MongoDB 聚合查询 $unwind

在aggregate中，常常会遇到一些字段属性是数组对象，然后又需要对这些数组对象进行统计。
这时候就需要用到$unwind操作符。这是一个常用的，又容易被忽略的一个操作。

定义

• field 版

  { $unwind: <field path> }
  

• document版

  {
    $unwind:
      {
        path: <field path>,
        includeArrayIndex: <string>, preserveNullAndEmptyArrays: <boolean> } } 

1. 你要打散的字段
2. includeArrayIndex，分配一个存该数组索引的字段
3. preserveNullAndEmptyArrays，是否输出空内容。

1、插入数据

MongoDB Enterprise > db.mytestcol.insert({user_id:"A_id",bonus:[{ type:"a" ,amount:1000 },{ type:"b" ,amount:2000 },{ type:"b" ,amount:3000 }]}

MongoDB Enterprise > db.mytestcol.insert({user_id:"B_id",bonus:[{ type:"a" ,amount:1000 },{ type:"b" ,amount:2000 },{ type:"b" ,amount:3000 }]})

2、查看数据

1)普通查看

MongoDB Enterprise > db.mytestcol.find().pretty()
{
        "_id" : ObjectId("5e1d66b240741afd9cfdee9d"),
        "user_id" : "B_id",
        "bonus" : [
                {
                        "type" : "a",
                        "amount" : 1000
                },
                {
                        "type" : "b",
                        "amount" : 2000
                },
                {
                        "type" : "b",
                        "amount" : 3000
                }
        ]
}
{
        "_id" : ObjectId("5e1d66d540741afd9cfdee9e"),
        "user_id" : "A_id",
        "bonus" : [
                {
                        "type" : "a",
                        "amount" : 1000
                },
                {
                        "type" : "b",
                        "amount" : 2000
                },
                {
                        "type" : "b",
                        "amount" : 3000
                }
        ]
}

2）$unwind查看，将数组拆开

MongoDB Enterprise > db.mytestcol.aggregate([{"$unwind":"$bonus"}])
{ "_id" : ObjectId("5e1d66b240741afd9cfdee9d"), "user_id" : "B_id", "bonus" : { "type" : "a", "amount" : 1000 } }
{ "_id" : ObjectId("5e1d66b240741afd9cfdee9d"), "user_id" : "B_id", "bonus" : { "type" : "b", "amount" : 2000 } }
{ "_id" : ObjectId("5e1d66b240741afd9cfdee9d"), "user_id" : "B_id", "bonus" : { "type" : "b", "amount" : 3000 } }
{ "_id" : ObjectId("5e1d66d540741afd9cfdee9e"), "user_id" : "A_id", "bonus" : { "type" : "a", "amount" : 1000 } }
{ "_id" : ObjectId("5e1d66d540741afd9cfdee9e"), "user_id" : "A_id", "bonus" : { "type" : "b", "amount" : 2000 } }
{ "_id" : ObjectId("5e1d66d540741afd9cfdee9e"), "user_id" : "A_id", "bonus" : { "type" : "b", "amount" : 3000 } }

3.查询数据：查询不同user_id下，数据元素type为b的amount之和与交易笔数

MongoDB Enterprise > db.mytestcol.aggregate([{"$unwind":"$bonus"},{$match:{"bonus.type":"b"}},{$group:{"_id":"$user_id","amount":{$sum:"$bonus.amount"}}}])
{ "_id" : "B_id", "amount" : 5000 }
{ "_id" : "A_id", "amount" : 5000 }

MongoDB Enterprise > db.mytestcol.aggregate([{"$unwind":"$bonus"},{$match:{"bonus.type":"b"}},{$group:{"_id":"$user_id","amount":{$sum:1}}}])
{ "_id" : "B_id", "amount" : 2 }
{ "_id" : "A_id", "amount" : 2 }