Programming is not just about writing code. It is important to verify that the code does what it should (and about fixing it if needed). The process of verification that the program works as expected is called testing.
You have probably tested your programs by trying to execute them, entered some input data and looked if the results are correct. It is harder to do it for bigger programs. Bigger programs have more possibilities what to do and it is harder to verify that all possibilities do what they should.
That is why developers write code that verifies their program instead of testing their programs manually.
Automated tests are functions that check that the program works correctly. You can execute the tests anytime and verify that the code works. The main benefit is that you can change the code in the future and let the tests verify that the change does not break existing functionalitye.
Up to now, we have used only the modules that come installed with Python,
for example, modules such as
There are many more libraries that are not included in Python
but you can install them to your Python environment and use them.
The library for testing in Python is called
It is quite difficult to use this library so we will use a better one.
We will install the library
pytest which is easy to use and very popular.
You install libraries into your active virtual environment. We have learned how to create and activate a virtual environment in the lesson about Python installation. Make sure that you have activated a virtual environment.
Submit the following command. (It is a command-line command,
mkdir; do not enter it into the Python console.)
(venv)$ python -m pip install pytest
What does Pip do?
python -m pip calls Python and tells it to execute the
pip module. This module can install and uninstall libraries.
(Similarly, when you created a virtual environment, you used the
python -m venv – the
venv module can create virtual environments.)
And the arguments
install pytest tell Pip to install
You can display the help for the Pip module using the command
python -m pip --help.
For Windows users
If you use Windows, it is important to run Python programs using
python program.py, and not just
Although we always show
python in our lessons,
it could work without it so far.
If you do not use the command
python in the beginning, the program
could start in a different Python and different virtual environment,
pytest module might not have been installed.
We will show testing thtough a very simple example.
There is a function
add that can add two numbers.
There is another function that tests if the
add function returns correct results for specific numbers.
Make a copy of the code into a file named
in a new empty directory.
The naming of files and test functions is important for
pytest (with default settings).
It is important for names of files containing tests and test functions
to start with
def add(a, b): return a + b def test_add(): assert add(1, 2) == 3
What does the test function do?
assert statement evaluates the expression that follows it.
If the result is not true then it raises an exception
and it makes the test fail.
You can imagine that
assert a == b does following:
if a != b: raise AssertionError('Test failed!')
Do not use
assert outside of test functions for now.
For "regular" code, the
assert has functionality that
we will not explain now.
You execute tests with the command
python -m pytest -v,
followed by the name of the file containing the tests.
By using this command you are telling: Python:
Execute the module named pytest,
in verbose mode, for the given file.
$ python -m pytest -v test_addition.py ============= test session starts ============= platform darwin -- Python 3.6.5, pytest-3.9.1, py-1.7.0, pluggy-0.8.0 -- rootdir: learn, inifile: collecting ... collected 1 items test_addition.py::test_add PASSED ============= 1 passed in 0.01 seconds =============
This command scans the given file and calls all functions that start
test_. It checks that they do not raise any exceptions,
for example, an exception raised by
If an exception occurs,
pytest shows a red message with
additional details that can help you find the bug and fix it.
You can omit the argument with the filename:
python -m pytest -v
In this case,
pytest scans the current directory and runs tests
in all files whose names start with
test_. You can use a path to
a directory and
pytest finds tests in it.
Try to change the
add function (or its test) and see what happens
if a test fails.
You do not usually write tests in the same file with the regular code. Typically, you write tests in another file. This way, your code is easier to read, and it makes it possible to distribute only the code, without the tests, to someone who is interested only in executing the program.
test_addition.py file: Move the
add function to a new module
test_addition.py file, keep only the test.
test_addition.py file, add
from addition import add to the top
so the test can call the tested function.
The test should pass again.
Automated tests have to run "unattended". They are usually executed automatically and the failures are reported automatically (e.g. by email) and the code that passes all tests can be automatically released (installed to a system where it runs or is made available to customers).
What does this mean to us?
input function will not work in tests. There is no-one who can reply.
This can make your work harder sometimes. Let's look at a more complex project: 1D (one-dimensional) tic-tac-toe.
If you do not have the 1D tic-tac-toe program, the following sections are only theoretical. If you study at home, complete the 1D tic-tac-toe lesson before continuing. The homework assignment is in PyLadies projects on page 2. Unfortunately, so far it is only in Czech.
The structure of your 1D tic-tac-toe code could look like this:
import random # (and possibly other import statements that are needed) def move(board, space_number, mark): """Returns the board with the specified mark placed in the specified position""" ... def player_move(board): """Asks the player what move should be done and returns the board with the move played. """ ... input('What is your move? ') ... def tic_tac_toe_1d(): """Starts the game It creates an empty board and runs player_move and computer_move alternately until the game is finished. """ while ...: ... player_move(...) ... # Start the game: tic_tac_toe_1d()
If you import this module, Python executes all commands in it from top to bottom.
The first command,
import, makes some functions and variables available.
Imports do not usually have any side-effects.
The definitions of functions (
def statements and everything in them)
just define the functions (but they do not execute the functions).
tic_tac_toe_1d function starts the game.
tic_tac_toe_1d calls the
player_move() function which calls
If you import this module from tests, the
and the module is not imported.
If you want to import such a module from elsewhere – for example, you would like
move in another game – the import of the module requires the user to
play 1D tic-tac-toe!
The calling of
tic_tac_toe_1d is a side-effect and we need to remove it.
Yeah but you cannot start the game without it! What about it?
You can create a new module.
game.py and put just this call into it:
import tic_tac_toe tic_tac_toe.tic_tac_toe_1d()
You cannot test this module because it calls
But you can execute it if you want to play.
Since you do not have tests for this module, it should be very simple:
one import and one statement.
You can import the original module from tests or other modules.
A test for the original module could look like this:
import tic_tac_toe def test_move_to_empty_space(): board = tic_tac_toe.computer_move('--------------------') assert len(board) == 20 assert board.count('x') == 1 assert board.count('-') == 19
The tests that verify that the program works correctly under correct conditions are called positive tests. But you can test what your program does under unexpected conditions.
The tests that check the behavior in case of "invalid" input
are called negative tests.
They can check for a specific negative result (for example
that a call like
or that a "reasonable" exception is raised.
For example, the
computer_move function should raise an error
ValueError) when the board is full.
It is much better to raise an exception than doing nothing and silently letting the program get stuck. You can use such function in a more complex program and be sure that you will get an understandable error when it is called under bad conditions. Then you can easily fix it.
with statement and the
to test that your code raises the expected exception.
raises function is imported from the
We will explain the
with statement later.
You just need to know that it checks that the block of code below
raises the specified exception:
import pytest import tic_tac_toe def test_move_failure(): with pytest.raises(ValueError): tic_tac_toe.computer_move('oxoxoxoxoxoxoxoxoxox')