On Fri, Oct 31, 2008 at 6:51 PM, Rickard Öberg <[EMAIL PROTECTED]> wrote:
>> But if I did uowf.currentUnitOfWork().newUnitOfWork() > > This I think is a bug. Currently the nested UoW uses the visibility of > the original module. It should be the visibility of the module that > creates it, I think. This takes a bit of trickery to accomplish, but > it's doable. When we discussed this in the office yesterday, I also came to the conclusion that perhaps the problem lied in 'nested UoW should be able to have different visibility'. uowf.currentUnitOfWork().newUnitOfWork( Module module ); Could be a straight forward solution. Another would be to 'tie' the uowf.newUnitOfWork() to any existing running ones, so they get nested automagically. However, that requires the app developer to be aware of any existing UoW and suspend() that, before creating an independent one. To me that sounds more troublesome, than the above with explicitly passing Module. Cheers Niclas _______________________________________________ qi4j-dev mailing list [email protected] http://lists.ops4j.org/mailman/listinfo/qi4j-dev
