Re: vacuum_truncate configuration parameter and isset_offset

В письме от вторник, 25 марта 2025 г. 17:57:46 MSK пользователь Nathan Bossart
написал:

> In any case, AFAICT the votes are somewhat evenly divided at the moment, so
> I do not intend to proceed with this patch for now.

Counting votes does not lead anywhere, as I can ask friends and colleagues to
vote for me for example. This is wrong way.

There are people who involved it supporting reloption engine. If Alvaro said
"isset_offset" is good, let's have it, I will shut keep my mouth shut, right
after saying "As you say". Because he is the author of that part of the code,
I respect this.

I am also the person that dedicated lot's of efforts to work with reloptions. I
will do it for may years ahead, I think. This isset_offset I will have to
support from now on, if we do not revert it. I do not like it, I see there is
no logic in it, at least the way you suggest it.

If you want to dedicate part of your life to reloptions, you are welcome. I
need help with a code to review. If you really need isset_offset let's together
redisign options to use it instead of default_value. Go for it, but not
partly, but totally.

But if you want to fix one option and leave, please do not leave us with
isset_offset.

This is not about democracy, this is about who will deal with that part of the
code later. I guess it will be me and Alvaro, not you. Please let us have the
code the way we like it, since we support it most of the time.

--
Nikolay Shaplov aka Nataraj
Fuzzing Engineer at Postgres Professional
Matrix IM: @dhyan:nataraj.su