Re: Keep notnullattrs in RelOptInfo (Was part of UniqueKey patch series)

On Mon, 22 Nov 2021 at 02:14, Andy Fan <zhihui.fan1213@gmail.com> wrote:
>
> _I_ think there is no way to bypass that.  I guess Tom has a bigger plan on
> Var (not only for notnull), but no time to invest in them so far.  If
> that is the case,
> personally I think we can go ahead with my method first and continue the review
> process.

This discussion has gone on for two years now and meandered into
different directions. There have been a number of interesting
proposals and patches in that time but it's not clear now what patch
is even under consideration and what questions remain for it. And I
think this message from last November is the last comment on it so I
wonder if it's reached a bit of an impasse.

I think I would suggest starting a fresh thread with a patch distilled
from the previous discussions. Then once that's settled repeat with
additional patches, keeping the discussion focused just on the current
change.

Personally I think these kinds of optimizations are important because
they're what allow people to use SQL without micro-optimizing each
query.

-- 
greg