Logging Configuration Guide

The project uses structlog for structured logging and python-json-logger for JSON output in non‑development environments.

Configuration

In development, logs are formatted as human‑readable console output.
In production/staging, logs are JSON objects compatible with log aggregators (e.g. ELK, Datadog).

Two logging middlewares are active:

CorrelationIDMiddleware
- Reads X-Correlation-ID header or generates a UUID.
- Attaches the ID to the request state and response header.
RequestLoggingMiddleware
- Logs every request with method, path, status code, duration, and correlation ID.
- Uses structlog under the hood.

Use LogContext to temporarily bind key‑value pairs to all log messages emitted within a block:

from app.logging.context import LogContext

with LogContext(user_id=42, action="update"):
    logger.info("User update started")

Import the logger:

import structlog
logger = structlog.get_logger(__name__)
logger.info("Something happened", extra_field="value")