Lazy fixtures in PyTest - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is a lazy fixture in pytest?

A lazy fixture is a fixture that is not executed immediately but only when it is actually used in a test. It helps to delay setup until necessary, saving time and resources.

beginner

How do you declare a lazy fixture in pytest?

You declare a lazy fixture by using the @pytest.fixture decorator with scope and optionally autouse=False. The fixture code runs only when a test requests it.

intermediate

Why use lazy fixtures instead of eager fixtures?

Lazy fixtures run only when needed, which saves test execution time and avoids unnecessary setup. Eager fixtures run before every test regardless of need, which can slow tests down.

beginner

What happens if a lazy fixture is not used in any test?

If a lazy fixture is not used by any test, it will not run at all. This means no setup or teardown code inside it will execute, saving resources.

intermediate

Can lazy fixtures depend on other fixtures in pytest?

Yes, lazy fixtures can depend on other fixtures. The dependent fixtures will also be lazy and run only when needed by the test.

What triggers a lazy fixture to run in pytest?

A Before any test runs automatically

B When the fixture is declared

C When pytest starts

D When a test function requests it as a parameter

Lazy fixtures run only when a test function includes them as parameters.

Which decorator is used to create a lazy fixture in pytest?

A @pytest.fixture

B @pytest.test

C @pytest.lazy

D @pytest.setup

The @pytest.fixture decorator creates fixtures, which are lazy by default unless autouse=True.

If a fixture is marked with autouse=True, is it lazy?

A No, it runs before every test automatically

B No, it never runs

C Yes, but only for some tests

D Yes, it runs only when requested

Fixtures with autouse=True run automatically before each test, so they are not lazy.

What is a benefit of using lazy fixtures?

A They run faster because they run before all tests

B They reduce test time by running only when needed

C They run multiple times per test

D They replace test functions

Lazy fixtures save time by running only when a test actually uses them.

Can a lazy fixture depend on another fixture?

A Only if both are autouse

B No, fixtures cannot depend on each other

C Yes, dependencies are also lazy

D Only if declared in the same file

Fixtures can depend on other fixtures, and those dependencies are lazy too.

Explain what lazy fixtures are in pytest and why they are useful.

Think about when the fixture code runs during testing.

Describe how fixture dependencies work with lazy fixtures in pytest.

Consider how pytest manages fixture calls.

Practice

1. What is the main benefit of using lazy_fixture in pytest tests?

easy

A. It automatically retries failed tests using the fixture.

B. It runs all fixtures before any test starts, ensuring setup order.

C. It delays fixture setup until the test actually needs it, improving speed.

D. It converts fixtures into global variables accessible everywhere.

Solution

1. Step 1: Understand lazy_fixture purpose

   The lazy_fixture delays the setup of a fixture until the test that uses it actually runs, avoiding unnecessary setup.

2. Step 2: Compare options to this behavior

   Only It delays fixture setup until the test actually needs it, improving speed. correctly describes this benefit. Other options describe unrelated behaviors.

3. Final Answer:

   It delays fixture setup until the test actually needs it, improving speed. -> Option C

4. Quick Check:

   lazy_fixture delays setup = A [OK]

Hint: Lazy fixtures delay setup until needed, saving time [OK]

Common Mistakes:

• Thinking lazy_fixture runs all fixtures upfront
• Confusing lazy_fixture with test retries
• Assuming lazy_fixture makes fixtures global

2. Which of the following is the correct way to use lazy_fixture inside pytest.mark.parametrize?

easy

A. pytest.mark.parametrize('data', lazy_fixture('my_fixture'))

B. pytest.mark.parametrize('data', lazy_fixture(['my_fixture']))

C. pytest.mark.parametrize('data', ['lazy_fixture(my_fixture)'])

D. pytest.mark.parametrize('data', [lazy_fixture('my_fixture')])

Solution

1. Step 1: Recall correct syntax for lazy_fixture usage

   The lazy_fixture function must be called inside the list of parameters passed to pytest.mark.parametrize, like [lazy_fixture('fixture_name')].

2. Step 2: Check each option's syntax

   pytest.mark.parametrize('data', [lazy_fixture('my_fixture')]) correctly wraps lazy_fixture('my_fixture') inside a list. Options A and D misuse the function call or argument types. pytest.mark.parametrize('data', ['lazy_fixture(my_fixture)']) treats it as a string, which is incorrect.

