Hi Fabio,

>SELECT NVL(A.Item, b.Item), count(a.Item)+Count(b.Item) FROM Items A FULL JOIN Items2 B ;
>ON A.Item=b.Item AND .F. GROUP BY 1

I did not find this to be fast. A better way to use rushmore is something like

SELECT NVL(A.Item, b.Item), count(a.Item)+Count(b.Item) ;
    FROM Items A ;
         FULL JOIN Items2 B ON A.Item=b.Item * 0 -1 ;
    GROUP BY 1

It seems to do the trick and is fast even with a lot of records (if the item columns of both tables are indexed)

Walter,