Blob Class and Modularity - Object Oriented Programming

Welcome to part 16 of the intermediate Python programming tutorial series. In this part of the series, we're going to discuss the modularity of the class we wrote.

Recall in the beginning of this series that I proclaimed that intermediate programming was:

• Maintainable
• Scale-able
• Modular

So far, we've not introduced anything that will make this code too hard to maintain or scale over time, at least within the scope of what we can do with PyGame. What about making it modular? There's a really easy test for this, let's try to import it!

To do this, let's have two files. Let's copy the Blob class and random, and make a new file: blob.py

import random


class Blob:

    def __init__(self, color):
        self.x = random.randrange(0, WIDTH)
        self.y = random.randrange(0, HEIGHT)
        self.size = random.randrange(4,8)
        self.color = color

    def move(self):
        self.move_x = random.randrange(-1,2)
        self.move_y = random.randrange(-1,2)
        self.x += self.move_x
        self.y += self.move_y

        if self.x < 0: self.x = 0
        elif self.x > WIDTH: self.x = WIDTH
        
        if self.y < 0: self.y = 0
        elif self.y > HEIGHT: self.y = HEIGHT

Back to our original file, let's remove the Blob class, and then import Blob from blob.py.

import pygame
import random
from blob import Blob

STARTING_BLUE_BLOBS = 10
STARTING_RED_BLOBS = 3

WIDTH = 800
HEIGHT = 600
WHITE = (255, 255, 255)
BLUE = (0, 0, 255)
RED = (255, 0, 0)

game_display = pygame.display.set_mode((WIDTH, HEIGHT))
pygame.display.set_caption("Blob World")
clock = pygame.time.Clock()

def draw_environment(blob_list):
    game_display.fill(WHITE)

    for blob_dict in blob_list:
        for blob_id in blob_dict:
            blob = blob_dict[blob_id]
            pygame.draw.circle(game_display, blob.color, [blob.x, blob.y], blob.size)
            blob.move()

    pygame.display.update()
    

def main():
    blue_blobs = dict(enumerate([Blob(BLUE) for i in range(STARTING_BLUE_BLOBS)]))
    red_blobs = dict(enumerate([Blob(RED) for i in range(STARTING_RED_BLOBS)]))
    while True:
        for event in pygame.event.get():
            if event.type == pygame.QUIT:
                pygame.quit()
                quit()
        draw_environment([blue_blobs,red_blobs])
        clock.tick(60)

if __name__ == '__main__':
    main()

Immediately, we're given an error in the blob.py file, regarding our Blob class, where we have some undefined variables. This is definitely a problem with writing classes, you should try to avoid using constants or variables outside of the class. Let's add these values to the __init__ method, then modify all of the parts where we used the constants.

blob.py:

import random


class Blob:

    def __init__(self, color, x_boundary, y_boundary):
        self.x_boundary = x_boundary
        self.y_boundary = y_boundary
        self.x = random.randrange(0, self.x_boundary)
        self.y = random.randrange(0, self.y_boundary)
        self.size = random.randrange(4,8)
        self.color = color
        

    def move(self):
        self.move_x = random.randrange(-1,2)
        self.move_y = random.randrange(-1,2)
        self.x += self.move_x
        self.y += self.move_y

        if self.x < 0: self.x = 0
        elif self.x > self.x_boundary: self.x = self.x_boundary
        
        if self.y < 0: self.y = 0
        elif self.y > self.y_boundary: self.y = self.y_boundary

Next, within our original file, when we call the Blob class, it's expecting some values for those arguments, so you'd add those in the main function:

def main():
    blue_blobs = dict(enumerate([Blob(BLUE,WIDTH,HEIGHT) for i in range(STARTING_BLUE_BLOBS)]))
    red_blobs = dict(enumerate([Blob(RED,WIDTH,HEIGHT) for i in range(STARTING_RED_BLOBS)]))
    while True:
        ...

Great, so now our Blob class can at least be imported, so it's modular by nature already! Another good idea is to attempt to give the developer that is using your code as much power as possible, and to make your class as generalize-able as possible. At least one example where we could definitely give more to the programmer using this class is in the definition of the blob's size:

        self.size = random.randrange(4,8)

Is there any reason why we wouldn't want to give the programmer an easy way to change these? I don't think so. Unlike x_boundary and y_boundary, however, we do not necessarily *need* the programmer to provide us a value for the size, since we can at least use a reasonable starting default. Thus, we can do something like:

class Blob:

    def __init__(self, color, x_boundary, y_boundary, size_range=(4,8)):
        self.x_boundary = x_boundary
        self.y_boundary = y_boundary
        self.x = random.randrange(0, self.x_boundary)
        self.y = random.randrange(0, self.y_boundary)
        self.size = random.randrange(size_range[0],size_range[1])
        self.color = color

Now, if the programmer wants to change the size, they can, otherwise they don't have to. We might also want to allow the programmer to modify the speed of the blob if they want to:

import random


class Blob:

    def __init__(self, color, x_boundary, y_boundary, size_range=(4,8), movement_range=(-1,2)):
        self.size = random.randrange(size_range[0],size_range[1])
        self.color = color
        self.x_boundary = x_boundary
        self.y_boundary = y_boundary
        self.x = random.randrange(0, self.x_boundary)
        self.y = random.randrange(0, self.y_boundary)
        self.movement_range = movement_range

    def move(self):
        self.move_x = random.randrange(self.movement_range[0],self.movement_range[1])
        self.move_y = random.randrange(self.movement_range[0],self.movement_range[1])
        self.x += self.move_x
        self.y += self.move_y

        if self.x < 0: self.x = 0
        elif self.x > self.x_boundary: self.x = self.x_boundary
        
        if self.y < 0: self.y = 0
        elif self.y > self.y_boundary: self.y = self.y_boundary

Now we've opened up the class quite a bit. Does anything else jump out at you? At least for me, the line where we force the blob to remain in-bounds. Might there be examples where we'd like the blobs to be able to roam freely out of view? Certainly! Is this bounding code useful though? Is it likely that programmers will want to make use of this quite often? Certainly! I would argue, however, that it makes more sense to either not have the code at all, or to give it its own method, like so:

import random

class Blob:

    def __init__(self, color, x_boundary, y_boundary, size_range=(4,8), movement_range=(-1,2)):
        self.size = random.randrange(size_range[0],size_range[1])
        self.color = color
        self.x_boundary = x_boundary
        self.y_boundary = y_boundary
        self.x = random.randrange(0, self.x_boundary)
        self.y = random.randrange(0, self.y_boundary)
        self.movement_range = movement_range

    def move(self):
        self.move_x = random.randrange(self.movement_range[0],self.movement_range[1])
        self.move_y = random.randrange(self.movement_range[0],self.movement_range[1])
        self.x += self.move_x
        self.y += self.move_y

    def check_bounds(self):
        if self.x < 0: self.x = 0
        elif self.x > self.x_boundary: self.x = self.x_boundary
        
        if self.y < 0: self.y = 0
        elif self.y > self.y_boundary: self.y = self.y_boundary

Now, the programmer can decide to use it or not. You could also give some sort of argument in the move method, where, if True, then boundaries would be enforced.

In the next object oriented programming tutorial, we will be discussing inheritance.

The next tutorial: Inheritance - Object Oriented Programming