[ 
https://issues.apache.org/jira/browse/GERONIMODEVTOOLS-580?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12722519#action_12722519
 ] 

David Jencks commented on GERONIMODEVTOOLS-580:
-----------------------------------------------

A really really long time ago we had the idea of enhancing the in-place 
deployment so that it didn't need a properly structured javaee app but instead 
had some kind of location map that could point into the internal structure of 
an eclipse project (or idea, or maven....)  I think the original GEP developer 
worked on this a bit but I don't know how far he got.

If we can do this then jasper ought to be able to notice whenever a jsp is 
changed and recompile it.

For classes, I think I heard there is a java6 feature that lets you relead 
modified classes.  This obviously won't help if you add or remove annotations 
or interface members but for small code changes it ought to be able to help 
speed up debugging.  I have no idea how practical this class reloading actually 
is :-)

Hot deploy won't help as it is just a slow unreliable way of triggering a 
normal full deployment.

> Servlet/JSP update forces a complete redeploy of a WAR
> ------------------------------------------------------
>
>                 Key: GERONIMODEVTOOLS-580
>                 URL: 
> https://issues.apache.org/jira/browse/GERONIMODEVTOOLS-580
>             Project: Geronimo-Devtools
>          Issue Type: Improvement
>    Affects Versions: 2.0.x, 2.2.0
>            Reporter: Kevan Miller
>            Assignee: Tim McConnell
>             Fix For: 2.2.0
>
>
> I've had several users comment to me that application development with GEP is 
> not as fast as it should be. Basic scenario is create an application, deploy, 
> test, update, test, update, etc... IIUC, for each update, GEP is repackaging 
> the WAR and running a redeploy. Repackaging/redeploy can take some time (for 
> large apps > 30 seconds). IIUC, we could be updating the appropriate 
> JSP/Servlet classes directly and greatly improving this behavior. I don't 
> know how this works, however. Anybody interested in looking at this? I'd like 
> to see this improved in 2.2. If it's something that could be fixed in 2.1.x, 
> that would be great. But that might depend on how extensive the changes 
> are...  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to