GROUP BY / HAVING clauses problem

I'm trying to set up my adhoc query to return just one single record,
which is aliased as 'foreign' in my sql statement (which is just the
total amount of foreign overseas orders for just one day. All
Sale_Type_Ids over 2 [integer datatype] are foreign orders):
SELECT SUM(CASE WHEN Orders.Sale_Type_Id > 2 THEN
Orders.Sale_Type_Id ELSE NULL END) AS foreign
FROM Orders INNER JOIN
Processing ON Orders.ID = Processing.Order_ID
WHERE (Processing.Orderdate = '20050915') AND (Processing.status =
1)
GROUP BY CASE WHEN Orders.Sale_Type_Id > 2 THEN Orders.Sale_Type_Id
ELSE NULL END
HAVING (SUM(CASE WHEN Orders.Sale_Type_Id > 2 THEN
Orders.Sale_Type_Id ELSE NULL END) >= 0)
..but my resultset is returning two records. If I remove the HAVING
clause, it will return three records, with one being blank.
?
.netsportsIf I understand correctly, your GROUP BY is on:
All Orders.Sale_Type_ID greater than 2
NULL
So, without the HAVING I would expect one row returned for each Sale_Type_ID
> 2 and one for all the rest which become NULL. Do you have two ID>2 within
the rows covered by your WHERE clause.
The HAVING apparently is able to prune out the NULL value.
Perhaps all you wanted was:
SELECT SUM(Orders.Sale_Type_Id) AS foreign
FROM Orders INNER JOIN
Processing ON Orders.ID = Processing.Order_ID
WHERE (Processing.Orderdate = '20050915') AND (Processing.status =1) AND
Orders.Sale_Type_ID > 2
RLF
".Net Sports" <ballz2wall@.cox.net> wrote in message
news:1127848228.326875.56120@.g49g2000cwa.googlegroups.com...
> I'm trying to set up my adhoc query to return just one single record,
> which is aliased as 'foreign' in my sql statement (which is just the
> total amount of foreign overseas orders for just one day. All
> Sale_Type_Ids over 2 [integer datatype] are foreign orders):
> SELECT SUM(CASE WHEN Orders.Sale_Type_Id > 2 THEN
> Orders.Sale_Type_Id ELSE NULL END) AS foreign
> FROM Orders INNER JOIN
> Processing ON Orders.ID = Processing.Order_ID
> WHERE (Processing.Orderdate = '20050915') AND (Processing.status =
> 1)
> GROUP BY CASE WHEN Orders.Sale_Type_Id > 2 THEN Orders.Sale_Type_Id
> ELSE NULL END
> HAVING (SUM(CASE WHEN Orders.Sale_Type_Id > 2 THEN
> Orders.Sale_Type_Id ELSE NULL END) >= 0)
> ..but my resultset is returning two records. If I remove the HAVING
> clause, it will return three records, with one being blank.
> ?
> .netsports
>|||If you only want a single row then remove the GROUP BY clause. GROUP BY
returns one row per group.
David Portas
SQL Server MVP
--|||Thanks. Looks like i'm getting the desired resultset. This sql
statement was sort of a permutation of an extensive one that would
bring back multiple records, but yes, I was starting to think if the
Group By was really necessary.|||I'm inferring your schema to be something like this:
CREATE TABLE Orders
(
ID INT PRIMARY KEY ?
, Sale_Type_Id INT
, Order_Total MONEY ?maybe
)
CREATE TABLE Processing
(
OrderID INT REFERENCES Orders(OrderID)
, Orderdate DATETIME
, Status INT
)
Your question sounds like you're either trying to find a count of
orders, or total amount purchased (which is why I made up that
ordertotal field). If either of those are what you're looking for,
there are much syntactically simpler solutions.
e.g.
SELECT COUNT(*) CountOfForeignOrders
, SUM(Order_Total) TotalOfForeignOrders
FROM Orders
WHERE Sale_Type_Id > 2 AND EXISTS(SELECT * FROM Processing WHERE
OrderID = Orders.ID AND OrderDate = '20050915' AND Status = 1)
When performing aggregates with join clauses, it's possible to
aggregate the same value more than once if a join causes the same row
to appear multiple times in the resultset. GROUP BY and HAVING aren't
necessary unless you're selecting your data based on your aggregate,
and if you'd return multiple aggregate sets.
-Alan