Identifying and correcting program errors

What you'll learn

• How to tell the difference between syntax, runtime, and logic errors.
• How to identify the type of error from what the program does.
• How to locate the faulty line or condition using error messages, testing, and trace tables.
• How to correct the error and retest your Python 3 code.

The basic idea: bugs and debugging

A program rarely works perfectly the first time. Part of developing code is finding mistakes, understanding them, fixing them, and checking that the fix has actually worked.

For GCSE, the key skill is not just saying “there is an error”. You need to be able to:

• Identify the type of error.
• Locate where the error is in the code.
• Correct the code so it matches the intended behaviour.

Before you can debug properly, you need to know what the program is meant to do. The expected output is what should happen. The actual output is what really happens when the program runs.

A practical debugging workflow

When you run or inspect a Python program, the symptoms usually guide you towards the type of error.

The three main types of program error

Syntax error

Syntax means the grammar and punctuation rules of a programming language.

In Python, common syntax errors include:

• Missing : after an if, while, for, or function header.
• Missing quotation marks around a string.
• Unmatched brackets.
• Incorrect indentation where Python expects an indented block.

Python usually gives you an error message and a line number. The line number is a strong clue, but it may show where Python noticed the problem rather than where the problem began.

score = int(input("Score: "))
if score >= 50
    print("Pass")

score = int(input("Score: "))
if score >= 50:
    print("Pass")

Runtime errors

Runtime means the period when the program is actually executing.

A runtime error may only appear for certain inputs. For example, a program might work for 4 but crash for 0.

Common Python runtime errors include:

• ZeroDivisionError — trying to divide by zero.
• ValueError — trying to convert unsuitable data, such as int("cat").
• IndexError — trying to access a list position that does not exist.
• NameError — using a variable name that has not been defined.

A traceback is Python’s runtime error report. It shows the line where the crash happened and the type of error.

people = int(input("Number of people: "))
sweets_each = 24 // people
print(sweets_each)

people = int(input("Number of people: "))

if people > 0:
    sweets_each = 24 // people
    print(sweets_each)
else:
    print("Number of people must be more than 0")

Logic errors

Logic means the sequence of decisions, calculations, and steps used by an algorithm.

Logic errors are often the hardest to spot because Python may not display any error message. The program looks as if it worked, but the answer is wrong.

Common logic errors include:

• Using > instead of >=.
• Adding when you should subtract.
• Updating the wrong variable.
• Putting statements in the wrong order.
• Repeating a loop one time too many or one time too few.

An off-by-one error is a common logic error where a loop includes one extra value or misses one value.

Locating logic errors with dry runs and trace tables

A dry run means manually stepping through code without running it on the computer.

A trace table records the values of variables as each line of code is followed. It helps you compare what the program actually does with what you expected it to do.

def total_to(limit):
    total = 0

    for number in range(1, limit):
        total = total + number

    return total

print(total_to(5))

def total_to(limit):
    total = 0

    for number in range(1, limit + 1):
        total = total + number

    return total

print(total_to(5))

Correcting errors carefully

When you correct a program, avoid making lots of changes at once. If the program still fails afterwards, you will not know which change caused the new problem.

A good correction process is:

1. Reproduce the error using the same test data.
2. Identify the error type from the symptoms.
3. Locate the most likely line, condition, loop, or calculation.
4. Make one focused correction.
5. Retest the original case and at least one other relevant case.

Boundary data means data at the edge of what should be accepted, such as 0, 1, or a maximum allowed value. Boundary tests are especially useful for finding > / >= and loop-limit mistakes.

Typical exam tasks

In Paper 2-style programming questions, you may be asked to:

• State the type of error in a short piece of code.
• Identify the line containing the error.
• Explain why the error happens.
• Rewrite the faulty line or small section correctly.
• Use test data to show that the corrected program works.

Be precise. “There is an error in the loop” is weaker than “the loop uses range(1, limit), so it excludes limit; it should use range(1, limit + 1).”