Mock return values and side effects in PyTest - Build an Automation Script

Mock a function to test return values and side effects

Preconditions (2)

Step 1: Create a mock for the external service function

Step 2: Set the mock to return a specific value when called

Step 3: Call the main function that uses the external service

Step 4: Verify the main function returns the mocked value

Step 5: Set the mock to raise an exception as a side effect

Step 6: Call the main function again and verify it handles the exception

✅ Expected Result: The test passes verifying the function returns the mocked value and correctly handles the side effect exception

Automation Requirements - pytest with unittest.mock

Assertions Needed:

Assert the function returns the mocked return value

Assert the function handles the side effect exception properly

Best Practices:

Use patch decorator or context manager to mock dependencies

Set return_value and side_effect on mocks explicitly

Keep tests isolated and independent

Use clear and descriptive assertion messages

Automated Solution

PyTest

from unittest.mock import patch
import pytest

# Example external service function to be mocked
def external_service():
    # Imagine this calls a real external API
    return "real data"

# Function under test that uses the external service
def process_data():
    try:
        data = external_service()
        return f"Processed {data}"
    except Exception:
        return "Error handled"

@patch('__main__.external_service')
def test_process_data(mock_service):
    # Mock return value
    mock_service.return_value = "mocked data"
    result = process_data()
    assert result == "Processed mocked data", f"Expected 'Processed mocked data' but got {result}"

    # Mock side effect to raise exception
    mock_service.side_effect = Exception("Service failure")
    result = process_data()
    assert result == "Error handled", f"Expected 'Error handled' but got {result}"

This test uses patch to replace external_service with a mock.

First, it sets return_value to simulate a successful call returning "mocked data". The test calls process_data and asserts the returned string includes the mocked data.

Next, it sets side_effect to raise an exception to simulate a failure. The test calls process_data again and asserts it returns the error handling message.

This approach isolates the function from the real external service, allowing controlled testing of different scenarios.

Common Mistakes - 3 Pitfalls

Not using patch to mock the external function

Setting return_value but forgetting to reset side_effect

{'mistake': 'Asserting on the mock instead of the function output', 'why_bad': 'Tests do not verify actual function behavior, only mock calls', 'correct_approach': "Assert on the function's return value or side effects, not just mock calls"}

Bonus Challenge

Now add data-driven testing with 3 different mocked return values and verify the function output for each

Show Hint

Practice

1. What does setting mock.return_value do in a pytest mock object?

easy

A. It delays the mock response by a set time.

B. It defines a fixed value that the mock returns when called.

C. It records the number of times the mock was called.

D. It raises an exception when the mock is called.

Solution

1. Step 1: Understand the role of return_value

   The return_value attribute sets a fixed return value for the mock when it is called.

2. Step 2: Differentiate from other mock features

   Unlike side_effect which can raise exceptions or return different values, return_value always returns the same fixed value.

3. Final Answer:

   It defines a fixed value that the mock returns when called. -> Option B

4. Quick Check:

   return_value = fixed return [OK]

Hint: return_value sets fixed output for mock calls [OK]

Common Mistakes:

• Confusing return_value with side_effect
• Thinking return_value raises exceptions
• Assuming return_value tracks call count

2. Which of the following is the correct way to set a mock to raise an exception when called?

easy

A. mock.side_effect = Exception('Error')

B. mock.return_value = Exception('Error')

C. mock.raise_exception = Exception('Error')

D. mock.throw = Exception('Error')

Solution

1. Step 1: Identify how to raise exceptions with mocks

   Setting side_effect to an exception causes the mock to raise that exception when called.

2. Step 2: Check syntax correctness

   Only mock.side_effect = Exception('Error') is valid syntax to raise an exception on call.

3. Final Answer:

   mock.side_effect = Exception('Error') -> Option A

4. Quick Check:

   side_effect raises exceptions [OK]

Hint: Use side_effect to raise exceptions in mocks [OK]

Common Mistakes:

• Using return_value to raise exceptions
• Using non-existent attributes like raise_exception
• Confusing syntax for exception raising

3. What will be the output of this pytest mock code?

from unittest.mock import Mock
mock = Mock()
mock.side_effect = [1, 2, 3]
print(mock())
print(mock())
print(mock())

medium

A. 1 2 3

B. [1, 2, 3] [1, 2, 3] [1, 2, 3]

C. None None None

D. Error on third call

Solution

1. Step 1: Understand side_effect as a list

   When side_effect is a list, each call to the mock returns the next item in the list.

2. Step 2: Trace each call output

   First call returns 1, second returns 2, third returns 3, so printed outputs are 1, 2, 3.

3. Final Answer:

   1 2 3 -> Option A

4. Quick Check:

   side_effect list returns items in order [OK]

Hint: side_effect list returns next item each call [OK]

Common Mistakes:

• Expecting entire list returned each call
• Confusing return_value with side_effect list
• Assuming calls after list ends raise error

4. Identify the error in this pytest mock setup:

from unittest.mock import Mock
mock = Mock()
mock.side_effect = Exception('Fail')
mock.side_effect = [1, 2, 3]
print(mock())

medium

A. side_effect cannot be assigned twice

B. Exception object should be raised, not assigned to side_effect

C. The second assignment overwrites the first, no error occurs

D. Mock must be called before setting side_effect

Solution

1. Step 1: Analyze multiple side_effect assignments

   The second assignment to side_effect replaces the first one; no error occurs.

2. Step 2: Understand mock call behavior

   After setting side_effect = [1, 2, 3], the mock returns 1 on first call, so print(mock()) outputs 1.

3. Final Answer:

   The second assignment overwrites the first, no error occurs -> Option C

4. Quick Check:

   Last side_effect assignment wins [OK]

Hint: Last side_effect assignment replaces previous ones [OK]

Common Mistakes:

• Thinking multiple side_effect assignments cause errors
• Expecting exception to be raised immediately
• Confusing assignment with raising exceptions

5. You want a mock to return 10 on the first call, raise a ValueError on the second, and return 20 on the third call. Which setup achieves this?

hard

A. mock.side_effect = [10, ValueError('bad'), 20]

B. mock.return_value = 10
mock.side_effect = ValueError('bad')
mock.return_value = 20

C. mock.side_effect = lambda: [10, ValueError('bad'), 20]

D. mock.side_effect = [10, lambda: (_ for _ in ()).throw(ValueError('bad')), 20]

Solution

1. Step 1: Understand side_effect list behavior

   When side_effect is a list, the mock returns successive items from the list on each call. Non-callable items like numbers or exception instances are returned directly.

2. Step 2: Recognize exception raising in side_effect list

   Placing an exception instance directly returns it as a value, not raises it. To raise, use a generator expression that throws the exception.

3. Step 3: Analyze options

   mock.side_effect = [10, ValueError('bad'), 20] returns the exception object instead of raising it. mock.side_effect = [10, lambda: (_ for _ in ()).throw(ValueError('bad')), 20] uses a generator expression to raise ValueError on second call, returning 10 and 20 on first and third calls respectively.

4. Final Answer:

   mock.side_effect = [10, lambda: (_ for _ in ()).throw(ValueError('bad')), 20] -> Option D

5. Quick Check:

   Use generator throw to raise exceptions in side_effect list [OK]

Hint: Use generator throw to raise exceptions in side_effect list [OK]

Common Mistakes:

• Putting exception instance directly in side_effect list
• Trying to set multiple return_values
• Using lambda returning list instead of generator throw