消息应用程序的 MongoDB 聚合查询

Question

我正在开发一个具有消息功能（聊天）的应用程序，我想获取我拥有或向我发送消息的所有用户，并按照最近通信的联系人到达的顺序对所有这些用户进行分组最佳。 并计算我从其他用户那里收到但我尚未阅读的所有消息。

Mongo 消息模型

const userChat = new Schema({
  senderUserId: {
    type: Schema.Types.ObjectId,
    ref: "User"
  },
  message: {
    type: String
  },
  room: {
    type: Array
  },
  files: {
    type: Array
  },
  read: {
    type: Boolean,
    default: false
  },
  receiverUserId: {
    type: Schema.Types.ObjectId,
    ref: "User"
  }
}, {
  timestamps: true
})

我的 MongoDB 聚合

const user = await UserChat.aggregate([{
  $match: {
    $and: [{
      $or: [{
          receiverUserId: req.user._id
        }, // User is in recepients
        {
          senderUserId: req.user._id
        }, // or the sender
      ],
    }, ],
  },
},
{
  $sort: {
    createdAt: -1,
  },
},
{
  // if the message is send by me then i group the "$receiverUserId"
  // if not that mean i am the receiver so i group the "$senderUserId"
  $group: {
    _id: {
      $cond: {
        if: {
          $eq: ["$senderUserId", req.user._id]
        },
        then: "$receiverUserId",
        else: "$senderUserId"
      }
    },
    count: {
      $sum: {
        $cond: ["$read", 0, 1]
      },
    },
    createdAt: {
      $first: "$createdAt",
    },
  },
},

$lookup: {
  from: "users",
  localField: "_id",
  foreignField: "_id",
  as: "user",
},
}, {
$sort: {
  createdAt: -1,
},
}, {
$project: {
  _id: 1,
  count: 1,
  createdAt: 1,
  user: 1,
},
}, ]);

我想获取的数据示例

比如说我们有两个用户互相写信。 用户 1 叫Papi Johnson ，另一个叫Prince Michel 。

米歇尔王子向帕皮约翰逊发送了两条信息。 由于Papi johnson 还没有阅读米歇尔王子的消息，所以当Papi johnson 请求他发送或接收的消息时，结果应该是：

{
_id: 5f8cf7e3c68dd730b5edb8a9, // the ID of Prince Michel
count: 2, // number of message unread send by Prince Michel to Papi johnson
createdAt: 2020-10-19T02:27:39.759Z, // the last message receive
user: [ [Object] ] // in then the user info of Prince Michel 
}

当米歇尔王子请求获取他发送或接收的消息时，结果应该是：

{
_id: 5f8cf8f5b76f3e3d5e6e249f,// the ID of Papi johnson
count: 0, // number of message unread send by Papi johnson to Prince Michel  
createdAt: 2020-10-19T02:27:39.759Z,
user: [ [Object] ]// in then the user info of Papi johnson
}

这就是我收到的结果：

照片中的例子

收到的回复例如 Papi 和 Prince

**问题是我收到的两个用户的未读消息数量相同。 我认为问题可能是“$ group”和“$ sum”查询不仅统计用户收到的未读消息，还统计用户发送或接收的所有未读消息。 自从我在这个级别上写博客已经有 10 多天了，我不知道该怎么做。 **

当 Papi johnson 请求他发送或接收的消息时，结果是：

{
_id: 5f8cf7e3c68dd730b5edb8a9, // the ID of Prince Michel
count: 2, 
// I think here the "$ group" and "$ sum" 
//query does not count only unread message received by the user,
// but rather all unread message that is sent or received by the user.
createdAt: 2020-10-19T02:27:39.759Z, // the last message receive
user: [ [Object] ] // in then the user info of Prince Michel 
}

当米歇尔王子请求获取他发送或接收的消息时，结果是：

{
_id: 5f8cf8f5b76f3e3d5e6e249f,// the ID of Papi johnson
count: 2, 
// I think here the "$ group" and "$ sum" 
//query does not count only unread message received by the user,
// but rather all unread message that is sent or received by the user.
createdAt: 2020-10-19T02:27:39.759Z,
user: [ [Object] ]// in then the user info of Papi johnson
}