力扣數據庫題目176第二高的薪水

力扣數據庫題目176第二高的薪水

題目

編寫一個 SQL 查詢，獲取 Employee 表中第二高的薪水（Salary） 。

+----+--------+
| Id | Salary |
+----+--------+
| 1 | 100 |
| 2 | 200 |
| 3 | 300 |
+----+--------+
例如上述 Employee 表，SQL查詢應該返回 200 做爲第二高的薪水。若是不存在第二高的薪水，那麼查詢應返回 null。

+---------------------+
| SecondHighestSalary |
+---------------------+
| 200 |
+---------------------+

來源：力扣（LeetCode）

方案一

SELECT salary SecondHighestSalary
FROM (		
	SELECT DISTINCT salary FROM test.employee 
    UNION ALL 
    SELECT NULL  
    UNION ALL 
	SELECT NULL         
) t ORDER BY salary DESC LIMIT 1,1

方案二

SELECT salary SecondHighestSalary 
FROM (		
	SELECT salary,DENSE_RANK() OVER(ORDER BY salary DESC) num
    FROM test.employee 
    UNION ALL 
    SELECT NULL,2
    UNION ALL 
	SELECT NULL,2    
) t
WHERE num = 2
LIMIT 1;

方案三

SELECT t.salary SecondHighestSalary
FROM (
	SELECT salary  
	FROM test.employee 	
    WHERE salary < (
		SELECT MAX(salary)
        FROM test.employee 
	) 
    UNION ALL 
    SELECT NULL   
) t
ORDER BY t.salary DESC LIMIT 1

方案四

SELECT t.salary SecondHighestSalary
FROM (
	SELECT salary  
	FROM test.employee 	
    WHERE salary < ANY(
		SELECT salary
        FROM test.employee 
	) 
    UNION ALL 
    SELECT NULL   
) t
ORDER BY t.salary DESC LIMIT 1

方案五

SELECT (SELECT salary 
FROM (		
	SELECT DISTINCT salary FROM test.employee        
) t ORDER BY salary DESC LIMIT 1,1)SecondHighestSalary

方案六

SELECT (SELECT salary 
FROM (   
  SELECT salary,DENSE_RANK() OVER(ORDER BY salary DESC) num
  FROM test.employee 
) t
WHERE num = 2
LIMIT 1) SecondHighestSalary

方案七

SELECT (SELECT t.salary 
FROM (
	SELECT salary  
	FROM test.employee 	
    WHERE salary < (
		SELECT MAX(salary)
        FROM test.employee 
	)  
) t
ORDER BY t.salary DESC LIMIT 1) SecondHighestSalary

方案八

SELECT (SELECT t.salary 
FROM (
	SELECT salary  
	FROM test.employee 	
    WHERE salary < ANY(
		SELECT salary
        FROM test.employee 
	) 
) t
ORDER BY t.salary DESC LIMIT 1) SecondHighestSalary

分析

題目要求的結果是一行一列，並保證兩個條件

一、列別名SecondHighestSalary

二、行至少一行NULL,爲了維持這個問題

前四個方案是製造出來一個NULL行，後四個方案是0行使用select （）子查詢後造出NULL行

爲了保證條件2，廢了不少事。

方案一五使用distinct，二六使用的開窗，三七使用的子查詢，四八使用的ANY

建表

CREATE TABLE `employee` (
  `id` int NOT NULL AUTO_INCREMENT,
  `salary` int NOT NULL DEFAULT '0',
  PRIMARY KEY (`id`)
) ENGINE=InnoDB  COLLATE=utf8mb4_0900_ai_ci;

數據初始化

INSERT INTO `test`.employee(`id`,`salary`)
VALUES(1,100),(2,200),(3,300)