1caseclass.md

Wrapper case class

Solution

Keeping it simple, wrap the value in a case class, for example:

case class EnumLike(value: String)

object EnumLike {
    final val One = EnumLike("One")
    final val Two = EnumLike("Two")
    final val Three = EnumLike("Three")
}

Source

Pros

• simple
• works as expected also with pattern matching

Cons

• wrapper is creating an overhead over the primitive value (impacting the performance)
• you can't actually separate logic for different cases, since there is only one main class

Source

• Enums in Spark Datasets