Re: counting distinct rows on more than one column

In 7.0.3, I believe the following would work:

SELECT count(distinct(a || b)) FROM t;

if subselects in from were supported in 7.0.3 as they are in 7.1, you
could do:

SELECT count(*) FROM (SELECT DISTINCT a,b FROM t) FROM x

Michael Fork - CCNA - MCP - A+
Network Support - Toledo Internet Access - Toledo Ohio

On Wed, 28 Mar 2001, Dirk Lutzebaeck wrote:

> 
> Hi,
> 
> on 7.0.3 want to COUNT
> 
>   SELECT DISTINCT a,b FROM t;
> 
> I can't find a solution because any combination with
> count with more than one column gives syntax errors.
> 
> One solution would be to set a view:
> 
> CREATE VIEW v AS SELECT DISTINCT a,b FROM t;
> 
> and then
> 
> SELECT count(a) FROM v
> 
> but views don't support distinct in v7.0.3
> 
> Ok I could use a temporary table but my select distinct tends to give
> large results.
> 
> Any clues?
> 
> Dirk
> 
> 
> 
> ---------------------------(end of broadcast)---------------------------
> TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org
>