Why mocking isolates code under test in PyTest - Test Execution Impact

Test Overview

This test checks how mocking helps isolate the code under test by replacing a real dependency with a mock object. It verifies that the function behaves correctly when the dependency returns a controlled value.

Test Code - pytest

PyTest

import pytest
from unittest.mock import Mock

def fetch_data(api_client):
    response = api_client.get_data()
    return response['value'] * 2

def test_fetch_data_with_mock():
    mock_api_client = Mock()
    mock_api_client.get_data.return_value = {'value': 10}
    result = fetch_data(mock_api_client)
    assert result == 20

Execution Trace - 5 Steps

Step	Action	System State	Assertion	Result
1	Test starts	Test environment is ready with pytest and unittest.mock imported	—	PASS
2	Creates a mock object to replace the real api_client	mock_api_client is a Mock instance with get_data method mocked	—	PASS
3	Sets mock_api_client.get_data to return {'value': 10}	Calling mock_api_client.get_data() returns {'value': 10}	—	PASS
4	Calls fetch_data with mock_api_client	fetch_data calls mock_api_client.get_data(), receives {'value': 10}	Result should be 10 * 2 = 20	PASS
5	Asserts that result equals 20	Result is 20 as expected	assert result == 20	PASS

Failure Scenario

Failing Condition: If mock_api_client.get_data does not return the expected dictionary or is not mocked

Execution Trace Quiz - 3 Questions

What does mocking do in this test?

A Replaces the real api_client with a controlled fake object

B Runs the real api_client to get live data

C Skips the fetch_data function

D Changes the return value of fetch_data directly

Key Result

Mocking replaces real dependencies with controlled fakes, isolating the code under test so tests are reliable and fast.

Practice

1. Why do we use mocking in pytest when testing a function that depends on an external API?

easy

A. To make the test code longer and more complex

B. To replace the external API with a fake version so tests run faster and are reliable

C. To avoid writing any assertions in the test

D. To connect to the real API every time the test runs

Solution

1. Step 1: Understand the purpose of mocking

   Mocking replaces real parts like external APIs with fake versions to isolate the code under test.

2. Step 2: Identify benefits in testing context

   This makes tests faster and more reliable because they don't depend on the real API's availability or speed.

3. Final Answer:

   To replace the external API with a fake version so tests run faster and are reliable -> Option B

4. Quick Check:

   Mocking isolates code = Replace real parts with fakes [OK]

Hint: Mocking swaps real parts with fakes for fast, reliable tests [OK]

Common Mistakes:

• Thinking mocking makes tests more complex
• Believing mocking skips assertions
• Assuming mocking calls the real API

2. Which of the following is the correct way to mock a function get_data in pytest using the monkeypatch fixture?

easy

A. monkeypatch.setattr('module.get_data', lambda: 'mocked')

B. monkeypatch.mock('module.get_data', 'mocked')

C. monkeypatch.patch('module.get_data', lambda: 'mocked')

D. monkeypatch.replace('module.get_data', 'mocked')

Solution

1. Step 1: Recall monkeypatch syntax

   The correct method to replace an attribute is setattr with the target path and replacement function.

2. Step 2: Verify the correct usage

   monkeypatch.setattr('module.get_data', lambda: 'mocked') uses monkeypatch.setattr('module.get_data', lambda: 'mocked'), which correctly mocks the function.

3. Final Answer:

   monkeypatch.setattr('module.get_data', lambda: 'mocked') -> Option A

4. Quick Check:

   Use monkeypatch.setattr to mock functions [OK]

Hint: Use monkeypatch.setattr to replace functions in pytest [OK]

Common Mistakes:

• Using non-existent monkeypatch methods like mock or patch
• Passing string instead of a callable for function replacement
• Confusing monkeypatch with unittest.mock syntax

3. Given the code below, what will be the output of the test?

import pytest

def fetch_data():
    return 'real data'

def process():
    return fetch_data()

def test_process(monkeypatch):
    monkeypatch.setattr('__main__.fetch_data', lambda: 'mocked data')
    result = process()
    assert result == 'mocked data'
    print(result)

medium

A. real data

B. AssertionError

C. mocked data

D. NameError

Solution

1. Step 1: Understand monkeypatch effect on fetch_data

   The monkeypatch replaces fetch_data with a lambda returning 'mocked data'.

2. Step 2: Trace the process function call

   When process() calls fetch_data(), it gets 'mocked data' instead of 'real data'.

3. Final Answer:

   mocked data -> Option C

4. Quick Check:

   Mocked function returns mocked data [OK]

Hint: Monkeypatch changes function output to mocked value [OK]

Common Mistakes:

• Expecting original function output despite mocking
• Confusing assertion failure with output
• Assuming NameError due to monkeypatch

4. In the following test, what is the main reason the assertion fails?

def get_value():
    return 10

def test_value(monkeypatch):
    monkeypatch.setattr('__main__.get_value', lambda: 20)
    assert get_value() == 10

medium

A. The assertion expects the original value, but mocking changed it to 20

B. The monkeypatch syntax is incorrect

C. The function get_value is not defined

D. The test is missing a return statement

Solution

1. Step 1: Analyze the monkeypatch effect

   The monkeypatch replaces get_value to return 20 instead of 10.

2. Step 2: Check the assertion

   The assertion expects 10, but the mocked function returns 20, causing failure.

3. Final Answer:

   The assertion expects the original value, but mocking changed it to 20 -> Option A

4. Quick Check:

   Mock changes output, assertion expects old value [OK]

Hint: Assertion fails if expected value ignores mocking effect [OK]

Common Mistakes:

• Thinking monkeypatch syntax is wrong
• Assuming function is undefined
• Believing test needs a return statement

5. You have a function send_email() that sends real emails. You want to test notify_user() which calls send_email()</>. How does mocking send_email() help isolate notify_user() in pytest?

hard

A. It disables notify_user completely

B. It makes notify_user send multiple emails automatically

C. It connects notify_user to the real email server faster

D. It prevents real emails from sending and lets you test notify_user's logic only

Solution

1. Step 1: Identify the problem with real send_email

   Calling real send_email() during tests can send unwanted emails and slow tests.

2. Step 2: Understand mocking benefit

   Mocking send_email() replaces it with a fake that does nothing or simulates success, isolating notify_user() logic.

3. Final Answer:

   It prevents real emails from sending and lets you test notify_user's logic only -> Option D

4. Quick Check:

   Mocking avoids side effects, isolates tested code [OK]

Hint: Mock side-effect functions to isolate main logic in tests [OK]

Common Mistakes:

• Thinking mocking triggers real side effects
• Assuming mocking disables the tested function
• Believing mocking speeds up real connections