentsIndex);
                contentsIndexList.add(contentsIndex);
            });
            template.save(contentsIndexList);
        }
        // ID查询
        ContentsIndex contentsIndex = template.get("10",ContentsIndex.class);
        log.info("contents-index-10:{}",contentsIndex);
    }
}

3、仓储接口

继承ElasticsearchRepository接口,可以对ES这种特定类型的存储库进行通用增删改查操作;在测试类中对该接口的方法进行测试;

// 1、接口定义
public interface ContentsIndexRepository extends ElasticsearchRepository<ContentsIndex,Long> {
}

// 2、接口测试
public class ContentsIndexRepositoryTest {
    @Autowired
    private ContentsIndexRepository contentsIndexRepository;

    @Test
    public void testAdd (){
        // 单个新增
        contentsIndexRepository.save(buildOne());
        // 批量新增
        contentsIndexRepository.saveAll(buildList()) ;
    }

    @Test
    public void testUpdate (){
        // 根据ID查询后再更新
        Optional<ContentsIndex> contentsOpt = contentsIndexRepository.findById(14L);
        if (contentsOpt.isPresent()){
            ContentsIndex contentsId = contentsOpt.get();
            System.out.println("id=14:"+contentsId);
            contentsId.setContent("update-content");
            contentsId.setCreateTime(new Date());
            contentsIndexRepository.save(contentsId);
        }
    }

    @Test
    public void testQuery (){
        // 单个ID查询
        Optional<ContentsIndex> contentsOpt = contentsIndexRepository.findById(1L);
        if (contentsOpt.isPresent()){
            ContentsIndex contentsId1 = contentsOpt.get();
            System.out.println("id=1:"+contentsId1);
        }
        // 批量ID查询
        Iterator<ContentsIndex> contentsIterator = contentsIndexRepository
                                        .findAllById(Arrays.asList(10L,12L)).iterator();
        while (contentsIterator.hasNext()){
            ContentsIndex contentsIndex = contentsIterator.next();
            System.out.println("id="+contentsIndex.getId()+":"+contentsIndex);
        }
    }

    @Test
    public void testDelete (){
        contentsIndexRepository.deleteById(15L);
        contentsIndexRepository.deleteById(16L);
    }
}

4、查询语法

无论是ElasticsearchTemplate类还是ElasticsearchRepository接口,都是对ES常用的简单功能进行封装,在实际使用时,复杂的查询语法还是依赖ElasticsearchClient和原生的API封装;

这里主要演示七个查询方法,主要涉及:ID查询,字段匹配,组合与范围查询,分页与排序,分组统计,最大值查询和模糊匹配;更多的查询API还是要多看文档中的案例才行;

public class ElasticsearchClientTest {

    @Autowired
    private ElasticsearchClient client ;

    @Test
    public void testSearch1 () throws IOException {
        // ID查询
        GetResponse<ContentsIndex> resp = client.get(
                getReq ->getReq.index("contents_index").id("7"), ContentsIndex.class);
        if (resp.found()){
            ContentsIndex contentsIndex = resp.source() ;
            System.out.println("contentsIndex-7:"+contentsIndex);
        }
    }

    @Test
    public void testSearch2 () throws IOException {
        // 指定字段匹配
        SearchResponse<ContentsIndex> resp = client.search(searchReq -> searchReq.index("contents_index")
                        .query(query -> query.match(field -> field
                        .field("createName").query("张三"))),ContentsIndex.class);
        printResp(resp);
    }

    @Test
    public void testSearch3 () throws IOException {
        // 组合查询:姓名和时间范围
        Query byName = MatchQuery.of(field -> field.field("createName").query("王五"))._toQuery();
        Query byTime = RangeQuery.of(field -> field.field("createTime")
                        .gte(JsonData.of("2023-07-10T00:00:00"))
                        .lte(JsonData.of("2023-07-12T00:00:00")))._toQuery();
        SearchResponse<ContentsIndex> resp = client.search(searchReq -> searchReq.index("contents_index")
                        .query(query -> query.bool(boolQuery -> boolQuery.must(byName).must(byTime))),ContentsIndex.class);
        printResp(resp);
    }

    @Test
    public void testSearch4 () throws IOException {
        // 排序和分页,在14条数据中,根据ID倒序排列,从第5条往后取4条数据
        SearchResponse<ContentsIndex> resp = client.search(searchReq -> searchReq.index("contents_index")
                .from(5).size(4)
                .sort(sort -> sort.field(sortField -> sortField.field("id").order(SortOrder.Desc))),ContentsIndex.class);
        printResp(resp);
    }

    @Test
    public void testSearch5 () throws IOException {
        // 根据createId分组统计
        SearchResponse<ContentsIndex> resp = client.search(searchReq -> searchReq.index("contents_index")
                .aggregations("createIdGroup",agg -> agg.terms(term -> term.field("createId"))),ContentsIndex.class);
        Aggregate aggregate = resp.aggregations().get("createIdGroup");
        LongTermsAggregate termsAggregate = aggregate.lterms();
        Buckets<LongTermsBucket> buckets = termsAggregate.buckets();
        for (LongTermsBucket termsBucket : buckets.array()) {
            System.out.println(termsBucket.key() + " : " + termsBucket.docCount());
        }
    }

    @Test
    public void testSearch6 () throws IOException {
        // 查询最大的ID
        SearchResponse<ContentsIndex> resp = client.search(searchReq -> searchReq.index("contents_index")
                .aggregations("maxId",agg -> agg.max(field -> field.field("id"))),ContentsIndex.class);
        for (Map.Entry<String, Aggregate> entry : resp.aggregations().entrySet()){
            System.out.println(entry.getKey()+":"+entry.getValue().max().value());
        }
    }

    @Test
    public void testSearch7 () throws IOException {
        // 模糊查询title字段,允许1个误差
        Query byContent = FuzzyQuery.of(field -> field.field("title").value("设计").fuzziness("1"))._toQuery();
        SearchResponse<ContentsIndex> resp = client.search(
                searchReq -> searchReq.index("contents_index").query(byContent),ContentsIndex.class);
        printResp(resp);
    }

    private void printResp (SearchResponse<ContentsIndex> resp){
        TotalHits total = resp.hits().total();
        System.out.println("total:"+total);
        List<Hit<ContentsIndex>> hits = resp.hits().hits();
        for (Hit<ContentsIndex> hit: hits) {
            ContentsIndex contentsIndex = hit.source();
            System.out.println(hit.id()+":"+contentsIndex);
        }
    }
}

 

 

 

转载至:https://www.cnblogs.com/cicada-smile/p/17632955.html

最后修改于
上一篇