MongoDB Applied Design Patterns

Chapter 2. Polymorphic Schemas

MongoDB is sometimes referred to as a “schemaless” database, meaning that it does not enforce a particular structure on documents in a collection. It is perfectly legal (though of questionable utility) to store every object in your application in the same collection, regardless of its structure. In a well-designed application, however, it is more frequently the case that a collection will contain documents of identical, or closely related, structure. When all the documents in a collection are similarly, but not identically, structured, we call this a polymorphic schema.

In this chapter, we’ll explore the various reasons for using a polymorphic schema, the types of data models that they can enable, and the methods of such modeling. You’ll learn how to use polymorphic schemas to build powerful and flexible data models.

Polymorphic Schemas to Support Object-Oriented Programming

In the world of object-oriented (OO) programming, developers have become accustomed to the ability to have different classes of objects that share data and behavior through inheritance. In particular, OO languages allow functions to manipulate child classes as though they were their parent classes, calling methods that are defined in the parent but may have been overridden with different implementations in the child. This feature of OO languages is referred to as polymorphism.

Relational databases, with their focus on tables with a fixed schema, don’t support this feature all that ...