I've got a non-normalized sales database like this:
Invoice Item P开发者_如何学Crice Total_Sale
------- ---- ----- ----------
565 Nails 1.25 6.25
565 Hammer 5.00 6.25
566 Paint 7.95 7.95
...
As you can see, the "total_sale" is repeated for each line item in the invoice. Invoice #565 has two items, for a total of $6.25. Invoice #566 only has one item, the total is $7.95. The sum(total_sale) for both invoices should be $14.20.
Q: Is there any easy way to get all the columns* for just one row for each different invoice? Regardless if the invoice has one, two or one hundred line items?
Q: Is there any easy way to get all the columns* for just one row for each different invoice?
WITH T AS
(
SELECT
Invoice,
Item,
Price,
Total_Sale,
ROW_NUMBER() OVER (PARTITION BY Invoice ORDER BY (SELECT 0)) AS RN
FROM YourTable
)
SELECT
Invoice,
Item,
Price,
Total_Sale
FROM T
WHERE RN=1;
A simple query would be:
Select SUM(Price), Invoice, Total_Sale From InvoicesTable
Group By Invoice, Total_Sale
Order By Invoice
The item column cannot be included as the group by would yield multiple rows.
精彩评论