Springboot 整合 Elasticsearch（五）：使用RestHighLevelClient操作ES ②

Springboot 整合 Elasticsearch（四）：使用RestHighLevelClient操作ES ②

程序员大禹

1509人浏览 · 2024-03-12 14:51:50

程序员大禹 · 2024-03-12 14:51:50 发布

📁 前情提要：

Springboot 整合 Elasticsearch（三）：使用RestHighLevelClient操作ES ①

一、Springboot 整合 Elasticsearch

1、RestHighLevelClient API介绍

一、Springboot 整合 Elasticsearch

1、RestHighLevelClient API介绍

目前的测试数据有：

1.1、全查询 & 分页 & 排序

    @Test
    public void queryAllDoc() {
        try {
            String indexName = "forest";
            SearchRequest request = new SearchRequest();
            request.indices(indexName);

            // 构建查询的请求体
            SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();
            sourceBuilder.query(QueryBuilders.matchAllQuery());
            // 设置分页
            sourceBuilder.from(1);
            sourceBuilder.size(3);
            // 设置排序
            sourceBuilder.sort("price",  SortOrder.DESC);

            request.source(sourceBuilder);

            SearchResponse search = esConfig.restHighLevelClient().search(request, RequestOptions.DEFAULT);
            if (200 != search.status().getStatus()) {
                log.error("查询失败");
            }
            System.out.println("--------查询结果-----------");
            SearchHits hits = search.getHits();
            for (SearchHit hit : hits) {
                System.out.println(hit.getSourceAsString());
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

1.2、单条件查询

1.2.1、termQuery

term属于精确匹配，而且只能查单个词（一个汉字或者一个英文单词），不会对用户输入的内容进行分词；

    @Test
    public void queryDocByKey() {
        try {
            String indexName = "forest";
            SearchRequest request = new SearchRequest();
            request.indices(indexName);
            // 构建查询的请求体
            SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();
            sourceBuilder.query(QueryBuilders.termQuery("price", "300"));
            request.source(sourceBuilder);

            SearchResponse search = esConfig.restHighLevelClient().search(request, RequestOptions.DEFAULT);
            if (200 != search.status().getStatus()) {
                log.error("查询失败");
            }
            System.out.println("--------查询结果-----------");
            SearchHits hits = search.getHits();
            for (SearchHit hit : hits) {
                System.out.println(hit.getSourceAsString());
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

1.2.2、matchQuery

match进行搜索的时候，会先进行分词拆分，拆完后，再来匹配；例如输入“枫树”，match会拆分成“枫”，“树”两个字，只要包含其中一个字的都会被查出来。

    @Test
    public void queryDocByKey() {
        try {
            String indexName = "forest";
            SearchRequest request = new SearchRequest();
            request.indices(indexName);
            // 构建查询的请求体
            SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();
            sourceBuilder.query(QueryBuilders.matchQuery("title", "枫树"));
            request.source(sourceBuilder);

            SearchResponse search = esConfig.restHighLevelClient().search(request, RequestOptions.DEFAULT);
            if (200 != search.status().getStatus()) {
                log.error("查询失败");
            }
            System.out.println("--------查询结果-----------");
            SearchHits hits = search.getHits();
            for (SearchHit hit : hits) {
                System.out.println(hit.getSourceAsString());
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

1.2.3、短语检索

要求所有的分词必须同时出现在文档中，同时位置必须紧邻一致

    @Test
    public void queryDocByPhraseKey() {
        try {
            String indexName = "forest";
            SearchRequest request = new SearchRequest();
            request.indices(indexName);
            // 构建查询的请求体
            SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();
            sourceBuilder.query(QueryBuilders.matchPhraseQuery("title", "枫树"));
            request.source(sourceBuilder);

            SearchResponse search = esConfig.restHighLevelClient().search(request, RequestOptions.DEFAULT);
            if (200 != search.status().getStatus()) {
                log.error("查询失败");
            }
            System.out.println("--------查询结果-----------");
            SearchHits hits = search.getHits();
            for (SearchHit hit : hits) {
                System.out.println(hit.getSourceAsString());
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

1.3、组合查询

must：返回的文档必须满足must子句的条件,并且参与计算分值；
mustNot：返回的文档必须不满足定义的条件；
should：在一个bool查询中，如果没有must或者filter，有一个或者多个should子句，那么只要满足一个就可以返回；
filter：返回的文档必须满足filter子句的条件，但是不会像must一样，参与计算分值；

