Debugging Strategies for Finding Errors in Your Code

When Running Code Is Not Enough

A program can run from beginning to end and still be wrong. This happens when the code follows Python's grammar but does not express the meaning you intended. Such a mistake is called a semantic error. Unlike an error that stops execution, a semantic error may produce ordinary-looking output, so you must compare that output with your expectation.

The Comma Trap

print(1,000,000)

1 0 0

The commas are not formatting marks for one large integer. They separate three integer values: 1, 0, and 0. Those values become three arguments to print(), and print() displays multiple arguments with spaces between them. The statement is syntactically valid, so Python runs it. Its meaning is nevertheless different from the intended meaning of one million.

Correct Numeric Literals

To represent one million as one integer, write the digits without commas or spaces. Python also allows underscores in numeric literals for readability. The underscores do not change the value.

print(1000000)
print(1_000_000)

1000000
1000000

A Reliable Debugging Trace

Debugging a semantic error begins with the difference between what you expected and what the program actually produced. First state the expected result. Then inspect the actual output carefully. Next, trace that output back to the statement that supplied the value or values. In the comma example, the first divergence appears in the written expression 1,000,000: the commas create separate values before print() displays anything. Finally, change the statement and run the test again to verify that the output now matches the expectation.

Mistakes That Hide in Plain Sight

• Assuming that familiar written number formatting works unchanged in Python.

  Python treats the commas as separators between values, so print() receives three arguments.

  Fix: Write print(1000000) or print(1_000_000).

• Treating successful execution as proof that the program is correct.

  Semantic errors produce no warning even though the result is incorrect.

  Fix: Always compare the actual output with the expected output.

• Looking only for error messages.

  Syntax errors are reported immediately, but semantic errors can run to completion.

  Fix: Inspect unexpected output and trace it back to the statement that produced it.

• Confusing a syntax error with a semantic error.

  A missing colon violates Python's grammar, so Python refuses to run the code.

  Fix: Separate errors that prevent execution from errors that produce an incorrect result.

Practice the Trace

print(1,000,000)

Debugging Checklist

1. A syntax error violates Python's grammar and prevents the program from running.
2. A semantic error runs without an error message but produces a result different from the intended result.
3. In Python, commas separate values; they do not format the digits of one large integer.
4. Use 1000000 or 1_000_000 for the integer one million.
5. Debug semantic errors by stating the expectation, inspecting the actual output, tracing the first divergence to its statement, correcting the code, and testing again.