3. Final Answer:

   pytest.mark.parametrize('data', [lazy_fixture('my_fixture')]) -> Option D

4. Quick Check:

   lazy_fixture inside list = B [OK]

Hint: Use lazy_fixture inside a list in parametrize [OK]

Common Mistakes:

• Passing lazy_fixture call directly without list
• Using string quotes around lazy_fixture call
• Passing list inside lazy_fixture instead of outside

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

import pytest
from pytest_lazyfixture import lazy_fixture

@pytest.fixture
def number():
    print('Setup number')
    return 42

@pytest.mark.parametrize('value', [lazy_fixture('number')])
def test_value(value):
    print(f'Test got {value}')
    assert value == 42

medium

A. Setup number\nTest got 42

B. Test got 42\nSetup number

C. Setup number only

D. No output, test skipped

Solution

1. Step 1: Understand lazy_fixture execution timing

   The fixture number is only set up when the test runs because of lazy_fixture. So 'Setup number' prints before the test body.

2. Step 2: Trace test execution output

   First, 'Setup number' prints from fixture setup, then 'Test got 42' prints from the test function. The assertion passes.

3. Final Answer:

   Setup number Test got 42 -> Option A

4. Quick Check:

   Fixture setup before test print = A [OK]

Hint: Fixture prints before test body when lazy_fixture used [OK]

Common Mistakes:

• Assuming test prints before fixture setup
• Thinking fixture runs before parametrize
• Expecting no output due to lazy_fixture

4. Identify the error in the following pytest code using lazy_fixture:

import pytest
from pytest_lazyfixture import lazy_fixture

@pytest.fixture
def data():
    return [1, 2, 3]

@pytest.mark.parametrize('input', lazy_fixture('data'))
def test_sum(input):
    assert sum(input) == 6

medium

A. The test function must not use parameter named 'input'.

B. lazy_fixture must be inside a list in parametrize, not passed directly.

C. Fixture 'data' must return a single integer, not a list.

D. lazy_fixture cannot be used with parametrize.

Solution

1. Step 1: Check lazy_fixture usage in parametrize

   The lazy_fixture call must be inside a list when passed to pytest.mark.parametrize. Here it is passed directly, which is incorrect syntax.

2. Step 2: Validate other code parts

   The fixture returns a list correctly, parameter name 'input' is allowed, and lazy_fixture is designed for parametrize usage.

3. Final Answer:

   lazy_fixture must be inside a list in parametrize, not passed directly. -> Option B

4. Quick Check:

   lazy_fixture inside list required = C [OK]

Hint: Wrap lazy_fixture call in a list for parametrize [OK]

Common Mistakes:

• Passing lazy_fixture call directly without list
• Misunderstanding fixture return types
• Thinking parameter names are restricted

5. You want to test a function with two different fixtures, but only one fixture should be set up per test run to save time. How can you use lazy_fixture with pytest.mark.parametrize to achieve this?

hard

A. Use pytest.mark.parametrize('arg', [lazy_fixture('fix1'), lazy_fixture('fix2')]) so only the needed fixture runs per test.

B. Call both fixtures inside the test and use lazy_fixture to delay both setups.

C. Use pytest.mark.parametrize with a list of fixture names as strings, then call them inside the test.

D. Set both fixtures as autouse=True to run only one at a time.

Solution

1. Step 1: Understand lazy_fixture with parametrize

   Using lazy_fixture inside the list passed to pytest.mark.parametrize allows pytest to run only the fixture needed for each test case.

2. Step 2: Evaluate options for efficiency

   Use pytest.mark.parametrize('arg', [lazy_fixture('fix1'), lazy_fixture('fix2')]) so only the needed fixture runs per test. correctly uses lazy_fixture in the parametrize list, so only one fixture runs per test. Call both fixtures inside the test and use lazy_fixture to delay both setups. runs both fixtures regardless. Use pytest.mark.parametrize with a list of fixture names as strings, then call them inside the test. uses strings incorrectly. Set both fixtures as autouse=True to run only one at a time. misuses autouse and does not control fixture setup per test.

3. Final Answer:

   Use pytest.mark.parametrize('arg', [lazy_fixture('fix1'), lazy_fixture('fix2')]) so only the needed fixture runs per test. -> Option A

4. Quick Check:

   lazy_fixture in parametrize list runs one fixture per test = D [OK]

Hint: Parametrize with lazy_fixture list to run one fixture per test [OK]

Common Mistakes:

• Calling both fixtures inside test causing both setups
• Passing fixture names as strings without lazy_fixture
• Misusing autouse to control fixture runs