# No Null Values

Functional programming is like writing a series of algebraic equations, and because you don’t use null values in algebra, you don’t use null values in FP. That brings up an interesting question: In the situations where you might normally use a null value in Java/OOP code, what do you do?

Scala’s solution is to use constructs like the Option/Some/None classes. We’ll provide an introduction to the techniques in this lesson.

## A first example

While this first Option/Some/None example doesn’t deal with null values, it’s a good way to demonstrate the Option/Some/None classes, so we’ll start with it.

Imagine that you want to write a method to make it easy to convert strings to integer values, and you want an elegant way to handle the exceptions that can be thrown when your method gets a string like "foo" instead of something that converts to a number, like "1". A first guess at such a function might look like this:

def toInt(s: String): Int = {
try {
Integer.parseInt(s.trim)
} catch {
case e: Exception => 0
}
}


The idea of this function is that if a string converts to an integer, you return the converted Int, but if the conversion fails you return 0. This might be okay for some purposes, but it’s not really accurate. For instance, the method might have received "0", but it may have also received "foo" or "bar" or an infinite number of other strings. This creates a real problem: How do you know when the method really received a "0", or when it received something else? The answer is that with this approach, there’s no way to know.

## Using Option/Some/None

Scala’s solution to this problem is to use a trio of classes known as OptionSome, and None. The Some and None classes are subclasses of Option, so the solution works like this:

• You declare that toInt returns an Option type
• If toInt receives a string it can convert to an Int, you wrap the Int inside of a Some
• If toInt receives a string it can’t convert, it returns a None

The implementation of the solution looks like this:

def toInt(s: String): Option[Int] = {
try {
Some(Integer.parseInt(s.trim))
} catch {
case e: Exception => None
}
}


This code can be read as, “When the given string converts to an integer, return the integer wrapped in a Some wrapper, such as Some(1). When the string can’t be converted to an integer, return a None value.”

Here are two REPL examples that demonstrate toInt in action:

scala> val a = toInt("1")
a: Option[Int] = Some(1)

scala> val a = toInt("foo")
a: Option[Int] = None


As shown, the string "1" converts to Some(1), and the string "foo" converts to None. This is the essence of the Option/Some/None approach. It’s used to handle exceptions (as in this example), and the same technique works for handling null values.

You’ll find this approach used throughout Scala library classes, and in third-party Scala libraries.

## Being a consumer of toInt

Now imagine that you’re a consumer of the toInt method. You know that the method returns a subclass of Option[Int], so the question becomes, how do you work with these return types?

• Use a match expression
• Use a for-expression

There are other approaches, but these are the two main approaches, especially from an FP standpoint.

### Using a match expression

One possibility is to use a match expression, which looks like this:

toInt(x) match {
case Some(i) => println(i)
case None => println("That didn't work.")
}


In this example, if x can be converted to an Int, the first case statement is executed; if x can’t be converted to an Int, the second case statement is executed.

### Using for/yield

Another common solution is to use a for-expression — i.e., the for/yield combination that was shown earlier in this book. To demonstrate this, imagine that you want to convert three strings to integer values, and then add them together. The for/yield solution looks like this:

val y = for {
a <- toInt(stringA)
b <- toInt(stringB)
c <- toInt(stringC)
} yield a + b + c


When that expression finishes running, y will be one of two things:

• If all three strings convert to integers, y will be a Some[Int], i.e., an integer wrapped inside a Some
• If any of the three strings can’t be converted to an inside, y will be a None

You can test this for yourself in the Scala REPL. First, paste these three string variables into the REPL:

val stringA = "1"
val stringB = "2"
val stringC = "3"


Next, paste the for-expression into the REPL. When you do that, you’ll see this result:

scala> val y = for {
|     a <- toInt(stringA)
|     b <- toInt(stringB)
|     c <- toInt(stringC)
| } yield a + b + c
y: Option[Int] = Some(6)


As shown, y is bound to the value Some(6).

