I agree, 1.1 is the best version.  I'm not sure I want to try to get the 
appropriate tck tests running in 2.2, but if anyone else is game.....

thanks
david jencks

On Apr 22, 2010, at 10:10 AM, Donald Woods wrote:

> So why not release it as version=1.1 and we'll spin a 1.1.1 if more
> updates are required?  Not sure why we need to version it as a Milestone
> release.... Is there a TCK required?  If so, why not upgrade 2.1.5 or
> 2.2.1-SNAPSHOT to use the new level and verify it still passes the Java
> EE 5 TCKs?
> 
> -Donald
> 
> 
> On 4/21/10 6:33 PM, David Jencks wrote:
>> I think I've done  everything I want to for now with yoko as far as getting 
>> it running in osgi.  It would be great if someone wanted to review it, 
>> otherwise I think its time to start trying to release it.
>> 
>> BTW, we're going to have problems with osgi and milestones, since e.g. 
>> 3.0.M1 is after 3.0.  Similarly for yoko, 1.1.M! is after 1.1.  So I think 
>> this release should be 1.1.M! and when we get all the interop stuff to work 
>> we can release 1.2.
>> 
>> Basically what I did is to steal Rick's locator code from the specs, but 
>> register stuff through blueprint rather than scanning bundles.  I also added 
>> a priority to make it easier to predict which class you'd end up with.  I 
>> also tried to locate every place a class is loaded dynamically in yoko and 
>> have it use the locator.
>> 
>> This seems to allow configuring your orb completely through these blueprint 
>> beans, rather than supplying a properties file with the service name to 
>> class mapping.  This has been one of my goals for yoko for a long time, 
>> although I don't know if anyone else thinks it's valuable.
>> 
>> thanks
>> david jencks

Reply via email to