Re: Improving avg performance for numeric

On 24.9.2013 17:57, Pavel Stehule wrote:
>
>
>
> 2013/9/24 Robert Haas <robertmhaas(at)gmail(dot)com
> <mailto:robertmhaas(at)gmail(dot)com>>
>
> On Mon, Sep 23, 2013 at 4:15 PM, Tomas Vondra <tv(at)fuzzy(dot)cz
> <mailto:tv(at)fuzzy(dot)cz>> wrote:
>> Seems "ready for commiter" to me. I'll wait a few days for others
>> to comment, and then I'll update the commitfest page.
>
> Some thoughts:
>
> The documentation doesn't reflect the restriction to type internal.
> On a related note, why restrict this to type internal?
>
>
>
> Now, for almost all types Postgres well estimate size of state
> value. Only arrays with unknown size can be a different. When we
> enable this value for all types, then users can specify some bad
> values for scalar buildin types. Next argument is simply and bad - I
> don't see a good use case for customization this value for other than
> types than internal type.
>
> I have no strong position here - prefer joining with internal type
> due little bit higher robustness.

I share this oppinion. I was not able to come up with a single use case
benefiting from allowing this for types other than internal. Seems like
a footgun to me, with no potential benefit.

So +1 to keeping the patch 'type internal only' from me.

With the formatting issues now fixed, I believe the patch is ready for
committer (and someone already switched it to that state).

Tomas