Re: [PATCH] Proof of concept for GUC improvements

> On Mar 21, 2022, at 7:53 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>
> Andres Freund <andres(at)anarazel(dot)de> writes:
>> My impression is that there's not a lot of enthusiasm for the concept? If
>> that's true we maybe ought to mark the CF entry as rejected?
>
> Yeah, I'm kind of leaning that way too. I don't see how we can
> incorporate the symbolic values into any existing display paths
> without breaking applications that expect the old output.
> That being the case, it seems like we'd have "two ways to do it"
> indefinitely, which would add enough confusion that I'm not
> sure there's a net gain. In particular, I foresee novice questions
> along the lines of "I set foo to disabled, why is it showing
> as zero?"

Yeah, my main motivation here was about trying to have less special values in the config files, but I guess it would effectively be making everything effectively an enum and still relies on knowing just what the specific magic values are, no not really a net gain in this department.

For the record, I thought this would have a fairly low chance of getting in, was mainly curious what level of effort it would take to get something like this going and get some feedback on the actual idea.

> If we'd done it like this from the beginning, it'd have been
> great, but retrofitting it now is a lot less appealing.

Yeah, agreed on this. As far as I’m concerned we can reject.

Thanks,

David