At 10:03 AM 8/25/2003 -0500, Jacob Kjome wrote:

Well, given the fact that I use Tomcat exclusively and haven't used Tomcat-3.x.x for a couple of years, this doesn't matter to me in the least. Heck, tomcat-5 supporting Servlet-2.4 pretty stable and once the 2.4 spec is finalized, Tomcat-5 will be released officially. Sometimes people get stuck on old technology. Let's move forward. Anyway, if someone wanted to set this up at the container level and register context's with the servlet context log appender, then there would be no need for a servlet context listener in the first place. However, a servlet context listener provides one the ability to do this without having to wait for the container to agree to it. It is merely a convenience.

I mentioned the 2.3 spec to say that ServletContextListener was a relativelyrecent addition.


[snip]

Isn't o.a.l.appender too generic?

Maybe. What do you suggest? We have o.a.l.servlet and o.a.l.selector in log4j-sandbox. I was just going with that naming scheme.

I suggest o.a.l.servlet for both files.


As you can see, I changed the package and the names of your classes to be a better fit for the log4j-sandbox. Let me know if you think they should be different.

If no one else has objects to these classes going into the log4j-sandbox, I'll check them in as-is. Let me know if any changes to packages, class names, or design are desired.

Please do. Thank you.

yep, after we finalize the packages and class names and any other details that are pertinent.

I think the only minor issue was the package names.


Oh, and should Aleksei get committer access to the log4j-sandbox so he can modify/update his contribution? There isn't any explanation of how to use this stuff in the Javadoc. The extra info he provided at the top of his email should probably be added to the Javadoc of these classes. Giving Aleksei committer access would allow him to add the appropriate documentation after this stuff has been committed without having to resubmit modifications.

It is still a bit too early to grant Aleksei committer access to the sand-box. Assuming Aleksei continues to contribute, then by all means...

Yeah, I probably jumped the gun there but, then again, it is the sandbox and not log4j proper.

Indeed. I was thinking of the infrastructure guys who have to maintain all the accounts. It is probably better to wait a little before creating accounts in order to filter users who do not use their commit privileges.


Jake

-- Ceki For log4j documentation consider "The complete log4j manual" ISBN: 2970036908 http://www.qos.ch/shop/products/clm_t.jsp




--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to