cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Torsten Curdt <>
Subject Re: [RT] user-components.xconf
Date Sun, 10 Mar 2002 15:57:58 GMT
guys, could you share your visions or RT on this... I don't want this 
thread to be just dropped without any result.

More RT from me:

Another possible sollution for a per cocoon application configuration 
would be to tie an optional xconf to (the mount point of) a sitemap. 
Although I'm quite sure if chaining of both of this is a valid 
assumption I am sure that every application will come with it's own 
sitemap. So dropping in the application currently means dropping a 
sitemap mount. Stefano, would you like to see this change, too?
If not I see two ways of doing this.

1) specify a xconf where the sitemap is mounted

    <map:mount uri-prefix="myapp"


    Problem: I remember there was a snippet on the list mounting sitemaps
             with a pattern. So each sitemap could only have the same
             xconf. So this feels a bit like a kludge, too

2) let the sitemap point to the configuration and roles

    a)<map:sitemap xmlns:map=""
    b)<map:sitemap xmlns:map="">

    But thinking this way we shouldn't define the cocoon.xconf as serlvet
    parameter but the root sitemap.

    So we would remove/deprecate


    And have



     <map:sitemap xmlns:map="">

On the other hand using the sitemap this way makes it more of a 
application descriptor than just a plain sitemap. For SoC we maybe 
should mount *applications* (or what you guys called it:blocks) not 

So we would need something like a "Cocoon Application Descriptor"

   mounting the first one via WEB-INF/web.xml


   with the cocoon demo application


    and then mount inside the sitemap (the right place?) another

    <map:mount uri-prefix="myapp" src="myapp/myapp.cad" />

      <sitemap reload-method="synchron"

Stefano, is this more the direction you thought of?
Please, I can't wait hear your RT on this!!

To unsubscribe, e-mail:
For additional commands, email:

View raw message