Functions as Objects

We have seen that Scala's numeric types and the Boolean type can be implemented like normal classes.

But what about functions?

In fact function values are treated as objects in Scala.

The function type A => B is just an abbreviation for the class scala.Function1[A, B], which is defined as follows.

package scala
trait Function1[A, B] {
  def apply(x: A): B
}

So functions are objects with apply methods.

There are also traits Function2, Function3, ... for functions which take more parameters (currently up to 22).

Expansion of Function Values

An anonymous function such as

(x: Int) => x * x

is expanded to:

{
  class AnonFun extends Function1[Int, Int] {
    def apply(x: Int) = x * x
  }
  new AnonFun
}

or, shorter, using anonymous class syntax:

new Function1[Int, Int] {
  def apply(x: Int) = x * x
}

Expansion of Function Calls

A function call, such as f(a, b), where f is a value of some class type, is expanded to:

f.apply(a, b)

So the OO-translation of:

val f = (x: Int) => x * x
f(7)

would be:

val f = new Function1[Int, Int] {
  def apply(x: Int) = x * x
}
f.apply(7)

Functions and Methods

Note that a method such as

def f(x: Int): Boolean = …

is not itself a function value.

But if f is used in a place where a Function type is expected, it is converted automatically to the function value

(x: Int) => f(x)

for expressions

You probably noticed that several data types of the standard library have methods named map, flatMap and filter.

These methods are so common in practice that Scala supports a dedicated syntax: for expressions.

map

Thus, instead of writing the following:

xs.map(x => x + 1)

You can write:

for (x <- xs) yield x + 1

You can read it as “for every value, that I name ‘x’, in ‘xs’, return ‘x + 1’”.

filter

Also, instead of writing the following:

xs.filter(x => x % 2 == 0)

You can write:

for (x <- xs if x % 2 == 0) yield x

The benefit of this syntax becomes more apparent when it is combined with the previous one:

for (x <- xs if x % 2 == 0) yield x + 1

// Equivalent to the following:
xs.filter(x => x % 2 == 0).map(x => x + 1)

flatMap

Finally, instead of writing the following:

xs.flatMap(x => ys.map(y => (x, y)))

You can write:

for (x <- xs; y <- ys) yield (x, y)

You can read it as “for every value ‘x’ in ‘xs’, and then for every value ‘y’ in ‘ys’, return ‘(x, y)’”.

Putting Things Together

Here is an example that puts everything together:

for {
  x <- xs if x % 2 == 0
  y <- ys
} yield (x, y)

The equivalent de-sugared code is the following:

xs.filter { x =>
  x % 2 == 0
}.flatMap { x =>
  ys.map { y =>
    (x, y)
  }
}

Method’s Parameters

Named Parameters

It can sometimes be difficult to figure out what is the meaning of each parameter passed to a function. Consider for instance the following expression:

Range(1, 10, 2)

What does it mean? We can improve the readability by using named parameters.

Based on the fact that the Range constructor is defined as follows:

case class Range(start: Int, end: Int, step: Int)

We can rewrite our expression as follows:

Range(start = 1, end = 10, step = 2)

It is now clearer that this expression defines a range of numbers from 1 to 10 by increments of 2.

Default Values

Methods’ parameters can have default values. Let’s refine the Range constructor:

case class Range(start: Int, end: Int, step: Int = 1)

Here, we say that the step parameter has a default value, 1.

Then, at use site we can omit to supply this parameter and the compiler will supply it for us, by using its default value:

Repeated Parameters

You can define a function that can receive an arbitrary number of parameters (of the same type) as follows:

This section introduces several syntactic sugars supported by the language.

String Interpolation

To splice values into constant String at runtime, you can use string interpolation:

After having prefixed the string literal with s you can introduce dynamic values in it with $.

If you want to splice a complex expression (more than just an identifier), surround it with braces:

In the above example, the type (Int, String) represents a pair whose first element is an Int and whose second element is a String.

Similarly, the value (i, s) is a pair whose first element is i and whose second element is s.

More generally, a type (T1, …, Tn) is a tuple type of n elements whose i^th element has type Ti.

And a value (t1, … tn) is a tuple value of n elements.

Manipulating Tuples

You can retrieve the elements of a tuple by using a tuple pattern:

Tuples

We saw earlier that case classes are useful to aggregate information. However, sometimes you want to aggregate information without having to define a complete case class for it. In such a case you can use tuples:

The average function takes at least one Int parameter and then an arbitrary number of other values and computes their average. By forcing users to supply at least one parameter, we make it impossible for them to compute the average of an empty list of numbers.

Sometimes you want to supply each element of a list as many parameters. You can do that by adding a : _* type ascription to your list:

val xs: List[Int] = …
average(1, xs: _*)

Type Aliases

In the same way as you can give meaningful names to expressions, you can give meaningful names to type expressions: