elasticsearch 7.9.3知识归纳整理（四）之 java实现深分页scroll

java实现深分页scroll

ES对from+size是有限制的，from和size的和不能超过1w。超过后效率十分低下。 ES查询过程：
①将用户的指定的关键字进行分词
②将词汇去分词库中进行检索，得到多个文档的id
③去各个分片中拉去数据指定的数据(耗时长)
④将数据根据score进行排序(耗时长)
⑤根据from的值将查询到的数据舍弃一部分
⑥返回结果
scroll在ES的数据查询方式：
①将用户的指定的关键字进行分词
②将词汇去分词库中进行检索，得到多个文档的id
③将文档的id存放在es的上下文
④根据指定的size去ES中检索指定的数据。拿完数据的文档id，会从上下文移除
⑤如果需要下一页数据，直接去ES的上下文中找后续数据
⑥循环第4、5步 scroll方式，不适合做适时查询：

1.导包

<dependency>
    <groupId>org.elasticsearch.client</groupId>
    <artifactId>elasticsearch-rest-high-level-client</artifactId>
    <version>7.4.2</version>
</dependency>

2.配置

@Configuration
public class MyEsSearchConfig {
	/**
	* es在yml中的配置(这里只有ip没有端口)
	*es:
  	*	host1: es-node1
  	*	host2: es-node2
  	*	host3: es-node3
	**/
    @Value("${es.host1}")
    private String host1;
    @Value("${es.host2}")
    private String host2;
    @Value("${es.host3}")
    private String host3;

    public static final RequestOptions COMMON_OPTIONS;
    static {
        RequestOptions.Builder builder = RequestOptions.DEFAULT.toBuilder();
        COMMON_OPTIONS = builder.build();
    }
    /**
     *方式二 使用账号密码连接
     **/
    @Bean
    public RestHighLevelClient esRestClient(){
        RestClientBuilder builder = RestClient.builder(
                new HttpHost(host1,9200,"http"),
                new HttpHost(host2,9200,"http"),
                new HttpHost(host3,9200,"http"));
        CredentialsProvider credentialsProvider = new BasicCredentialsProvider();
        credentialsProvider .setCredentials(AuthScope.ANY,new UsernamePasswordCredentials("elastic","123456"));
        builder.setHttpClientConfigCallback(f->f.setDefaultCredentialsProvider(credentialsProvider ));
        RestHighLevelClient client = new RestHighLevelClient (builder);
        return client;
    }
}

3.实现深分页

@RestController
@RequestMapping("/esScroll")
public class EsScrollController {
    /**
     * es对from + size 是有限制的 from+size二者之和超过1w会报错
     */
    @Autowired
    private RestHighLevelClient client;

    @PostMapping("/getEsScroll")
    public R getEsScroll(){
        List<Map<String,Object>> list = new ArrayList<>();
        list.clear();
        String indexName = "esIndexAccessControl";
        //1、创建SearchRequest
        SearchRequest request = new SearchRequest(indexName);
        //2、指定scroll的生存时间
        request.scroll(TimeValue.timeValueMinutes(2L));

        SearchSourceBuilder builder = new SearchSourceBuilder();
        ///3、指定每页查询条数
        builder.size(20);
        //3.1、排序
        builder.sort("createTimeMilliss", SortOrder.DESC);
        //3.2、查询全部
        builder.query(QueryBuilders.matchAllQuery());
        request.source(builder);

        try {
            //4、获取返回结果 scrollId,source
            SearchResponse response = client.search(request, RequestOptions.DEFAULT);
            String scrollId = response.getScrollId();
            for(SearchHit hit:response.getHits().getHits()){
                //4.1、首页数据
                Map<String, Object> sourceAsMap = hit.getSourceAsMap();
                list.add(sourceAsMap);
            }
            while(true){
                // 5、循环创建SearchScrollRequest
                SearchScrollRequest scrollRequest = new SearchScrollRequest(scrollId);
                //6、再指定scroll的生存时间，若不指定它会归零
                scrollRequest.scroll(TimeValue.timeValueMinutes(2L));
                //7、执行查询获取结果
                SearchResponse scrollResp = client.scroll(scrollRequest, RequestOptions.DEFAULT);
                //8、判断是否查询到了数据输出
                SearchHit[] hits = scrollResp.getHits().getHits();
                if(hits != null && hits.length>0){
                    for(SearchHit hit : hits){
                        //循环输出
                        Map<String, Object> sourceAsMap = hit.getSourceAsMap();
                        list.add(sourceAsMap);
                    }
                }else {
                    //9、若无数据则退出
                    break;
                }
            }
            //10、创建ClearScrollRequest
            ClearScrollRequest clearScrollRequest = new ClearScrollRequest();
            //11、指定scrollId
            clearScrollRequest.addScrollId(scrollId);
            //12、删除scrollId
            ClearScrollResponse clearScrollResponse = client.clearScroll(clearScrollRequest, RequestOptions.DEFAULT);
           //13、根据它返回判断删除成功没
            boolean succeeded = clearScrollResponse.isSucceeded();
        } catch (IOException e) {
            e.printStackTrace();
        }
        return R.ok("查询成功",list);
    }