function - Python Alternatives to Global Variables

Question

import random  

#----------------------------------------------#
def main():  
    create_list_and_find_max_and_min(10)  
    the_smart_way()
#----------------------------------------------#
def create_list_and_find_max_and_min(n):
    global my_array
    my_array = []
    n = input("How many numbers do you want in your array?:")
    for i in range(n):
        my_array.append(random.randint(1,n))
    print "My array is:", my_array
    #----------------------------------------------#
    min = my_array[0]
    for number in my_array:
    if min > number:
        min = number
    print "The minimum value in the array is:", min
    #----------------------------------------------#
    max = my_array[0]
    for number in my_array:
        if max < number:
        max = number
    print "The maximum value in the array is:", max
#----------------------------------------------#
def the_smart_way():
    # "This one uses the built-in Python functions for min/max..."
    min_my_array = min(my_array)
    max_my_array = max(my_array)
    return min_my_array, max_my_array
#----------------------------------------------#
if __name__ == '__main__':
    main()

I'm fairly new to using Python. I've generally heard that using global variables is a bad practice when programming.. So are there any alternatives to what I'm trying to do here. Which is to use the variable "my_array" from inside another function? Thanks!

Answer 1

Yes, there are two alternatives.

First, you can pass the values around instead of using globals. For example, create_list_and_find_max_and_min can create the array locally and return it, then you can pass it in to the_smart_way:

import random  

#----------------------------------------------#
def main():  
    my_array = create_list_and_find_max_and_min(10)  
    print the_smart_way(my_array)
#----------------------------------------------#
def create_list_and_find_max_and_min(n):
    my_array = []
    n = input("How many numbers do you want in your array?:")
    for i in range(n):
        my_array.append(random.randint(1,n))
    print "My array is:", my_array
    #----------------------------------------------#
    min = my_array[0]
    for number in my_array:
        if min > number:
            min = number
    print "The minimum value in the array is:", min
    #----------------------------------------------#
    max = my_array[0]
    for number in my_array:
        if max < number:
            max = number
    print "The maximum value in the array is:", max
    return my_array
#----------------------------------------------#
def the_smart_way(my_array):
    # "This one uses the built-in Python functions for min/max..."
    min_my_array = min(my_array)
    max_my_array = max(my_array)
    return min_my_array, max_my_array
#----------------------------------------------#
if __name__ == '__main__':
    main()

Second, you can create a class that encapsulates the data and the functions that operate on that data:

import random  

#----------------------------------------------#
class MyArrayClass(object):
    def create_list_and_find_max_and_min(self, n):
        self.my_array = []
        n = input("How many numbers do you want in your array?:")
        for i in range(n):
            self.my_array.append(random.randint(1,n))
        print "My array is:", self.my_array
        #----------------------------------------------#
        min = self.my_array[0]
        for number in self.my_array:
            if min > number:
                min = number
        print "The minimum value in the array is:", min
        #----------------------------------------------#
        max = self.my_array[0]
        for number in self.my_array:
            if max < number:
                max = number
        print "The maximum value in the array is:", max
    #----------------------------------------------#
    def the_smart_way(self):
        # "This one uses the built-in Python functions for min/max..."
        min_my_array = min(self.my_array)
        max_my_array = max(self.my_array)
        return min_my_array, max_my_array
#----------------------------------------------#
def main():
    my_array = MyArrayClass()
    my_array.create_list_and_find_max_and_min(10)  
    print my_array.the_smart_way()
#----------------------------------------------#
if __name__ == '__main__':
    main()

You should probably understand the reasons global variables are bad practice.

Imagine that you want to create two arrays. With global variables, the second one will replace the first one, which will be gone forever.

create_list_and_fix_max_and_min(10)
create_list_and_fix_max_and_min(20)
# No way to operate on the original array!

With a local variable, you can store both of them:

my_array_1 = create_list_and_fix_max_and_min(10)
my_array_2 = create_list_and_fix_max_and_min(20)
the_smart_way(my_array_1)

Using an object provides the same benefit; the difference between the two ultimately comes down to whether the operations are part of the meaning of the data, or whether the data stand alone and the operations are generic. (Or, sometimes, whether you're more of a functional snob or an OO snob…)