Day 43 of 50 Days of Python: Logging & Debugging Best Practices

Part of Week 7: Python in Production

Welcome back to the series! Today we’ll be covering logging and debugging techniques to improve your error handling and resolving best practices.

---

What We’ll Cover

→ Why using logging over print() is best practice.

→ Tips for reducing log noise and protecting sensitive data.

→ Structured logging with structlog and FastAPI middleware.

---

Key Concepts

→ Level: The importance of messages, which can range from INFO, DEBUG, CRITICAL, ERROR.

→ Handler: Where logs end up.

→ Formatter: What the log structure lookslike.

→ Structured: Making sure they are machine-parseable (JSON)

→ ID: Unique ID of a request/job to link back to a log.

---

Hands‑On: Configure Logging

# logging_config.py
import logging.config

LOGGING = {
    "version": 1,
    "disable_existing_loggers": False,
    "formatters": {
        "default": {
            "format": "%(asctime)s %(levelname)s [%(name)s] %(message)s",
        },
    },
    "handlers": {
        "console": {
            "class": "logging.StreamHandler",
            "formatter": "default",
        },
        "file": {
            "class": "logging.handlers.RotatingFileHandler",
            "filename": "logs/app.log",
            "maxBytes": 10_000_000,   # 10 MB
            "backupCount": 5,
            "formatter": "default",
        },
    },
    "root": {"level": "INFO", "handlers": ["console", "file"]},
}

logging.config.dictConfig(LOGGING)
logger = logging.getLogger(__name__)
logger.info("Logging ready...")

Structured Logging with structlog

import structlog, logging
logging.basicConfig(format="%(message)s", level="INFO")
structlog.configure(
    wrapper_class=structlog.make_filtering_bound_logger(logging.INFO),
    processors=[
        structlog.processors.UUID4Adder("corr_id"),
        structlog.processors.JSONRenderer(),
    ],
)
log = structlog.get_logger()

log.info("started", service="calihouse-api")

Output

{"event":"started","service":"calihouse-api","corr_id":"b3f6...","timestamp":"2025-07-02T09:15:12.345Z"}

FastAPI Middleware for Correlation IDs

from fastapi import FastAPI, Request
import uuid, structlog

app = FastAPI()
log = structlog.get_logger()

@app.middleware("http")
async def add_correlation_id(request: Request, call_next):
    cid = request.headers.get("x-correlation-id", str(uuid.uuid4()))
    structlog.contextvars.bind_contextvars(correlation_id=cid)
    response = await call_next(request)
    response.headers["x-correlation-id"] = cid
    return response

Debugging with pdp and VS Code

# drop into pdb REPL
import pdb; pdb.set_trace()

import debugpy

debugpy.listen(("0.0.0.0", 5678))
debugpy.wait_for_client()

Then → VS Code > Run & Debug > Python: Remote Attach.

---

---

logging.info() is something you’ll likely use at every point. If you want to see how many rows you’ve inserted, then you would save that metric to a rows = df.count() and then in your logging.info(f”Rows inserted: {rows})

The main reasoning for using this over print is because it doesn’t print it in memory, it stores it within a file located elsewhere. The more prints you do in a run the less efficient the process becomes. You want your process doing only the necessary writes and reads, nothing more.

---

TL;DR

• Configure logs with dictConfig, rotate files, and set root level to INFO.

• Structure logs via structlog → JSON for painless searching.

• Trace requests by injecting correlation‑IDs in middleware.

• Debug quickly with pdb, ipdb, or VS Code remote attach.

---

Next Up: Day 44 - Writing Unit and Integration Tests in Python.

I’ll explain the reasoning behind unit and integration tests. As well as help you understand how they bolster your CI/CD pipelines.

See you for the next one and as always… Happy coding!