Why test?

A check you write once

Lesson 5.8 turned that table into a loop: eight scores, each with the letter it should get, and a line that printed only when an answer was wrong. That loop was a testtestCode that runs your code on an input you chose and compares the result with the answer it should give.: code that runs your code on an input you chose and compares the result with the right answer.

Python has a statement for the comparing part. assert is followed by a condition: if it's true, nothing happens; if it's false, the program stops with an error. Each line is an assertionassertionA statement that something must be true, like assert grade_for(85) == 'A'. When it's false, Python raises AssertionError.. grades.py is lesson 5.8's grade_for in its own file, imported as in lesson 8.5:

from grades import grade_for

assert grade_for(90) == "A"
assert grade_for(85) == "A"
assert grade_for(49) == "F"
print("All checks passed")

def grade_for(score):
    if score >= 85:
        return "A"
    elif score >= 70:
        return "B"
    elif score >= 50:
        return "C"
    else:
        return "F"

Three true assertions print nothing, so the last line runs. In lesson 5.8 you broke grade_for on purpose, changing >= 85 to > 85. Here's that version. Predict what the assertions show:

from grades import grade_for

assert grade_for(90) == "A"
assert grade_for(85) == "A"
print("All checks passed")

def grade_for(score):
    if score > 85:
        return "A"
    elif score >= 70:
        return "B"
    elif score >= 50:
        return "C"
    else:
        return "F"

Traceback (most recent call last):
  File "main.py", line 4, in <module>
    assert grade_for(85) == "A"
           ^^^^^^^^^^^^^^^^^^^^
AssertionError

Read the last line first (lesson 7.1): just AssertionError. If you expected something like got B, expected A, that's the most common guess, and plain assert doesn't give it. You learn which check failed, then you'd need a print to see what grade_for(85) returned. The next lesson's tool shows both values for you.

A bug that comes back

Change > 85 back to >= 85 and the assertion passes. Now keep it. Months later you tidy grades.py, type > again by accident, and the bug is back. That has a name: a regressionregressionA bug that comes back after it was fixed, usually when other code changes..

A test you keep runs every time you run your tests, so the bug it checks can't come back quietly. Try it in the first runner: change >= 85 to > 85 in the grades.py tab and run. Line 4 stops it with AssertionError.

In lesson 7.4's six steps, a test is step 4, "check the guess", written down instead of done once. A failing test also does step 1, reproducing the bug on every run. And it does step 6 for you: the check stays in the file, so remembering is the computer's job.

One mistake makes a test that can never fail. You can add a message after a comma, assert condition, "message", and it's shown after AssertionError:. Brackets around both turn them into one tuple (lesson 4.3):

What a passing test tells you

Here's grade_for with a different bug. Every assertion passes:

from grades import grade_for

assert grade_for(90) == "A"
assert grade_for(85) == "A"
assert grade_for(75) == "B"
assert grade_for(49) == "F"
print("All checks passed")

def grade_for(score):
    if score >= 85:
        return "A"
    elif score > 70:
        return "B"
    elif score >= 50:
        return "C"
    else:
        return "F"

A score of 70 gets a C, and no assertion tried 70. Add assert grade_for(70) == "B" above the print and run: now it fails.