    @Test
    public void queryByCombinationKey() {
        try {
            // 声明搜索条件对象
            String indexName = "forest";
            SearchRequest request = new SearchRequest();
            request.indices(indexName);
            // 构建查询的请求体
            SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();
            BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();

            boolQueryBuilder.must(QueryBuilders.matchQuery("price", 1000));
            boolQueryBuilder.should(QueryBuilders.matchQuery("id", 3));
            sourceBuilder.query(boolQueryBuilder);
            request.source(sourceBuilder);
            SearchResponse search = esConfig.restHighLevelClient().search(request, RequestOptions.DEFAULT);

            if (200 != search.status().getStatus()) {
                log.error("查询失败");
            }
            System.out.println("--------查询结果-----------");
            SearchHits hits = search.getHits();
            for (SearchHit hit : hits) {
                System.out.println(hit.getSourceAsString());
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

[注]：当使用should查询时，如果包含了must或者filter查询，那么should的查询语句就不是或者的意思了,而是有或者没有都行的含义。

1.4、范围查询

闭区间查询：QueryBuilders.rangeQuery("${fieldName}").from(${fieldValue1}).to(${fieldValue2});
开区间查询：QueryBuilders.rangeQuery("${fieldName}").from(${fieldValue1}, false).to(${fieldValue2}, false);
大于：QueryBuilders.rangeQuery("${fieldName}").gt(${fieldValue});
大于等于：QueryBuilders.rangeQuery("${fieldName}").gte(${fieldValue});
小于：QueryBuilders.rangeQuery("${fieldName}").lt(${fieldValue});
小于等于：QueryBuilders.rangeQuery("${fieldName}").lte(${fieldValue});

 @Test
    public void queryByCombinationKey() {
        try {
            // 声明搜索条件对象
            String indexName = "forest";
            SearchRequest request = new SearchRequest();
            request.indices(indexName);
            // 构建查询的请求体
            SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();
            BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();

            boolQueryBuilder.must(QueryBuilders.rangeQuery("price").from(300).to(500));
            boolQueryBuilder.must(QueryBuilders.rangeQuery("inventory").from(500).to(1000));

            sourceBuilder.query(boolQueryBuilder);
            request.source(sourceBuilder);
            SearchResponse search = esConfig.restHighLevelClient().search(request, RequestOptions.DEFAULT);

            if (200 != search.status().getStatus()) {
                log.error("查询失败");
            }
            System.out.println("--------查询结果-----------");
            SearchHits hits = search.getHits();
            for (SearchHit hit : hits) {
                System.out.println(hit.getSourceAsString());
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

1.5、模糊查询

    @Test
    public void queryFuzzyDocByKey() {
        try {
            String indexName = "forest";
            SearchRequest request = new SearchRequest();
            request.indices(indexName);
            // 构建查询的请求体
            SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();
            sourceBuilder.query(QueryBuilders.fuzzyQuery("title", "松").fuzziness(Fuzziness.AUTO));
            request.source(sourceBuilder);

            SearchResponse search = esConfig.restHighLevelClient().search(request, RequestOptions.DEFAULT);
            if (200 != search.status().getStatus()) {
                log.error("查询失败");
            }
            System.out.println("--------查询结果-----------");
            SearchHits hits = search.getHits();
            for (SearchHit hit : hits) {
                System.out.println(hit.getSourceAsString());
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

1.6、分组查询

按价格分组后求每组的条数

    @Test
    public void queryGroupByKey() {
        try {
            // 声明搜索条件对象
            String indexName = "forest";
            SearchRequest request = new SearchRequest();
            request.indices(indexName);
            // 构建查询的请求体
            SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();

            sourceBuilder.aggregation(AggregationBuilders.terms("price_groupBy").field("price"));

            request.source(sourceBuilder);
            SearchResponse search = esConfig.restHighLevelClient().search(request, RequestOptions.DEFAULT);

            if (200 != search.status().getStatus()) {
                log.error("查询失败");
            }
            List<KeyAndCount> list = new ArrayList<>();
            System.out.println("--------查询结果-----------");
            Terms terms = search.getAggregations().get("price_groupBy");
            List<? extends Terms.Bucket> buckets = terms.getBuckets();
            for (Terms.Bucket bucket : buckets) {
                KeyAndCount keyAndCount = new KeyAndCount();
                keyAndCount.setKey(bucket.getKey().toString());
                keyAndCount.setCount(bucket.getDocCount());
                list.add(keyAndCount);
                System.out.println(keyAndCount);
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }