Python 中的浅拷贝和深拷贝

1. 列表和字典，直接赋值，都是浅拷贝，即赋值双方指向同一地址，因为 Python 对可变对象按引用传递。

>>> a = [1, 2, 3]
>>> b = a
>>> b
[1, 2, 3]
>>> a[0] = 'apple'
>>> a
['apple', 2, 3]
>>> b
['apple', 2, 3]
>>>
>>>
>>> dic = {'a': 'apple', 'b': 'banbana'}
>>> dic2 = dic
>>> dic2
{'a': 'apple', 'b': 'banbana'}
>>> dic['b'] = 'blueberry'
>>> dic
{'a': 'apple', 'b': 'blueberry'}
>>> dic2
{'a': 'apple', 'b': 'blueberry'}

2. 对非嵌套的列表，使用完全切片[:]， 工厂函数 list()，或者 copy.copy() 都是深拷贝。

>>> a = [1, 2, 3]
>>> b = list(a)
>>> c = a[:]
>>> import copy
>>> d = copy.copy(a)
>>>
>>> a[0] = 'apple'
>>> a
['apple', 2, 3]
>>> b
[1, 2, 3]
>>> c
[1, 2, 3]
>>> d
[1, 2, 3]

3. 对于非嵌套的字典，使用工厂函数 dict()，或者 copy.copy() 都是深拷贝。

>>> dic = {'a': 'apple', 'b': 'banana'}
>>> dic2 = dict(dic)
>>> import copy
>>> dic3 = copy.copy(dic)
>>>
>>> dic['b'] = 'blueberry'
>>> dic
{'a': 'apple', 'b': 'blueberry'}
>>> dic2
{'a': 'apple', 'b': 'banana'}
>>> dic3
{'a': 'apple', 'b': 'banana'}

4. 对于嵌套的列表，嵌套的字典，只能使用 copy.deepcopy() 实现深拷贝。

>>> import copy
>>>
>>> a = [1, 2, ['apple']]
>>> b = copy.deepcopy(a)
>>> a[2][0] = 'watermelon'
>>> a
[1, 2, ['watermelon']]
>>> b
[1, 2, ['apple']]
>>>
>>>
>>> dic = {'a': 'apple', 'b': {'b1': 'banana', 'b2': 'blueberry'}}
>>> dic2 = copy.deepcopy(dic)
>>> dic['b']['b1'] = 'bukeneng'
>>> dic
{'a': 'apple', 'b': {'b1': 'bukeneng', 'b2': 'blueberry'}}
>>> dic2
{'a': 'apple', 'b': {'b1': 'banana', 'b2': 'blueberry'}}

5. 对于嵌套可变对象的元组，只有 copy.deepcopy() 实现深拷贝。

>>> import copy
>>>
>>> t = (1, 2, ['apple'])
>>> t2 = t
>>> t3 = tuple(t)
>>> t4 = copy.copy(t)
>>> t5 = copy.deepcopy(t)
>>>
>>> t[2][0] = 'watermelon'
>>> t
(1, 2, ['watermelon'])
>>> t2
(1, 2, ['watermelon'])
>>> t3
(1, 2, ['watermelon'])
>>> t4
(1, 2, ['watermelon'])
>>> t5
(1, 2, ['apple'])

完。