Python Logging

The logging module is Python's built-in solution for recording program events. Replacing print() with logging gives you control over what gets recorded, where it goes, and how it's formatted — without changing your code.

Why Use logging Instead of print()

print() goes to stdout and has no built-in way to filter by severity, route to files, or disable in production. The logging module provides all of this:

Log Levels

Python defines five standard log levels, in increasing severity:

logging.basicConfig()

The simplest setup — configures the root logger. Call this once, early in your program:

Important: basicConfig() only works if the root logger has no handlers yet. It has no effect if called after logging has already been configured.

Format String Variables

logging.getLogger(name)

In any module, get a logger named after the module:

Using __name__ creates a hierarchy that mirrors your package structure. You can configure logging for myapp and it automatically applies to myapp.database, myapp.api, etc.

Handlers — Where Logs Go

A handler sends log records to a destination. One logger can have multiple handlers:

Formatters — How Logs Look

Attach a Formatter to a handler to control the output format:

Logging Exceptions

Log exceptions with their full traceback:

Logger Propagation

By default, log records propagate up to parent loggers. The root logger is the ultimate parent:

Logging Context with extra={}

Pass extra fields to add contextual information to log records:

Best Practices

1. Never use print() in library code — it pollutes user output. Use logging.
2. Use logging.getLogger(__name__) — not the root logger directly.
3. Set levels on handlers, not messages — one codebase, multiple deployment configurations.
4. Log the why, not the what — "Connection failed after 3 retries" not "Error occurred".
5. Use lazy formatting: logger.debug("Value: %s", expensive_repr) — the string isn't formatted if DEBUG is disabled.