Re: GROUP BY DISTINCT

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Tuesday, March 2, 2021 5:51 PM, Vik Fearing <vik(at)postgresfriends(dot)org> wrote:

> On 3/2/21 4:06 PM, Georgios Kokolatos wrote:
>
> > As a minor gripe, I would note the addition of list_int_cmp.
> > The block
> >
> > - /* Sort each groupset individually */
> >
> >
> > - foreach(cell, result)
> >
> >
> > - list_sort(lfirst(cell), list_int_cmp);
> >
> >
> >
> > Can follow suit from the rest of the code, and define a static cmp_list_int_asc(), as
> > indeed the same patch does for cmp_list_len_contents_asc.
> > This is indeed point of which I will not hold a too strong opinion.
>
> I did it this way because list_int_cmp is a general purpose function for
> int lists that can be reused elsewhere in the future. Whereas
> cmp_list_len_contents_asc is very specific to this case so I kept it local.

Of course. I got the intention and I have noted my opinion.
>
> I'm happy to change it around if that's what consensus wants.

As before, I will not hold a too strong opinion.

>
> > Overall :+1: from me.
>
> Thanks for looking at it!
>
> > I will be bumping to 'Ready for Committer' unless objections.
>
> In that case, here is another patch that fixes a typo in the docs
> mentioned privately to me by Erik. The typo (and a gratuitous rebase)
> is the only change to what you just reviewed.

Thank you. The typo was indistiguishable to me too.

My :+1: stands for version 3 of the patch. Updating status in the
commitfest to reflect that.

//Georgios -- https://www.vmware.com

>
> ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Vik Fearing