logging-log4j-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jim Moore <jim.mo...@veritas.com>
Subject RE: JSR47 vs. log4j (take two)
Date Fri, 03 Aug 2001 17:40:59 GMT
The reason given at the presentation at JavaOne was primarily that you can't
have static methods in interfaces.  IMO, for example, it sure would be nice
if the Category and Priority classes in Log4j was an interface instead of
classes -- it sure would be easier to extend them.  Unfortunately, there are
practical "issues" with doing that, and Sun faces the same issues.

-Jim Moore

-----Original Message-----
From: Kent Yang [mailto:yang@nosc.mil]
Sent: Tuesday, July 31, 2001 9:16 PM
To: LOG4J Users Mailing List
Subject: Re: JSR47 vs. log4j (take two)

I am encourage by the latest developments.  However, I wonder why
Sun didn't just specify an interface and allow different logging backends
to be plug-in.  That way developers can decide what implementation
they want to use.  Sun did this with a lot of their APIs (Security, Sockets,
etc...)  Can this be done?  Is there some reason log4j can't be the engine
behind a specified interface?  I am still a newbie with the log4j API (I
you can plug in different appenders) but I am no log4j expert and I
don't know enough about the innerds to know whether this is possible
or not.

Mabye someone else can comment?


----- Original Message -----
From: "Ceki Gülcü" <cgu@qos.ch>
To: <log4j-user@jakarta.apache.org>
Sent: Tuesday, July 31, 2001 5:18 PM
Subject: JSR47 vs. log4j (take two)

For the latest developments on the JSR47 front you might want to read:


Ceki Gülcü - http://qos.ch

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

View raw message