Home > Java > javaTutorial > body text

An in-depth analysis of Java functional programming

WBOY
Release: 2022-11-10 16:17:22
forward
1296 people have browsed it

This article brings you relevant knowledge about java, which mainly introduces relevant content about functional programming. Java did not support functional programming at the beginning, but in java8 In order to support functional programming in this major version, Java has introduced many important features. Let’s take a look at them together. I hope it will be helpful to everyone.

An in-depth analysis of Java functional programming

Recommended study: "java Video Tutorial"

Java did not support functional programming at the beginning, it’s a good idea Understand, because Class is the first-class citizen in Java, which makes it not easy to implement programming in Java. However, although it is difficult, we already know the result. In the major version of Java 8 In order to support functional programming, Java has introduced many important features. In the previous articles, we have studied Lambda expressions and various stream operations in the Stream API. In today's article, we will sort out the built-in Java Provide us with a functional interface.

The outline of this article is as follows:

Java abstracts several built-in functional interfaces for developers to use based on use cases of common demand scenarios, such as Function, Supplier Wait, the parameters or return value types of various operation methods in Stream are often these built-in functional interfaces.

For example, the parameter type of the map operation method in Stream is Function

<R> Stream<R> map(Function<? super T, ? extends R> mapper);
Copy after login
Copy after login

Then why do we never see this type declared when we usually use the map method of Stream operation? What about the parameters? You can review the example of using the map method in our Stream API operation article, such as the following example of converting each element in the stream to uppercase through the map method.

List<String> list = new ArrayList<String>();
Stream<String> stream = list.stream();

Stream<String> streamMapped = stream.map((value) -> value.toUpperCase());
Copy after login

The parameter of the map method is directly a Lambada expression:

(value) -> value.toUpperCase()
Copy after login

This Lambda expression is the implementation of the Function interface.

The carrier of a functional interface is usually a Lambda expression. Through the Lambda expression, the compiler will infer which functional interface it implements based on the parameters and return value of the Lambda expression. Using Lambda expressions to implement interfaces, we don't have to specify the interface to be implemented by the class like anonymous inner classes. Therefore, although many parameters or return value types in Stream operations are Java's built-in functional interfaces, we do not display them. Implement them using anonymous classes.

Although Lambda expressions are very convenient to use, it also causes us to be a little confused when we see those built-in functional interface types in Java: "What is this thing? What is this thing?" What?" feeling.

Let’s first talk about functional programming, Java’s functional interfaces, and why Lambda can only implement functional interfaces. After clarifying these things, we will sort out what functional interfaces Java provides built-in .

Functional programming

Functional programming contains the following two key concepts:

  • Functions are first-class citizens
  • The function must satisfy the following constraints
    • The return value of the function only depends on the input parameters passed to the function.
    • The execution of the function has no side effects.

#Even if we don’t always follow all of these rules when writing programs, we can still benefit a lot from writing programs using functional programming ideas.

Next, let’s take a look at the implementation of these two key concepts in Java functional programming.

Functions are first-class citizens

In the functional programming paradigm, functions are first-class citizens in the language. This means that you can create "instances" of functions, and variable references to function instances, just like references to Strings, Maps, or any other objects. Functions can also be passed as arguments to other functions.

In Java, functions are obviously not first-class citizens, classes are. That’s why Java introduced Lambda expressions. This syntactic sugar allows Java to have functions from the presentation layer, allowing functions to be used as references to variables, parameters of methods, etc. Why do you say it is from the presentation layer? Because actually the Java compiler will still compile Lambda expressions into classes during compilation.

Pure function

In functional programming, there is the concept of pure function. If a function meets the following conditions, it is a pure function:

  • The execution of this function has no side effects.
  • The return value of a function depends only on the input parameters passed to the function.

The following is an example of a pure function (method) in Java

public class ObjectWithPureFunction{    public int sum(int a, int b) {        return a + b;
    }
}
Copy after login

The return value of the above sum() method only depends on its input parameters, And sum() has no side effects, it will not modify any state (variable) outside the function anywhere.

