KaiGai Kohei wrote:
>> We are going to need to come up with specific answers to these issues
>> soon.
>
> The origion of issue is simple.
>
> Whether we should support to activate (not only compile) two or more
> security
> mechanism in same time, or not.
> In my opinion, it is not a frequent situation, and it gives us several big
> pains, but benefit is smaller than the pains.
With all respect, you were asked to divide up the issues so we don't
have to deal with them all at once.
For instance, a separate patch that implements SQL-level row level
security would be fairly uncontroversial and issue-free at this point,
but it would be completely useful on its own and it would build
confidence in the developer community about your other plans.
Most committers have expressed the viewpoint in one way or another that
having this available is a prerequisite for accepting further work.
Yet, I am not aware of even an interface proposal for this.
Meanwhile, we are busy worrying about what system columns the follow-up
features will have.
