add : 176

h-j-13 · h-j-13 · commit 26a3bc609c5b · 2017-12-11T16:05:45.000+08:00
diff --git a/Soulution/LeetCode176. Second Highest Salary.md b/Soulution/LeetCode176. Second Highest Salary.md
@@ -0,0 +1,47 @@
+# LeetCode176. Second Highest Salary
+
+# 题目
+Write a SQL query to get the second highest salary from the ```Employee``` table.
+```
++----+--------+
+| Id | Salary |
++----+--------+
+| 1  | 100    |
+| 2  | 200    |
+| 3  | 300    |
++----+--------+
+```
+For example, given the above Employee table, the query should return 200 as the second highest salary. If there is no second highest salary, then the query should return null.
+```
++---------------------+
+| SecondHighestSalary |
++---------------------+
+| 200                 |
++---------------------+
+```
+
+题目大意 : 找出第二高的薪水
+
+## 解题思路
+
+首先很容易想到,先WHERE salary != 最高的薪水 然后在做排序 或者 取最大值就可获得第二高的薪水,但是若是第三 第N 高的薪水则要迭代此方法数次 造成效率下降,在只select 一次表的情况 排序后使用limit选择第N条记录 不失为一个更高效的选择
+
+### LIMIT 
+```sql
+SELECT Salary AS 'SecondHighestSalary' FROM Employee ORDER BY Salary DESC LIMIT 1,1;
+```
+![](http://upload-images.jianshu.io/upload_images/5617720-be6e1556f0609a31.png?imageMogr2/auto-orient/strip%7CimageView2/2/w/1240)
+
+结果在只有一条记录的时候出错了...
+
+因为从第一条记录向下偏移了一条, 导致本该的null的变成了无结果
+再加一层select 筛选即可 因为 select 的无结果默认值为null
+最后别忘了加上DISTINCT 因为同一工资的人可能有很多
+
+进行修改
+```sql
+SELECT (
+    SELECT DISTINCT Salary AS 'SecondHighestSalary' FROM Employee ORDER BY Salary DESC LIMIT 1 OFFSET 1
+) AS 'SecondHighestSalary';  
+```
+
