Assignment and copy in Python

Assignment

In python, a value assignment is to create an object reference, instead of storing the object as another copy. For example:

>>> a=[1,2,3]>>> b=a>>> c=a

The object is [1, 2, 3], which is referenced by three variables a, B, and c respectively. The three variables do not exclusively occupy the object [1, 2, 3], or you can use any variable to modify the object [1, 2, 3.

>>> c.append(4)>>> c[1, 2, 3, 4]>>> a[1, 2, 3, 4]>>> b[1, 2, 3, 4]>>> b.append("from b")>>> b[1, 2, 3, 4, 'from b']>>> a[1, 2, 3, 4, 'from b']>>> c[1, 2, 3, 4, 'from b']

Copy

The copy function can be divided into shortest copy and deep copy. First, let's look at the example below.

>>> A = [1, 2, 3] # establish a reference connection between a and the object >>> B = a # by assigning values, B Also creates a reference with the object >>> import copy >>> c = copy. copy (a) # create a copy of [1, 2, 3], that is, a new object >>> d = copy. deepcopy (a) # creates a copy of [1, 2, 3.

The relationships between a, B, c, and d are as follows:

A -- |
| --> [1, 2, 3]
B -- |

C -----> [1, 2, 3]

D -----> [1, 2, 3]

Modify

>>>. Append ('A') # change the original object through variable a to [1, 2, 3, 'a'] >>> B # B also changes [1, 2, 2, 3, 'a'] >>> c # c corresponds to a copy, which is not affected [1, 2, 3] >>>> d # d same as above [1, 2, 3]

The relationships between a, B, c, and d are shown in:

A -- |
| --> [1, 2, 3, 'a']
B -- |

C -----> [1, 2, 3]

D -----> [1, 2, 3]

Modify c:

>>> C. append ("cc") # modify the object of c> a # a is not affected, because a and c point to different objects [1, 2, 3, 'A'] >>> B [1, 2, 3, 'a'] >>> c [1, 2, 3, 'cc']> d [1, 2, 3]

The relationships between a, B, c, and d are shown in:

A -- |
| --> [1, 2, 3, 'a']
B -- |

C -----> [1, 2, 3, 'cc']

D -----> [1, 2, 3]

In the above example, there seems to be no difference between copy and deepcopy, and they both create another copy. See the following example:

>>> Q = [1, 2, 3, ['A', 'B'] # note that this object is a list with a list element, q [3] = ['A', 'B'] >>> w = copy. copy (q) # w and e are references of the copy objects of the shortest copy and the deep copy respectively> e = copy. deepcopy (q) >>> q. append ('4q') # [1, 2, 3, ['A', 'B'] corresponding to q is changed to [1, 2, 3, ['A ', 'B'], '4q']> q [1, 2, 3, ['A', 'B'], '4q'] >>> w # w, e is the same as before, and is not affected [1, 2, 3, ['A ', 'B']> e [1, 2, 3, ['A', 'B']

Modify the corresponding list of w

>>> W. append (4) >>> w # adds an integer 4 [1, 2, 3, ['A', 'B'], 4] >>> q # q and e are not affected [1, 2, 3, ['A', 'B'] >>> e [1, 2, 3, ['A', 'B']> q [1, 2, 3, ['A', 'B']

Modify the elements of w [3]. The prompt is that w is a shortest copy of q.

>>> w[3].append('w3c')>>> q[1, 2, 3, ['a', 'b', 'w3c']]>>> w                               [1, 2, 3, ['a', 'b', 'w3c'], 4]>>> e[1, 2, 3, ['a', 'b']]

Carefully observe the above results and find that:

Q and w. After the list elements in the list are modified, the two are modified simultaneously. E is not affected

That is, the shortest copy creates only the outer copy, but not the inner copy. The deep copy creates a complete copy. This explains the meaning of the Chinese translation name "Shallow". Its "Shallow" means copying a layer (outer layer ).

Further modify e to see the effect:

>>> e.append('5e')>>> q[1, 2, 3, ['a', 'b', 'w3c']]>>> w[1, 2, 3, ['a', 'b', 'w3c'], 4]>>> e[1, 2, 3, ['a', 'b'], '5e']>>> e[3].append('e3c')>>> q[1, 2, 3, ['a', 'b', 'w3c']]>>> w[1, 2, 3, ['a', 'b', 'w3c'], 4]>>> e[1, 2, 3, ['a', 'b', 'e3c'], '5e']

Programmers who think about it will ask a question here. Why do we need to have a shortest copy and a deep copy? How do they work? To answer this question on the magic network, please refer to the following two connection content:

Http://blog.csdn.net/llg8212/article/details/22782387http://blog.csdn.net/yueguanghaidao/article/details/25613887