We recommend using the available can{severity}() methods to determine if we can log at a specific log level before actually writing the logging method line. This is done as best practice in order to avoid processing of messages that will never be logged anyways. So let's look at a very simple example of what NOT to do:
This will call the logger's debug() method, execute the lines of code and then the logger determines if it can be logged or not. This is ok, but we all love performance and best practice, so we encourage you to do the following:
This way, the logger determines if it can send debug log messages, and only IF IT CAN does it. This is faster and cleaner, but you will type more. Sorry!