Java中的深拷贝(深复制)和浅拷贝(浅复制)介绍

时间:2022-06-26 11:13:07

深拷贝(深复制)和浅拷贝(浅复制)是两个比较通用的概念,尤其在C++语言中,若不弄懂,则会在delete的时候出问题,但是我们在这幸好用的是Java。虽然java自动管理对象的回收,但对于深拷贝(深复制)和浅拷贝(浅复制),我们还是要给予足够的重视,因为有时这两个概念往往会给我们带来不小的困惑。

浅拷贝是指拷贝对象时仅仅拷贝对象本身(包括对象中的基本变量),而不拷贝对象包含的引用指向的对象。深拷贝不仅拷贝对象本身,而且拷贝对象包含的引用指向的所有对象。举例来说更加清楚:对象A1中包含对B1的引用,B1中包含对C1的引用。浅拷贝A1得到A2,A2 中依然包含对B1的引用,B1中依然包含对C1的引用。深拷贝则是对浅拷贝的递归,深拷贝A1得到A2,A2中包含对B2(B1的copy)的引用,B2 中包含对C2(C1的copy)的引用。

若不对clone()方法进行改写,则调用此方法得到的对象即为浅拷贝,下面我们着重谈一下深拷贝。

运行下面的程序,看一看浅拷贝:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
class Professor0 implements Cloneable {
  String name;
  int age;
 
  Professor0(String name, int age) {
    this.name = name;
    this.age = age;
  }
 
  public Object clone() throws CloneNotSupportedException {
    return super.clone();
  }
}
 
class Student0 implements Cloneable {
  String name;// 常量对象。
  int age;
  Professor0 p;// 学生1和学生2的引用值都是一样的。
 
  Student0(String name, int age, Professor0 p) {
    this.name = name;
    this.age = age;
    this.p = p;
  }
 
  public Object clone() {
    Student0 o = null;
    try {
      o = (Student0) super.clone();
    } catch (CloneNotSupportedException e) {
      System.out.println(e.toString());
    }
 
    return o;
  }
}
 
public class ShallowCopy {
  public static void main(String[] args) {
    Professor0 p = new Professor0("wangwu", 50);
    Student0 s1 = new Student0("zhangsan", 18, p);
    Student0 s2 = (Student0) s1.clone();
    s2.p.name = "lisi";
    s2.p.age = 30;
    s2.name = "z";
    s2.age = 45;
    System.out.println("学生s1的姓名:" + s1.name + "\n学生s1教授的姓名:" + s1.p.name + "," + "\n学生s1教授的年纪" + s1.p.age);// 学生1的教授
  }
}

s2变了,但s1也变了,证明s1的p和s2的p指向的是同一个对象。这在我们有的实际需求中,却不是这样,因而我们需要深拷贝:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
class Professor implements Cloneable {
  String name;
  int age;
 
  Professor(String name, int age) {
    this.name = name;
    this.age = age;
  }
 
  public Object clone() {
    Object o = null;
    try {
      o = super.clone();
    } catch (CloneNotSupportedException e) {
      System.out.println(e.toString());
    }
    return o;
  }
}
 
class Student implements Cloneable {
  String name;
  int age;
  Professor p;
 
  Student(String name, int age, Professor p) {
    this.name = name;
    this.age = age;
    this.p = p;
  }
 
  public Object clone() {
    Student o = null;
    try {
      o = (Student) super.clone();
    } catch (CloneNotSupportedException e) {
      System.out.println(e.toString());
    }
    o.p = (Professor) p.clone();
    return o;
  }
}
 
public class DeepCopy {
  public static void main(String args[]) {
    long t1 = System.currentTimeMillis();
    Professor p = new Professor("wangwu", 50);
    Student s1 = new Student("zhangsan", 18, p);
    Student s2 = (Student) s1.clone();
    s2.p.name = "lisi";
    s2.p.age = 30;
    System.out.println("name=" + s1.p.name + "," + "age=" + s1.p.age);// 学生1的教授不改变。
    long t2 = System.currentTimeMillis();
    System.out.println(t2-t1);
  }
}

当然我们还有一种深拷贝方法,就是将对象串行化:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
import java.io.*;
//Serialization is time-consuming
class Professor2 implements Serializable {
  /**
   *
   */
  private static final long serialVersionUID = 1L;
  String name;
  int age;
 
  Professor2(String name, int age) {
    this.name = name;
    this.age = age;
  }
}
 
class Student2 implements Serializable {
  /**
   *
   */
  private static final long serialVersionUID = 1L;
  String name;// 常量对象。
  int age;
  Professor2 p;// 学生1和学生2的引用值都是一样的。
 
  Student2(String name, int age, Professor2 p) {
    this.name = name;
    this.age = age;
    this.p = p;
  }
 
  public Object deepClone() throws IOException, OptionalDataException,
      ClassNotFoundException {
    // 将对象写到流里
    ByteArrayOutputStream bo = new ByteArrayOutputStream();
    ObjectOutputStream oo = new ObjectOutputStream(bo);
    oo.writeObject(this);
    // 从流里读出来
    ByteArrayInputStream bi = new ByteArrayInputStream(bo.toByteArray());
    ObjectInputStream oi = new ObjectInputStream(bi);
    return (oi.readObject());
  }
 
}
 
public class DeepCopy2 {
 
  /**
   * @param args
   */
  public static void main(String[] args) throws OptionalDataException,
      IOException, ClassNotFoundException {
    long t1 = System.currentTimeMillis();
    Professor2 p = new Professor2("wangwu", 50);
    Student2 s1 = new Student2("zhangsan", 18, p);
    Student2 s2 = (Student2) s1.deepClone();
    s2.p.name = "lisi";
    s2.p.age = 30;
    System.out.println("name=" + s1.p.name + "," + "age=" + s1.p.age); // 学生1的教授不改变。
    long t2 = System.currentTimeMillis();
    System.out.println(t2-t1);
  }
 
}

但是串行化却很耗时,在一些框架中,我们便可以感受到,它们往往将对象进行串行化后进行传递,耗时较多。