Generators and tests

Generators

def number_sequence_generator():
    a = 0
    while True:
        yield a
        a += 1

Generators are a simple tool for creating iterators

They are just functions with yield instead of return statement whenever they want to return data

Generators

if __name__ == '__main__':
    generator = number_sequence_generator()
    next(generator) # 0
    next(generator) # 1
    next(generator) # 2

Each time next() is called on it, the generator resumes where it left off. It remembers all the data values and which statement was last executed

Finite generators

def finite_generator():
    for i in range(4):
        yield i ** 2

if __name__ == '__main__':
    generator = finite_generator()
    next(generator) # 0
    next(generator) # 1
    next(generator) # 4
    next(generator) # 9
    next(generator) # StopIteration error

Infinite generators

def finite_generator():
    a = 0
    while True:
        yield a ** 2
        a += 1

if __name__ == '__main__':
    generator = finite_generator()
    next(generator) # 0
    next(generator) # 1
    next(generator) # 4
    next(generator) # 9
    next(generator) # 16
    next(generator) # 25
    # ...

Generators

They should be lazy whenever it's possible

for i in range(100000000000):
    print(i)

e.g. this will not cause your computer to die in python 3.x. Cause all numbers are generated one after another, not all at once

though it will in python 2.x, therefore xrange should be used in python 2.x

Generator expressions

for i_square in (i ** 2 for i in range(100000000000)):
    print(i_square)

Yet more concise way to write iterators

What to read

Why do we need tests

To be sure that out code does what it's supposed to
To alleviate the pain of refactoring
To show other people that our code is reliable
...

There are many ways to categorize tests

Unit testing
Integration testing
System testing
Sanity testing
Smoke testing
Regression testing
Stress testing

Unit testing - test the smallest pieces of functionality, e.g. functions

Libraries for unit tests

unittest
doctest
pytest (not in standard libraries)

unittest

from unittest import TestCase

class SimpleTests(TestCase):
    
    @classmethod
    def setUpClass(cls):
        print("I'm executed once before all tests in class")

    def setUp(self):
        print("I'm executed before each test in class")

    def test_something(self):
        self.assertEqual(1, 2, "Some test comment")
    
    @classmethod
    def tearDownClass(cls):
        print("I'm executed once after all tests in class")

    def setUp(self):
        print("I'm executed after each test in class")

Repo with example

https://gitlab.com/AleksandrSl/bi_python_course