sql - 如何选择以使行可以是列？

Question

我有一张这样的表：

   Item      Name             Name_location   Price

    1      item1_london        london         10
    1      item1_beijing       bejing         10
    2      item2_london        london         20
    2      item2_beijing       bejing         20

基本上，这个表意味着我有很多项目each item will have a different name in different location (two locations: london and beijing)。

如何查询以便我可以像这样获取表格：

  Item     london             bejing          Price
    1      item1_london       item1_beijing    10
    2      item2_london       item2_beijing    20

我正在使用MS SQL Server。

编辑：更新了两个表，添加了另一列

Answer 1

如果您只有两个位置，那么您可以使用带有聚合的CASE语句：

CASE with Aggregate ：

select item,
  max(case when Name_location = 'london' then name end) london,
  max(case when Name_location = 'bejing' then name end) bejing,
  sum(price) price
from yourtable
group by item

请参阅SQL Fiddle with Demo

或

select item,
  max(case when Name_location = 'london' then name end) london,
  max(case when Name_location = 'bejing' then name end) bejing,
  price
from yourtable
group by item, price

使用PIVOT函数，有两种静态/动态方式：

静态PIVOT

select item, [london], [bejing], price
from 
(
  select item, name, name_location, price
  from yourtable
) x
pivot
(
  max(name)
  for name_location in ([london], [bejing])
) p

请参阅SQL Fiddle with Demo

如果你有一个已知数量的值，静态PIVOT版本会很好用，如果你有一个未知的数字，那么你可以使用动态的sql：

动态数据透视：

DECLARE @cols AS NVARCHAR(MAX),
    @query  AS NVARCHAR(MAX)

select @cols = STUFF((SELECT distinct ',' + QUOTENAME(name_location) 
                    from yourtable
            FOR XML PATH(''), TYPE
            ).value('.', 'NVARCHAR(MAX)') 
        ,1,1,'')

set @query = 'SELECT item,' + @cols + ', price from 
             (
                select item, name, name_location, price
                from yourtable
            ) x
            pivot 
            (
                max(name)
                for name_location in (' + @cols + ')
            ) p '

execute(@query)

请参阅SQL Fiddle with Demo

Answer 2

使用pivot

select * 
from YourTable src
pivot (max(name) for name_location in ([london], [beijing]) ) p

Answer 3

SELECT DISTINCT o.Item, l.itemName as London, b.ItemName as Beijing
FROM myTable o inner join myTable l ON o.item = l.item
INNER JOIN myTable b ON o.item = b.item
WHERE l.location = 'london' AND b.location = 'beijing'

SQLFiddle链接：http://sqlfiddle.com/#!3/93c2a/8

Answer 4

select item,london,beijing
from t1
pivot(max(Name) for Name_location in (london, beijing)) pvt

select item,london,beijing,price
from t1
pivot(max(Name) for Name_location in (london, beijing)) pvt