https://issues.apache.org/bugzilla/show_bug.cgi?id=52563
--- Comment #6 from Ivan <xhh...@gmail.com> 2012-02-06 03:07:32 UTC --- Thanks for your response. I copied the comment from the r287546 ---> Update the algortithm, weighting it towards trying the context CL association first (since that's what is used now in Catalina). In some very special cases, the thread binding will be checked, as well as the parent classloader. <--- >From the comment, I am thinking that it did not state that the initial purpose is to check the classloader hierarchy firstly, and then the thread binding. Also, from the code logic now, if it really means to check the classloader hierarchy firstly, why not put those codes in the end of the get method ? >From another side, think that in most application server, including Tomcat, the context classloader should be the web application classloader, and it always has parent classloader, which may be library classloader or server runtime classloader, so the value binding on the thread will never have chance to be used, as it always throw an IllegalStateException. -- Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
