#distinct 不显示重复的查询结果
(1) 对于表中一些没有唯一性约束的字段，可能存在重复的值，这时可以使用distinct来消除那些查询结果中的重复值
select cust_id from orders; +---------+ | cust_id | +---------+ | 10001 | | 10001 | | 10003 | | 10004 | | 10005 | +---------+ select distinct cust_id from orders; +---------+ | cust_id | +---------+ | 10001 | | 10003 | | 10004 | | 10005 | +---------+
#order by 查询结果排序
(1) order by 属性名 [desc|asc]
这个语句的意义是:对于查询结果，指定根据这个属性来进行升序|降序的排列
(2) 排序时对于字段中的NULL值：
asc升序时，NULL在最前面
desc降序时，NULL在最后面
select distinct cust_id from orders order by cust_id desc; +---------+ | cust_id | +---------+ | 10005 | | 10004 | | 10003 | | 10001 | +---------+
#limit 限制查询结果数目
(1) limit n: 表示显示前n条记录
(2) limit n,m: 表示从第n位置的记录开始，往后显示m条(注意:第一条记录的位置是0,往后依次类推)
select * from orders where order_num>2006; +-----------+---------------------+---------+ | order_num | order_date | cust_id | +-----------+---------------------+---------+ | 20005 | 2005-09-01 00:00:00 | 10001 | | 20006 | 2005-09-12 00:00:00 | 10003 | | 20007 | 2005-09-30 00:00:00 | 10004 | | 20008 | 2005-10-03 00:00:00 | 10005 | | 20009 | 2005-10-08 00:00:00 | 10001 | +-----------+---------------------+---------+ select * from orders where order_num>2006 limit 2; +-----------+---------------------+---------+ | order_num | order_date | cust_id | +-----------+---------------------+---------+ | 20005 | 2005-09-01 00:00:00 | 10001 | | 20006 | 2005-09-12 00:00:00 | 10003 | +-----------+---------------------+---------+ select * from orders where order_num>2006 limit 0,2; +-----------+---------------------+---------+ | order_num | order_date | cust_id | +-----------+---------------------+---------+ | 20005 | 2005-09-01 00:00:00 | 10001 | | 20006 | 2005-09-12 00:00:00 | 10003 | +-----------+---------------------+---------+ select * from orders where order_num>2006 limit 2,2; +-----------+---------------------+---------+ | order_num | order_date | cust_id | +-----------+---------------------+---------+ | 20007 | 2005-09-30 00:00:00 | 10004 | | 20008 | 2005-10-03 00:00:00 | 10005 | +-----------+---------------------+---------+
#union合并查询结果
(1) union all: 把所有查询结果合并
(2) union : 把所有查询结果合并且去除重复行
select vend_id from vendors ; +---------+ | vend_id | +---------+ | 1001 | | 1002 | | 1003 | | 1004 | | 1005 | | 1006 | +---------+ select vend_id from products; +---------+ | vend_id | +---------+ | 1001 | | 1001 | | 1001 | | 1002 | | 1002 | | 1003 | | 1003 | | 1003 | | 1003 | | 1003 | | 1003 | | 1003 | | 1005 | | 1005 | +---------+ select vend_id from vendors union all select vend_id from products; +---------+ | vend_id | +---------+ | 1001 | | 1002 | | 1003 | | 1004 | | 1005 | | 1006 | | 1001 | | 1001 | | 1001 | | 1002 | | 1002 | | 1003 | | 1003 | | 1003 | | 1003 | | 1003 | | 1003 | | 1003 | | 1005 | | 1005 | +---------+ select vend_id from vendors union select vend_id from products; +---------+ | vend_id | +---------+ | 1001 | | 1002 | | 1003 | | 1004 | | 1005 | | 1006 | +---------+
原文：http://www.cnblogs.com/snsdzjlz320/p/5737749.html