ElasticSearch入门

简介

类比mysql等关系型数据库，elasticsearch这种nosql的搜索引擎效率更快，通常的案例有百度淘宝等即时搜索功能
Relational DB -> Databases -> Tables -> Rows -> Columns
关系型数据库 数据库 表 行 列
Elasticsearch -> Indices -> Types -> Documents -> Fields
Elasticsearch 索引 类型 文档 域（字段）

elasticsearch安装(windows)

1. 上官网下载压缩包
2. 解压缩
3. 运行bin目录下的elasticsearch.bat批处理文件
4. cd到对应目录在dos命令行elasticsearch-service.bat install安装elasticsearch服务
5. 访问localhost:9200(默认端口)

kibana可视化工具安装(windows)

1. 上官网下载压缩包
2. 解压缩
3. 运行bin目录下的kibana.bat（git-bash可用cmd.exe /C kibana运行）
4. 访问localhost:5601(默认端口)

elasticsearch在java中的使用方式

RestHighLevelClient的使用

RestHighLevelClient即是Elasticsearch的一个高级客户端，对Elasticsearch集群的所有操作都要经过该客户端来操作。其在java中可通过低级客户端RestClient来获得。

public RestHighLevelClient(RestClientBuilder restClientBuilder) {
    this(restClientBuilder, Collections.emptyList());
}

对索引Index的操作

创建索引

在elasticsearch中索引就是类似数据库的结构，我们可通过创建一个CreateIndexRequest，让客户端来执行创建请求

CreateIndexRequest request=new CreateIndexRequest("new_index");
CreateIndexResponse response = client.indices().create(request, RequestOptions.DEFAULT);
//获得创建结果:boolean
System.out.println(response.isAcknowledged());

同时也可对CreateIndexRequest的各种设置参数的方法来对所创建的索引结构进行设置（如分词器，过滤器等），此处不再一一说明。

判断索引是否存在

public boolean existIndex(String index) {
    GetIndexRequest request=new GetIndexRequest(index);
    try {
        return client.indices().exists(request, RequestOptions.DEFAULT);
    } catch (IOException e) {
        e.printStackTrace();
    }
    return false;
}

删除索引

public boolean deleteIndex(String index) {
    //先判断是否存在
    if(!existIndex(index)){
        return false;
    }
    DeleteIndexRequest request=new DeleteIndexRequest(index);
    try {
        AcknowledgedResponse response = client.indices().delete(request,
                RequestOptions.DEFAULT);
        return response.isAcknowledged();
    } catch (IOException e) {
        e.printStackTrace();
    }
    return false;
}

对文档Document的操作

文档document类比数据库中的行，元组，通常搜索出来的元数据结果就是以每个文档为单位。每个文档都有特定的id，所以以下与特定文档相关的方法都会包含id这个参数

创建文档

/**
*index:索引的名称
*id:在索引中设置的id(如果不手动设也可以，会自动分配id)
*content:向index中添加的document元组,格式为json
**/
public boolean createDocument(String index, String id, String content) {
    //查看文档是否存在
    if(existDocument(index,id)){
        return false;
    }
    IndexRequest request=new IndexRequest(index).id(id);
    request.source(content, XContentType.JSON);
    IndexResponse response = null;
    try {
        response = client.index(request, RequestOptions.DEFAULT);
    } catch (IOException e) {
        e.printStackTrace();
    }
    return (response.status()==RestStatus.CREATED);
}

判断文档是否存在

public boolean existDocument(String index, String id){
    GetRequest request=new GetRequest(index,id);
    try {
        return client.exists(request, RequestOptions.DEFAULT);
    } catch (IOException e) {
        e.printStackTrace();
    }
    return false;
}

获取文档

/**
*这里获得的返回结果为json字符串
**/
public String getDocument(String index, String id) {
    GetRequest request=new GetRequest(index,id);
    GetResponse response = null;
    try {
        response = client.get(request, RequestOptions.DEFAULT);
    } catch (IOException e) {
        e.printStackTrace();
    }
    return response.getSourceAsString();
}

更新文档

public boolean updateDocument(String index, String id, String content) {
    //判断文档是否存在
    if(!existDocument(index,id)){
        return false;
    }
    UpdateRequest request=new UpdateRequest(index,id);
    request.doc(content,XContentType.JSON);
    UpdateResponse response = null;
    try {
        response = client.update(request, RequestOptions.DEFAULT);
    } catch (IOException e) {
        e.printStackTrace();
    }
    return (response.status()==RestStatus.OK);
}

删除文档

public boolean deleteDocument(String index, String id) {
    if(!existDocument(index,id)){
        return false;
    }
    DeleteRequest request=new DeleteRequest(index,id);
    DeleteResponse response = null;
    try {
        response = client.delete(request, RequestOptions.DEFAULT);
    } catch (IOException e) {
        e.printStackTrace();
    }
    return (response.status()==RestStatus.OK);
}

