Re: thoughts on the camel-websocket component

2013-09-12 Thread Aki Yoshida
Hi Charles, If you will be working on the atmosphere component, I will be interested in collaborating with you. I'm currently more interested in the websocket part than in the atmosphere component itself, as I mentioned in my initial mail. We can use this list and irc for communication. thanks. re

Re: thoughts on the camel-websocket component

2013-09-11 Thread Charles Moulliard
You are right, we need something which is web container agnostic. As atmosphere was in my todo list, maybe it is time to develop the camel-atmosphere component. If you are interested we can collaborate on that ? On Tue, Sep 10, 2013 at 3:23 PM, Aki Yoshida wrote: > Hi Charles, > > If we only su

Re: thoughts on the camel-websocket component

2013-09-10 Thread Aki Yoshida
Hi Charles, If we only support jetty, that will be fine. But to support tomcat, we need a component that avoids using jetty's websocket API directly. In this aspect, camel-cxf is different because its servlet is not web-container specific and can also be used in a gemini/tomcat based OSGi environm

Re: thoughts on the camel-websocket component

2013-09-10 Thread Charles Moulliard
Hi Aki, As the camel-websocket component uses Jetty + WebSocket servlet of Jetty, that should not be a big change to add a property for the endpoint to request that we use Jetty deployed in Karaf, Felix, ... instead of creating a local jetty instance. This is what we do with camel-cxf endpoint Re

thoughts on the camel-websocket component

2013-09-10 Thread Aki Yoshida
I have been looking into the websocket component and I was wondering about a few things. First of all, I would like to make the component use the framework's servlet if it's running in e.g., an OSGi container like karaf that has jetty or another container that uses geminiweb/tomcat. I was using A