elasticsearch聚合aggregations

聚合常见的有三类：

• 桶（Bucket）聚合：用来对文档做分组

  • TermAggregation：按照文档字段值分组，例如按照品牌值分组、按照国家分组

  • Date Histogram：按照日期阶梯分组，例如一周为一组，或者一月为一组

• 度量（Metric）聚合：用以计算一些值，比如：最大值、最小值、平均值等

  • Avg：求平均值

  • Max：求最大值

  • Min：求最小值

  • Stats：同时求max、min、avg、sum等

• 管道（pipeline）聚合：其它聚合的结果为基础做聚合

========================================================

Bucket聚合语法

GET /hotel/_search
{
  "size": 0,  // 设置size为0，结果中不包含文档，只包含聚合结果
  "aggs": { // 定义聚合
    "brandAgg": { //给聚合起个名字
      "terms": { // 聚合的类型，按照品牌值聚合，所以选择term
        "field": "brand", // 参与聚合的字段
        "size": 20 // 希望获取的聚合结果数量
      }
    }
  }
}

=============================

聚合结果排序

GET /hotel/_search
{
  "size": 0,
  "aggs": {
    "brandAgg": {
      "terms": {
        "field": "brand",
        "order": {
          "_count": "asc" // 按照_count升序排列
        },
        "size": 20
      }
    }
  }
}

=======================

限定聚合范围

GET /hotel/_search
{
  "query": {
    "range": {
      "price": {
        "lte": 200 // 只对200元以下的文档聚合
      }
    }
  },
  "size": 0,
  "aggs": {
    "brandAgg": {
      "terms": {
        "field": "brand",
        "size": 20
      }
    }
  }
}

===============================

Metric聚合语法

GET /hotel/_search
{
  "size": 0,
  "aggs": {
    "brandAgg": {
      "terms": {
        "field": "brand",
        "size": 20
      },
      "aggs": { // 是brands聚合的子聚合，也就是分组后对每组分别计算
        "score_stats": { // 聚合名称
          "stats": { // 聚合类型，这里stats可以计算min、max、avg等
            "field": "score" // 聚合字段，这里是score
          }
        }
      }
    }
  }
}

===============================

GET /hotel/_search
{
  "query": {
    "term": {
      "city": {
        "value": "上海"
      }
    }
  },
  "size": 0,
  "aggs": {
    "brandressult": {
      "terms": {
        "field": "brand",
        "size": 10,
        "order": {
           "socer_stats.avg": "asc"
        }
        
      },
      "aggs": {
        "socer_stats": {
          "stats": {
            "field": "price"
          
          }
      
        }
        
      }
    }
    
  }
}

============================

RestAPI实现聚合

@Override
public Map<String, List<String>> filters(RequestParams params) {
    try {
        // 1.准备Request
        SearchRequest request = new SearchRequest("hotel");
        // 2.准备DSL
        // 2.1.query
        buildBasicQuery(params, request);
        // 2.2.设置size
        request.source().size(0);
        // 2.3.聚合
        buildAggregation(request);
        // 3.发出请求
        SearchResponse response = client.search(request, RequestOptions.DEFAULT);
        // 4.解析结果
        Map<String, List<String>> result = new HashMap<>();
        Aggregations aggregations = response.getAggregations();
        // 4.1.根据品牌名称，获取品牌结果
        List<String> brandList = getAggByName(aggregations, "brandAgg");
        result.put("品牌", brandList);
        // 4.2.根据品牌名称，获取品牌结果
        List<String> cityList = getAggByName(aggregations, "cityAgg");
        result.put("城市", cityList);
        // 4.3.根据品牌名称，获取品牌结果
        List<String> starList = getAggByName(aggregations, "starAgg");
        result.put("星级", starList);

        return result;
    } catch (IOException e) {
        throw new RuntimeException(e);
    }
}

private void buildAggregation(SearchRequest request) {
    request.source().aggregation(AggregationBuilders
                                 .terms("brandAgg")
                                 .field("brand")
                                 .size(100)
                                );
    request.source().aggregation(AggregationBuilders
                                 .terms("cityAgg")
                                 .field("city")
                                 .size(100)
                                );
    request.source().aggregation(AggregationBuilders
                                 .terms("starAgg")
                                 .field("starName")
                                 .size(100)
                                );
}

private List<String> getAggByName(Aggregations aggregations, String aggName) {
    // 4.1.根据聚合名称获取聚合结果
    Terms brandTerms = aggregations.get(aggName);
    // 4.2.获取buckets
    List<? extends Terms.Bucket> buckets = brandTerms.getBuckets();
    // 4.3.遍历
    List<String> brandList = new ArrayList<>();
    for (Terms.Bucket bucket : buckets) {
        // 4.4.获取key
        String key = bucket.getKeyAsString();
        brandList.add(key);
    }
    return brandList;
}

===========================

获取数据

@Test
void buildAggregation2() throws IOException {
    // 1.准备Request
    SearchRequest request = new SearchRequest("hotel");

    request.source().query(QueryBuilders.termQuery("city", "上海"));
    request.source().size(10);
    request.source().aggregation(AggregationBuilders
            .terms("brandressult")
            .field("brand")
            .size(5)
            .order(BucketOrder.count(true))
            .subAggregation(AggregationBuilders.stats("priceResult").field("price")
            )

    );
    SearchResponse response = client.search(request, RequestOptions.DEFAULT);
    Aggregations aggregations = response.getAggregations();
    Map<String, Aggregation> stringAggregationMap = aggregations.asMap();
    ParsedStringTerms brandressult = (ParsedStringTerms) stringAggregationMap.get("brandressult");
    List<? extends Terms.Bucket> buckets = brandressult.getBuckets();
    System.out.println(buckets.size());
    for (Terms.Bucket bucket : buckets) {
        String key = (String) bucket.getKey();
        long docCount = bucket.getDocCount();
        System.out.println("key=" + key);
        System.out.println("docCount=" + docCount);

        Stats priceResult = bucket.getAggregations().get("priceResult");
        double avg = priceResult.getAvg();
        priceResult.getCount();
        priceResult.getMax();
        priceResult.getMin();
        System.out.println(avg);
        
    }


    handleResponse(response);
}

private void handleResponse(SearchResponse response) {
    // 4.解析响应
    SearchHits searchHits = response.getHits();
    // 4.1.获取总条数
    long total = searchHits.getTotalHits().value;
    System.out.println("共搜索到" + total + "条数据");
    // 4.2.文档数组
    SearchHit[] hits = searchHits.getHits();
    // 4.3.遍历
    for (SearchHit hit : hits) {
        // 获取文档source
        String json = hit.getSourceAsString();
        // 反序列化
        HotelDoc hotelDoc = JSON.parseObject(json, HotelDoc.class);
        System.out.println("hotelDoc = " + hotelDoc);
    }
}