Hi Matt, 2017-06-01 1:03 GMT+02:00 Matt Ryan <[email protected]>: > I'm curious as to the status of this. Is it being worked on? >
Please excuse me a bit late response to that topic. Unfortunately I'm not working *actively* on this anymore at least for now but I'm happy to help and contribute. What I see is that some unification work has been done in oak-run by Chetan in [0] so I guess we should now follow that pattern and unify it for all Oak OSGi services (if possible). I've created an issue [1] to track our status so this way it will be much easier to discuss and when we have some doubts we can come back again to a broader discussion here. Please note that by purpose I haven't mentioned the solution for issue resolution (whether this is nstab or not) as I agree with Michael (in earlier discussion in [2]) that we should first do this on codebase side (unify factories) and then think about the proper format. It might be nstab but it can be also anything else similar, especially that on oak-run side most of the work has been already done. > From the perspective of an AbstractDataStoreService subclass, will the > config still be passed into the "createDataStore()" method as a Map<String, > Object> after having been read from the nstab-formatted config file? I > assume so but wonder if that is the case. I guess we should move the question to the corresponding JIRA issue [1]. Cheers, Arek [0] https://issues.apache.org/jira/browse/OAK-6210 [1] https://issues.apache.org/jira/browse/OAK-6312 [2] http://markmail.org/message/xrtajittrccwddd7
