Lately, I had to write unit tests using Pytest for a Python module. The module contains a class where other classes are initialize within its constructor.
As usual I created a fixture for this class to make it easy to write a test for each class method. At this point I ran into some issues when I tried to mock the different classes initiated in the constructor. The mocking didn't work, and instances of these classes were still being created.
After some research and combining a few different solutions I found online, I want to share how I managed to mock the classes.
Solution
Here is an example of the class I tried to mock:
class ClassA: def __init__(self): self.class_b = ClassB() self.class_c = ClassC() self.count = 0
We want to set a value for every field of this class during tests. This value can be None or a class mock, but we don't want initiations of the classes ClassB and ClassC.
In our case, let's decide that self.class_b and self.class_c should be mocks:
@pytest.fixture def mock_class_b(): class_b = Mock(spec=ClassB) return class_b @pytest.fixture def mock_class_c(): class_c = Mock(spec=ClassC) return class_c
So a fixture for this class that serves our goal looks like this:
@pytest.fixture def class_a_mock(mock_class_b, mock_class_c): with patch.object(target=ClassA, attribute="__init__", return_value=None) as mock_init: class_a = ClassA() class_a.class_b = mock_class_b class_a.class_c = mock_class_c class_b.count = 0 return class_a
The important part is how to use the patch.object function, which is from unittest.mock module in Python. It is used in testing to temporarily replace an attribute of a given object wit a mock or another value.
Arguments
- target=ClassA: the object (usually a class) whose attribute we want to patch.
- attribute="__init__": the name of the attribute we want to patch.
- return_value=None: replacing the __init__ method with a function that does nothing
In this way we can create mocked variables in our fixture.
Read more about patch.object
Testing Tips
I wrote this tutorial for this kind of cases where, for any reason, we cannot change the code of Class A. However, I generally recommend modifying the code if possible, not to change the logic, but to make it more testable.
Here are some examples of how to modify Class A to make it more testable:
Option 1: Pass instances of class B and class C as parameters.
This way, when we write the fixture, we can pass mocks instead of instances.
class ClassA: def __init__(self, class_b_instance, class_c_instance): self.class_b = class_b_instance self.class_c = class_c_instance self.count = 0
Option 2: Create a Boolean variable that indicates test mode.
This way we can decide which fields of class A will or will not get a value when it is initiated.
class ClassA: def __init__(self, test_mode=False): if not test_mode: self.class_b = ClassB() self.class_c = ClassC() self.count = 0
Option 3: Make class initiations in a separate method.
This approach gives us the choice to avoid calling set_class_variables in the test module.
class ClassA: def __init__(self): self.class_b = None self.class_c = None self.count = None def set_class_variables(self): self.class_b = ClassB() self.class_c = ClassC() self.count = 0
Hope this helps! :)
The above is the detailed content of Mocking Python Classes. For more information, please follow other related articles on the PHP Chinese website!

Solution to permission issues when viewing Python version in Linux terminal When you try to view Python version in Linux terminal, enter python...

This article explains how to use Beautiful Soup, a Python library, to parse HTML. It details common methods like find(), find_all(), select(), and get_text() for data extraction, handling of diverse HTML structures and errors, and alternatives (Sel

This article compares TensorFlow and PyTorch for deep learning. It details the steps involved: data preparation, model building, training, evaluation, and deployment. Key differences between the frameworks, particularly regarding computational grap

When using Python's pandas library, how to copy whole columns between two DataFrames with different structures is a common problem. Suppose we have two Dats...

This article guides Python developers on building command-line interfaces (CLIs). It details using libraries like typer, click, and argparse, emphasizing input/output handling, and promoting user-friendly design patterns for improved CLI usability.

The article discusses popular Python libraries like NumPy, Pandas, Matplotlib, Scikit-learn, TensorFlow, Django, Flask, and Requests, detailing their uses in scientific computing, data analysis, visualization, machine learning, web development, and H

The article discusses the role of virtual environments in Python, focusing on managing project dependencies and avoiding conflicts. It details their creation, activation, and benefits in improving project management and reducing dependency issues.

Regular expressions are powerful tools for pattern matching and text manipulation in programming, enhancing efficiency in text processing across various applications.


Hot AI Tools

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Undress AI Tool
Undress images for free

Clothoff.io
AI clothes remover

AI Hentai Generator
Generate AI Hentai for free.

Hot Article

Hot Tools

SublimeText3 English version
Recommended: Win version, supports code prompts!

SublimeText3 Chinese version
Chinese version, very easy to use

WebStorm Mac version
Useful JavaScript development tools

SublimeText3 Mac version
God-level code editing software (SublimeText3)

SublimeText3 Linux new version
SublimeText3 Linux latest version