批量操作

我们可以通过BulkRequest来对多个DocWriteRequest(DeleteRequest,IndexRequest,UpdateRequest)操作进行添加集成，然后统一让RestHighLevelClient执行对文档的批量操作，下面封装了一个进行批量操作的方法

public boolean bulkRequest(String index, Iterable<DocWriteRequest<?>> requests) {
    BulkRequest bulkRequest=new BulkRequest();
    requests.forEach(x->{
        bulkRequest.add(x);
    });
    BulkResponse responses = null;
    try {
        responses = client.bulk(bulkRequest, RequestOptions.DEFAULT);
    } catch (IOException e) {
        e.printStackTrace();
    }
    BulkItemResponse[] items = responses.getItems();
    for(BulkItemResponse tmp:items){
        System.out.println("操作:"+tmp.getOpType().name()+"-主键id:"+tmp.getId()+"-是否成功:"+(!tmp.isFailed()));
    }
    return !responses.hasFailures();
}

使用用例

@Test
void testBulkOperation(){
    //添加document操作
    IndexRequest indexRequest=new IndexRequest("desktop").id("testId")
        .source(XContentType.JSON,"testField1","testField1");
    //修改document操作
    JSONObject jsonObject=new JSONObject();
    jsonObject.put("programId","5");
    jsonObject.put("programName","网易云音乐-updated");
    UpdateRequest updateRequest=new UpdateRequest("desktop","1006")
        .doc(jsonObject.toJSONString(),XContentType.JSON);
    //操作集合
    List<DocWriteRequest<?>>requests=new ArrayList<>();
    requests.add(indexRequest);
    requests.add(updateRequest);
    System.out.println(service.bulkRequest("desktop",requests));
}

对多个索引中特定的字段进行关键字查找

/**
*要设置查找的字段也可以当成参数传递进来，懒得改了
**/
public List<JSONObject> searchByKeyword(String keyword,String...indices) {
    List<JSONObject>results=null;
    SearchRequest request=new SearchRequest();
    request.indices(indices);
    SearchSourceBuilder builder=new SearchSourceBuilder();
    //设置查找字段
    builder.query(QueryBuilders.multiMatchQuery(keyword,"programName","tag","username")
                  .fuzziness(Fuzziness.AUTO));
    request.source(builder);
    SearchResponse response = null;
    try {
        response = client.search(request, RequestOptions.DEFAULT);
    } catch (IOException e) {
        e.printStackTrace();
    }
    SearchHit[] hits = response.getHits().getHits();
    if(hits.length>0){
        results=new ArrayList<>();
        for(SearchHit hit:hits){
            JSONObject jsonObject=JSONObject.parseObject(hit.getSourceAsString());
            results.add(jsonObject);
        }
    }
    return results;
}

使用账号密码连接elasticsaerch

通常默认情况下es本地是不设账号密码的，可直接通过localhost:9200访问，但是也可以设置密码来加强es的安全性

修改elasticsearch.yml配置文件

使用xpack给es添加安全支持，同时允许http请求跨域，在elasticsearch.yml文件中添加以下配置

http.cors.enabled: true
http.cors.allow-origin: "*"
http.cors.allow-headers: Authorization
xpack.security.enabled: true
xpack.security.transport.ssl.enabled: true

执行自定义密码的命令

cd到es的bin目录下，执行elasticsearch-setup-passwords interactive，按照提示分别给几个系统都设置密码

通过kibana或localhost:9200访问测试

启动kibana/elasticsearch服务，在访问lcalhost:5601/localhost:9200时输入es账号密码，默认账号为elastic

修改密码的curl命令

POST /_security/user/{username}/_password
{
  "password": "your password"
}

在Java中通过api连接

我们通常使用的是RestHighLevelClient，其可通过低级客户端的构造器RestClientBuilder获得，根据官方文档的指引，我们可通过设置http客户端的配置来设置账号密码最后构建RestClientBuilder

@Bean
public RestClientBuilder restClientBuilder(){
    final CredentialsProvider credentialsProvider =
        new BasicCredentialsProvider();
    credentialsProvider.setCredentials(AuthScope.ANY,
                                       new UsernamePasswordCredentials(username, 	password));
    RestClientBuilder builder=RestClient.builder(new HttpHost(host,port,protocol));
    builder.setHttpClientConfigCallback(new RestClientBuilder.HttpClientConfigCallback() {
        @Override
        public HttpAsyncClientBuilder customizeHttpClient(HttpAsyncClientBuilder httpAsyncClientBuilder) {
            httpAsyncClientBuilder.disableAuthCaching();
            return httpAsyncClientBuilder.setDefaultCredentialsProvider(credentialsProvider);
        }
    });
    return builder;
}

@Bean(name = "restHighLevelClient")
public RestHighLevelClient restHighLevelClient(@Autowired RestClientBuilder restClientBuilder){
    return new RestHighLevelClient(restClientBuilder);
}