Why monkeypatch.setattr in PyTest? - Purpose & Use Cases

The Big Idea

What if you could change how your code behaves just for tests, without touching the real code?

The Scenario

Imagine you have a function that calls another function to get data from the internet. You want to test your function without actually making the slow network call every time.

Doing this manually means changing the original code or running the real network call during tests.

The Problem

Manually changing code to avoid network calls is risky and time-consuming.

Running real network calls in tests makes tests slow and flaky because of internet issues.

It's easy to forget to undo changes, causing bugs later.

The Solution

Using monkeypatch.setattr lets you temporarily replace parts of your code during tests.

You can swap the real network call with a fake function that returns quick, predictable data.

This keeps your tests fast, reliable, and safe without changing your real code.

Before vs After

✗ Before

def test_fetch():
    # manually change code or call real network
    result = fetch_data()
    assert result == expected

✓ After

def test_fetch(monkeypatch):
    monkeypatch.setattr('module.fetch_data', lambda: 'fake data')
    result = module.fetch_data()
    assert result == 'fake data'

What It Enables

You can test code parts in isolation by swapping real functions with fake ones easily and safely.

Real Life Example

Testing a weather app without calling the real weather service every time by replacing the data fetch function with a fixed response.

Key Takeaways

Manual code changes for testing are slow and risky.

monkeypatch.setattr swaps functions temporarily during tests.

This makes tests faster, safer, and more reliable.

Practice

1. What is the main purpose of monkeypatch.setattr in pytest?

easy

A. To temporarily replace an attribute or function during a test

B. To permanently change code in the source files

C. To skip a test function

D. To run tests in parallel

Solution

1. Step 1: Understand monkeypatch.setattr usage

   It is used to temporarily replace or modify attributes or functions during a test to isolate behavior.

2. Step 2: Differentiate from other pytest features

   It does not permanently change code or skip tests; it only affects the test scope temporarily.

3. Final Answer:

   To temporarily replace an attribute or function during a test -> Option A

4. Quick Check:

   Temporary replacement = A [OK]

Hint: Think 'temporary change' during test only [OK]

Common Mistakes:

• Confusing temporary patching with permanent code changes
• Thinking it skips tests
• Assuming it runs tests in parallel

2. Which of the following is the correct syntax to patch the function get_data in module my_module using monkeypatch in pytest?

easy

A. monkeypatch.setattr('get_data', my_module, lambda: 42)

B. monkeypatch.setattr('my_module.get_data', lambda: 42)

C. monkeypatch.setattr(my_module, 'get_data', lambda: 42)

D. monkeypatch.setattr('my_module', 'get_data', 42)

Solution

1. Step 1: Identify correct monkeypatch.setattr signature

   The correct usage is monkeypatch.setattr(target_object, attribute_name, new_value).

2. Step 2: Match parameters to patch get_data function

   We pass the module object my_module, the attribute name 'get_data', and the replacement function lambda: 42.

3. Final Answer:

   monkeypatch.setattr(my_module, 'get_data', lambda: 42) -> Option C

4. Quick Check:

   Object, attribute, new value = A [OK]

Hint: Pass object, attribute name, then new value [OK]

Common Mistakes:

• Passing attribute as a string path instead of object and attribute
• Swapping parameters order
• Passing value instead of function for patching

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

import my_module

def test_func(monkeypatch):
    def fake_func():
        return 10
    monkeypatch.setattr(my_module, 'get_value', fake_func)
    result = my_module.get_value()
    print(result)

medium

A. None

B. Original value from my_module.get_value()

C. AttributeError

D. 10

Solution

1. Step 1: Understand monkeypatch effect on get_value

   The original my_module.get_value is replaced temporarily by fake_func which returns 10.

2. Step 2: Evaluate the call and print output

   Calling my_module.get_value() inside the test calls fake_func(), so it prints 10.

3. Final Answer:

   10 -> Option D

4. Quick Check:

   Patched function returns 10 [OK]

Hint: Patched function output replaces original during test [OK]

Common Mistakes:

• Expecting original function output instead of patched
• Thinking patch is permanent outside test
• Assuming print outputs None

4. What is wrong with the following pytest test using monkeypatch?

def test_example(monkeypatch):
    monkeypatch.setattr('os.path.exists', lambda x: True)
    assert os.path.exists('anyfile.txt') == True

medium

A. The lambda function should accept no arguments

B. The first argument to setattr should be the module object, not a string

C. os.path.exists cannot be patched

D. The assert statement is incorrect

Solution

1. Step 1: Check monkeypatch.setattr usage

   The first argument must be the module or object, not a string path.

2. Step 2: Identify correct usage

   It should be monkeypatch.setattr(os.path, 'exists', lambda x: True) to patch the function properly.

3. Final Answer:

   The first argument to setattr should be the module object, not a string -> Option B

4. Quick Check:

   Object, not string path = B [OK]

Hint: Pass object, not string path, as first argument [OK]

Common Mistakes:

• Passing string path instead of module object
• Incorrect lambda parameters
• Assuming some functions cannot be patched

5. You want to test a function fetch_data() that calls requests.get. How can you use monkeypatch.setattr to replace requests.get with a fake function that returns a mock response with status_code=200 and text='OK'?

hard

A. Define a fake function returning a mock object, then patch with monkeypatch.setattr(requests, 'get', fake_func)

B. Patch requests.get with a string 'mock' using monkeypatch.setattr

C. Use monkeypatch.setattr('requests.get', fake_func) with a lambda returning 200

D. Replace fetch_data with a fake function instead of patching requests.get

Solution

1. Step 1: Create a fake function returning a mock response

   Define a function that returns an object with status_code=200 and text='OK'.

2. Step 2: Patch requests.get with monkeypatch.setattr

   Use monkeypatch.setattr(requests, 'get', fake_func) to replace the real HTTP call during the test.

3. Final Answer:

   Define a fake function returning a mock object, then patch with monkeypatch.setattr(requests, 'get', fake_func) -> Option A

4. Quick Check:

   Patch requests.get with fake function = D [OK]

Hint: Patch requests.get with a fake function returning mock response [OK]

Common Mistakes:

• Passing string path instead of module object
• Patching with wrong types like strings
• Replacing the wrong function instead of requests.get