logging-log4j-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Curt Arnold <carn...@apache.org>
Subject Re: Is it me or FilterBasedTriggeringPolicy has a bug!
Date Sat, 02 Apr 2011 03:49:26 GMT
I believe that the behavior is designed to mimic the logic of the filter's in AppenderSkeleton.
 If no filter either chooses to accept or deny the request it is assumed to be accepted. 
If the desired behavior is to deny any request that is not explicitly accepted or denied,
then the last filter in the chain should be a DenyAllFilter.

FilterBasedTriggeringPolicy is final since the intended method of changing its behavior is
by changing the filters in the policy.  Marking a class final should be the default since
it is not possible to add final after a class is published without breaking compatibility,
however it is always allowable to remove final and preserve compatibility.  See Effective
Java's discussions on designing for extension or prohibiting it.


---------------------------------------------------------------------
To unsubscribe, e-mail: log4j-dev-unsubscribe@logging.apache.org
For additional commands, e-mail: log4j-dev-help@logging.apache.org


Mime
View raw message