Best‑Practices & Debugging

Best‑Practices & Debugging

Even after you know how to write code, writing good code and fast debugging are what make a professional developer.

1️⃣ Code‑style & Linting

• PEP 8 is the official style guide. Enforce it with flake8.
• Black automatically formats code (black .).
• isort sorts imports (isort .).

pip install flake8 black isort
flake8 src/   # shows style warnings
black src/    # rewrites files in‑place
isort src/   # tidy import order

2️⃣ Debugging with pdb

Insert breakpoint() (Python 3.7+) anywhere:

def compute(x):
    breakpoint()   # drops you into the REPL
    return x * 2

You can inspect variables, step (n), continue (c) or quit (q).

For a richer experience install ipdb:

pip install ipdb

Replace breakpoint() with import ipdb; ipdb.set_trace().

3️⃣ Profiling & Performance

• cProfile gives you a call‑graph of where time is spent.

python -m cProfile -s cumulative myscript.py

• For line‑by‑line profiling use line_profiler.

pip install line_profiler
kernprof -l -v myscript.py

4️⃣ Logging vs. print

import logging
logging.basicConfig(level=logging.INFO,
                    format='%(asctime)s - %(levelname)s - %(message)s')

logging.debug('Debug details')
logging.info('Normal operation')
logging.warning('Something odd')
logging.error('Error occurred')

You can switch the level without editing code, pipe logs to a file, and add rotating file handlers for production.

5️⃣ Checklist for Production‑Ready Code

• PEP 8 compliance (flake8/black).
• Typed where possible (mypy --strict).
• Unit tests with ≥80 % coverage (pytest --cov).
• Logging set to INFO+, with optional DEBUG flag.
• No hard‑coded secrets – use environment variables or .env files (python‑dotenv).
• CI (GitHub Actions, GitLab CI) runs lint, tests and type‑check on every PR.

Following this checklist will keep your code clean, maintainable, and easier to debug when things go wrong.