Python | Avoiding class data shared among the instances
Last Updated :
15 Jul, 2021
Class attributes belong to the class itself and they will be shared by all the instances and hence contains same value of each instance. Such attributes are defined in the class body parts usually at the top, for legibility.
Suppose we have the following code snippet :
Python3
class Geek_Class:
geek = []
x = Geek_Class()
y = Geek_Class()
x.geek.append( 1 )
y.geek.append( 2 )
x.geek.append( 3 )
y.geek.append( 4 )
print (x.geek)
print (y.geek)
|
Output:
[1, 2, 3, 4]
[1, 2, 3, 4]
It prints [1, 2, 3, 4] for x and [1, 2, 3, 4] for y. Suppose the output we want is [1, 3] for x and [2, 4] for y. We can get the desired output by the following ways:
Method #1: By declaring them inside the __init__
Declaring the variables inside the class declaration makes them class members and not instance members. Declaring them inside the __init__ method ensures that a new instance of the members is created alongside every new instance of the object, which is what we need.
Python3
class Geek_Class:
def __init__( self ):
self .geek = []
x = Geek_Class()
y = Geek_Class()
x.geek.append( 1 )
y.geek.append( 2 )
x.geek.append( 3 )
y.geek.append( 4 )
print (x.geek)
print (y.geek)
|
In the original code no value is assigned to list attribute after instantiation; so it remains a class attribute. Defining list inside __init__ works because __init__ is called after instantiation.
Method #2: By creating the new list and storing the values in that.
Python3
class Geek_Class:
geek = []
x = Geek_Class()
y = Geek_Class()
x.geek = []
y.geek = []
x.geek.append( 1 )
y.geek.append( 2 )
x.geek.append( 3 )
y.geek.append( 4 )
print (x.geek)
print (y.geek)
|
Share your thoughts in the comments
Please Login to comment...