tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Remy Maucherat <r...@apache.org>
Subject Re: cvs commit: jakarta-tomcat-catalina/catalina/src/share/org/apache/catalina/loader WebappClassLoader.java
Date Wed, 28 Sep 2005 20:43:40 GMT
Jan Luehe wrote:
> No, I did.

Cool, there's one, at least :)

> Yes, but with lazy resolution, it will be loaded when the
> IntrospectionUtils symbol is first encountered, which may
> be inside WebappClassLoader.stop().

Normally, it's used by plenty of things, like the digester. Who knows 
anyway, I find the classloading behavior to be weird sometimes.

>>So I am ok with your fix, but I don't understand how it can occur 
>>in regular Tomcat.
> 
> It's probably not occurring in standalone Tomcat, but only
> in "embedded" Tomcat. In standalone Tomcat, IntrospectionUtils is
> probably getting resolved (and its static initializer invoked)
> prior to calling WebappClassLoader.stop(),
> whereas in "embedded" mode, IntrospectionUtils is first referenced
> and loaded by WebappClassLoader.stop().
> 
>>The big comment block is quite pointless, as it tries to be 
>>informational, but doesn't correspond to reality (I am personally 
>>against this kind of "commit message duplication" comment).
> 
> Sure, I just thought this line might be an easy candidate for
> being moved around if there was no comment.

Stuff won't get moved around for no reason ;)

It would actually be good to move the 3 cleanup calls to 
StandardContext.stop.

Rémy

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


Mime
View raw message