Check the behavior when assigning Python

Summary

• A variable is a name for referencing an object, and the object referenced by the variable can be confirmed by id (hoge).
• The meaning of y = x is that the object referenced by the variable y of the assignment destination is the object referenced by the variable x of the assignment source.
• This is a common behavior for all variables, whether x is an int, a list, or a tuple.
• Passing by reference means passing a reference to the variable itself, so so-called passing by reference is not possible.
• The same is true when substituting an element of a list, but since the reference destination of the list itself is not changed, it looks like passing by reference at first glance.

Behavior confirmation when substituting

Assign an int variable

• The reference destination of variable a changes to the reference destination of variable b by assignment.

a = 1
b = 2
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))
a = b   #id of a replaces id of b
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))

result

a: 1 b: 2
id_a: 93893417024192 id_b: 93893417024224
a: 2 b: 2
id_a: 93893417024224 id_b: 93893417024224

Substitute a new variable 3 for b

• Substituting 3 for b causes b to reference a new object called 3.
• At this time, a remains the original reference destination, so it is not affected by the change.

b = 3   #The id of b is replaced by the id of 3,The id of a remains the same
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))

result

a: 2 b: 3
id_a: 93893417024224 id_b: 93893417024256

If the variable is list

• Same as int

a = [1,1]
b = [2,2]
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))
a = b  #id of a replaces id of b
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))
b = [3,3]  #The id of b is replaced by the id of 3,The id of a remains the same
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))

result

a: [1, 1] b: [2, 2]
id_a: 140361960710984 id_b: 140361960208392
a: [2, 2] b: [2, 2]
id_a: 140361960208392 id_b: 140361960208392
a: [2, 2] b: [3, 3]
id_a: 140361960208392 id_b: 140361960256584

Assignment of list to element

• When assigning to a list element, the reference destination of the list element changes, but the reference destination of the list itself does not change.
• a and b still refer to the same object, so a is affected by the change in b
• Since the reference destination of the list element has changed, it can be seen that it is not passed by reference.

a = [1,1]
b = [2,2]
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))
print("id_a[0]:", id(a[0]), "id_b[0]:", id(b[0]))
a = b  #id of a replaces id of b
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))
print("id_a[0]:", id(a[0]), "id_b[0]:", id(b[0]))
b[0] = 3  # b[0]Id is replaced by id 3 and at the same time a[0]Id is also replaced by id of 3,The ids of a and b remain the same
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))
print("id_a[0]:", id(a[0]), "id_b[0]:", id(b[0]))

result

a: [1, 1] b: [2, 2]
id_a: 140361961620360 id_b: 140361958272904
id_a[0]: 93893417024192 id_b[0]: 93893417024224
a: [2, 2] b: [2, 2]
id_a: 140361958272904 id_b: 140361958272904
id_a[0]: 93893417024224 id_b[0]: 93893417024224
a: [3, 2] b: [3, 2]
id_a: 140361958272904 id_b: 140361958272904
id_a[0]: 93893417024256 id_b[0]: 93893417024256

Assignment of dict type elements

When assigning directly inside the dict, the id of the dict does not change and the id of the assigned dict element is replaced. At this time, if there is a variable that shares the id of dict, the id of dict does not change, so the dict element is changed.

a = {0:1}
b = {0:2}
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))
print("id_a[0]:", id(a[0]), "id_b[0]:", id(b[0]))
a = b  #id of a replaces id of b
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))
print("id_a[0]:", id(a[0]), "id_b[0]:", id(b[0]))
b[0] = 3  # b[0]Id is replaced by id 3 and at the same time a[0]Id is also replaced by id of 3,The ids of a and b remain the same
print("a:",a,"b:",b)
print("id_a:", id(a), "id_b:", id(b))
print("id_a[0]:", id(a[0]), "id_b[0]:", id(b[0]))

result

a: {0: 1} b: {0: 2}
id_a: 140362018772456 id_b: 140361986516168
id_a[0]: 93893417024192 id_b[0]: 93893417024224
a: {0: 2} b: {0: 2}
id_a: 140361986516168 id_b: 140361986516168
id_a[0]: 93893417024224 id_b[0]: 93893417024224
a: {0: 3} b: {0: 3}
id_a: 140361986516168 id_b: 140361986516168
id_a[0]: 93893417024256 id_b[0]: 93893417024256

Reference: [Programming FAQ — Python 3.8.1rc1 Documentation](https://docs.python.org/ja/3/faq/programming.html#how-do-i-write-a-function-with-output-parameters-call -by-reference) Pass by value and pass by reference in C #-Qiita Why shouldn't we use the word "pass by reference" in Java, JavaScript, Python, Ruby, etc.? --Qiita I can't say it's passed by reference anymore-Qiita