SQLite HAVING子句

SQLite HAVING子句用于指定过滤分组的结果，并作为最终查询结果的条件。 WHERE子句将条件放在选定的列上，而HAVING子句指定的条件是由GROUP BY子句创建的列分组上(使用HAVING子句条件一定要作用在由GROUP BY子句指定列上)。

SELECT查询中HAVING子句的位置：

SELECT  
FROM  
WHERE  
GROUP BY  
HAVING  
ORDER BY

语法

SELECT column1, column2  
FROM table1, table2  
WHERE [ conditions ]  
GROUP BY column1, column2  
HAVING [ conditions ]  
ORDER BY column1, column2

下面举个例子来演示如何使用HAVING子句。假设有一个名为student的表，并具有以下数据：

sqlite> select * from student;
1|Maxsu|27|Shengzhen|20000.0
2|Minsu|25|Beijing|15000.0
3|Avgsu|23|Shanghai|2000.0
4|Linsu|25|Guangzhou|65000.0
5|Sqlsu|26|Haikou|25000.0
6|Javasu|21|Shengzhen|18000.0
7|Linsu|27|Haikou|10000.0
8|Minsu|23|Guangzhou|5000.0
9|Maxsu|23|Shenzhen|9000.0
sqlite>

示例1：

下面查询name的数量小于2的所有记录，在查询之前，先来查询看看每个名字的数量 -

-- 名字的数量
SELECT name, count(name) as total_number FROM student GROUP BY name;

-- `name`的数量小于`2`的所有记录
SELECT name, count(name) as total_number FROM student GROUP BY NAME HAVING COUNT(NAME) < 2;

执行上面语句，得到结果如下 -

-- 所有记录
sqlite> select * from student;
1|Maxsu|27|Shengzhen|20000.0
2|Minsu|25|Beijing|15000.0
3|Avgsu|23|Shanghai|2000.0
4|Linsu|25|Guangzhou|65000.0
5|Sqlsu|26|Haikou|25000.0
6|Javasu|21|Shengzhen|18000.0
7|Linsu|27|Haikou|10000.0
8|Minsu|23|Guangzhou|5000.0
9|Maxsu|23|Shenzhen|9000.0
sqlite>
sqlite> 
-- 每个名字的数量
sqlite> SELECT name, count(name) as total_number FROM student GROUP BY name;
Avgsu|1
Javasu|1
Linsu|2
Maxsu|2
Minsu|2
Sqlsu|1
-- 查询数量小于2的名字
sqlite> SELECT name, count(name) as total_number FROM student GROUP BY NAME HAVING total_number < 2;
Avgsu|1
Javasu|1
Sqlsu|1
sqlite>

示例2：

下面查询address的数量大于等于2的所有记录，在查询之前，先来查询看看每个地址的数量 -

sqlite> SELECT address, count(address) as total_number FROM student GROUP BY address ;
Beijing|1
Guangzhou|2
Haikou|2
Shanghai|1
Shengzhen|2
Shenzhen|1
sqlite> SELECT address, count(address) as total_number FROM student GROUP BY address HAVING total_number >= 2;
Guangzhou|2
Haikou|2
Shengzhen|2
sqlite>