Mongodb 如何使用聚合对文档进行计数

我有以下文件：

{ uid:2, tid: 100 },
{ uid:2, tid: 100 },
{ uid:2, tid: 103 }, 
{ uid:3, tid: 103 }, 
{ uid:3, tid: 100 }, 
{ uid:3, tid: 100 },
{ uid:8, tid: 103, total: 1 }

我想统计相同uid和tid的计数

{ uid: 2, tid: 100, total: 2},
{ uid: 2, tid: 103, total: 1},
{ uid: 3, tid: 100, total: 2},
{ uid: 3, tid: 103, total: 1 }

我试着用

db.collection.aggregate([
    { $match: { uid: { $in: [ "2", "3" ] } } },
    { $group: { _id: "$tid", total: { $sum: 1 } } },
    { $sort: { total: -1 } }
])

然后我得到的数据如下

{
    "result" : [ 
          {
              "_id" : "100",
              "uid" : "2",
              "total" : 2.0000000000000000
          }, 
          {
              "_id" : "103",
              "student" : "2",
              "total" : 1.0000000000000000
          }
      ],
      "ok" : 1.0000000000000000 }

---

某些数据未显示。

如果需要计算

总计

字段的总计，正确的合计应为：

db.collection.aggregate([
    { $match: { uid: { $in: ["2","3"] } } },
    { $group: { 
        _id: { tid: "$tid", uid: "$uid" }, 
        total: { $sum: "$total" } 
        } 
    },
    { $sort: { total: -1 } }
])

这将按

uid

筛选集合，然后计算按

tid

和

uid

分组的所有

total

字段的总和，并按

total

向后排序