Serialization of deep copy (deep clone) and shallow copy (shallow clone) of objects in Java

1. The concepts of shallow copy and deep copy

⑴Shallow copy (shallow clone)

All variables of the copied object contain the same values ​​as the original object, and all references to other objects still point to the original object. In other words, a shallow copy only copies the object in question, not the objects it refers to.

Example:

There are many common ways to clone List. Let’s list several common methods of shallow copying of List:

public static void main(String []args){

  List<Map<String,String>> list1 = new ArrayList<Map<String, String>>();

   Map<String,String> map = new HashMap<String, String>();

   map.put("name", "xiaoming");

   map.put("age", "28");

   list1.add(map);

   //克隆方法1：利用原list1作为参数直接构造方法生成。

   List<Map<String,String>> list2 = new ArrayList<Map<String, String>>(list1);

    //克隆方法2：手动遍历将原list1中的元素全部添加到复制表中。

   for(int i = 0, l = list1.size(); i < l; i++)

       list2.add(list1.get(i));

//克隆方法3：调用Collections的静态工具方法 Collections.copy


//克隆方法4：使用System.arraycopy方法进行复制


}

List itself It is an object. When storing class types, it is only responsible for storing addresses. When storing basic types, what is stored is the actual value. Even if you have thousands of Lists, there are still only a few elements. Whether it is reconstruction, the copy method of Collections, the copy method of System, or manual traversal, the result is the same. These methods only change the ArrayList object itself, simply adding a few addresses pointing to the old elements. No deep copying was done. (And there is no operation of new new object at all.) Sometimes we really need to copy these elements instead of just using the elements in the original memory. The List layer implements this problem. The Java language was taken into consideration from the beginning to avoid operating these data buried in the heap memory. All operations are directed to the address where they can be found. If the address loses itself, it will be killed by GC. So I have to traverse bit by bit, new creates a new object and assigns the original value. It is said that you may feel that the above approach has been slightly adjusted, so you can use serialized objects to make the data run around in the IO stream and copy it. In fact, the Java language is really compromised when it comes to serializing objects into streams. After all, you can't just throw the address in, right? Besides, the io flow needs to interact with other systems. You send someone an address and ask them to find it in which heap? So it goes without saying that the heap memory must be newly allocated.

⑵Serialization of Deep Copy (Deep Clone)

All variables of the copied object contain the same values ​​as the original object, except those variables that refer to other objects. Variables that reference other objects will point to the copied new objects, rather than the original referenced objects. In other words, deep copy copies all the objects referenced by the object to be copied.

Use serialization to do deep copying (deep cloning) in Java (to avoid rewriting the clone() method of deep copying of more complex objects, you can also program to implement breakpoint resumption and other functions)

The object The process of writing to the stream is a serialization process, but in Java it is also very vividly called a "freezing" or "pickling" process; and the parallelization of reading objects from the stream ( Deserialization) process is called "thawing" or "depicking" process. It should be pointed out that what is written in the stream is a copy of the object, and the original object still exists in the JVM. Therefore, what is "pickled into pickles" is only a copy of the object, and Java pickles can still be fresh. To deep copy an object in the Java language, you can often first make the object implement the Serializable interface, then write the object (actually just a copy of the object) into a stream (pickle it into a pickle), and then read it out from the stream (put the pickle back into the stream). fresh), the object can be reconstructed. The following is a deep copy of the source code.

public List<Map<String,String>> deClone(Object obj) throws IOException,OptionalDataException,ClassNotFoundException{
//将对象写到流里
ByteArrayOutoutStream bo=new ByteArrayOutputStream();
ObjectOutputStream oo=new ObjectOutputStream(bo);
oo.writeObject(obj);//从流里读出来
ByteArrayInputStream bi=new ByteArrayInputStream(bo.toByteArray());
ObjectInputStream oi=new ObjectInputStream(bi);
return(oi.readObject());
}

The premise for this is that the object and all objects referenced within the object are serializable. Otherwise, you need to carefully examine whether those non-serializable objects or attributes can be set to transient, thereby excluding them. outside the copying process.