Graph lookup for recursive data in MongoDB - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ query_result

intermediate

Find all subordinates of an employee using $graphLookup

Given a collection employees with documents containing _id, name, and manager_id, which query correctly finds all employees under the manager with _id: 1?

MongoDB

db.employees.aggregate([
  {
    $graphLookup: {
      from: "employees",
      startWith: "$_id",
      connectFromField: "_id",
      connectToField: "manager_id",
      as: "subordinates"
    }
  },
  {
    $match: { _id: 1 }
  }
])

A

db.employees.aggregate([
  {
    $graphLookup: {
      from: "employees",
      startWith: "manager_id",
      connectFromField: "_id",
      connectToField: "manager_id",
      as: "subordinates"
    }
  },
  {
    $match: { _id: 1 }
  }
])

B

db.employees.aggregate([
  {
    $graphLookup: {
      from: "employees",
      startWith: 1,
      connectFromField: "_id",
      connectToField: "manager_id",
      as: "subordinates"
    }
  }
])

C

db.employees.aggregate([
  {
    $match: { _id: 1 }
  },
  {
    $graphLookup: {
      from: "employees",
      startWith: "$manager_id",
      connectFromField: "_id",
      connectToField: "manager_id",
      as: "subordinates"
    }
  }
])

D

db.employees.aggregate([
  {
    $match: { _id: 1 }
  },
  {
    $graphLookup: {
      from: "employees",
      startWith: "$_id",
      connectFromField: "_id",
      connectToField: "manager_id",
      as: "subordinates"
    }
  }
])

💡 Hint

Remember to start the graph lookup from the manager's _id and filter before the lookup.

Option D first filters the document with _id:1, then uses $graphLookup starting from that _id to find all subordinates recursively. Other options either start from a literal 1 without filtering or use wrong fields for startWith.

📝 Syntax

intermediate

Identify the syntax error in $graphLookup usage

Which option contains a syntax error in the $graphLookup stage?

MongoDB

db.collection.aggregate([
  {
    $graphLookup: {
      from: "collection",
      startWith: "$parent_id",
      connectFromField: "_id",
      connectToField: "parent_id",
      as: "descendants"
    }
  }
])

A { $graphLookup: { from: "collection", startWith: "$parent_id", connectFromField: "_id", connectToField: "parent_id", as: "descendants" } }

B { $graphLookup: { from: "collection", startWith: parent_id, connectFromField: "_id", connectToField: "parent_id", as: "descendants" } }

C $graphLookup: { from: "collection", startWith: "$parent_id", connectFromField: "_id", connectToField: "parent_id", as: "descendants" } }

D } } "stnadnecsed" :sa ,"di_tnerap" :dleiFoTtcennoc ,"di_" :dleiFmorFtcennoc ,"di_tnerap$" :htiWtrats ,"noitcelloc" :morf { :pukooLhparg$ {

💡 Hint

Check for missing commas between fields in the object.

Option A is missing a comma between connectToField and as fields, causing a syntax error. Other options have correct syntax except B where startWith is missing quotes but is valid if parent_id is a variable (but in MongoDB aggregation it must be a string path).

❓ optimization

advanced

Optimize recursive lookup to limit depth

You want to find all descendants of a node but limit recursion to 3 levels deep. Which $graphLookup option achieves this?

A { $graphLookup: { from: "nodes", startWith: "$_id", connectFromField: "_id", connectToField: "parent_id", as: "descendants", maxDepth: "3" } }

B { $graphLookup: { from: "nodes", startWith: "$_id", connectFromField: "_id", connectToField: "parent_id", as: "descendants", depthField: 3 } }

C { $graphLookup: { from: "nodes", startWith: "$_id", connectFromField: "_id", connectToField: "parent_id", as: "descendants", maxDepth: 3 } }

D { $graphLookup: { from: "nodes", startWith: "$_id", connectFromField: "_id", connectToField: "parent_id", as: "descendants", maxDepth: 0 } }

💡 Hint

maxDepth expects a number indicating maximum recursion depth.

Option C correctly uses maxDepth: 3 to limit recursion to 3 levels. Option C uses depthField incorrectly (it names a field to store depth, not limit it). Option C uses a string instead of number for maxDepth, which is invalid. Option C sets maxDepth to 0, meaning no recursion.

🧠 Conceptual

advanced

Understanding $graphLookup's connectFromField and connectToField

In a $graphLookup stage, what is the role of connectFromField and connectToField?

A connectFromField is the field in the 'from' collection to match, connectToField is the field in the local collection to match against.

B connectFromField is the field in the local collection to match, connectToField is the field in the 'from' collection to match against.

C connectFromField is the field in the 'from' collection to match against, connectToField is the field in the 'from' collection to match.

D connectFromField and connectToField are both fields in the local collection used for matching.

💡 Hint

Think about which collection is being searched and which is the starting point.

connectFromField is the field in the 'from' collection whose values are matched against the values in connectToField of the local collection documents. This defines how documents connect recursively.

🔧 Debug

expert

Why does this $graphLookup return an empty array?

Given the collection categories with documents having _id and parent_id, the following query returns an empty children array for _id: 10. Why? Query:

db.categories.aggregate([
  { $match: { _id: 10 } },
  { $graphLookup: {
      from: "categories",
      startWith: "$_id",
      connectFromField: "parent_id",
      connectToField: "_id",
      as: "children"
    }
  }
])

A The 'as' field name 'children' is reserved and cannot be used.

B connectFromField and connectToField are reversed; connectFromField should be '_id' and connectToField 'parent_id'.

C The $match stage should come after $graphLookup, not before.

D startWith should be a literal value 10, not '$_id'.

💡 Hint

Check which fields represent parent and child in the recursive relation.

The query uses connectFromField: 'parent_id' and connectToField: '_id', which is backwards. To find children, connectFromField should be '_id' (child's id) and connectToField should be 'parent_id' (parent's id). This reversal causes no matches and an empty array.