Back to DocumentationMQL Documentation

$redact

The `$redact` stage in aggregation pipeline is used to filter fields of the documents in a collection dynamically based on access rights or other conditions. It processes each document and removes or retains fields based on the specified logic.

Syntax

{
  $redact: <expression>
}

Parameters

expressionobjectrequired

A valid MongoDB expression that evaluates to one of: $$DESCEND, $$PRUNE, or $$KEEP. These variables determine whether to keep, remove, or traverse deeper into the document.

Examples

Sample Data

{
  "_id": "0fcc0bf0-ed18-4ab8-b558-9848e18058f4",
  "name": "First Up Consultants | Beverage Shop",
  "promotionEvents": [
    {
      "discounts": [
        { "categoryName": "Whiskey", "discountPercentage": 7 }
      ]
    }
  ],
  "tag": ["#MembershipDeals"]
}

Redacting sensitive information

Filter the promotionEvents field for documents where the discountPercentage exceeds 15%.

This query uses $redact with conditional logic to prune documents with high discounts.

Query:

db.stores.aggregate([
  {
    $redact: {
      $cond: {
        if: { $gt: ["$promotionEvents.discounts.discountPercentage", 15] },
        then: "$$PRUNE",
        else: "$$DESCEND"
      }
    }
  }
])

Output:

[
  {
    "_id": "new-store-001",
    "name": "Adatum Corporation - Downtown Branch",
    "sales": { "totalSales": 5000 }
  }
]

Restricting access based on tags

Remove all documents that contain the tag #MembershipDeals.

This query prunes documents with specific tags using conditional logic.

Query:

db.stores.aggregate([
  {
    $redact: {
      $cond: {
        if: { $in: ["#MembershipDeals", "$tag"] },
        then: "$$PRUNE",
        else: "$$DESCEND"
      }
    }
  }
])

Output:

[
  {
    "_id": "gaming-store-mall-001",
    "name": "Trey Research | Gaming Paradise",
    "status": "active"
  }
]

Related