Re: [VOTE] Adding a couple user APIs dealing with RegionFactory.

2019-12-17 Thread Mark Hanson
Hi All, In working with Udo, I moved the needed public API into a helper class. The RegionFactory copy constructor still remains in RegionFactory but is protected. The only way to use the method now is by creating a helper class that extends RegionFactory. Thanks for you all your help. I thin

Re: [VOTE] Adding a couple user APIs dealing with RegionFactory.

2019-12-16 Thread Owen Nichols
A -1 vote on a code change should be framed as a “request for change”. Udo, you’ve made it clear what you don’t want, but not what it would take to make PR #4409 acceptable to you. “Management V2 API” is unlikely to solve all problems in the near term, and even to do so, it needs a sound under

Re: [VOTE] Adding a couple user APIs dealing with RegionFactory.

2019-12-16 Thread Owen Nichols
Thanks, but it isn’t clear to me exactly what is at stake here. If this is a “design” level proposal, perhaps it should go through the RFC process rather than straight to a vote. Either way, a short summary of the problem description, the possible paths forward, and the advantages/disadvantage

Re: [VOTE] Adding a couple user APIs dealing with RegionFactory.

2019-12-16 Thread Mark Hanson
It has been said I have a negative vote which is counter intuitive. VOTE SUBJECT: Should we continue migrating from AttributesFactory usage to RegionFactory usage and merge the RegionFactory copy constructor. +1 to Migrate to RegionFactory from AttributesFactory and merge the RegionFactory co

[VOTE] Adding a couple user APIs dealing with RegionFactory.

2019-12-16 Thread Mark Hanson
Actually, I would say that it would not be necessary to have a copy constructor if it were not for the way the tests are written that assume an AttributesFactory. I think the discussion boils down to this… Do we migrate to the RegionFactory API from AttributesFactory or do we wait for the Manag