Speaking of CLASS.class, this method of obtaining a logger will yield a different (i.e. parent) class when utilizing inheritance. Therefore, without having to change the code in the parent everytime you subclass, I recommend
Logger.getLogger(getClass()) at least, whether or not you use the string name is of no consequence.
-Kevin
Vladimir R. Bossicard wrote:
+1 on getClass().getName(), everywhere (not just the server).
I would prefer CLASS.class to take advantage of the LogFactory:getLog(Class cls) method. Why use the class's name transformed into a String when the class itself is a valid option?
-Vladimir
