Hands-on experience on Python logging

07.01.2022 — 2 min read

import logging

logging.basicConfig()
logging.info("messages")

What does logging.basicConfig() do? => creates a StreamHandler with a default Formatter and adding it to the root logger.

*LogRecord
*Filter
- Both Handler and Logger can add filters
Formatter: different formatters for different handlers
Handler
- StreamHandler
- FileHandler
Logger
- can link to multiple (shared) handlers (many-to-many relation)
- has a manager
See Logging Flow for details

Multiple Loggers arranged in a namespace hierarchy using dots as separators.
- When propagate is true, foo.bar logger will propagate the messages to its parent, the foo logger.
Multiple destinations

For simple projects/scripts, simply use the root logger, i.e. config via logging.basicConfig() and call logging.xxx("message")
For packages/libraries, create logger under the package name, but do not add any handler by default. The idea is to let users to decide the way to log. If the package can be used via cli, then assign wanted handlers inside __main__ .
- find more at It is strongly advised that you do not add any handlers other than NullHandler to your library’s loggers. This is because the configuration of handlers is the prerogative of the application developer who uses your library.
For complex projects, create different handers based on our needs, and add them to the named loggers, e.g. current project name, third party packages, etc.