[jbosstools-issues] [JBoss JIRA] Updated: (JBIDE-6346) New Struts/JSF Project Wizards slowdowns while typing the name for the project

[ https://jira.jboss.org/browse/JBIDE-6346?page=com.atlassian.jira.plugin.s... ] Rob Stryker updated JBIDE-6346: ------------------------------- Attachment: JBIDE-6346.patch The following patch makes the UI a lot more efficient, and I'm in favor of 90% of the patch. I did, however, comment out 5 lines which I'm not sure why they were there at all, and it may be that someone needs to review the removal of those 5 lines. Basically, instead of using static methods that each time traverse the module tree, I added a constructor and a cache that can be used if someone wants. However the most troublesome method of the entire batch is the call to server.getRootModules(m, null); The reason this is troublesome is that if your workspace has a lot of projects, and you're asking it to find the root / parent of project A, it will get Project B, get *it's* list of children, see if they match, get project C, get *it's* list of children, see if htey match, etc. Right now, on the project level in wtp, the cache is not enabled. It is set to be enabled in 3.2.1. Angel was scared to add it in. So parsing the same project twice in a row will take a long time. It may be the call to server.getRootModules(m,null) is there for a reason, but it is not commented and I am not sure what (if any) benefit it provides. I ran some manual (non-scientific) tests with just my cache added, and not commenting out the get parent module call. My cache helps a lot, but the majority of the gains were by removing the get parent module call. Hope this helps. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira