cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Daniel Fagerstrom <>
Subject Re: [Fwd: Cocoon and Felix]
Date Sun, 19 Feb 2006 21:15:49 GMT
We also need the ability to use unpacked bundles for fast prototyping 
(available in Equinox). We need the possibility to embed the OSGi 
framework in a servlet container 
And we need a framework that work together with some declarative 
services implementation (IIUC the coupling between declarative services 
and the framework is not standardized).

As a first step I think it is more important to get Cocoon working under 
OSGi at all than supporting our local OSGi provider ;) As soon as we 
have got to that point I am all for making Cocoon work with all OSGi R4 
compliant frameworks, and maybe having Felix as default choice.


Upayavira skrev:
> I just received this message from Richard Hall. Seems like Felix just go
> that bit closer to meeting our needs, which is great. Is there anything
> else we specifically need in order to use Felix instead of Eclipse?
> Upayavira
> -------- Original Message --------
> Subject: Cocoon and Felix
> Date: Sun, 19 Feb 2006 12:58:42 -0500
> From: Richard S. Hall <>
> To:,
> Hey guys,
> I wanted to follow up with Cocoon and Felix...
> I can't remember who said it, but someone mentioned at ApacheCon that
> Cocoon needed Bundle.getEntry()/getEntryPaths() to start using Felix. I
> just wanted to let you guys know these methods are now implemented in Felix.
> It was actually a complicated process because I realized that Felix'
> module layer abstraction was not really sufficient for implementing the
> methods. While I could have hacked them in without much difficulty, I
> decided to refactor and rewrite the module layer abstractions, since it
> would also help me with some other issues I wanted to solve. Needless to
> say, after a few weeks worth of work, I am pretty satisfied with the new
> module still needs some tweaking, but for the most part it is
> significantly improved.
> All of this work is committed and ready for experimentation.
> Were there other specific show stoppers for Cocoon? We should enter them
> into JIRA if so, so that we can try to pick them off one by one.
> -> richard

View raw message