Python-Design-Patterns builder

Table of Contents

  1. builder Model
  2. Python-Design-Patterns builder
  3. builder Test

builder Model

Python-Design-Patterns builder

# -*- coding : utf-8 -*-

*What is this pattern about?
It decouples the creation of a complex object and its representation,
so that the same process can be reused to build objects from the same
This is useful when you must separate the specification of an object
from its actual representation (generally for abstraction).

*What does this example do?

The first example achieves this by using an abstract base
class for a building, where the initializer (__init__ method) specifies the
steps needed, and the concrete subclasses implement these steps.

In other programming languages, a more complex arrangement is sometimes
necessary. In particular, you cannot have polymorphic behaviour in a constructor in C++ -
- which means this Python technique will not work. The polymorphism
required has to be provided by an external, already constructed
instance of a different class.

In general, in Python this won't be necessary, but a second example showing
this kind of arrangement is also included.

*Where is the pattern used practically?


Decouples the creation of a complex object and its representation.

# Abstract Building
class Building(object):
    def __init__(self):

    def build_floor(self):
        raise NotImplementedError

    def build_size(self):
        raise NotImplementedError

    def __repr__(self):
        return 'Floor: {0.floor} | Size: {0.size}'.format(self)

# Concrete Buildings
class House(Building):
    def build_floor(self):
        self.floor = 'One'

    def build_size(self):
        self.size = 'Big'

class Flat(Building):
    def build_floor(self):
        self.floor = 'More than One'

    def build_size(self):
        self.size = 'Small'

# In some very complex cases, it might be desirable to pull out the building
# logic into another function (or a method on another class), rather than being
# in the base class '__init__'. (This leaves you in the strange situation where
# a concrete class does not have a useful constructor)

class ComplexBuilding(object):
    def __repr__(self):
        return 'Floor: {0.floor} | Size: {0.size}'.format(self)

class ComplexHouse(ComplexBuilding):
    def build_floor(self):
        self.floor = 'One'

    def build_size(self):
        self.size = 'Big and fancy'

def construct_building(cls):
    building = cls()
    return building

# Client
if __name__ == "__main__":
    house = House()
    flat = Flat()

    # Using an external constructor function:
    complex_house = construct_building(ComplexHouse)

### OUTPUT ###
# Floor: One | Size: Big
# Floor: More than One | Size: Small
# Floor: One | Size: Big and fancy

builder Test

#!/usr/bin/env python
# -*- coding: utf-8 -*-
import unittest
from creational.builder import construct_building, House, Flat, ComplexHouse

class TestSimple(unittest.TestCase):
    def test_house(self):
        house = House()
        self.assertEqual(house.size, 'Big')
        self.assertEqual(house.floor, 'One')

    def test_flat(self):
        flat = Flat()
        self.assertEqual(flat.size, 'Small')
        self.assertEqual(flat.floor, 'More than One')

class TestComplex(unittest.TestCase):
    def test_house(self):
        house = construct_building(ComplexHouse)
        self.assertEqual(house.size, 'Big and fancy')
        self.assertEqual(house.floor, 'One')