Re: Add default role 'pg_access_server_files'

On Sun, Mar 25, 2018 at 09:43:25PM -0400, Stephen Frost wrote:
> * Michael Paquier (michael@paquier.xyz) wrote:
>> On Thu, Mar 08, 2018 at 10:15:11AM +0900, Michael Paquier wrote:
>> > Other than that the patch looks in pretty good shape to me.
>>
>> The regression tests of file_fdw are blowing up because of an error
>> string patch 2 changes.
>
> Fixed in the attached.

Thanks for the updated version.  This test is fixed.

Patch 2 includes the documentation changes from patch 1, which would
matter only if you decide to keep things splitted.  As far as my brain
sees the patch is logically correct.

> Note that it'll be a bit more complicated since we can't just remove the
> checks from the existing functions- we'll need to have new functions
> where the checks are removed and a new extension version that updates to
> the new functions and then REVOKE's access to them.  Not a big deal,
> just pointing out that it's not quite as straight-forward since it's an
> extension and we need to deal with environments where the server's been
> upgraded and the .so changed, but the existing functions are still in
> place with their current public-execute rights.

Yeah, that's basically what you did for pgstattuple in fd321a1.  I am
not sure that I would have time to double-check what you code and the
commit fest ends in 5 days.  There are many other patches in need of
attention, so I would be incline to just do this portion in the future
and keep the proposal as-is.  My 2c.
--
Michael