In the domain of functional programming, our main goal is referential transparency, which we define in the following way:

An expression e is referentially transparent if we can replace any occurrences of e in any given program with the result of the evaluation of e without changing the behavior of the program.

Building on that, we need pure functions which are:

Only dependent on their input

Have no side effects

This means that our functions will be referentially transparent.

However, the above-mentioned libraries are built upon Future from Scala, which uses eager evaluation and breaks referential transparency.

Let’s look at an example.