I will second the notion that our classloading is inconsistent, at best. Especially when OpenJPA is integrated into other programming models such as Spring or OSGi. If you have the time to simplify the classloading configuration and mechanism, then all the more power to you. Thanks!
Kevin On Tue, Oct 4, 2011 at 11:18 AM, Pinaki Poddar <[email protected]> wrote: > OpenJPA class loading for persistent domain classes and plug-ins (both > native > and custom) uses several mechanics. Sometimes, a classloader is set on a > class to use, sometimes it is ClassResolver attached to the configuration, > sometimes it is the classloader of the persistence unit etc. > I am planning to bring these different mechanics under Configuration. > 1. Configuration will have a MultiClassLoader. > 2. The facade can add other loaders to this MultiClassLoader. > 3. Everywhere this Configuration's MultiClassLoader will be used uniformly. > 4. ClassResolver notion will be deprecated. > > Comments/Thoughts? > > > > ----- > Pinaki Poddar > Chair, Apache OpenJPA Project > -- > View this message in context: > http://openjpa.208410.n2.nabble.com/Classloading-needs-to-be-refurbished-tp6859197p6859197.html > Sent from the OpenJPA Developers mailing list archive at Nabble.com. >
