python deep copy and shallow copy

Python中对于对象的赋值都是引用，而不是拷贝对象（Assignment statements in Python do not copy objects, they create bindings between a target and an object.）。对于可变对象来说，当一个改变的时候，另外一个不用改变，因为是同时的，也就是说是保持同步的。

此时不想让他们同步的话可以使用copy模块，copy.copy()浅拷贝，copy.deepcopy()深拷贝。

前者是对整个对象的拷贝，对子对象仍然是引用，后者是对对象以及子对象的拷贝。具体一些可以下面例子：

In [26]: import copy                                                                                                                                                                       [6/165]

In [27]: a = [1, 2, [3, 4]]                                                                     

In [28]: b = a  # 引用

In [29]: c = copy.copy(a)　　# shallow copy, 对１，２，【３，４】拷贝，对【３，４】中的元素是引用

In [30]: d = copy.deepcopy(a)    # deep copy　对１，２，【３，４】都是拷贝，【３，４】中的元素也是拷贝

In [31]: b
Out[31]: [1, 2, [3, 4]]

In [32]: c
Out[32]: [1, 2, [3, 4]]

In [33]: d
Out[33]: [1, 2, [3, 4]]

In [34]: a.append(5)

In [35]: a
Out[35]: [1, 2, [3, 4], 5]

In [36]: b
Out[36]: [1, 2, [3, 4], 5]

In [37]: c
Out[37]: [1, 2, [3, 4]]

In [38]: d
Out[38]: [1, 2, [3, 4]]

In [39]: a[2].append(6)

In [40]: a
Out[40]: [1, 2, [3, 4, 6], 5]

In [41]: b
Out[41]: [1, 2, [3, 4, 6], 5]

In [42]: c
Out[42]: [1, 2, [3, 4, 6]]

In [43]: d
Out[43]: [1, 2, [3, 4]]

 

一些对象的浅拷贝可以通过其他方式实现，如字典可以使用 dict.copy()，列表使用listb = lista[:]，如下:

In [46]: a = {'name': 'wang',}

In [47]: b = dict.copy(a)

In [48]: b
Out[48]: {'name': 'wang'}

In [49]: a.update({'age': 13})

In [50]: a
Out[50]: {'age': 13, 'name': 'wang'}

In [51]: b
Out[51]: {'name': 'wang'}

In [52]: a = {'name': {'first_name': 'wang'}}                                                                                                                                                     

In [53]: b = dict.copy(a)

In [54]: b
Out[54]: {'name': {'first_name': 'wang'}}

In [55]: a['name'].update({'last_name': 'Emma'})                                                                                                                                                  

In [56]: a
Out[56]: {'name': {'first_name': 'wang', 'last_name': 'Emma'}}

In [57]: b
Out[57]: {'name': {'first_name': 'wang', 'last_name': 'Emma'}}

In [58]: lista = [1, 2]

In [59]: listb = lista[:]

In [60]: lista.append(3)

In [61]: lista
Out[61]: [1, 2, 3]

In [62]: listb
Out[62]: [1, 2]

In [63]: lista = [1, 2, [3, 4]]

In [64]: listb = lista[:]

In [65]: lista[2].append(5)

In [66]: listb
Out[66]: [1, 2, [3, 4, 5]]

 

注意：

深拷贝浅拷贝都不能拷贝文件，模块等类型（This module does not copy types like module, method, stack trace, stack frame, file, socket, window, array, or any similar types. It does “copy” functions and classes (shallow and deeply), by returning the original object unchanged;）

同时可以自定义这两个函数～　　

 

参考：

https://docs.python.org/2/library/copy.html

http://www.cnblogs.com/coderzh/archive/2008/05/17/1201506.html

http://blog.csdn.net/sharkw/article/details/1934090