mongoDB group documents after unwind-CodePudding

In my mongodb books collection I have documents that look like:

    {
      _id: ObjectId(625efa44f1ba751c8275ea51),
      contributors:[ObjectId(625efa44f1ba751c8275ea52), ObjectId(625efa44f1ba751c8275ea53)]
      //other fields
    }

And I want to do a query that returns me documents like:

    {
      _id: ObjectId(625efa44f1ba751c8275ea51),
      contributors:[
                     {
                       _id: ObjectId(625efa44f1ba751c8275ea52),
                       first_name: 'Luigi'
                       //many other fields
                     },
                     {
                       _id: ObjectId(625efa44f1ba751c8275ea53),
                       first_name: 'Mario'
                       //many other fields
                     },
                   ]
      //other fields
    }

I did an unwind on contributors and a lookup with my users collection and now I need to group them. I haven't used it before but I did something like:

    {
        $group:{
                 _id: '_id'
               }
    }

But I don't know what to do next in order to preserve all the fields from books and also from users.

Do you have any idea?

CodePudding user response：

If the $lookup result is small enough (<16MB document size limit), you can simply do a $lookup.

db.books.aggregate([
  {
    "$lookup": {
      "from": "users",
      "localField": "contributors",
      "foreignField": "_id",
      "as": "contributors"
    }
  }
])

Here is the Mongo Playground for your reference.

If the $lookup result will exceed 16 MB limit, you can still $unwind. Just use $firstto regroup the other fields after the$unwind`

db.books.aggregate([
  {
    "$lookup": {
      "from": "users",
      "localField": "contributors",
      "foreignField": "_id",
      "as": "contributors"
    }
  },
  {
    "$unwind": "$contributors"
  },
  {
    "$group": {
      "_id": "_id",
      bookName: {
        $first: "$bookName"
      },
      "contributors": {
        "$push": "$contributors"
      }
    }
  }
])

Here is the Mongo playground for your reference.