Java基础(十三)--深拷贝和浅拷贝

时间:2021-07-23 19:49:36

在上篇文章:Java基础(十二)--clone()方法,我们简单介绍了clone()的使用

clone()对于基本数据类型的拷贝是完全没问题的,但是如果是引用数据类型呢?

@Data
@NoArgsConstructor
@AllArgsConstructor
@ToString
public class Student implements Cloneable{

    private int id;
    private String name;
    private int sex;
    private Score score;

    @Override
    protected Object clone() throws CloneNotSupportedException {
        return super.clone();
    }
} 
@Data
@NoArgsConstructor
@AllArgsConstructor
@ToString
public class Score {

    private int math;
    private int chinese;
}
public static void main(String[] args) throws Exception{
	Student student = new Student(1001, "sam", 1, new Score(78, 91));
	Student student1 = (Student)student.clone();
	System.out.println(student == student1);

	student1.getScore().setChinese(99);
	System.out.println(student.toString());
}

结果:

false
Student(id=1001, name=sam, sex=1, score=Score(math=78, chinese=99))

从结果上看,clone默认实现的是浅拷贝,并没有达到我们的预期。那么什么是深拷贝和浅拷贝?

深拷贝:在浅拷贝的基础上,引用变量也进行了clone,并指向clone产生的新对象

浅拷贝:被复制对象的所有值属性都含有与原来对象的相同,但是对象引用属性仍然指向原来的对象

clone()如何实现深拷贝?

1、引用成员变量Score需要实现Cloneable接口,并且重写Object的Clone()

2、自定义Student的Clone()

@Override
protected Object clone() throws CloneNotSupportedException {
	Student student = (Student) super.clone();
	Score score = student.getScore();
	student.setScore((Score)score.clone());
	return student;
}

结果:

false
Student(id=1001, name=sam, sex=1, score=Score(math=78, chinese=91))

结论:

  想要通过Clone()实现深拷贝,该对象必须要实现Cloneable接口,实现并且自定义clone(),把引用变量进行clone,然后引用变量对应的类

也要实现Cloneable接口并且实现clone方法。

  假如这个对象有很多个引用变量,都要实现clone接口,并且重写clone(),而且该对象自定义clone(),真的不是太方便,我们还可以序列化来

实现深拷贝

序列化实现深拷贝

可以写一个序列化实现深拷贝的工具类,兼容所有对象

public class DeepCloneUtils {

    public static <T extends Serializable> T deepClone(T object) {
        T cloneObject = null;
        try {
            ByteArrayOutputStream outputStream = new ByteArrayOutputStream();
            ObjectOutputStream objectOutputStream = new ObjectOutputStream(outputStream);
            objectOutputStream.writeObject(object);
            objectOutputStream.close();

            ByteArrayInputStream inputStream = new ByteArrayInputStream(outputStream.toByteArray());
            ObjectInputStream objectInputStream = new ObjectInputStream(inputStream);
            cloneObject = (T)objectInputStream.readObject();
            objectInputStream.close();

        }  catch (ClassNotFoundException e) {
            e.printStackTrace();
        } catch (IOException e) {
            e.printStackTrace();
        }
        return cloneObject;
    }
}

当前类和引用对象对应的类都要实现序列化

public static void main(String[] args) throws Exception{
	Student student = new Student(1001, "sam", 1, new Score(78, 91));
	Student student1 = (Student)DeepCloneUtils.deepClone(student);
	System.out.println(student == student1);

	student1.getScore().setChinese(99);
	System.out.println(student.toString());
}

结果:

false
Student(id=1001, name=sam, sex=1, score=Score(math=78, chinese=91))

PS:

  把当前对象写入到一个字节流中,再从字节流中将其读出来,这样就可以创建一个新的对象了,并且该新对象与源对象引用之间没有关联

  序列化实现方式,省略了clone()内部自定义的过程,但是还是要实现序列化的(当前类及引用类)

  现在有一个问题,如果这个引用对象是第三方jar包呢,我们如果让它实现Serializable和Cloneable接口,上述两种解决方案没法使用了

,我们需要新的解决方案

以下通过第三方jar包实现对象拷贝,不需要实现Serializable和Cloneable接口:

modelMapper、Spring中的BeanUtils、Commons-BeanUtils、cglib、orika等,那么哪些才是深拷贝?

