Schema Validation

Schema Validation

By default MongoDB has a flexible schema - documents in a collection can have different fields and types. Sometimes you want to enforce structure, and that's what schema validation is for.

$jsonSchema

MongoDB uses the $jsonSchema operator to define validation rules. You pass it to db.createCollection() in the validator option:

db.createCollection("posts", {
  validator: {
    $jsonSchema: {
      bsonType: "object",
      required: [ "title", "body" ],
      properties: {
        title: { bsonType: "string", description: "Title of post - Required." },
        body: { bsonType: "string", description: "Body of post - Required." },
        category: { bsonType: "string", description: "Category of post - Optional." },
        likes: { bsonType: "int", description: "Post like count. Must be an integer - Optional." },
        tags: { bsonType: ["string"], description: "Must be an array of strings - Optional." },
        date: { bsonType: "date", description: "Must be a date - Optional." }
      }
    }
  }
})

The keywords

• bsonType: enforces the data type of a field (string, int, date, etc.).
• required: lists fields that MUST be present.
• properties: describes the rules for each field.
• description: human-readable notes about the field.

When validation runs

Validation runs on insert and update operations. By default, invalid documents are rejected (validationAction: "error").

TL;DR

• Flexible schema by default; validation makes it strict.
• $jsonSchema defines rules with bsonType, required, properties.
• Set up with db.createCollection(name, { validator: {...} }).