Debug with small checks and logging

When output is wrong, reproduce the problem with a small input. Check intermediate values to find where the result first differs from what you expect. Temporary print() calls work for simple scripts; logging provides named severity levels for longer-running programs.

import logging

logging.basicConfig(level=logging.INFO, format="%(levelname)s: %(message)s")
sessions = [20, 25, 30]
total = 0
for minutes in sessions:
    total += minutes
    logging.info("Added %s minutes; total is %s", minutes, total)
print(f"Final total: {total}")

Final total: 75

INFO: Added 20 minutes; total is 20
INFO: Added 25 minutes; total is 45
INFO: Added 30 minutes; total is 75

By default, logging writes to standard error, so its INFO lines may appear separately from print() output. Log useful values, but avoid putting passwords or other private information in logs.

Pause in the debugger

For a closer look, save the logging example as debugging.py, add breakpoint() immediately before its final print, and run the file from an interactive terminal. At the (Pdb) prompt, type p total to inspect the current value, n to move to the next line, and c to continue. Remove the breakpoint() call after the exercise so later runs do not pause.

Why this works

A small input with a known expected total lets you compare each step. Logging records the current value and the running total as the loop progresses. Python logging writes to standard error by default, so the INFO messages are diagnostic output separate from the final print on standard output.

Practice

Change the last duration to 35. Predict the final total, run the script, and use the INFO messages to inspect each addition. The new total should be 80.

Try another case

If the three durations become 20, 25, and 35, what should the final total be?