[jira] Updated: (SHINDIG-51) Overhaul servlet initialization.

["Kevin Brown (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/SHINDIG-51?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Kevin Brown updated SHINDIG-51:
-------------------------------

    Attachment: servlet-changes.patch

> Overhaul servlet initialization.
> --------------------------------
>
>                 Key: SHINDIG-51
>                 URL: https://issues.apache.org/jira/browse/SHINDIG-51
>             Project: Shindig
>          Issue Type: Improvement
>            Reporter: Kevin Brown
>            Assignee: Kevin Brown
>            Priority: Blocker
>         Attachments: servlet-changes.patch
>
>
> While doing work to implement the RpcServlet that has been discussed before (see https://issues.apache.org/jira/browse/SHINDIG-25), it became quite apparent that we can't do a whole lot more without fixing the gaping hole of servlet configuration (essentially, there's no way to use your own implementations of most of the core classes or inject your own javascript without either modifying GadgetRenderingServlet itself, or by using a servlet engine that allows for programmatic servlet creation).
> I've attached a patch with a proposed solution (it also contains some minor performance tweaks and bug fixes that I was working on at the time, but nothing major). I don't really know if there's a cleaner way to do this or not since I'm not that familiar with the Servlet spec, but it seems to fit well within the model and will finally give us a place to inject implementation specific components without resorting to ugly hacks and workarounds.
> This patch will break any existing hacky workaround type solutions, because it eliminates all of the constructors from all of the servlets, and instead configures the servlets the "proper" way by passing shared data through the ServletContext. This means that if I apply this patch, anyone who has done this sort of thing will need to change some code to make it work again. With that in mind, I'd prefer to send it here and let everyone review it before committing. This also removes some previously deprecated methods from GadgetServer which simply won't be needed in this model.
> All feedback is appreciated.

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