Pagination of facet pipelines-CodePudding

I'm not very experienced with mongodb, so maybe someone can help me. I have a collection kind of like this:

 { user: 1, type: 0 , ... },
 { user: 1, type: 1, ... },
 { user: 1, type: 1, ... },
 ...

Now I want for a specific user documents grouped by type, then sorted offset & limit. ie. like:

[ { $match: { user: 1 } },
  { $facet: {
      type0: [
         { $match: { type: 0 },
         { $sort: ... },
         { $skip: offset0 },
         { $limit: limit0 }
      ],
      type1: [
         { $match: { type: 1 },
         { $sort: ... },
         { $skip: offset1 },
         { $limit: limit1 }
      ]
  }
]

This works, but: Is there a way to also get the total count of every type? $facet within $facet is not allowed, but maybe some completly different way? Or will a have to make an aggregate call for every single type in the end?

CodePudding user response：

Something like this:

  db.collection.aggregate([
  {
  $match: {
  user: 1
 }
 },
 {
 $facet: {
  type0: [
    {
      $match: {
        type: 0
      }
    },
    {
      $sort: {
        type: -1
      }
    },
    {
      $skip: 1
    },
    {
      $limit: 2
    }
  ],
  type1: [
    {
      $match: {
        type: 1
      }
    },
    {
      $sort: {
        type: -1
      }
    },
    {
      $skip: 1
    },
    {
      $limit: 2
    }
    ],
    counts: [
    {
      $group: {
        "_id": "$type",
        cnt: {
          $sum: 1
        }
      }
     },
     {
      $project: {
        type: "$_id",
        cnt: 1,
        _id: 0
      }
    }
   ]
   }
   }
   ])

Explained: You add one more stage named "counts" in the faced to group the total count of the two different types.

playground