Hi Martin As Thiago has already pointed out, it would make more sense to have new annotations instead of changing @CommitAfter which is already in use.
I would suggest either create a new library which can add these new annotations to the existing tapestry-hibernate without breaking the code or submit a patch(with tests of course) and some committer would look into it. Although implementing @Transactional annotation would be great but even @WriteTransaction and @ReadOnlyTransaction (I think that you have pointed out) are not bad. Anyways. You have done some useful work and it would be great if you can share it as a library! regards Taha On Aug 30, 2013, at 6:56 PM, Martin Kersten wrote: > I didnt started the project I am only maintaining it. I would love to > donate this code since this is so dangerous. I wont think about the money > this might have cost to find these bugs in a live system. Especially if the > database gots mixed up over these. > > But you are right annotating actions and pages might be a better way > anyhow. But my standpoint is that the framework should protect me and my > coworker for doing such harmful stuff. I will look into the code again and > commit / provide some code to the jira. Maybe one can pick it up. > > It also supports hibernates read only transactions to allow the utilization > of slave db nodes which is currently not supported by tapestry. > > > 2013/8/30 Lance Java <lance.j...@googlemail.com> > >> I never annotate my services or dao's with @CommitAfter and only annotate >> the actions in my pages / components. Is there a reason why you can't use >> this approach? >>
