How do I run one unittest test while debugging a function?

Give unittest the test’s fully qualified name: python -m unittest test_pricing.TestPrice.test_discount. For your example, that asks unittest to run test_discount from TestPrice in the importable test_pricing module. This is the most direct choice when you know the exact method. The documented command-line form supports selecting a module, class, or individual method. [S1]

While debugging, keep the test focused: change one part of the function, run that command again, and compare the result with what the test expects. For instance, if apply_discount is the function under investigation, the test can call it with a small, fixed input and check the expected result. The names in the command must match your actual test module, class, and method; the example names here are placeholders.

Use -k with a substring or pattern to select tests whose fully qualified names match it. For example: python -m unittest -k discount. This can select matching tests without spelling out one complete dotted method name. [S1]

A practical approach is to choose a distinctive term, rather than a very broad one that might match unrelated tests. If the selection doesn’t identify the test you intended, check the test names in your code and use the exact dotted form instead. The command selects by names; it does not establish that the chosen test covers every behavior of the function.

Here is an illustrative example; adapt the function’s inputs and expected result to your own requirements:

import unittest
from pricing import apply_discount

class TestPrice(unittest.TestCase):
    def test_discount(self):
        self.assertEqual(apply_discount(100, 10), 90)

Assuming the module can be imported from your project’s test-running location, run python -m unittest test_pricing.TestPrice.test_discount. The example is not a claim about how your function should behave: 100, 10, and 90 are placeholders for a case whose expected result you have decided.

If the test fails, use that small case to inspect the inputs, the function’s intermediate calculations, and the actual versus expected result. Make one change at a time and rerun the same command. A passing focused test tells you that this selected case passes; by itself, it does not show that other cases or other parts of the project are correct.

Run python -m unittest to use discovery, or explicitly run python -m unittest discover. These are the documented options for returning to a broader suite. [S1] The available commands depend on importable test modules and the project layout, so if discovery does not find your tests, check how your project organizes and imports them. [S1]

A reusable sequence is: run one named test; investigate with a small, understood case; rerun that test after a change; then run discovery before treating the investigation as finished. Keep the focused command handy, but don’t treat it as a replacement for the broader check.