From: | Alban Hertroys <dalroi(at)solfertje(dot)student(dot)utwente(dot)nl> |
---|---|
To: | Jon Smark <jon(dot)smark(at)yahoo(dot)com> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Record with a field consisting of table rows |
Date: | 2011-01-13 19:21:30 |
Message-ID: | 8C29A75C-3589-4993-9A70-DFE14900BB31@solfertje.student.utwente.nl |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On 13 Jan 2011, at 17:22, Jon Smark wrote:
> create type page_t AS
> (
> total int4,
> users user_t[]
> );
>
> create function get_page ()
> returns page_t
> language plpgsql as
> $$
> declare
> _page page_t;
> begin
> _page.total := select count (*) from users;
> select * into _page.users from users limit 10;
> return _page;
> end
> $$;
I think it would be easier to rewrite that to a set-returning function returning TABLE (...).
Something like this (untested):
create function get_page ()
returns setof table (total int, user users)
language plpgsql as
$$
declare
_total int;
begin
_total := select count (*) from users;
return query select _total AS total, u from users AS u limit 10;
end
$$;
In general it is considered a bad idea to rely on what * returns though, it's better to return the columns explicitly.
Which makes me wonder, is table cloning supported for these cases? For example:
create function get_page ()
returns setof table (LIKE users, total int)
...
Alban Hertroys
--
Screwing up is an excellent way to attach something to the ceiling.
!DSPAM:737,4d2f50bf11877227918328!
From | Date | Subject | |
---|---|---|---|
Next Message | Alban Hertroys | 2011-01-13 19:38:10 | Re: Record with a field consisting of table rows |
Previous Message | Pavel Stehule | 2011-01-13 19:21:23 | Re: Record with a field consisting of table rows |