SQL PIVOT without aggregate columns

Question

create table Product_Price
    (
      id int,
      dt date,
      SellerName varchar(20),
      Product varchar(10),
      ShippingTime varchar(20),
      Price money
    )

    insert into Product_Price values (1, '2012-01-16','Sears','AA','2 days',32)
    insert into Product_Price values (2, '2012-01-16','Amazon', 'AA','4 days', 40)
    insert into Product_Price values (3, '2012-01-16','eBay','AA','1 days', 27)
    insert into Product_Price values (4, '2012-01-16','Walmart','AA','Same day', 28)
    insert into Product_Price values (5, '2012-01-16','Target', 'AA','3-4 days', 29)
    insert into Product_Price values (6, '2012-01-16','Flipcart','AA',NULL, 30)


select *
from
(select dt, product, SellerName, sum(price) as price 
from product_price group by  dt, product, SellerName) t1

pivot (sum(price) for SellerName in ([amazon],[ebay]))as bob
) 

I want 2 more columns in output (One is AmazonShippinTime another is eBayshippintime ). How can I get these? Fiddle : http://sqlfiddle.com/#!3/2210d/1

Answer 1

Since you need to pivot on two columns and use different aggregates on both columns, I would use aggregate functions with a CASE expression to get the result:

select
  dt, 
  product,
  sum(case when SellerName = 'amazon' then price else 0 end) AmazonPrice,
  max(case when SellerName = 'amazon' then ShippingTime end) AmazonShippingTime,
  sum(case when SellerName = 'ebay' then price else 0 end) ebayPrice,
  max(case when SellerName = 'ebay' then ShippingTime end) ebayShippingTime
from product_price
group by dt, product;

See SQL Fiddle with Demo . This gives a result:

|         DT | PRODUCT | AMAZONPRICE | AMAZONSHIPPINGTIME | EBAYPRICE | EBAYSHIPPINGTIME |
|------------|---------|-------------|--------------------|-----------|------------------|
| 2012-01-16 |      AA |          40 |             4 days |        27 |           1 days |