Debugging in Python

Find and fix issues with interactive debuggers, logging, and systematic techniques.

Print and logging

Start with targeted prints; prefer structured logging for real apps.

import logging
logging.basicConfig(level=logging.INFO)
logging.info("value=%s", value)

Built-in debugger (pdb)

Drop into a debugging session at runtime.

import pdb; pdb.set_trace()
# or Python 3.7+: breakpoint()

Common commands:

• n next, s step in, c continue
• l list, p expr print, q quit

Post-mortem and tracing

Run post-mortem on exceptions.

import pdb, sys
try:
    risky()
except Exception:
    _, _, tb = sys.exc_info()
    pdb.post_mortem(tb)

IDE/debug tools

Use IDEs (VS Code, PyCharm) for breakpoints, watches, and step-by-step execution.

Profiling vs debugging

Don’t confuse performance issues with logic bugs; use cProfile/line_profiler to find hotspots.

Reproducing issues

• Create small, deterministic repro cases
• Control randomness with fixed seeds
• Isolate environment differences (OS, Python version, installed packages)

Summary

• Use logging for insight; pdb/breakpoint() for interactive debugging
• IDE tools help visualize state; create repros and distinguish perf from logic issues