Logging

Logging is implemented via a wrapper around Python’s logging module in src/logger.py.

API

Class: wrapped_logging(debug: bool)

• log_info(message: str): log at INFO.
• log_error(message: str): log at ERROR.
• log_warning(message: str): log at WARNING.

Behaviour:

• Default level is INFO.
• When debug=True, logging.basicConfig(filename="debug.log", filemode="w") writes a debug file locally (useful during development).

Usage

from logger import wrapped_logging

logger = wrapped_logging(False)  # use True to write debug.log locally

logger.log_info("Start address book run")
logger.log_warning("No verified emails for user: alice")
logger.log_error("Failed to write to S3: access denied")

In the Lambda handler (src/lambda_function.py), the logger is created once and passed to helpers:

logger = wrapped_logging(False)

CloudWatch Logs

• Lambda automatically forwards stdout/stderr from Python logging to CloudWatch Logs for the function.
• Use logs to track: start/end of runs, GitHub API calls, S3 writes, and errors.

Best Practices

• Prefer INFO for milestones, WARNING for recoverable anomalies, ERROR for failures.
• Add context to messages (org, bucket, key) to aid debugging.
• Avoid writing secrets or sensitive personal data to logs.
• For local dev, consider debug=True to capture a debug.log, but disable before committing.