Brief example of MongoDB graphLookup to work with tree structure such as e-commerce product categories

graphLookup.js

// Insert some data
db.product_categories.insertMany([
    {
        _id: 1,
        name: 'Products',
        parent_id: null
    },
    {
        _id: 2,
        name: 'Digital & Electronics',
        parent_id: 1
    },
    {
        _id: 3,
        name: 'Clothing',
        parent_id: 1
    },
    {
        _id: 4,
        name: 'Books',
        parent_id: 1
    },
    {
        _id: 5,
        name: 'Mobile Phone',
        parent_id: 2
    },
    {
        _id: 6,
        name: 'Mobile Phone Accessories',
        parent_id: 5
    },
    {
        _id: 7,
        name: 'Mobile Phone Pouch covers',
        parent_id: 6
    },
    {
        _id: 8,
        name: 'Mobile Phone Power banks',
        parent_id: 6
    }
]);

// Get specific node children
db.product_categories.aggregate([
    {
        $match: {
            _id: 1
        }
    },
    {
        $graphLookup: {
            from: "product_categories",
            startWith: "$_id",
            connectFromField: "_id",
            connectToField: "parent_id",
            as: "children"
        }
    },
    {
        $project: {
            'name': 1,
            'children._id': 1,
            'children.name': 1
        }
    }
])

// Get specific node parents
db.product_categories.aggregate([
    {
        $match: {
            _id: 6
        }
    },
    {
        $graphLookup: {
            from: "product_categories",
            startWith: "$parent_id",
            connectFromField: "parent_id",
            connectToField: "_id",
            as: "parents"
        }
    },
    {
        $project: {
            'name': 1,
            'parents._id': 1,
            'parents.name': 1
        }
    }
])

Thanks, help a lot. Easier to understand than official docs.

From what I can see this will not give the desired nested structure that most people are looking for in $graphlookup which is:

{
  "_id": {
    "$oid": "641841c3ae5e0d1f81c21df4"
  },
  "name": "backup-power",
  "slug": "backup-power",
  "label": "Back-up Power",
  "items": [
    {
      "_id": {
        "$oid": "6418bf4cae5e0d1f81c21dfc"
      },
      "name": "inverters",
      "slug": "backup-power/inverters",
      "label": "Inverters",
      "parent": "641841c3ae5e0d1f81c21df4",
      "items": [
         {
          "_id": {
            "$oid": "641b058aae5e0d1f81c21e0d"
          },
          "name": "hybrid-inverters",
          "slug": "backup-power/inverters/hybrid-inverters",
          "label": "Hybrid Inverters",
          "parent": "6418bf4cae5e0d1f81c21dfc"
        }
     ]
    },
  ]
}

It will only give one level deep:

{
  "_id": {
    "$oid": "641841c3ae5e0d1f81c21df4"
  },
  "name": "backup-power",
  "slug": "backup-power",
  "label": "Back-up Power",
  "items": [
    {
      "_id": {
        "$oid": "6418bf4cae5e0d1f81c21dfc"
      },
      "name": "inverters",
      "slug": "backup-power/inverters",
      "label": "Inverters",
      "parent": "641841c3ae5e0d1f81c21df4",
    },
  ]
}

To get the complete list of parent child, we can set parent id in $match to null.

// Get specific node children db.product_categories.aggregate([ { $match: { parent_id: null } }, { $graphLookup: { from: "product_categories", startWith: "$_id", connectFromField: "_id", connectToField: "parent_id", as: "children" } }, { $project: { 'name': 1, 'children._id': 1, 'children.name': 1 } } ])