SQL How to Pivot this table? - sql

Question

I have a very hard time understanding how to pivot something.
I have this simple query
select
year
,AVG(Quantity) Quantity
,AVG(Price) Price
,CAST(Datepart(wk,Date) as nvarchar) + '-' + RIGHT(CAST(year(Date) as NVARCHAR),2) Week
from Yearly
GROUP BY Year, CAST(Datepart(wk,Date) as nvarchar) + '-' + RIGHT(CAST(year(Date) as NVARCHAR),2)
Which results in this table
+------+----------+---------------+------+
| year | Quantity | Price | Week |
+------+----------+---------------+------+
| 16 | 877814 | 68636081.39 | 6-20 |
| 17 | 436029 | 2635873.72 | 6-20 |
| 18 | 3793464 | 65971353.61 | 6-20 |
| 19 | 23552519 | 478741292.122 | 6-20 |
| 20 | 6973687 | 34658140.815 | 6-20 |
| Z01 | 7776508 | 54949609.221 | 6-20 |
+------+----------+---------------+------+
Right now I only have the one week, but as the days go by, I have a job that is going to build those 6 rows for 7-20, 8-20, 9-20, etc.
I want my table to look like
+------+--------+-------------+--------+------------+---------+-------------+----------+-------------+---------+-------------+---------+-------------+----------+-------------+
| | 16 | 17 | 18 | 19 | 20 | Z01 | Total | |
+------+--------+-------------+--------+------------+---------+-------------+----------+-------------+---------+-------------+---------+-------------+----------+-------------+
| Week | Qty | Price | Qty | Price | Qty | Price | Qty | Price | Qty | Price | Qty | Price | Qty | Price |
| 6-20 | 877814 | 68636081.39 | 436029 | 2635873.72 | 3793464 | 65971353.61 | 23552519 | 478741292.1 | 6973687 | 34658140.82 | 7776508 | 54949609.22 | 43410021 | 705592350.9 |
| 7-20 | | | | | | | | | | | | | | |
| 8-20 | | | | | | | | | | | | | | |
+------+--------+-------------+--------+------------+---------+-------------+----------+-------------+---------+-------------+---------+-------------+----------+-------------+
Should I use Pivot or is there a better way to do this? If

Answer 1

This is a variation on pwilcox's answer, but more concise:
select v.week,
avg(case when year = 16 then quantity end) as quantityYr16,
avg(case when year = 16 then price end) as priceYr16,
avg(case when year = 17 then quantity end) as quantityYr17,
avg(case when year = 17 then price end) as priceYr17,
. . .
sum(quantity) as totalQuantity,
sum(price) as totalPrice
from yearly cross apply
(values (concat(datename(week, date), '-', datename(year, date)))
) v(week)
group by v.week
order by v.week;
Notes:
Never use varchar() without a length. The default length varies by context and may not be long enough.
datename() is a convenient function that returns strings and not numbers.
When using the date part functions, spell out the full names of the date parts -- week, year. This makes the code easier to read.

Answer 2

For a multi column pivot of the sort that you're wanting, you're going to have to take advantage of the fact that aggregate operations don't consider null values. So place case statements inside your averages that give the quantity or price value if associated with any given year, and null otherwise.
select ap.week,
quantityYr16 = avg(case when year = 16 then quantity end),
priceYr16 = avg(case when year = 16 then price end),
quantityYr17 = avg(case when year = 17 then quantity end),
priceYr17 = avg(case when year = 17 then price end),
...
from yearly
cross apply (select week =
cast(datepart(wk,date) as nvarchar) + '-' +
right(cast(year(date) as nvarchar),2)
) ap
group by ap.week
However, this structure is for reporting. SQL doesn't handle it as well as reporting tools such as HTML, SSRS or Excel. I would do this operation with whatever reporting tool you ultimately report this with.

Answer 3

Here is a PIVOT. Assumed you did not need Dynamic
Example
Select *
From (
Select A.Week
,B.*
From (
-- YOUR ORIGINAL QUERY HERE (without the Order By) ---
) A
Cross Apply ( values (concat(year,'_Qty') ,[Quantity])
,(concat(year,'_Price'),[Price])
,(concat('Total','_Qty'),[Quantity])
,(concat('Total','_Price'),[Price])
) B(item,value)
) src
Pivot (sum(Value) for Item in ([16_Qty],[16_Price],[17_Qty],[17_Price],[18_Qty],[18_Price],[19_Qty],[19_Price],[20_Qty],[20_Price],[Z01_Qty],[Z01_Price],[Total_Qty],[Total_Price]) ) pvt
Returns