SELECT p.turno,
p.data,
COALESCE(q.PezziBuono,0) AS PezziBuono,
COALESCE(q.PezziCattivi,0) AS PezziCattivi
FROM
(
SELECT *
FROM (SELECT DISTINCT Turno FROM sottostante) m
CROSS JOIN (SELECT DISTINCT data FROM sottostante)n
)p
LEFT JOIN (SELECT Turno,data,SUM(PezziBuoni) AS PezziBuoni, SUM(PezziCattivi) AS PezziCattivi
           FROM sottostante
           GROUP BY Turno,data
           ) q
ON q.Turno = p.Turno
AND q.data = p.data

quote:

---

Originally posted by giannilionetti

Sorry i miss this is an english forum.
I've an PRODUCTION table as :

CodProd Turn Date Good NoGood
xxxxx 1 20/12/2012 1 0
xxxxx 1 20/12/2012 1 0
xxxxx 1 20/12/2012 0 1
xxxxx 1 20/12/2012 1 0
xxxxx 2 19/12/2012 1 0
xxxxx 2 19/12/2012 0 1
xxxxx 2 19/12/2012 0 1
xxxxx 1 19/12/2012 1 0
xxxxx 1 18/12/2012 1 0
xxxxx 2 18/12/2012 1 0
xxxxx 2 18/12/2012 1 0
xxxxx 2 18/12/2012 1 0
Output
I need group by following data Output of Good and NoGood for Date and Turn as:
Cod.Prod Turn Date Good NoGood
xxx 1 20/12/2012 3 1
xxx 2 20/12/2012 0 0
' 1 19/12/2012 1 0
' 2 19/12/2012 1 2
' 1 18/12/2012 1 0
xxx 2 18/12/2012 3 0

Thanks

---