If you’re a Java programmer, then you must have heard about or experienced NullPointerExceptions in your programs.

NullPointerExceptions are Runtime Exceptions which are thrown by the jvm at runtime. Null checks in programs are often overlooked by developers causing serious bugs in code.

Java 8 introduced a new type called Optional<T> to help developers deal with null values properly.

The concept of Optional is not new and other programming languages have similar constructs. For example – Scala has Optional[T], Haskell has Maybe type.

In this blog post, I’ll explain about Java 8’s Optional type and show you how to use it by giving simple examples.

What is Optional?

Optional is a container type for a value which may be absent. Confused? Let me explain.

Consider the following function which takes a user id, fetches the user’s details with the given id from the database and returns it –

User findUserById(String userId) { ... };

If userId is not present in the database then the above function returns null. Now, let’s consider the following code written by a client –

User user = findUserById("667290");
System.out.println("User's Name = " + user.getName());

A common NullPointerException situation, right? The developer forgot to add the null check in his code. If userId is not present in the database, then the above code snippet will throw a NullPointerException.

Now, let’s understand how Optional will help you mitigate the risk of running into NullPointerException here –

Optional<User> findUserById(String userId) { ... };

By returning Optional<User> from the function, we have made it clear to the clients of this function that there might not be a User with the given userId. Now the clients of this function are explicitly forced to handle this fact.

The client code can now be written as –

Optional<User> optional = findUserById("667290");

optional.ifPresent(user -> {
    System.out.println("User's name = " + user.getName());    
})

Once you have an Optional object, you can use various utility methods to work with the Optional. The ifPresent() method in the above example calls the supplied lambda expression if the user is present, otherwise it does nothing.

Well! You get the idea here right? The client is now forced by the type system to write the Optional check in his code.

Creating an Optional object

1. Create an empty Optional

An empty Optional Object describes the absence of a value.

Optional<User> user = Optional.empty();

2. Create an Optional with a non-null value –

User user = new User("667290", "Rajeev Kumar Singh");
Optional<User> userOptional = Optional.of(user);

If the argument supplied to Optional.of() is null, then it will throw a NullPointerException immediately and the Optional object won’t be created.

3. Create an Optional with a value which may or may not be null –

Optional<User> userOptional = Optional.ofNullable(user);

If the argument passed to Optional.ofNullable() is non-null, then it returns an Optional containing the specified value, otherwise it returns an empty Optional.

Checking the presence of a value

1. isPresent()

isPresent() method returns true if the Optional contains a non-null value, otherwise it returns false.

if(optional.isPresent()) {
    // value is present inside Optional
    System.out.println("Value found - " + optional.get());
} else {
    // value is absent
    System.out.println("Optional is empty");
}	

2. ifPresent()

ifPresent() method allows you to pass a Consumer function that is executed if a value is present inside the Optional object.

It does nothing if the Optional is empty.

optional.ifPresent(value -> {
    System.out.println("Value found - " + value);
});

Note that I have supplied a lambda expression to the ifPresent() method. This makes the code more readable and concise.

Retrieving the value using get() method

Optional’s get() method returns a value if it is present, otherwise it throws NoSuchElementException.

User user = optional.get()

You should avoid using get() method on your Optionals without first checking whether a value is present or not, because it throws an exception if the value is absent.

Returning default value using orElse()

orElse() is great when you want to return a default value if the Optional is empty. Consider the following example –

// return "Unknown User" if user is null
User finalUser = (user != null) ? user : new User("0", "Unknown User");

Now, let’s see how we can write the above logic using Optional’s orElse() construct –

// return "Unknown User" if user is null
User finalUser = optionalUser.orElse(new User("0", "Unknown User"));

Returning default value using orElseGet()

Unlike orElse(), which returns a default value directly if the Optional is empty, orElseGet() allows you to pass a Supplier function which is invoked when the Optional is empty. The result of the Supplier function becomes the default value of the Optional –

User finalUser = optionalUser.orElseGet(() -> {
    return new User("0", "Unknown User");
});

Throw an exception on absence of a value

You can use orElseThrow() to throw an exception if Optional is empty. A typical scenario in which this might be useful is – returning a custom ResourceNotFound() exception from your REST API if the object with the specified request parameters does not exist.

@GetMapping("/users/{userId}")
public User getUser(@PathVariable("userId") String userId) {
    return userRepository.findByUserId(userId).orElseThrow(
	    () -> new ResourceNotFoundException("User not found with userId " + userId);
    );
}

Filtering values using filter() method

Let’s say you have an Optional object of User. You want to check its gender and call a function if it’s a MALE. Here is how you would do it using old school method –

if(user != null && user.getGender().equalsIgnoreCase("MALE")) {
    // call a function
}

Now, let’s use Optional along with filter to achieve the same –

userOptional.filter(user -> user.getGender().equalsIgnoreCase("MALE"))
.ifPresent(() -> {
    // Your function
})

The filter() method takes a predicate as an argument. If the Optional contains a non-null value and the value matches the given predicate, then filter() method returns an Optional with that value, otherwise it returns an empty Optional.

So, the function inside ifPresent() in the above example will be called if and only if the Optional contains a user and user is a MALE.

Extracting and transforming values using map()

Let’s say that you want to get the address of a user if it is present and print it if the user is from India.

Considering the following getAddress() method inside User class –

Address getAddress() {
    return this.address;
}

Here is how you would achieve the desired result –

if(user != null) {
    Address address = user.getAddress();
    if(address != null && address.getCountry().equalsIgnoreCase("India")) {
	    System.out.println("User belongs to India");
    }
}

Now, let’s see how we can get the same result using map() method –

userOptional.map(User::getAddress)
.filter(address -> address.getCountry().equalsIgnoreCase("India"))
.ifPresent(() -> {
    System.out.println("User belongs to India");
});

You see how concise and readable the above code is? Let’s break the above code snippet and understand it in detail –

// Extract User's address using map() method.
Optional<Address> addressOptional = userOptional.map(User::getAddress)

// filter address from India
Optional<Address> indianAddressOptional = addressOptional.filter(address -> address.getCountry().equalsIgnoreCase("India"));

// Print, if country is India
indianAddressOptional.ifPresent(() -> {
    System.out.println("User belongs to India");
});

In the above example, map() method returns an empty Optional in the following cases – 1. user is absent in userOptional. 2. user is present but getAdderess() returns null.

otherwise, it returns an Optional<Address> containing user’s address.

Cascading Optionals using flatMap()

Let’s consider the above map() example again. You might ask that if user’s address can be null then why the heck aren’t you returning an Optional<Address> instead of plain Address from getAddress() method?

And, You’re right! Let’s correct that, let’s now assume that getAddress() returns Optional<Address>. Do you think that above code will still work?

The answer is no! The problem is the following line –

Optional<Address> addressOptional = userOptional.map(User::getAddress)

Since getAddress() returns Optional<Address>, the return type of userOptional.map() will be Optional<Optional<Address>>

Optional<Optional<Address>> addressOptional = userOptional.map(User::getAddress)

Oops! We certainly don’t want that nested Optional. Let’s use flatMap() to correct that –

Optional<Address> addressOptional = userOptional.flatMap(User::getAddress)

Cool! So, Rule of thumb here – if the mapping function returns an Optional, use flatMap() instead of map() to get the flattened result from your Optional

Conclusion

Thank you for reading. If you Optional<Liked> this blog post. Give an Optional<High Five> in the comment section below.