Logging (optional)#

graphlow uses the standard library logging module. Library code in modules such as graphlow.core.mesh is wired for logging, but no handlers are attached when you import graphlow. That keeps library behavior aligned with the usual logging rule.

When to use graphlow.configure_logging()#

configure_logging() is a small convenience helper for situations where you want readable graphlow output without writing a full logging config:

  • Jupyter notebooks and interactive sessions

  • Python REPL

  • Short one-off scripts while debugging

It loads the bundled logging.conf (or a path you pass in) and sets up handlers for the graphlow logger. It does not run on import so you call it explicitly when you want that behavior.

When not to rely on it#

For production services, libraries you ship, or any project with a defined logging strategy, configure logging yourself. Do not treat configure_logging() as the backbone of your application’s logging. It is only a shortcut for interactive and exploratory use.

See also#