理解Java中的深层复制构造函数

Question

我有一个名为Bar的主类调用类Foo,我认为我正确地放入了一个深层构造函数

深拷贝构造函数的目的是将一个对象的内容复制到另一个对象,并且更改复制的对象不应该更改原始内容,对吗？

我的代码那样做,但我不明白为什么当我设置原始对象变量时,复制对象不包含该set变量,它只包含默认的构造函数变量.

public class Bar
{
    public static void main(String[] args)
    {
        Foo object = new Foo();
        object.setName1("qwertyuiop");



//the below line of code should copy object to object2?
        Foo object2 = new Foo(object);          
        System.out.println(object.getName1());

//shouldn't the below line of code should output qwertyuiop since object2 is a copy of object? Why is it outputting the default constructor value Hello World?
        System.out.println(object2.getName1()); 

//changes object2's name1 var to test if it changed object's var. it didn't, so my deep copy constructor is working

        object2.setName1("TROLL");
        System.out.println(object2.getName1()); 
        System.out.println(object.getName1());
    }


}

public class Foo
{

    //instance variable(s)
    private String name1;

    public Foo()
    {
        System.out.println("Default Constructor called");
        name1= "Hello World";

    }
    //deep copy constructor
    public Foo(Foo deepCopyObject)
    {   

        name1 = deepCopyObject.name1; 

    }
    public String getName1() {
    return name1;
}
public void setName1(String name1) {
    this.name1 = name1;
}
}

Answer 1

这不是深刻的副本.Java 不是 C++.您可以自由编写一个复制构造函数,该构造函数接受Foo实例并使用另一个Foo初始化它,但是没有语言支持来帮助您实现.这完全取决于你.

您还应该知道Java不像C++那样需要复制构造函数.Java对象存在于堆上.传递给方法的是对堆上对象的引用,而不是对象的副本.

您可以编写一个复制构造函数,但这取决于它的行为方式.你必须非常小心:

public class Foo {
    private Map<String, Bar> barDictionary;

    public Foo() {
        this.barDictionary = new HashMap<String, Bar>();
    }

    public Foo(Foo f) { 
        // What happens here is up to you.  I'd recommend making a deep copy in this case.
        this.barDictionary = new HashMap<String, Bar>(); 
        this.barDictionary.putAll(f.barDictionary);  // Question: What about the Bar references?  What happens to those?
    }
}