To see the failure case, change any of those strings to something that won’t convert to an integer. When you do that, you’ll see that y is a None:

y: Option[Int] = None


## Options can be thought of as a container of 0 or 1 items

One good way to think about the Option classes is that they represent a container, more specifically a container that has either zero or one item inside:

• Some is a container with one item in it
• None is a container, but it has nothing in it

If you prefer to think of the Option classes as being like a box, None is a little like getting an empty box for a birthday gift.

## Using foreach

Because Some and None can be thought of containers, they can be further thought of as being like collections classes. As a result, they have all of the methods you’d expect from a collection class, including mapfilterforeach, etc.

This raises an interesting question: What will these two values print, if anything?

toInt("1").foreach(println)
toInt("x").foreach(println)


The answer is that the first example prints the number 1, and the second example doesn’t print anything. The first example prints 1 because:

• toInt(“1”) evaluates to Some(1)
• The expression evaluates to Some(1).foreach(println)
• The foreach method on the Some class knows how to reach inside the Some container and extract the value (1) that’s inside it, so it passes that value to println

Similarly, the second example prints nothing because:

• toInt("x") evaluates to None
• The foreach method on the None class knows that None doesn’t contain anything, so it does nothing

Again, None is just an empty container.

Somewhere in Scala’s history, someone noted that the first example (the Some) represents the “Happy Path” of Option/Some/None approach, and the second example (the None) represents the “Unhappy Path.” But, despite having two different possible outcomes, the cool thing about the approach is that the code you write to handle an Option looks exactly the same in both cases. The foreach examples look like this:

toInt("1").foreach(println)
toInt("x").foreach(println)


And the for-expression looks like this:

val y = for {
a <- toInt(stringA)
b <- toInt(stringB)
c <- toInt(stringC)
} yield a + b + c


You only have to write one piece of code to handle both the Happy and Unhappy Paths, and that simplifies your code. The only time you have to think about whether you got a Some or a None is when you finally handle the result value in a match expression, like this:

toInt(x) match {
case Some(i) => println(i)
case None => println("That didn't work.")
}


## Using Option to replace null values

Another place where a null value can silently creep into your code is with a class like this:

class Address (
var street1: String,
var street2: String,
var city: String,
var state: String,
var zip: String
)


While every address on Earth has a street1 value, the street2 value is optional. As a result, that class is subject to this type of abuse:

val santa = new Address(
"1 Main Street",
null,               // <-- D'oh! A null value!
"North Pole",
"99705"
)


To handle situations like this, developers tend to use null values or empty strings, both of which are hacks to work around the main problem: street2 is an optional field. In Scala — and other modern languages — the correct solution is to declare up front that street2 is optional:

class Address (
var street1: String,
var street2: Option[String],
var city: String,
var state: String,
var zip: String
)


With that definition, developers can write more accurate code like this:

val santa = new Address(
"1 Main Street",
None,
"North Pole",
"99705"
)


or this:

val santa = new Address(
"123 Main Street",
Some("Apt. 2B"),
"Talkeetna",
"99676"
)


Once you have an optional field like this, you work with it as shown in the previous examples: With match expressions, for expressions, and other built-in methods like foreach.

## Option isn’t the only solution

This lesson focused on the Option/Some/None solution, but Scala has a few other alternatives. For example, a trio of classes known as Try/Success/Failure work in the same manner, but a) you primarily use these classes when code can throw exceptions, and b) the Failure class gives you access to the exception message. For example, Try/Success/Failure is commonly used when writing methods that interact with files, databases, and internet services, as those functions can easily throw exceptions. These classes are demonstrated in the Functional Error Handling lesson that follows.

## Key points

This lesson was a little longer than the others, so here’s a quick review of the key points:

• Functional programmers don’t use null values
• A main replacement for null values is to use the Option/Some/None classes
• Common ways to work with Option values are match and for expressions
• Options can be thought of as containers of one item (Some) and no items (None)
• You can also use Options when defining constructor parameters
Page structure