Approximate comparisons (pytest.approx) - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of approximate float comparison with pytest.approx

What is the result of running this pytest assertion code snippet?

PyTest

import pytest

def test_approx():
    assert 0.1 + 0.2 == pytest.approx(0.3)

A Test passes (no assertion error)

B AssertionError: 0.1 + 0.2 is not approximately equal to 0.3

C SyntaxError due to incorrect pytest.approx usage

D TypeError because pytest.approx cannot compare floats

💡 Hint

Remember that floating point arithmetic can cause small precision errors, pytest.approx helps with that.

The sum 0.1 + 0.2 is not exactly 0.3 due to floating point precision, but pytest.approx allows approximate equality, so the assertion passes.

❓ assertion

intermediate

Choosing the correct pytest.approx usage for relative tolerance

Which assertion correctly tests that variable x is approximately 10 with a relative tolerance of 1%?

A assert x == pytest.approx(10, rel=1)

B assert x == pytest.approx(10, rel=0.01)

C assert x == pytest.approx(10, abs=0.01)

D assert x == pytest.approx(10, tol=0.01)

💡 Hint

Check the parameter name for relative tolerance in pytest.approx.

The rel parameter sets relative tolerance. abs sets absolute tolerance. tol is not a valid parameter.

🔧 Debug

advanced

Identify the error in pytest.approx usage

What error will this test raise when run?

PyTest

import pytest

def test_value():
    assert 5 == pytest.approx(5, rel=-0.1)

A Test passes without error

B AssertionError because 5 is not approximately equal to 5 with negative tolerance

C TypeError due to invalid argument type

D ValueError: relative tolerance must be non-negative

💡 Hint

Check the allowed range for relative tolerance values in pytest.approx.

pytest.approx requires rel to be zero or positive. Negative values cause ValueError.

🧠 Conceptual

advanced

Understanding pytest.approx behavior with sequences

Given two lists a = [1.0, 2.0, 3.0] and b = [1.0, 2.01, 3.0], which pytest assertion will pass?

PyTest

import pytest
a = [1.0, 2.0, 3.0]
b = [1.0, 2.01, 3.0]

A assert a == pytest.approx(b, rel=0.005)

B assert a == pytest.approx(b, abs=0.005)

C assert a == pytest.approx(b, abs=0.02)

D assert a == pytest.approx(b, rel=0.01)

💡 Hint

Consider the difference between 2.0 and 2.01 and the tolerance values.

Difference is 0.01. abs=0.02 allows difference up to 0.02, so assertion passes only with option C.

❓ framework

expert

Configuring pytest to globally change default tolerance for approx

How can you globally set pytest.approx to use a relative tolerance of 0.001 for all tests without changing each assertion?

A Override pytest.approx in conftest.py to wrap original approx with rel=0.001

B Use a pytest fixture that monkeypatches pytest.approx to set rel=0.001 by default

C Set an environment variable PYTEST_APPROX_REL=0.001 before running tests

D Create a pytest.ini file with [pytest] and add approx-rel = 0.001

💡 Hint

Think about how to customize pytest behavior programmatically.

pytest does not support config files or env vars for approx tolerance. Overriding approx in conftest.py is the way to globally change behavior.