I am using a older version of Postgres 8.2.11. Can anyone tell me the equivalent of MySql's
The "not quite duplicate" in the comments should point you in the right direction: create your own aggregate function. First you'll need a non-aggregate string concatenation function, something like this:
create function concat(t1 text, t2 text) returns text as $$ begin return t1 || t2; end; $$ language plpgsql;
Then you can define your own aggregate version of that function:
create aggregate group_concat( sfunc = concat, basetype = text, stype = text, initcond = '' );
Now you can
group_concat all you want:
select group_concat(s) from t group by g
I dug this out of my archives but I think it should work in 8.2.
Keep in mind that 8.2 is no longer supported so you might want to upgrade to at least 8.4 as soon as possible.