Instead, here is an example of a non-pure function:

public class ObjectWithNonPureFunction{    private int value = 0;    public int add(int nextValue) {        this.value += nextValue;        return this.value;
    }
}
Copy after login

add()方法使用成员变量value来计算其返回值,并且它还修改了value成员变量的状态,这代表它有副作用,这两个条件都导致add方法不是一个纯函数

正如我们看到的,函数式编程并不是解决所有问题的银弹。尤其是“函数是没有副作用的”这个原则就使得在一些场景下很难使用函数式编程,比如要写入数据库的场景,写入数据库就算是一个副作用。所以,我们需要做的是了解函数式编程擅长解决哪些问题,把它用在正确的地方。

函数式接口

Java中的函数式接口在 Lambda 表达式那篇文章里提到过,这里再详细说说。函数式接口是只有一个抽象方法的接口(抽象方法即未实现方法体的方法)。一个 Interface 接口中可以有多个方法,其中默认方法和静态方法都自带实现,但是只要接口中有且仅有一个方法没有被实现,那么这个接口就可以被看做是一个函数式接口

下面这个接口只定义了一个抽象方法,显然它是一个函数式接口:

public interface MyInterface {    public void run();
}
Copy after login

下面这个接口中,定义了多个方法,不过它也是一个函数式接口:

public interface MyInterface2 {
    public void run();

    public default void doIt() {
        System.out.println("doing it");
    }

    public static void doItStatically() {
        System.out.println("doing it statically");
    }
}
Copy after login

因为doIt方法在接口中定义了默认实现,静态方法也有实现,接口中只有一个抽象方法run没有提供实现,所以它满足函数式接口的要求。

这里要注意,如果接口中有多个方法没有被实现,那么接口将不再是函数式接口,因此也就没办法用 Java 的 Lambda 表达式实现接口了

编译器会根据 Lambda 表达式的参数和返回值类型推断出其实现的抽象方法,进而推断出其实现的接口,如果一个接口有多个抽象方法,显然是没办法用 Lambda 表达式实现该接口的。

@FunctionalInterface 注解

这里扩充一个标注接口是函数式接口的注解@FunctionalInterface

@FunctionalInterface 
// 标明接口为函数式接口
public interface MyInterface {    public void run(); 
//抽象方法}
Copy after login

一旦使用了该注解标注接口,Java 的编译器将会强制检查该接口是否满足函数式接口的要求:“确实有且仅有一个抽象方法”,否则将会报错。

需要注意的是,即使不使用该注解,只要一个接口满足函数式接口的要求,那它仍然是一个函数式接口,使用起来都一样。该注解只起到--标记接口指示编译器对其进行检查的作用。

Java 内置的函数式接口

Java 语言内置了一组为常见场景的用例设计的函数式接口,这样我们就不必每次用到Lambda 表达式、Stream 操作时先创建函数式接口了,Java 的接口本身也支持泛型类型,所以基本上 Java 内置的函数式接口就能满足我们平时编程的需求,我自己在开发项目时,印象里很少见过有人自定义函数式接口。

在接下来的部分中,我们详细介绍下 Java 内置为我们提供了的函数式接口。

Function

Function接口(全限定名:java.util.function.Function)是Java中最核心的函数式接口。 Function 接口表示一个接受单个参数并返回单个值的函数(方法)。以下是 Function 接口定义的:

@FunctionalInterface
public interface Function<T, R> {
    R apply(T t);

    default <V> Function<V, R> compose(Function<? super V, ? extends T> before) {
        Objects.requireNonNull(before);
        return (V v) -> apply(before.apply(v));
    }

    default <V> Function<T, V> andThen(Function<? super R, ? extends V> after) {
        Objects.requireNonNull(after);
        return (T t) -> after.apply(apply(t));
    }

    static <T> Function<T, T> identity() {
            return t -> t;
    }
Copy after login

Function接口本身只包含一个需要实现的抽象方法apply,其他几个方法都已在接口中提供了实现,这正好符合上面我们讲的函数式接口的定义:“有且仅有一个抽象方法的接口”。

Function 接口中的其他三个方法中compseandThen 这两个方法用于函数式编程的组合调用,identity用于返回调用实体对象本身,我们之前在把对象 List 转换为 Map 的内容中提到过,可以回看前面讲 List 的文章复习。

Function接口用Java 的类这么实现

public class AddThree implements Function<Long, Long> {

    @Override
    public Long apply(Long aLong) {
        return aLong + 3;
    }

    public static void main(String[] args) {
        Function<Long, Long> adder = new AddThree();
		Long result = adder.apply(4L);
		System.out.println("result = " + result);
    }
}
Copy after login

不过现实中没有这么用的,前面说过 Lambda 表达式是搭配函数式接口使用的,用Lambda表达式实现上Function 接口只需要一行,上面那个例子用 Lambda 实现的形式是:

Function<Long, Long> adder = (value) -> value + 3;Long resultLambda = adder.apply(8L);
System.out.println("resultLambda = " + resultLambda);
Copy after login

是不是简洁了很多。后面的接口示例统一用 Lambda 表达式举例,不再用类实现占用太多篇幅。

Function接口的常见应用是 Stream API 中的 map 操作方法,该方法的参数类型是Function接口,表示参数是一个“接收一个参数,并返回一个值的函数”。

