Alexander Farber wrote:
And:
pref=> select * from (
select u.id,
u.first_name,
u.city,
m.money,
rank() over (order by money desc)
from pref_users u, pref_money m where
m.yw=to_char(current_timestamp, 'YYYY-IW')
) ar
where ar.id='OK138239987797';
works, but delivers 800 different ranks:
id | first_name | city | money | rank
----------------+------------+-------------+-------+---------
OK138239987797 | ÐÐÐÐ | ÐÐÐÐÑÐÐÐÑÑÐ | 2169 | 1
OK138239987797 | ÐÐÐÐ | ÐÐÐÐÑÐÐÐÑÑÐ | 1955 | 3479
OK138239987797 | ÐÐÐÐ | ÐÐÐÐÑÐÐÐÑÑÐ | 1948 | 6957
OK138239987797 | ÐÐÐÐ | ÐÐÐÐÑÐÐÐÑÑÐ | 1060 | 10435
OK138239987797 | ÐÐÐÐ | ÐÐÐÐÑÐÐÐÑÑÐ | 1034 | 13913
OK138239987797 | ÐÐÐÐ | ÐÐÐÐÑÐÐÐÑÑÐ | 1012 | 17391
OK138239987797 | ÐÐÐÐ | ÐÐÐÐÑÐÐÐÑÑÐ | 929 | 20869
I haven't created your tables, but it looks to me like the sub-select
needs something from u to m. When run by itself what does the
sub-select generate. (I'm tempted to recomment count(*) cause I think
you're getting a cross-product.) This could all be a cut / paste error
in my first response.
--
Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
