如何在TSQL拆分数据

Question

需要一些帮助搞清楚如何分割了一些数据

数据目前看起来像这样

╔═════════════════════════════════════════════╗ 
║ Name Data1 Data2 Data3 Field1 field2 Field3 ║ 
╠═════════════════════════════════════════════╣ 
║ a  1  2  3  x  y  z ║ 
╚═════════════════════════════════════════════╝ 

我需要，使其看起来像这样

分割数据

+-----------------+ 
| name data field | 
+-----------------+ 
| a  1  x | 
| a  2  y | 
| a  3  z | 
+-----------------+ 

任何人都可以帮我这个

Answer 1

有很多方法可以做到这一点（只是罗在bluefeet的回答OK），但可以使用UNION ALL一个简单的方法：

SELECT [Name], Data1 AS Data, Field1 AS Field 
FROM YourTable 
UNION ALL 
SELECT [Name], Data2 AS Data, Field2 AS Field 
FROM YourTable 
UNION ALL 
SELECT [Name], Data3 AS Data, Field3 AS Field 
FROM YourTable 

而且here is a fiddle与此演示（bluefeet提供）。

Answer 2

取决于您的SQL Server版本。在SQL Server 2008开始，您可以使用unpivot的数据CROSS APPLY：

SELECT t.name, 
    x.Data, 
    x.Field 
FROM YourTable t 
CROSS APPLY 
(
    VALUES 
     (t.Data1, t.Field1), 
     (t.Data2, t.Field2), 
     (t.Data3, t.Field3) 
) x (Data, Field); 

见SQL Fiddle with Demo。

这也可以使用UNPIVOT和SQL Server 2005+的PIVOT函数来完成：

select name, data, field 
from 
(
    select name, left(col, len(col) -1) col, value, 
    row_number() over(partition by left(col, len(col) -1) order by col) rn 
    from 
    (
    select name, 
     cast([Data1] as varchar(10)) Data1, 
     cast([Data2] as varchar(10)) Data2, 
     cast([Data3] as varchar(10)) Data3, 
     [Field1], [field2], [Field3] 
    from yourtable 
) src 
    unpivot 
    (
    value 
    for col in ([Data1], [Data2], [Data3], 
       [Field1], [field2], [Field3]) 
) unpiv 
) u 
pivot 
(
    max(value) 
    for col in (data,field) 
) piv 

见SQL Fiddle with Demo

均可以得到结果：

| NAME | DATA | FIELD | 
----------------------- 
| a | 1 |  x | 
| a | 2 |  y | 
| a | 3 |  z |