Testing Like a Professional: A Step-by-Step Information to Python’s Mock Library – KDnuggets

Testing software program is essential for guaranteeing reliability and performance throughout totally different situations. Nevertheless, if the code implementation is dependent upon exterior companies, it turns into fairly a problem. That is the place mocking is available in. Python’s mock library supplies instruments to create mock objects to switch actual objects, making your checks simpler to take care of. Mocking facilitates centered testing of elements and faster testing cycles.

What’s Mocking?

Mocking is a method utilized in software program testing to simulate actual objects. Actual objects are changed by mock objects to simulate their performance, permitting you to check code in several situations and isolation. Mocking is very helpful to check particular elements of the codebase with out counting on the interplay with exterior programs, databases, or different complicated companies.

Let me clarify this idea with an instance. Contemplate that you’ve an online utility that makes use of an exterior API to retrieve information. To check with out relying on the true API, you can also make a mock object that mimics the solutions of the API. This fashion, you possibly can check your utility’s performance with out being depending on the true API, which could be gradual, unreliable, or not even obtainable throughout growth.

Sounds attention-grabbing, proper? Let’s now go over an in depth how-to for truly utilizing this library.

Step-by-Step Information to Utilizing Mock

Step 1: Importing the Mock Library

The unittest.mock is the usual library in Python (3.3 and in all newer variations) that gives mock objects to manage the habits of actual objects. First it is advisable to import it the unittest.mock library.

from unittest.mock import Mock, patch

Step 2: Making a Mock Object

Making a mock object is simple. As soon as imported, you possibly can instantiate a mock object like this:

Now, my_mock is a mock object you can configure to simulate the habits of an actual object.

Step 3: Setting Return Values

The Mock library supplies varied methods to configure mock objects and management their habits. As an example, you possibly can specify what a way ought to return when known as:

my_mock.some_method.return_value="Hello, World!"
print(my_mock.some_method())

Output:

Step 4: Setting Aspect Results

Unwanted side effects are extra actions or behaviors triggered when a way of a mock object is known as, corresponding to elevating exceptions or executing features. Apart from return values, you may as well outline attributes or specify unwanted side effects like this:

def raise_exception():
    elevate ValueError("An error occurred")

my_mock.some_method.side_effect = raise_exception

# It will elevate a ValueError
attempt:
    my_mock.some_method()
besides ValueError as e:
    print(e)  

Output:

On this instance, ValueError raises each time some_method() is known as.

Step 5: Asserting Calls

Verifying the tactic calls is essential for thorough testing. You need to use assertions to specify whether or not a way was known as, when, and with what arguments.

my_mock.calculate_length('foo', 'bar')
my_mock.calculate_length.assert_called()
my_mock.calculate_length.assert_called_once()
my_mock.calculate_length.assert_called_with('foo', 'bar')
my_mock.calculate_length.assert_called_once_with('foo', 'bar')

• assert_called(): Returns True if calculate_length was known as at the least as soon as
• assert_called_once(): Returns True if calculate_length was known as precisely as soon as
• assert_called_with('foo', 'bar'): Returns True if calculate_length was known as with the identical arguments
• assert_called_once_with('foo', 'bar'): Returns True if calculate_length was known as precisely as soon as with the identical arguments

If any of those assertions fail on the mock object, an AssertionError shall be raised, indicating that the anticipated habits didn’t match the precise habits of the mock.

Step 6: Utilizing Patch

The patch operate permits you to change actual objects with mock objects throughout checks. As mentioned earlier, that is notably helpful for simulating third-party libraries or APIs, guaranteeing your checks stay remoted from precise implementations. To reveal patching, contemplate the next instance operate that fetches information from the URL.

# my_module.py
import requests

def fetch_data(url):
    response = requests.get(url)
    return response.json()

You may keep away from making actual HTTP requests by patching the ‘requests.get’ like this:

# test_my_module.py
import unittest
from unittest.mock import patch
import my_module

class TestFetchData(unittest.TestCase):
    @patch('my_module.requests.get')

    def test_fetch_data(self, mock_get):
        # Arrange the mock to return a particular response
        mock_get.return_value.json.return_value = {'key': 'worth'}
       
        # Name the operate to check
        outcome = my_module.fetch_data('http://instance.com')
       
        # Test the outcome
        self.assertEqual(outcome, {'key': 'worth'})
       
        # Confirm that requests.get was known as appropriately
        mock_get.assert_called_once_with('http://instance.com')

if __name__ == '__main__':
    unittest.primary()

The patch decorator is added simply above the test_fetch_data operate to switch the requests.get operate with a mock.

Step 7: Mocking Lessons

You may mock total courses and their strategies to simulate interactions between objects. As an example, you possibly can mock a database class to check your utility’s interplay with the database with out the necessity to arrange an actual database connection like this:

# database.py
class Database:
    def join(self):
        go

    def save_user(self, person):
        go

    def get_user(self, user_id):
        go


# test_database.py
from unittest.mock import Mock

# Making a mock database object
mock_db = Mock(spec=Database)

# Simulating methodology calls
mock_db.join()
mock_db.save_user({"id": 1, "name": "Alice"})
mock_db.get_user(1)

# Verifying that the strategies had been known as
mock_db.join.assert_called_once()
mock_db.save_user.assert_called_once_with({"id": 1, "name": "Alice"})
mock_db.get_user.assert_called_once_with(1)

Wrapping Up

 
That is it for right now’s article on unittest.mock, a robust library for testing in Python. It allows builders to check code, guaranteeing clean interactions between objects. With superior options like specifying unwanted side effects, asserting calls, mocking courses, and utilizing context managers, testing varied situations turns into simpler. Begin utilizing mocks in your checks right now to make sure higher-quality code and smoother deployments.

Kanwal Mehreen Kanwal is a machine studying engineer and a technical author with a profound ardour for information science and the intersection of AI with medication. She co-authored the e book “Maximizing Productivity with ChatGPT”. As a Google Technology Scholar 2022 for APAC, she champions range and tutorial excellence. She’s additionally acknowledged as a Teradata Range in Tech Scholar, Mitacs Globalink Analysis Scholar, and Harvard WeCode Scholar. Kanwal is an ardent advocate for change, having based FEMCodes to empower ladies in STEM fields.