> On Mon, 21 May 2018 at 15:46, Robert Haas <robertmhaas@gmail.com> wrote:
>
> On Sat, May 19, 2018 at 12:59 PM, Greg Stark <stark@mit.edu> wrote:
> > On 19 May 2018 at 01:13, Stephen Frost <sfrost@snowman.net> wrote:
> >> I'm not entirely sure about the varlena suggestion, seems like that
> >> would change a great deal more code and be slower, though perhaps not
> >> enough to matter; it's not like our aclitem arrays are exactly optimized
> >> for speed today.
> >
> > I don't actually understand the reason y'all are talking about
> > varlena.
>
> Because aclitem's typlen value in pg_type is currently "12".
This patch went through the last two commit fests without any noticeable
activity. As far as I can see, judging from the discussion, there isn't a
single opinion everyone would agree with, except that simply introducing a new
permission is probably not enough and we need to address how to do this
in an extendable way.
> On Sun, 18 Mar 2018 at 22:05, Isaac Morland <isaac.morland@gmail.com> wrote:
>
> Right now I'm really looking for whether anybody observes any problems with
> the basic idea. If it's considered to be at least in principle a good idea
> then I'll go and make a more complete patch.
There were at least two options suggested about how to address the questions
from the discussion (widening AclMode and introducing another type similar to
aclitem, but more flexible, to store an "extended" permissions). Maybe the
constructive approach here would be to try them out and propose a draft
implementation for one of them?