modelMapper实现对象拷贝:

1、首先引用maven依赖

<dependency>
  <groupId>org.modelmapper</groupId>
  <artifactId>modelmapper</artifactId>
  <version>1.1.0</version>
</dependency>
public static void main(String[] args) throws Exception{
	Student student = new Student(1001, "sam", 1, new Score(78, 91));
	ModelMapper modelMapper = new ModelMapper();
	Student student1 = new Student();
	modelMapper.map(student, student1);
	System.out.println(student == student1);

	student1.getScore().setChinese(99);
	System.out.println(student.toString());
}

结果:证明ModelMapper实现的是浅拷贝

false
Student(id=1001, name=sam, sex=1, score=Score(math=78, chinese=99))

Spring中的BeanUtils实现对象拷贝:

public static void main(String[] args) throws Exception{
	Student student = new Student(1001, "sam", 1, new Score(78, 91));
	Student student1 = new Student();
	BeanUtils.copyProperties(student, student1);
	System.out.println(student == student1);

	student1.getScore().setChinese(99);
	System.out.println(student.toString());
}

结果:Spring-BeanUtils实现的是浅拷贝

false
Student(id=1001, name=sam, sex=1, score=Score(math=78, chinese=99))

Commons-BeanUtils实现对象拷贝:

1、首先引用maven依赖

<dependency>
  <groupId>commons-beanutils</groupId>
  <artifactId>commons-beanutils</artifactId>
  <version>1.8.0</version>
</dependency>
public static void main(String[] args) throws Exception{
	Student student = new Student(1001, "sam", 1, new Score(78, 91));
	Student student1 = new Student();
	BeanUtils.copyProperties(student1, student);
	System.out.println(student == student1);

	student1.getScore().setChinese(99);
	System.out.println(student.toString());
}

结果:证明ModelMapper实现的是浅拷贝

false
Student(id=1001, name=sam, sex=1, score=Score(math=78, chinese=99))

Cglib实现对象拷贝:

1、首先引用maven依赖

<dependency>
  <groupId>cglib</groupId>
  <artifactId>cglib-nodep</artifactId>
  <version>3.1</version>
</dependency>
public static void main(String[] args) throws Exception{
	Student student = new Student(1001, "sam", 1, new Score(78, 91));
	Student student1 = new Student();
	BeanCopier beanCopier = BeanCopier.create(Student.class, Student.class, false);
	beanCopier.copy(student, student1, null);
	System.out.println(student == student1);

	student1.getScore().setChinese(99);
	System.out.println(student.toString());
}

结果:Cglib实现的依然是浅拷贝,感觉很扎心啊。。。

false
Student(id=1001, name=sam, sex=1, score=Score(math=78, chinese=99))

PS:网上有说cglib实现自定义转换器可以实现深拷贝,但是我试验下来还是不能,各位可以试验一下,如果可以,请留言。。。

orika实现对象拷贝:

<dependency>
  <groupId>ma.glasnost.orika</groupId>
  <artifactId>orika-core</artifactId>
  <version>1.5.0</version>
</dependency>
</dependencies>
public static void main(String[] args) throws Exception{
	MapperFactory mapperFactory = new DefaultMapperFactory.Builder().build();
	mapperFactory.classMap(Student.class, Student.class)
			.byDefault()
			.register();
	ConverterFactory converterFactory = mapperFactory.getConverterFactory();
	MapperFacade mapper = mapperFactory.getMapperFacade();

	Student student = new Student(1001, "sam", 1, new Score(78, 91));
	Student student1 = mapper.map(student, Student.class);
	System.out.println(student == student1);

	student1.getScore().setChinese(99);
	System.out.println(student.toString());
}

结果:可以实现深拷贝

false
Student(id=1001, name=sam, sex=1, score=Score(math=78, chinese=91))

 

参考:

https://blog.csdn.net/54powerman/article/details/64920431?locationNum=6&fps=1

https://blog.csdn.net/weixin_40581980/article/details/81388557