<R> Stream<R> map(Function<? super T, ? extends R> mapper);
Copy after login
Copy after login

所以我们在代码里常会见到这样使用 map 操作:

stream.map((value) -> value.toUpperCase())
Copy after login

Predicate

Predicate 接口 (全限定名:java.util.function.Predicate)表示一个接收单个参数,并返回布尔值 true 或 false 的函数。以下是 Predicate 功能接口定义:

public interface Predicate<T> {    boolean test(T t);
}
Copy after login

Predicate 接口里还有几个提供了默认实现的方法,用于支持函数组合等功能,这里不再赘述。 用 Lambda 表达式实现 Predicate 接口的形式如下:

Predicate predicate = (value) -> value != null;
Copy after login

Stream API 中的 filter 过滤操作,接收的就是一个实现了 Predicate 接口的参数。

Stream<T> filter(Predicate<? super T> predicate);
Copy after login

写代码时,会经常见到这样编写的 filter 操作:

Stream<String> longStringsStream = stream.filter((value) -> {    
// 元素长度大于等于3,返回true,会被保留在 filter 产生的新流中。
    return value.length() >= 3;
});
Copy after login

Supplier

Supplier 接口(java.util.function.Supplier),表示提供某种值的函数。其定义如下:

@FunctionalInterfacepublic interface Supplier<T> {
    T get();
}
Copy after login

Supplier接口也可以被认为是工厂接口,它产生一个泛型结果。与 Function 不同的是,Supplier 不接受参数。

Supplier<Integer> supplier = () -> new Integer((int) (Math.random() * 1000D));
Copy after login

上面这个 Lambda 表达式的 Supplier 实现,用于返回一个新的 Integer 实例,其随机值介于 0 到 1000 之间。

Consume

Consumer 接口(java.util.function.Consume)表示一个函数,该函数接收一个参数,但是不返回任何值。

@FunctionalInterfacepublic interface Consumer<T> {    void accept(T t);
}
Copy after login

Consumer 接口常用于表示:要在一个输入参数上执行的操作,比如下面这个用Lambda 表达式实现的 Consumer,它将作为参数传递给它的value变量的值打印到System.out标准输出中。

Consumer<Integer> consumer = (value) -> System.out.println(value);
Copy after login

Stream API 中的 forEach、peek 操作方法的参数就是 Consumer 接口类型的。

Stream<T> peek(Consumer<? super T> action);
void forEach(Consumer<? super T> action);
Copy after login

比如,Stream API 中的 forEach 操作,会像下面这样使用 Consume 接口的实现

Stream<String> stream = stringList.stream();
// 下面是Lambda 的简写形式
// 完整形式为:value -> System.out.println(value);
stream.forEach(System.out::println);
Copy after login

Optional

最后再介绍一下 Optional 接口,Optional 接口并不是一个函数式接口,这里介绍它主要是因为它经常在一些 Stream 操作中出现,作为操作的返回值类型,所以趁着学习函数式编程的契机也学习一下它。

Optional 接口是预防NullPointerException的好工具,它是一个简单的容器,其值可以是 null 或非 null。比如一个可能返回一个非空结果的方法,方法在有些情况下返回值,有些情况不满足返回条件返回空值,这种情况下使用 Optional 接口作为返回类型,比直接无值时返回 Null 要更安全。 接下来我们看看 Optional 怎么使用:

// of 方法用于构建一个 Optional 容器
Optional<String> optional = Optional.of("bam");
// 判断值是否为空
optional.isPresent();           // true
// 取出值,如果不存在直接取会抛出异常
optional.get();                 // "bam"
// 取值,值为空时返回 orElse 提供的默认值
optional.orElse("fallback");    // "bam"
// 如果只存在,执行ifPresent参数中指定的方法
optional.ifPresent((s) -> System.out.println(s.charAt(0)));// "b"
Copy after login

Stream 操作中像 findAny、 findFirst这样的操作方法都会返回一个 Optional 容器,意味着结果 Stream 可能为空,因此没有返回任何元素。我们可以通过 Optional 的 isPresent() 方法检查是否找到了元素。Java 编程那些绕不开的接口 这个子系列的文章已经更新完毕,感兴趣的请持续关注,后面还有更多实用、精彩的内容。


推荐学习:《java视频教程

The above is the detailed content of An in-depth analysis of Java functional programming. For more information, please follow other related articles on the PHP Chinese website!

source:juejin.im
Statement of this Website
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Popular Tutorials
More>
Latest Downloads
More>
Web Effects
Website Source Code
Website Materials
Front End Template