Hi, On 18 September 2017 at 13:51, Carsten Ziegeler <cziege...@apache.org> wrote:
> Julian Sedding wrote > > Hi Ian > > > > Would it make sense to remove the AdapterFactory implementation and > > instead pass along the URIProvider service(s) using the > > o.a.s.jcr.resource.internal.HelperData object instead? According to > > its javadoc, it is "used to pass several services/data to the > > resource". > > > > I assume the AdapterFactory is used because you don't have access to > > the OSGi service registry in JcrNodeResource where you need the > > URIProvider. > > > > Sounds like a good change to me, right a AdapterFactory shouldn't be > needed. > The patch is quite extensive and it would need care to ensure that the list that is passed is fully thread safe. So far this approach has only been used with an AtomicReference holding a single value. If a larger patch is wanted, I can track down everywhere the the JcrProviderState is created, as this holds the HelperData for the session, although the createProviderState method is already deprecated, which is the only way to create JcrProviderState afaict. See the JcrProviderStateFactory. Adapting to the URIConverter from the ResourceResolver seems a lot cleaner, assuming we believe the AdapterFactory pattern is a good pattern. Best Regards Ian > > Carsten > > > > -- > Carsten Ziegeler > Adobe Research Switzerland > cziege...@apache.org >
