Get rid of boilerplate code with Java 8

persons.stream()
    .filter(p -> p.getAge() > 20)
    .map(p -> fetchDetailsAsync(p))
    .map(pdf -> pdf.thenCompose(pd -> sendEmail(pd)))
    .map(emailF -> emailF.handle((email, e) -> complete(email, e)));

Matej Lazar [matejonnet@gmail.com](matejonnet@gmail.com)

???

Visit [http://matejonnet.github.io/get-rid-of-boilerplate-with-j8/] (http://matejonnet.github.io/get-rid-of-boilerplate-with-j8/) to see slides in presentation mode.

---

Agenda

- Lambda expressions

- Streams

- CompletableFuture

---

Lambda expressions

• Lambda is a function, function takes parameters and returns result

• Pre Java 8 functions could be implemented using methods

• A Lambda enables functions to be passed around

• Default method: method in interface (you don't break back compatibility)

• Method references "::" are used for lambda shortcuts

• Higher order function takes a function and creates new function

---

Anonymous inner vs Lambda expression

Anonymous Inner Class

Runnable r1 = new Runnable(){
    @Override
    public void run(){
        System.out.println("Hello world one!");
    }
};

Lambda expression

Runnable r2 = () -> System.out.println("Hello world two!");

---

@FunctionalInterface

• Is an interface where you have to implement single method (single abstract method)

• collections

  • Predicate
  • Comparator

• java.util.function

  • Supplier
  • Consumer
  • Function<T,R>

• https://docs.oracle.com/javase/8/docs/api/java/util/function/package-summary.html

---

Comparator with Lambda expression

Anonymous Inner Class

list.sort(new Comparator<String>() {
    @Override
    public int compare(String o1, String o2) {
        return o1.compareTo(o2);
    }
});

Lambda expression

list.sort((String o1, String o2) -> o1.compareTo(o2));
list.sort((o1, o2) -> o1.compareTo(o2));

---

Comparator with Higher order function

Higher order function

persons.sort((p1, p2) -> p1.getName().compareTo(p2.getName()));

//Higher order function
persons.sort(Comparator.comparing(p -> p.getName()));

//Method references
persons.sort(Comparator.comparing(Person::getName)); 

Using static import and multi level comparing

persons.sort(comparing(Person::getName));

persons.sort(comparing(Person::getAge).thenComparing(Person::getName));

---

New methods on List and Collection

Useful default methods added to list and collection interfaces

Removing element from list and avoiding ConcurrentModificationException

• no classical for loop -> iterator.remove

• collection.removeIf

  for (Person person : persons) { if (person.getName().startsWith("B")) { persons.remove(person); } }

  persons.removeIf(p -> p.getName().startsWith("B"));

Similar: list.replaceAll

---

Streams

• forEach
• stream
  • filter(predicate)
  • map
  • collect (collector / reducer)
    • java.util.stream.Collectors
    • SingletonCollector
    • string concatenating

---

ForEach

for (Person person : persons) {
    System.out.println(person.getName());
}

persons.forEach(p -> System.out.println(p.getName()));

//System.out.println(p)
persons.forEach(System.out::println);

---

Streams

Common stream usage

List<String> personNamesOver20 = persons.stream()
        .filter(p -> p.getAge() > 20)
        .map(Person::getName)
        .collect(Collectors.toList());

---

Streams :: map / reduce

Stream.reduce

int totalAge = persons.stream()
        .map(Person::getAge)
        .reduce(0, (a, b) -> a + b);

IntStream.sum

int totalAgeWithSum = persons.stream()
        .mapToInt(Person::getAge)
        .sum();

---

Streams :: filter

Remove elements form stream

    List<Person> personsOver20 = persons.stream()
            //.filter(p -> p.getAge() > 20)
            //.filter(filterAge20)
            .filter(filterAge(20))
            .collect(Collectors.toList());

---

Streams :: String concatenating

String personsString = persons.stream()
        .filter(p -> p.getAge() > 20)
        .sorted(Comparator.comparing(Person::getName))
        .map(Person::getName)
        .collect(Collectors.joining(", "));

