5. 聚合搜索 - 《Elastic Search》

在聚合查询之前，需要设置文本field的fielddata为true

PUT /person/_mapping
{
    "properties": {
    "tags": {
      "type": "text",
      "fielddata": true
    }
  }
}

需求：查询每个tag下的人数

GET /person/_search
{
"aggs": {
  "group_by_tags": {
    "terms": {
      "field": "tags"
    }
  }
}
}

group_by_tags 是自定义的分组名称。
搜索结果： ```json { //省略了查询结果 “aggregations” : { “group_by_tags” : {

"doc_count_error_upper_bound" : 0,
"sum_other_doc_count" : 0,
"buckets" : [
  {
    "key" : "cat",
    "doc_count" : 2
  },
  {
    "key" : "cute",
    "doc_count" : 2
  },
  {
    "key" : "beatiful",
    "doc_count" : 1
  },
  {
    "key" : "boy",
    "doc_count" : 1
  }
]

} } }


- **需求**：查询名称中包含“man”的用户，然后根据tag分组
```json
GET /person/_search
{
  "query": {
    "match": {
      "name": "man"
    }
  },
  "aggs": {
    "group_by_name": {
      "terms": {
        "field": "tags",
        "size": 10
      }
    }
  }
}

搜索结果

{
  //省略了查询结果
  "aggregations" : {
    "group_by_name" : {
      "doc_count_error_upper_bound" : 0,
      "sum_other_doc_count" : 0,
      "buckets" : [
        {
          "key" : "beatiful",
          "doc_count" : 1
        },
        {
          "key" : "boy",
          "doc_count" : 1
        },
        {
          "key" : "cat",
          "doc_count" : 1
        },
        {
          "key" : "cute",
          "doc_count" : 1
        }
      ]
    }
  }
}

需求：先分组，然后再计算平均年龄

GET /person/_search
{
"aggs": {
  "group_by_tags": {
    "terms": {
      "field": "tags",
      "size": 10
    },
    "aggs": {
      "avg_age": {
        "avg": {
          "field": "age"
        }
      }
    }
  }
}
}

{
//省略了查询结果
"aggregations" : {
  "group_by_tags" : {
    "doc_count_error_upper_bound" : 0,
    "sum_other_doc_count" : 0,
    "buckets" : [
      {
        "key" : "cat",
        "doc_count" : 2,
        "avg_age" : {
          "value" : 15.0
        }
      },
      {
        "key" : "cute",
        "doc_count" : 2,
        "avg_age" : {
          "value" : 22.5
        }
      },
      {
        "key" : "beatiful",
        "doc_count" : 1,
        "avg_age" : {
          "value" : 25.0
        }
      },
      {
        "key" : "boy",
        "doc_count" : 1,
        "avg_age" : {
          "value" : 10.0
        }
      }
    ]
  }
}
}

需求：计算每个person的平均年龄，然后按照平均年龄降序排序

GET /person/_search
{
"aggs": {
  "group_by_tags": {
    "terms": {
      "field": "tags",
      "order": {
        "avg_key": "desc"
      }
    },
    "aggs": {
      "avg_key": {
        "avg": {
          "field": "age"
        }
      }
    }
  }
}
}

avg_key是自定义的一个值，需要跟聚合的值保持一致。

{
//省略了查询结果
"aggregations" : {
  "group_by_tags" : {
    "doc_count_error_upper_bound" : 0,
    "sum_other_doc_count" : 0,
    "buckets" : [
      {
        "key" : "beatiful",
        "doc_count" : 1,
        "avg_key" : {
          "value" : 25.0
        }
      },
      {
        "key" : "cute",
        "doc_count" : 2,
        "avg_key" : {
          "value" : 22.5
        }
      },
      {
        "key" : "cat",
        "doc_count" : 2,
        "avg_key" : {
          "value" : 15.0
        }
      },
      {
        "key" : "boy",
        "doc_count" : 1,
        "avg_key" : {
          "value" : 10.0
        }
      }
    ]
  }
}
}

需求：根据指定年龄范围分组，然后每组内再根据tag进行分组，最后计算没组的平均年龄

GET /person/_search
{
"aggs": {
  "range_age": {
    "range": {
      "field": "age",
      "ranges": [
        {
          "from": 10,
          "to": 20
        },
        {
          "from": 20,
          "to": 30
        },
        {
          "from": 30,
          "to": 40
        }
      ]
    },
    "aggs": {
      "group_by_tags": {
        "terms": {
          "field": "tags",
          "size": 10
        },
        "aggs": {
          "avg_age": {
            "avg": {
              "field": "age"
            }
          }
        }
      }
    }
  }
} 
}

{
//省略了查询结果
"aggregations" : {
  "range_age" : {
    "buckets" : [
      {
        "key" : "10.0-20.0",
        "from" : 10.0,
        "to" : 20.0,
        "doc_count" : 1,
        "group_by_tags" : {
          "doc_count_error_upper_bound" : 0,
          "sum_other_doc_count" : 0,
          "buckets" : [
            {
              "key" : "boy",
              "doc_count" : 1,
              "avg_age" : {
                "value" : 10.0
              }
            },
            {
              "key" : "cat",
              "doc_count" : 1,
              "avg_age" : {
                "value" : 10.0
              }
            }
          ]
        }
      },
      {
        "key" : "20.0-30.0",
        "from" : 20.0,
        "to" : 30.0,
        "doc_count" : 2,
        "group_by_tags" : {
          "doc_count_error_upper_bound" : 0,
          "sum_other_doc_count" : 0,
          "buckets" : [
            {
              "key" : "cute",
              "doc_count" : 2,
              "avg_age" : {
                "value" : 22.5
              }
            },
            {
              "key" : "beatiful",
              "doc_count" : 1,
              "avg_age" : {
                "value" : 25.0
              }
            },
            {
              "key" : "cat",
              "doc_count" : 1,
              "avg_age" : {
                "value" : 20.0
              }
            }
          ]
        }
      },
      {
        "key" : "30.0-40.0",
        "from" : 30.0,
        "to" : 40.0,
        "doc_count" : 0,
        "group_by_tags" : {
          "doc_count_error_upper_bound" : 0,
          "sum_other_doc_count" : 0,
          "buckets" : [ ]
        }
      }
    ]
  }
}
}