Re: Pagination count strategies - Mailing list pgsql-general

From Leonardo M. Ramé
Subject Re: Pagination count strategies
Date
Msg-id 20140403154423.GA22638@leonardo-laptop
Whole thread Raw
In response to Re: Pagination count strategies  (Andrew Sullivan <ajs@crankycanuck.ca>)
Responses Re: Pagination count strategies  (Andrew Sullivan <ajs@crankycanuck.ca>)
List pgsql-general
On 2014-04-03 10:00:18 -0400, Andrew Sullivan wrote:
> On Thu, Apr 03, 2014 at 10:34:32AM -0300, Leonardo M. Ramé wrote:
> >
> > What strategy for showing the total number of records returned do you
> > recommend?.
>
> The best answer for this I've ever seen is to limit the number of rows
> you're counting (at least at first) to some reasonably small number --
> say 5000.  This is usually reasonably fast for a well-indexed query,
> and your pagination can say something like "First n of at least 5000
> results", unless you have fewer than 5000 results, in which case you
> know the number (and the count returned quickly anyway).  As you're
> displaying those first 5000 results, you can work in the background
> getting a more accurate number.  This is more work for your
> application, but it provides a much better user experience (and you
> can delay getting the detailed number until the user pages through to
> the second page of results, so you don't count everything needlessly
> in case the user just uses the first page, which IME happens a lot).
> Note that even Google doesn't give you an accurate number -- they just
> say "about ten trillion" or whatever.
>
> Hope that's useful,
>
> A
>

Sounds nice, is it possible to modify my "count(*) over()" to what you
suggest?.

--
Leonardo M. Ramé
Medical IT - Griensu S.A.
Av. Colón 636 - Piso 8 Of. A
X5000EPT -- Córdoba
Tel.: +54(351)4246924 +54(351)4247788 +54(351)4247979 int. 19
Cel.: +54 9 (011) 40871877



pgsql-general by date:

Previous
From: Kasahara Tatsuhito
Date:
Subject: Re: size of interval type
Next
From: Leonardo M. Ramé
Date:
Subject: Re: Pagination count strategies