ElasticSearch 基本使用

ElasticSearch 基本使用

上文说到 ElasticSearch的快速入门 ，本文将介绍ElasticSearch的基本使用

两种查询方式

https://www.elastic.co/guide/en/elasticsearch/reference/6.0/_the_search_api.html

REST 请求 + 搜索参数

GET /bank/_search?q=*&sort=account_number:asc

REST请求 + 请求体

DSL (Domain Specific Language)

GET /bank/_search
{

  "query": { "match_all": {} },

  "sort": [
    { "age" : "desc"},
    { "account_number": "asc"}
  ]

}

部分信息

QueryDSL基本语法

JSON风格得DSL语法结构

{
  QUERY_NAME:{   //功能名
     FIELD_NAME:{  //字段名
       ARGUMENT:VALUE, //参数-值
       ARGUMENT:VALUE,...
      }   
   }
}

match全文检索

match匹配

GET /bank/_search
{
  "query" : {
    "match": {
      "age": "22"
    }
  }
}

全文检索

分词全文检索,按照评分降序

GET /bank/_search
{
  "query" : {
    "match": {
      "address": "mill lane"
    }
  }
}

match_phrase短语匹配

使用match时会自动分词,如果不想分词,想查询一个完整的短语就可以使用短语匹配

GET /bank/_search
{
  "query": {
     "match_phrase": {
       "address": "mill lane"
     }
  }
}

multi_match多字段匹配

GET /bank/_search
{
  "query": {
    "multi_match": {
      "query": "mill Movico",
      "fields": ["address","city"]
    }
  }
}

注意: 查询内容会分词

bool复合查询

复合查询 可以理解为 与 , 多个查询条件都要一起满足

must,must_not,should

must: 必须满足

must_not : 必须不满足

should: 满不满足都可以,满足评分会更高

filer过滤查询

filer与must,must_not,should的不同: 是否满足都不会增加评分

GET /bank/_search
{
  "query": {
    "bool": {
      "must": [
        {"match": {
          "address": "mill"
        }}
      ],
      "filter": {
        "range": {
          "age": {
            "gte": 10,
            "lte": 40
          }
        }
      }
    }
  }
}

term查询

term查询与match查询类似

一般用term检索非文本的精确值

match检索文本

keyword匹配精确值

使用keyword后,文本精确值必须是匹配值才算匹配成功

GET /bank/_search
{
  "query": {
    "match": {
      "address.keyword": "198 Mill"
    }
  }
}

与match_phrase短语匹配区别

GET /bank/_search
{
  "query": {
    "match_phrase": {
      "address": "198 Mill"
    }
  }
}

聚合

语法格式

"aggregations" : {
    "<aggregation_name>" : {
        "<aggregation_type>" : {
            <aggregation_body>
        }
        [,"meta" : {  [<meta_data_body>] } ]?
        [,"aggregations" : { [<sub_aggregation>]+ } ]?
    }
    [,"<aggregation_name_2>" : { ... } ]*
}

• trem:对值进行分布统计

  GET /bank/_search
  {
    "query": {
      "match_all": {}
    },
    "aggs": {
      "ageaggs": {
        "terms": {
          "field": "age",
          "size": 10
        }
      }
    }
  }
  

• agg:对值进行平均统计

  GET /bank/_search
  {
    "query": {
      "match_all": {}
    },
    "aggs": {
      "ageaggs": {
        "terms": {
          "field": "age",
          "size": 10
        }
      },
      "balanceAvg":{
        "avg": {
          "field": "balance"
        }
      }
    }
  }
  

  

嵌套聚合

如果我想在上一个聚合出的结果上进行聚合,就要使用嵌套聚合

• 对年龄聚合,统计各年龄人数,并请求这些年龄段人的平均薪资

  GET /bank/_search
  {
    "query": {
      "match_all": {}
    },
    "aggs": {
      "ageTrems": {
        "terms": {
          "field": "age",
          "size": 10
        },
        "aggs": {
          "balanceAvg": {
            "avg": {
              "field": "balance"
            }
          }
        }
      }
    }
  }
  

  

• 对年龄聚合,统计各年龄人数,并请求这些年龄段中M的平均薪资和F的平均薪资,以及该年龄段总体平均薪资

  GET bank/_search
  {
    "query": {
      "match_all": {}
    },
    "aggs": {
      "ageTerms": {
        "terms": {
          "field": "age",
          "size": 10
        },
        "aggs": {
          "genderTrems": {
            "terms": {
              "field": "gender.keyword"
            },
            "aggs": {
              "balanceAvg": {
                "avg": {
                  "field": "balance"
                }
              }
            }
          },
          "balanceAgeAvg": {
            "avg": {
              "field": "balance"
            }
          }
        }
      }
    }
  }