Re: wip: functions median and percentile

On Wed, Oct 13, 2010 at 10:37 PM, Greg Stark <gsstark(at)mit(dot)edu> wrote:
> On Wed, Oct 13, 2010 at 5:54 PM, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
>> Or to put it more bluntly - what is the "problem with planner and hash
>> agg" that all of these functions need to solve?  And why does it need
>> a flag in pg_proc?  Why can't't we leave it to the individual
>> functions to perform a sort of one is needed?
>>
>
> So I think that's backwards. Why is the function doing data
> manipulations like sorts that we usually put in the plan? Is there
> some some key meta information that should be flagged in pg_proc and
> general functionality the executor should be providing so that this
> general class of problems can be solved efficiently?
>
> Otherwise I fear lots of things we would expect to be efficient such
> as calculating the top, median, and last items in the same sort order
> would require three separate sorts of the same data. We have a planner
> capable of comparing sort orders and estimating the costs of different
> plans, we should be using it.

Good point. I think you're right. I'm not sure what the best design
for it is, though.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company