| From: | Félix GERZAGUET <felix(dot)gerzaguet(at)gmail(dot)com> |
|---|---|
| To: | Mariel Cherkassky <mariel(dot)cherkassky(at)gmail(dot)com> |
| Cc: | pgsql-performance(at)lists(dot)postgresql(dot)org |
| Subject: | Re: distinct on extract returns composite type |
| Date: | 2019-09-29 10:34:30 |
| Message-ID: | CANVwZtuzT1w0aVQMyBi+U=d7e_HQmB70THnqZoxu=xooVRvhbA@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
Hello,
On Sun, Sep 29, 2019 at 11:46 AM Mariel Cherkassky <
mariel(dot)cherkassky(at)gmail(dot)com> wrote:
> I'm keep getting composite type (row) instead of two columns. Is there any
> sql way to convert the row type into two columns ? I want to get the first
> and last dates of each quarter with those columns and with this composite
> type I failed doing it
>
This seems to work as you expect:
select distinct extract(year from end_time) as year, extract(quarter from
end_time) quarter from generate_series
( '2017-09-01'::timestamp
, '2019-04-01'::timestamp
, '3 month'::interval) end_time
;
https://www.postgresql.org/docs/current/sql-select.html#SQL-DISTINCT
--
Félix
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2019-09-29 15:32:23 | Re: sequence depends on many tables |
| Previous Message | Mariel Cherkassky | 2019-09-29 09:46:31 | distinct on extract returns composite type |