A simple way I came up is to truncate the date. So if you have 2009-08-08, and you want a subset on month, then just truncate the day-part: 2009-08-00 on the whole column, and SELECT DISTINCT so you have a subset. You can use this subset then to join the dates, GROUP BY and aggregate.... An other way I found to do this is in Celko's 'SQL for smarties'. He uses modulo there. It seems powerful, but also tricky to implement. I was wondering if anyone knew some other way to create a subset of a timestamp column. Any input is welcome. Regards, Davor -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
