mongodb进阶聚合查询各种写法

mongodb聚合sql写法

1、聚合查询配上分页加排序

bash 复制代码
DBQuery.shellBatchSize = 30000
db.getCollection('svOrderRecordMo').aggregate([
{$match:{"actionType":9}},
{$group:{_id:"$svOrderId",total:{"$sum":1}}},
{$sort:{"createTime":1}},
{$skip:0},
{$limit:100}
])

2、在1的聚合查询基础上进行match过滤(过滤出total>1的数据)

bash 复制代码
DBQuery.shellBatchSize = 30000
db.getCollection('svOrderRecordMo').aggregate([
{$match:{"actionType":9}},
{$group:{_id:"$svOrderId",total:{"$sum":1}}},
{$match:{"total":{"$gt":1}}},
{$sort:{"createTime":1}},
{$skip:0},
{$limit:3000}
])

3、聚合查询时间范围并把Date的日期格式转成字符串(yyyyMMdd)然后聚合查询

bash 复制代码
db.getCollection('svRegisterRecordMo').aggregate([
  {$match:{"$and":[{"createTime":{"$gte":ISODate("2023-01-01T00:00:00.000+08:00")}}
    ,{"createTime":{"$lte":ISODate("2023-07-02T00:00:00.000+08:00")}}]   }},

  {
    $project: {
      convertedDate: {
        $dateToString: {
          format: "%Y%m%d",
          date: "$createTime"
        }
      }
    }
  },

  {
    $group: {
      _id: "$convertedDate",
      count: { $sum: 1 }
    }
  }
])

4、排序,聚合,sum求和同时应用

bash 复制代码
db.getCollection('reportAgentMo').aggregate([
{$match:{"dataType":"1.1","reportType":1,"platformType":2,"reportDate":{"$in":["2023-01-25","2023-02-25","2023-03-25","2023-04-25","2023-05-25","2023-06-25","2023-07-25","2023-08-25"]}}},
{$group:{_id:"$agentId",total:{"$sum":"$issuePercent"}}},
{$sort:{"total":-1}},
{$limit:10}
])

5、聚合对两字段求和,并把两个求和的值相除然后排序

bash 复制代码
db.getCollection('reportAgentMo').aggregate([
{$match:{"dataType":"1.1","reportType":1,"platformType":2,
    "reportDate":{"$in":["2023-01-25","2023-02-25","2023-03-25","2023-04-25","2023-05-25","2023-06-25","2023-07-25","2023-08-25"]}}},
{$group:{_id:"$agentId",total1:{"$sum":"$issueNum"},total2:{"$sum":"$svNum"}}},
{$match:{total2:{"$gt":0}}},
{$project:{_id:1,aver:{"$divide":["$total1","$total2"]}}},
{$sort:{aver:-1}},
{$limit:11}
])

6、聚合时使用$first关键字(根据customTextAnswer.text聚合,并获取第一条数据)

bash 复制代码
db.getCollection('orderExtraInfoMo').aggregate([
{$match:{"topic":"家庭年收入"
    ,"customTextAnswer.number":{"$ne":null}}},
{$group:{_id:"$customTextAnswer.text",svOrderId:{"$first":"$svOrderId"},"number":{"$first":"$customTextAnswer.number"}}},
{$sort:{"svOrderId":-1}},
{$skip:0},
{$limit:10000}
])

7、对targetValue字段进行聚合统计,key是cId

bash 复制代码
db.getCollection("channelTargetMo").aggregate([
{$match:{"targetDate":{"$gte":"2022-08-03"},"targetDate":{"$lte":"2022-09-02"},"targetDateType":"day","targetType":"deal_sv","cIdType":"subId","cId":"62eb25c3b6d6033cb3fd9174"}},
{$group:{_id:"$cId",total:{"$sum":"$targetValue"}}}
]);

8、mongodb的两表关联查询

如下所示:左表是reportCustDealInfoMo,右表是reportCustDataMo,连接条件是reportCustDealInfoMo.custId=reportCustDataMo.cId。因为默认是1对多 情况,所以aa是一个数组,

$unwind 是把aa数组展开

bash 复制代码
db.reportCustDealInfoMo.aggregate([
{
$match:{"isDeleted":false,"isDeal":true}
},

  {
    $lookup: {
      from: "reportCustDataMo",
      localField: "custId",
      foreignField: "cId",
      as: "aa"
    }
  },
  {
    $unwind: "$aa" 
  },

    {
    $match: {
      "aa.cIdType": "custId",
      "aa.reportType": "year",
      "aa.reportDate":"2021"
    }
  },
      {$group:{_id:null,
        
        dealInternetAmount:{"$sum":"$aa.dealInternetAmount"},
dealHighCriterionAmount:{"$sum":"$aa.dealHighCriterionAmount"},
dealCarCriterionAmount:{"$sum":"$aa.dealCarCriterionAmount"},
dealGroupCriterionAmount:{"$sum":"$aa.dealGroupCriterionAmount"},
dealLifeCriterionAmount:{"$sum":"$aa.dealLifeCriterionAmount"},
underwriteCdCriterionAmount:{"$sum":"$aa.underwriteCdCriterionAmount"}
    }}
    ,
    {$project:{
        dealInternetAmount:1,
        dealHighCriterionAmount:1,
        dealCarCriterionAmount:1,
        dealGroupCriterionAmount:1,
        dealLifeCriterionAmount:1,
	underwriteCdCriterionAmount:1,
        totalStaAmount:{"$add":["$dealInternetAmount","$dealHighCriterionAmount","$dealCarCriterionAmount","$dealGroupCriterionAmount","$dealLifeCriterionAmount"]}}}
])
相关推荐
薛定谔的算法13 小时前
phoneGPT:构建专业领域的检索增强型智能问答系统
前端·数据库·后端
Databend14 小时前
Databend 亮相 RustChinaConf 2025,分享基于 Rust 构建商业化数仓平台的探索
数据库
得物技术15 小时前
破解gh-ost变更导致MySQL表膨胀之谜|得物技术
数据库·后端·mysql
似水流年流不尽思念19 小时前
MongoDB 有哪些索引?适用场景?
后端·mongodb
Raymond运维20 小时前
MariaDB源码编译安装(二)
运维·数据库·mariadb
沢田纲吉20 小时前
🗄️ MySQL 表操作全面指南
数据库·后端·mysql
RestCloud1 天前
SQL Server到Hive:批处理ETL性能提升30%的实战经验
数据库·api
RestCloud1 天前
为什么说零代码 ETL 是未来趋势?
数据库·api
ClouGence2 天前
CloudCanal + Paimon + SelectDB 从 0 到 1 构建实时湖仓
数据库
DemonAvenger2 天前
NoSQL与MySQL混合架构设计:从入门到实战的最佳实践
数据库·mysql·性能优化