node.js - 如何使用聚合函数计算mongoDB中的唯一数据

Question

这个问题已经有答案了:

Mongo Aggregation : $group and $project array to object for counts (1 个回答)

已关闭 5 年前。

这是输入 JSON 数据，像这样存储在 mongoDB 中

输入

[{
  'name' : 'user1',
  'date' : '23/09/2017',
  'deviceType' : 'Android'
 },
{
 'name' : 'user2',
 'date' : '24/09/2017',
 'deviceType' :'ios'
},
{
'name' : 'user1',
'date' : '23/09/2017',
'deviceType' :'ios'
},
{
'name' : 'user2',
'date' : '23/09/2017',
'deviceType' :'ios'
},
{
'name' : 'user1',
'date' : '24/09/2017',
'deviceType' :'ios'
},
{
'name' : 'user2',
'date' : '25/09/2017',
'deviceType' :'Android'
}   
]

输出为

输出

   [{
    'date' : '23/09/2017',
    'iosCount': 2,
    'androidCount':1
   }, 
  {
   'date' : '24/09/2017',
   'iosCount': 2,
   'androidCount':0
  },
  {
   'date' : '25/09/2017',
  'iosCount': 0,
  'androidCount':1
 }
]

我使用聚合函数来执行任务。假设集合名称是 users 那么。

db.users.aggregate([{'$group':{
                               '_id':{'date':'$date','name':'$name'},
                              'count':{'$sum' :1}
                              }
                    }])

此聚合函数的输出未返回适当的结果。

Let us explain more I am intrested only date suppose 23/08/2017 how many users are there and which type of device they have used either ios or android not count of device. let's say example like

If date "23/09/2017" has input data like

[{'name' : 'user1','date' : '23/09/2017','deviceType' : 'Android'},

{'name' : 'user1','date' : '23/09/2017','deviceType' :'ios'},

{'name' : 'user1','date' : '23/09/2017','deviceType' : 'Android'},

{'name' : 'user2','date': '23/09/2017','deviceType':'ios'}]

output

[{'date' : 23/09/2017, totalIos : 2, totalAndroid: 1}]

Answer 1

$sum 要求您告诉它要求和的内容。这是您期望的输出吗:

db.users.aggregate([{'$group': {'_id': {'date':'$date','name':'$name'}, 'count': {'$sum': 1}}}])

{ "_id" : { "date" : "24/09/2017", "name" : "user1" }, "count" : 1 }
{ "_id" : { "date" : "23/09/2017", "name" : "user2" }, "count" : 1 }
{ "_id" : { "date" : "24/09/2017", "name" : "user2" }, "count" : 1 }
{ "_id" : { "date" : "25/09/2017", "name" : "user2" }, "count" : 1 }
{ "_id" : { "date" : "23/09/2017", "name" : "user1" }, "count" : 2 }