有下面的一段原数据：
| KS_Date | KS_MetalName | KS_TopLimit | KS_BottomLimit | KS_Average | KS_Change |
| 2011-12-14 00:00:00.000 | a | 100 | 10 | 55 | 54 |
| 2011-12-14 00:00:00.000 | b | 1 | 1 | 2 | 2 |
| 2011-12-14 00:00:00.000 | c | 1 | 1 | 10 | 1 |
| 2011-12-14 00:00:00.000 | d | 1 | 1 | 1 | 1 |
| 2011-12-14 00:00:00.000 | e | 1 | 1 | 1 | 1 |
| 2011-12-14 00:00:00.000 | f | 1 | 1 | 1 | 1 |
| 2011-12-14 00:00:00.000 | g | 1 | 1 | 1 | 1 |
| 2011-12-15 00:00:00.000 | a | 10 | 10 | 10 | -45 |
| 2011-12-16 00:00:00.000 | b | 2 | 10 | 6 | 4 |
| 2011-12-16 00:00:00.000 | c | 2653 | 2001 | 2327 | 2317 |
| 2011-12-16 00:00:00.000 | d | 300 | 10 | 155 | 154 |
| 2011-12-16 00:00:00.000 | e | 200 | 100 | 150 | 149 |
| 2011-12-16 00:00:00.000 | f | 500 | 300 | 400 | 399 |
| 2011-12-16 00:00:00.000 | g | 5000 | 200 | 2600 | 2599 |
| 2011-12-23 00:00:00.000 | a | 20 | 10 | 15 | 5 |
| 2011-12-24 | a | 22 | 3 | 3 | 3 |
| NULL | NULL | NULL | NULL | NULL | NULL |
1、求最新的日期并且KS_MetalName不重复
SELECT MAX(KS_Date) AS 最新日期, KS_MetalName
FROM KS_U_SpotPrice
GROUP BY KS_MetalName
缺陷是无法返回多行记录，只能返回最多现行：KS_Date和KS_MetalName不能满足我们的要求
2、现在变通方法如下：
SELECT KS_Date, KS_MetalName, KS_TopLimit, KS_BottomLimit, KS_Average, KS_Change
FROM KS_U_SpotPrice
WHERE (KS_Date IN
(SELECT MAX(KS_Date) AS AS最新日期
FROM KS_U_SpotPrice AS TempTable
GROUP BY KS_MetalName))
ORDER BY KS_MetalName
突然间发现上面这种方法也不能满足要求 ,但是下面的方法保证没有问题：
SELECT ID, KS_Date, KS_MetalName, KS_TopLimit, KS_BottomLimit, did, username, userleavel
FROM test AS a
WHERE (KS_Date IN
(SELECT MAX(KS_Date) AS Expr1
FROM test AS b
WHERE (a.KS_MetalName = KS_MetalName)))
ORDER BY KS_MetalName
3、如果要是求今天最新的日期的返回值方法如下：
select KS_Date,KS_MetalName,KS_TopLimit,KS_BottomLimit,KS_Average,KS_Change
from KS_U_SpotPrice
where (DATEDIFF("d",KS_Date,GETDATE())=0)
order by KS_MetalName ASC
