How to Read Kotlin Function Definitions

December 4th, 2017

In case you missed it, Kotlin is now the cool kid on the block.

It’s got all the advantages of Java and a great syntax with modern language features. Unfortunately, the documentation is a bit hard to parse if you’ve had limited experience with functional programming. Specifically, documentation for functions can look really weird.

So, let’s learn how to read function definitions/declarations! We’ll look at a few and break them down piece by piece.

The general pattern 1

Here’s how the documentation for functions normally looks. You can use this as reference as you continue the guide.

Optional elements of a function definition are in curly braces {}.


FUNCTION_TYPE defines what type of function is being declared. A non-exhaustive list of function types:

  • private
  • protected
  • public
  • open allows subclasses to override this function. 2
  • inline puts your function right into your Kotlin source rather than creating an object behind the scenes and managing references to the variables it’s closed over. You’d want to use this for performance reasons
  • infix makes it so that you can call functions without dots
  • tailrec converts tail-recursive (like normal recursion, but you go the other way) functions into regular loops, which is more efficient

fun is how you declare a function. It’s fun! Right?!

GENERIC_TYPES are things used in the rest of the function. They don’t have any meaning in themselves, just to say something is of one type or another but it doesn’t matter which type it is.

OBJECT is the object the function is defined on.

FUNCTION_NAME is self-explanatory.

PARAMETER_NAME also the above.

PARAMETER_TYPE is self explanatory, but parameters can get confusing fast if you’re passing functions around.

RETURN_TYPE is self-explanatory.

inline fun <T, R> Array<out T>.map(
    transform: (T) -> R
): List<R>

Here’s what we get from this definition, part by part:

  • it’s an inline function, so it doesn’t close over any variables
  • it uses two types, T and R, that will be Transformed and Returned
  • it returns an array with elements of type T3
  • it takes a function. That function takes a variable of type T and returns a variable of type R. This is the function that does the mapping from one element to another
  • it returns a list with elements of type R


inline fun <S, T : S> Array<out T>.reduce(
  operation: (acc: S, T) -> S
): S
  • it’s an inline function. No closure for you!
  • it uses two types, S and T, that represent the type of the accumulated value and the type of the elements the function is accumulating. T inherits from S.
  • it takes one parameter, a function named operation. The function takes a parameter named acc of type S, and another of type T and returns a value of type S
  • it returns a value of type S


fun <T : Comparable<T>> List<T?>.binarySearch(
    element: T?,
    fromIndex: Int = 0,
    toIndex: Int = size
): Int
  • it uses a type T that inherits from Comparable
  • it’s defined on a List of elements of type T that might be null (?)
  • it takes three parameters, the last two defaulting to 0 and List.size if not given
  • it returns an Int

Hope that helps!

Want to know something I didn’t cover? Was I incorrect? Do you understand compiler stuff and want to explain what out means?

Drop me a line — I’d love to hear from you!

  1. Read the documentation for functions for sure, just know it takes a bit of parsing to really understand it.

  2. See here. Presumably, this keyword is only allowed on functions in classes (it’s also allowed on classes themselves), but I couldn’t find any documentation to that effect.

  3. What’s out doing here? That’s a very good question! I’ve read the docs, and I get it, but I can’t explain it well so I’ll come back later and update this post when I do!

comments powered by Disqus