Mongo DB how to find orphans in a tree?-CodePudding

I have the trees collection with objects which have a tree like structure. I'm using child references pattern actually, but in my case the depth of each tree is undefined - it could be in range [0:100] approximately:

{
  "_id": "tree1",
  "tag_id": "1",
  "children": [
    {
      "_id": "tree2",
      "tag_id": "2",
      "children": [
        {
          "_id": "tree3",
          "tag_id": "3"
        }
      ]
    },
    {
      "_id": "tree4",
      "tag_id": "4"
    }
  ]
}

also I have the simple tags collection with objects like:

[
  {
    "_id": "1",
    "name": "one"
  },
  {
    "_id": "2",
    "name": "two"
  },
  {
    "_id": "3",
    "name": "three"
  },
  {
    "_id": "4",
    "name": "four"
  },
  {
    "_id": "5",
    "name": "five"
  }
]

As you can see, I have the orphan (5) which doesn't have a link inside trees collection:

{
    "_id": "5",
    "name": "five"
}

I want to find all such orphans and remove them from tags collection. I've tried to use aggregate method with $lookup:

db.tags.aggregate(
 [
  {$lookup:
    {
        from: "trees",
        localField: "_id",
        foreignField: "tag_id",
        as: "matched_docs"
    }
 }
 ]
);

But such approach joins and finds only "top" connections without looking to children array. How can I solve the task using mongo queries? Maybe, it's better to use different data-structure, pattern or other aggregation method?

CodePudding user response：

As per the documentation of latest version

Nested Depth for BSON Documents

MongoDB supports no more than 100 levels of nesting for BSON   
documents. Each object or array adds a level.

Mongodb is not fit for this schema. Hence, you cannot achieve this with huge dynamic levels of deeply nested arrays.

You can alter the schema or data store.