插入测试查询数据，注意这里已安装ik中文分词器

PUT mytest
{
  "mappings": {
    "properties": {
      "join_field": {
        "type": "join",
        "relations": {
          "parent": "child"
        }
      },
      "name": {
        "type": "keyword"
      },
      "age": {
        "type": "integer"
      },
      "city": {
        "type": "text",
        "analyzer": "ik_smart"
      },
      "hobbies":{
        "type": "keyword"
      },
      "desc": {
        "type": "text",
        "analyzer": "ik_smart"
      },
      "family":{
        "properties": {
            "father":{
              "type":"keyword"
            },
            "mather":{
              "type":"keyword"
            }
        }
      }
    }
  }
}

POST mytest/_bulk
{"index":{}}
{"name":"张三","age":"20","city":"广州","desc":"我是程序员，毕业于清华大学","hobbies":["唱歌","跳舞","rape"],"join_field": "parent"}
{"index":{}}
{"name":"李四","age":"27","city":"北京","desc":"我是宅男平时喜欢看动漫","hobbies":["动漫","cosplay","游戏"],"join_field": "parent"}
{"index":{}}
{"name":"王五","age":"31","city":"广州","desc":"我是快递员，每天都上门送快递","join": "parent","family":{"father":"王爸","mather":"王妈"}}
{"index":{}}
{"name":"小七","age":"40","city":"昆明","desc":"我是名医生，每天救死扶伤","join": "parent","family":{"father":"小爸","mather":"小妈"}}

// 子文档

POST /mytest/_doc/1?routing=1
{
  "join_field": {
    "name"  : "child",
    "parent": "BYELg4MBLEd9Buf4tGhT" // 父文档id 
  },
  "friends": "tom"
}

POST /mytest/_doc/2?routing=1
{
  "join_field": {
    "name"  : "child",
    "parent": "BoELg4MBLEd9Buf4tGhT"// 父文档id 
  },
  "friends": "king"
}

查询所有

GET /mytest/_doc/_search
{
  "query":{
    "match_all":{}
  }
}

一、单条件查询

1、模糊匹配

1.1 match

模糊查询，类似数据库的like ‘%三%’，查询出名字带有三的数据

GET /mytest/_doc/_search
{
  "query":{
    "match":{
      "name":"三"
    }
  }
}

1.2 prefix

前缀查询，查询前缀为王的数据

GET /mytest/_doc/_search
{
  "query":{
    "prefix":{
      "name":"王"
    }
  }
}

1.3 regexp

正则匹配，使用正则表达式去匹配，例子是匹配年龄小于二十岁的

GET /mytest/_doc/_search
{
  "query":{
    "regexp":{
      "age":"[0-2][0-9]"
    }
  }
}

 

2、精确匹配

2.1 term

精确匹配相当于=，需要与字段的值完全匹配，注意搜索的字段必须type=keyword，否则效果跟match一样

GET /mytest/_doc/_search
{
  "query":{
    "term":{
      "name":"张三"
    }
  }
}

2.2 terms

精确匹配多个值

GET /mytest/_doc/_search
{
  "query":{
    "terms":{
      "name":["张三","李四"]
    }
  }
}

2.3 range

字段属于某个范围内的值

GET /mytest/_doc/_search
{
  "query":{
    "range": { 
        "age": { 
            "gte":  20, 
            "lt":   30 
        } 
    }
  }
}

2.4 exists

某个字段的值是否存在，相当于mysql的is not null

 

GET /mytest/_doc/_search
{
  "query":{
    "exists": {
      "field":"hobbies"
    }
  }
}

二、多条件查询

1、bool 布尔查询

1.1 must

各个条件都必须满足，即各条件是and的 关系

GET /mytest/_doc/_search
{
  "query":{
    "bool": {
      "must": [
         {
           "match":{
              "city":"广州"
            }
         },
          { "range":{
              "age":{
                  "lt":"30"
              }
           }
          }
      ]
    }
  }
}

 

1.2 should

各个条件有一个满足即可，即各条件 是or的关系

GET /mytest/_doc/_search
{
  "query":{
    "bool": {
      "should": [
         {
           "match":{
              "city":"广州"
            }
         },
          { "range":{
              "age":{
                  "lt":"30"
              }
           }
          }
      ]
    }
  }
}

 

1.2 must_not

不满足所有条件，即各条件是not的 关系

GET /mytest/_doc/_search
{
  "query":{
    "bool": {
      "must_not": [
         {
           "match":{
              "city":"广州"
            }
         },
          { "range":{
              "age":{
                  "lt":"30"
              }
           }
          }
      ]
    }
  }
}

 

1.3 filter

不计算相关度评分，它不计算_score 即相关度评分，效率更高

GET /mytest/_doc/_search
{
  "query":{
    "bool": {
      "filter": [
         {
           "match":{
              "city":"广州"
            }
         },
          { "range":{
              "age":{
                  "lt":"30"
              }
           }
          }
      ]
    }
  }
}

2、constant_score

忽略TF/IDF打分，给搜索固定分数

GET /mytest/_doc/_search
{
  "query":{
    "bool": {
      "should": [
         {
           "constant_score": {
          "filter": { "match": { "desc": "wifi" }},
          "boost":   2 
        }
         },
          { "range":{
              "age":{
                  "lt":"30"
              }
           }
          }
      ]
    }
  }
}

3、取分数最高的，不考虑其他项的分数

取分数最高的，不考虑其他项的分数

GET /mytest/_doc/_search
{
  "query":{
    "dis_max": {
            "queries": [
                { "match": { "city": "广州" }},
                { "match": { "desc":  "我是" }}
            ]
      }
  }
}

三、链接查询

1、父子文档

查询子文档有某个字段

GET /mytest/_doc/_search
{
  "query": {
    "has_child": {
      "type" : "child",
      "query": {
        "match": {
          "friends": "tom"
        }
      }
    }
  }
}

2、嵌套文档

GET /mytest/_doc/_search
{
  "query": {
    "match": {
      "family.father" : "王"
    }
  }
}

四、聚合

1、查询平均年龄

GET /mytest/_doc/_search
{
  "query": {
    "match_all": {
    }
  },
  "aggs": {
    "my-agg-name": {
      "avg": {
        "field": "age"
      }
    }
  }
}

2、查询最小年龄

GET /mytest/_doc/_search
{
  "query": {
    "match_all": {
    }
  },
  "aggs": {
    "my-agg-name": {
      "min": {
        "field": "age"
      }
    }
  }
}

3、最大年龄

GET /mytest/_doc/_search
{
  "query": {
    "match_all": {
    }
  },
  "aggs": {
    "my-agg-name": {
      "max": {
        "field": "age"
      }
    }
  }
}

4、统计总年龄

GET /mytest/_doc/_search
{
  "query": {
    "match_all": {
    }
  },
  "aggs": {
    "my-agg-name": {
      "sum": {
        "field": "age"
      }
    }
  }
}