ElasticSearch 聚合查詢百分比

這里用的是es5.6.9

bucket_script ：它執行一個腳本，該腳本可以對多桶聚合中的指定度量執行每桶計算，指定的度量標准必須為數字，並且腳本必須返回數值。

官方語法

https://www.elastic.co/guide/en/elasticsearch/reference/master/search-aggregations-pipeline-bucket-script-aggregation.html

{
    "bucket_script": {
        "buckets_path": {
            "my_var1": "the_sum", 
            "my_var2": "the_value_count"
        },
        "script": "params.my_var1 / params.my_var2"
    }
}

給的一個例子

算出t-shirts賣出的錢占總錢數的百分比

POST /sales/_search
{
    "size": 0,
    "aggs" : {
        "sales_per_month" : {
            "date_histogram" : {
                "field" : "date",
                "interval" : "month"
            },
            "aggs": {
                "total_sales": {
                    "sum": {
                        "field": "price"
                    }
                },
                "t-shirts": {
                  "filter": {
                    "term": {
                      "type": "t-shirt"
                    }
                  },
                  "aggs": {
                    "sales": {
                      "sum": {
                        "field": "price"
                      }
                    }
                  }
                },
                "t-shirt-percentage": {
                    "bucket_script": {
                        "buckets_path": {
                          "tShirtSales": "t-shirts>sales",
                          "totalSales": "total_sales"
                        },
                        "script": "params.tShirtSales / params.totalSales * 100"
                    }
                }
            }
        }
    }
}

script：聚合運行的腳本
buckets_path：腳本變量的映射及其與我們希望用於變量的桶的關聯路徑
format:格式以應用於此聚合的輸出值
gap_policy:在數據中發現差距時應用的策略

其中"t-shirts"這個聚合里面又套了一層取值的聚合是因為buckets_path里引用的必須是一個值filter返回的不是一個值

也可以這樣寫

POST /sales/_search
{
    "size": 0,
    "aggs" : {
        "sales_per_month" : {
            "date_histogram" : {
                "field" : "date",
                "interval" : "month"
            },
            "aggs": {
                "total_sales": {
                    "sum": {
                        "field": "price"
                    }
                },
                "t-shirts": {
                  "filter": {
                    "term": {
                      "type": "t-shirt"
                    }
                  }
                },
                "t-shirt-percentage": {
                    "bucket_script": {
                        "buckets_path": {
                          "tShirtSales": "t-shirts>_count",
                          "totalSales": "total_sales"
                        },
                        "script": "params.tShirtSales / params.totalSales * 100"
                    }
                }
            }
        }
    }
}

在spring data 中使用 bucket_script 聚合

ValueCountAggregationBuilder valueCountAggregationBuilder=AggregationBuilders.count("聚合的名字:total_attendance").field("聚合的字段名");
FilterAggregationBuilder filterLate=AggregationBuilders
                .filter("聚合的名字:filter_count",QueryBuilders.termQuery("term的字段名","過濾的字段值"));
Map<String,String> scriptParams=new HashMap<>();
        lateScriptParams.put("total","total_attendance");
        lateScriptParams.put("count","filter_count>_count");
BucketScriptPipelineAggregationBuilder script= PipelineAggregatorBuilders
                .bucketScript("聚合的名字:result",scriptParams,new Script("params.count/params.total*100")).format("#.##");

然后就是把聚合放到查詢里面了

TermsAggregationBuilder usersResultAggs= AggregationBuilders.terms("users_term").field("userId.keyword")
        .subAggregation(result)
        .subAggregation(filter_count)
        .subAggregation(total);

取值

1.使用elasticsearchTemplate

elasticsearchTemplate.query(searchQuery, searchResponse -> {
            Map<String,Aggregation> aggregationMap=searchResponse.getAggregations().asMap();
            StringTerms stringTerms = (StringTerms)aggregationMap.get("取值的那個聚合的名字:users_term");

　　　　　　　　//取值
　　　　　　　return aggregationMap.get("取值的那個聚合的名字:result")).getValue()

　　 stringTerms.getBuckets().forEach(aggs->{
    Map<String,Aggregation> subAggMap=aggs.getAggregations().asMap();
　　　　//取百分比
　　　　Double result=((InternalSimpleValue)subAggMap.get("notsigendScript")).getValue();
 　　});

　　return 返回結果；
});