为什么classname $ 1.class在这种情况下生成？ [英] Why does classname$1.class generate in this situation?

问题描述

我写了以下代码来实现Singleton模式：

I wrote the following code to implement the Singleton pattern:

public final class Test {
     static final class TestHolder {
         private static final Test INSTANCE = new Test();
     }     

     private Test() {}

     public static Test getInstance() {
         return TestHolder.INSTANCE;
     }
}

当编译此文件时，应生成Test。类和Test $ TestHolder.class，但它也生成Test $ 1.class。这没有意义。

When I compile this file, it should generate Test.class and Test$TestHolder.class, but it also generates Test$1.class. This doesn't make sense. So why and how would this be?

推荐答案

类 TestHolder 需要调用 Test 中的私有构造函数。但它是私有的，并且实际上不能从另一个类调用。所以编译器扮演一个伎俩。 向 Test 添加了一个新的非私有构造函数，它只知道！构造函数接受此匿名类的一个（未使用）实例 Test $ 1 - 没有人知道存在。然后 TestHolder 创建 Test $ 1 的实例并调用构造函数，它的默认保护。）

Class TestHolder needs to call the private constructor in Test. But it's private, and can't actually be called from another class. So the compiler plays a trick. It adds a new non-private constructor to Test which only it knows about! That constructor takes an (unused) instance of this anonymous class Test$1 -- which nobody knows exists. Then TestHolder creates an instance of Test$1 and calls that constructor, which is accessible (it's default-protected.)

您可以使用 javap -c Test （和 javap -c Test\ $ 1 和 javap -c Test\ $ TestHolder ）来查看代码。这是相当聪明，实际上！

You can use javap -c Test (and javap -c Test\$1, and javap -c Test\$TestHolder) to see the code. It's quite clever, actually!

这篇关于为什么classname $ 1.class在这种情况下生成？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！