[comp.databases] SQL to select unique row on given fields

nacer@hpmcaa.mcm.hp.com (Abdenacer Moussaoui) (05/01/91)

I need help figuring out an SQL query to select unique rows with
respect to a set of fields.
Let T be a table with active unique index on k1, k2 field.
I would like to select rows such that combination of values
in fields  f1, f2 and f3 appear only once.
I don't care which row's keys (k1, k2) I get as long as I get one
and only one with for each given unique instance of f1, f2, f3.

When I use
		select count(*), f1, f2, f3 
		from T 
		group by f1, f2, f3
 I get a distribution of the different variations of f1..f3 fields,
now how do I include in the selection k1 and k2 fields?  I have looked
in C.J book on possibly using aliases but at this point I have no clue
how to do this.  Thank you so much.
-- 
Thank you.
--Abdenacer 	(nacer@hpmcaa.mcm.hp.COM)

hansene@hpcc01.HP.COM (Ed Hansen) (05/07/91)

The following query will retrieve all the distinct combination
of f1 , f2, f3 that occur in table T:

	SELECT DISTINCT f1 , f2, f3
	FROM T

JeanMarie