elastic search实践(2)

Elastic search查询

建模和准备数据

由于赛事机构的参加人员较多，且使用比较频繁，为了避免单一索引引起频繁操作，这里我们构建动态索引。索引的命名规则为：mock_data3_ + agencyId.

为了保证检索的效率，采用冗余的模型。在document中，建立nested的类型。（其原因如下：1. 对业务使用场景来说，对查询的效率要求较高；使用parent child的关联，其效率为单一文档的十分之一左右；2. 冗余的数据，在elastic search中保存在磁盘，都经过压缩，且磁盘空间代价较小；3. 对较大的文档，可能在建立index的耗时相对较长，这是业务可以接受的情形。）

相关的配置如下：

配置命令

PUT mock_data3_180103
{
  "mappings": {
    "doc": {
      "properties": {
        "registrations": {
          "type": "nested" 
        }
      }
    }
  }
}

建立index

PUT mock_data3_180103/doc/08C0B894-5CEE-4787-9DBB-00014256B7CE
{
  "agencyId": 180103,
  "addOnRevenue": 0.00,
  "openRate": 0,
  "emailRevenue": 0.00,
  "lastName":"Allabach",
  "address":{
    "city":"Centreville",
    "postalCode":"49032",
    "stateProvince":"MI",
    "line1":"27095 Marvin Rd"
   },
  "registrations": [
      {
        "eventName": "color run 2018",
        "priceCategoryName": "run 12k",
        "eventId": 12323,
        "priceCategoryId": 456,
        "id":123,
        "orderNumber": "C-0000119C",
        "registrationNumber": "R-00000DYB",
        "age": 18,
        "eventStartDate": "2009-07-20 20:07:52",
        "eventEndDate": "2018-07-21 20:07:52",
        "registrationDate": "2009-07-21 20:0:52",
        "activeUrl": "http://www.active.com",
        "location": "test location",
        "division": "division",
        "bib": "111111111111111111111",
        "wave": "tessdsdsds",
        "category": "registration category",
        "sport": "Running",
        "distance": "Marathon"
      },
      { 
        "eventName": "eventName",
        "priceCategoryName": "priceCategoryName",
        "eventId": 2382323,
        "priceCategoryId": 32322,
        "id":456,
        "orderNumber": "C-00002323C",
        "registrationNumber": "R-089DYB",
        "age": 20,
        "eventStartDate": "2010-07-20 20:07:52",
        "eventEndDate": "2018-07-21 20:07:52",
        "registrationDate": "2009-07-21 20:0:52",
        "activeUrl": "http://www.active.com",
        "location": "test location",
        "division": "division",
        "bib": "111111111111111111111",
        "wave": "tessdsdsds",
        "teamId": 123456,
        "teamName": "test team",
        "teamCaptainName": "test team captain",
        "category": "registration category",
        "sport": "Running",
        "distance": "5K"
      }
  ],  
  "gender":"FEMALE",
  "registrationRevenue": 29.99,
  "clickedCount":0,
  "clickedRate": 0,
  "dateOfBirth":"1975-09-11",
  "stars": 3,
  "sendCount":1,
  "engagementRate":"F",
  "firstName":"Shellie",
  "revenue": 29.99,
  "phoneNumber":"2697182152",
  "registrationCount":1,
  "openCount":0,
  "email":"MSSQLSERVERB128D6A6@null.activenetwork.com",
  "agencyName": "the color run"
}

Elastic search查询相关命令

简单的嵌套查询

GET mock_data3_180103/_search
{
  "query": {
    "nested": {
      "path": "registrations",
      "query": {
        "bool": {
          "must": [
            { "match": { "registrations.eventName": "color run 2018" }},
            { "match": { "registrations.eventId":  123232112 }} 
          ]
        }
      }
    }
  }
}

布尔过滤器编辑

一个 bool 过滤器由三部分组成：

{
   "bool" : {
      "must" :     [],
      "should" :   [],
      "must_not" : [],
   }
}

must
所有的语句都必须（must）匹配，与 AND 等价。

must_not
所有的语句都不能（must not）匹配，与 NOT 等价。

should
至少有一个语句要匹配，与 OR 等价。

范围

Elasticsearch 有 range 查询

gt: > 大于（greater than）

lt: < 小于（less than）

gte: >= 大于或等于（greater than or equal to）

lte: <= 小于或等于（less than or equal to）

缺失查询编辑

这个 missing 查询本质上与 exists 恰好相反：它返回某个特定无值字段的文档，与以下 SQL 表达的意思类似：

GET /my_store/products/_search
{
   "query" : {
      "filtered" : { 
         "filter" : {
            "bool" : {
              "should" : [
                 { "term" : {"price" : 20}}, 
                 { "term" : {"productID" : "XHDK-A-1293-#fJ3"}} 
              ],
              "must_not" : {
                 "term" : {"price" : 30} 
              }
           }
         }
      }
   }
}

案列

查询满足以下条件：

eventId= 2382323

stars between 3 to 30

first name like Shellie

GET mock_data3_180103/_search
{
  "query": {
    "bool": {
      "must": [
         {"nested": {
            "path": "registrations",
            "query": {
              "bool": {
                "must": [
                  { "match": { "registrations.eventId": 2382323 }}
                ]
              }
            }
           }
          },
          {"match": {
            "firstName": {
                  "query": "Shellie",
                  "type":  "phrase"
              }
            }
          },
          {"range": {
            "stars": {
              "gte": 3,
              "lte": 30
            }
          }
            
        }
      ] 
      
    }
  }
}