On Thu, Oct 30, 2008 at 11:19:15PM -0700, Jeff Davis wrote:
> If there are potential problems with the standard
> (where we don't want to implement a violation), we should just do
> array_accum(). If not, we might as well do the standard array_agg(),
> perhaps without the ORDER BY clause.
I've wanted an array_sort() function before; having this functionality
as a separate function also seems considerably prettier than some ad
hoc grammar, it also generalizes nicely to cases where the array isn't
coming from an aggregate.
In response to
pgsql-hackers by date
|Next:||From: Peter Eisentraut||Date: 2008-10-31 12:37:28|
|Subject: Distinct types|
|Previous:||From: Zdenek Kotala||Date: 2008-10-31 12:14:37|
|Subject: Re: PG_PAGE_LAYOUT_VERSION 5 - time for change|