mongoose, match using upper model field in deep population - mongodb

Question

I'm a newbie in MongoDB and mongoose. and I'm trying to create a simple messenger using MongoDB.
Subscription model:
const SubscriptionSchema = new Schema({
user: {
type: Schema.Types.ObjectId,
ref: 'User',
index: { unique: true }
},
conversations: [
{
conversation: {
type: Schema.Types.ObjectId,
ref: 'Conversation'
},
lastSeenMessage: {
type: Schema.Types.ObjectId,
ref: 'Message'
}
}
]
});
Conversation model:
const ConversationSchema = new Schema(
...,
{ timestamps: true, toJSON: { virtuals: true } }
);
ConversationSchema.virtual('messages', {
ref: 'Message',
localField: '_id',
foreignField: 'conversation'
});
Message model:
const MessageSchema = new Schema(
{
content: {
type: String,
maxlength: [1440, 'Text too long.']
},
type: {
type: String
},
conversation: {
type: Schema.Types.ObjectId,
ref: 'Conversation'
},
user: {
type: Schema.Types.ObjectId,
ref: 'User'
},
}
)
I need to retrieve m latest conversations of user and n messages within each conversation when they are written after last seen message. So I used mongoose like this:
let subscription = await Subscription.findOne(
{ user }
//{ ObjectArray: { $slice: [(page - 1) * limit, (page - 1) * limit + limit] } }
).populate({
path: 'conversations.conversation',
model: 'Conversation',
populate: {
path: 'messages',
match: { _id: { $gt: 'conversations.lastSeenMessage' } },
options: {
skip: 0,
limit: 50,
sort: { updatedAt: -1 }
}
}
});
the match condition doesn't works in above query. how can use upper model in match condition in the deep population ?
Is it possible or I must try another way?