如果您使用的是 SQL Server 2005+，那么您可以通过多种方式应用 PIVOT 函数.

If you are using SQL Server 2005+, then there are several ways that you can apply the PIVOT function.

您可以以静态枢轴的形式对值进行硬编码:

You can hard-code the values in the form of a static pivot:

select *
from
(
  select sname, 
    'hour_no_'+hr+'_'+convert(nvarchar(10), dt, 120) dt,
    att
  from stud
) st
pivot
(
  max(att)
  for dt in ([hour_no_1_2013-01-21], [hour_no_2_2013-01-21],
             [hour_no_1_2013-01-22], [hour_no_2_2013-01-22])
) piv

参见 SQL Fiddle with Demo

或者您可以使用动态 sql 在运行时生成 sql 语句.查询的动态版本是:

Or you can use dynamic sql to generate the sql statement at run-time. The dynamic version of the query is:

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

select @cols 
  = STUFF((SELECT ', ' + QUOTENAME('Hour_No_'+hr+'_'++convert(nvarchar(10), dt, 120)) 
           from stud
           group by hr, dt
           order by dt, hr
            FOR XML PATH(''), TYPE
            ).value('.', 'NVARCHAR(MAX)') 
        ,1,1,'')

set @query = 'SELECT sname,' + @cols + ' from 
             (
                select sname, 
                  ''hour_no_''+hr+''_''+convert(nvarchar(10), dt, 120) dt,
                  att
                from stud
            ) x
            pivot 
            (
                max(att)
                for dt in (' + @cols + ')
            ) p '

execute(@query)

参见 SQL Fiddle with Demo.

两者都给出结果:

| SNAME | HOUR_NO_1_2013-01-21 | HOUR_NO_2_2013-01-21 | HOUR_NO_1_2013-01-22 | HOUR_NO_2_2013-01-22 |
-----------------------------------------------------------------------------------------------------
|   Abi |                    a |                    p |                   ml |                   od |
|  bala |                    p |                    a |                   od |                   ml |