With formatting

String personsString = persons.stream()
        .filter(p -> p.getAge() > 20)
        .sorted(Comparator.comparing(Person::getName))
        .map(p -> p.getName() + " is " + p.getAge() + " years old.")
        //.map(formatString)
        .collect(Collectors.joining("\n"));

---

Streams :: SingletonCollector

Reduces result to single instance

Person person = persons.stream()
    .filter(p -> p.getId() == 1)
    .collect(StreamCollectors.singletonCollector());

---

CompletableFuture

• Reactive programming
• Extending Future
• Exception handling

---

Reactive programming

• Reactive programming is programming with asynchronous data streams
• A stream is a sequence of ongoing events ordered in time
• We capture these emitted events asynchronously, by defining a function that will execute when a value is emitted
• Observer Design Pattern

---

Extending java.util.concurrent.Future

• Extends Future by adding async support
• .complete(T value)
• .completeExceptionally(Throwable ex)
• .get() -> inherited from Future blocking
• .then* -> async
• .thenCompose(Function) -> function is called on complete
• .handle(BiFunction(T result, Throwable e))

---

CompletableFuture :: simple

CompletableFuture<String> future = new CompletableFuture<>();
new Thread(() -> longRunningTask(future)).start();

//blocks until future.complete is called
future.get();

private void longRunningTask(CompletableFuture<String> future) {
    //long running operations
    future.complete("42");
}

Calling complete unblocks all clients waiting for get and also calls functions that are referenced by .then* methods.

---

CompletableFuture :: Supplier

final CompletableFuture<String> future = 
    CompletableFuture.supplyAsync(new Supplier<String>() {
        @Override
        public String get() {
            //...long running...
            return "42";
        }
    }, executor);

With lambda

final CompletableFuture<String> future = CompletableFuture.supplyAsync(() -> {
    //...long running...
    return "42";
}, executor);

---

CompletableFuture :: acting

CompletableFuture<String> future = CompletableFuture.supplyAsync(() -> {
        sleepQuietly();
        return "42";
    }, executor);

future.thenApply(Integer::parseInt)
      .thenApply(r -> r * r * Math.PI)
      .thenAccept(p -> System.out.println("Result :" + p.toString()));

Thread is not blocked when we use .then* methods.

CompletableFuture :: exception handling

• Only runtime exceptions can be thrown out of lambda

• All exceptions must be wrapped in RuntimeException

• Main thread completed we cannot catch the exception

• The most flexible approach is using .handle()

• handle() takes a function receiving either correct result or exception

  future.thenApply(Integer::parseInt) .thenApply(r -> r * r * Math.PI) .thenAccept((p) -> System.out.println("Result :" + p.toString())) .handle((result, ex) -> { if (result != null) storeResult(result); else handleException(ex); });

---

CompletableFuture :: combining

Combining (chaining) two futures (thenCompose())

public void combining(Person person) {
    fetchDetailsAsync(person)
        .thenCompose(pd -> sendEmail(pd))
        .handle((email, e) -> complete(email, e));
}

private CompletableFuture<PersonWithDetails> fetchDetailsAsync(Person person) {
    return CompletableFuture.supplyAsync(() -> {
        return new PersonWithDetails(person);
    }, executor);
}

---

CompletableFuture :: with stream

persons.parallelStream()
    .filter(p -> p.getAge() > 20)
    .map(p -> fetchDetailsAsync(p))
    .map(pdf -> pdf.thenCompose(pd -> sendEmail(pd)))
    .map(emailF -> emailF.handle((email, e) -> complete(email, e)));

---

Links

References

Oracle Lambda-QuickStart
Java SE 8 - lntroduction to Lambda Expressions
http://www.nurkiewicz.com/2013/05/java-8-definitive-guide-to.html
https://gist.github.com/Unisay/ff1aefdbd840c574395c

Presentation tool used

https://github.com/gnab/remark/

Matej Lazar (matejonnet@gmail.com, mlazar@redhat.com)