业务场景说明:
假设待审核的订单表有三种状态,分别是“正常待审核”|“即将过期”|“已经过期”
订单提交后10小时内未处理:正常待审核
订单提交后10-20小时之间未处理:即将过期
订单提交后20小时后未处理:已经过期
表结构如图:
数据准备:
要解决的问题:
校验订单表中不同状态订单的数量与前台显示是否一致
解决思路:
1、通过存储过程,获取数据库中的不同时间段的待审核订单数量
2、将后台获取的数据与前台进行比较
实现步骤如下:
=============================================================================
1、连接数据库
mysql -uroot -hlocalhost -p123456 -P3306
=============================================================================
2、进入数据库
use sqltestdb;
=============================================================================
3、创建存储过程(直接复制到mysql命令行执行就可以了):
DELIMITER //
CREATE PROCEDURE order_state(IN will_timeout int,IN already_timeout int)
BEGIN
declare count_normal int;
declare count_will int;
declare count_already int;
set @current_time=now();
set @time1=date_sub(@current_time, interval will_timeout hour);
select count(*) into count_normal from lijuan where order_time>@time1 and new_db=0;
select count_normal;
set @current_time=now();
set @time2=date_sub(@current_time, interval already_timeout hour);
select count(*) into count_will from lijuan where order_time>@time2 and new_db=0;
set count_will=count_will-count_normal;
select count_will;
select count(*) into count_already from lijuan where new_db=0;
set count_already=count_already-(count_will+count_normal);
select count_already;
END //
DELIMITER ;
==============================================================================
4、调用存储过程
call order_state(10,20);
==============================================================================
5、删除存储过程
drop procedure order_state;
==============================================================================
存储过程代码解释说明:
1-1声明结束符为“//”
DELIMITER //
1-2声明存储过程名称及输入参数的数量和类型:
存储过程名称:order_state
输入参数的数量:2
IN:表示是入参
will_timeout:即将超时所用的时间(例:10小时)
already_timeout:已经超时所用的时间(例:20小时)
CREATE PROCEDURE order_state(IN will_timeout int,IN already_timeout int)
1-3表示开始
BEGIN
1-4声明变量count_normal表示“正常”状态的订单的数量
declare count_normal int;
1-5声明变量count_will表示“即将超时”状态的订单的数量
declare count_will int;
1-6声明变量count_already表示“已经超时”状态的订单的数量
declare count_already int;
1-7获取当前时间并赋值给变量@current_time
set @current_time=now();
1-8将当前时间减去“即将超时所用的时间”,然后再赋值给变量@time1
set @time1=date_sub(@current_time, interval will_timeout hour);
1-9查询出“正常”状态的订单数量,并赋值给变量count_normal
select count(*) into count_normal from lijuan where order_time>@time1 and new_db=0;
1-10输出“正常”状态的订单数量
select count_normal;
1-11将当前时间减去“已经超时所用的时间”,然后再赋值给变量@time2
set @time2=date_sub(@current_time, interval already_timeout hour);
1-12查询出“正常”状态+“即将超时”状态的订单数量之和,并赋值给变量count_will
select count(*) into count_will from lijuan where order_time>@time2 and new_db=0;
1-13将“正常”状态+“即将超时”状态的订单数量之和(即count_will)减去“正常”状态的订单数量(即count_normal),然后再赋值给count_will
set count_will=count_will-count_normal;
1-14输出“即将超时”状态的订单数量
select count_will;
1-15查询出所有状态的订单数量(通过new_db=0过滤旧数据库的数据)
select count(*) into count_invalid from lijuan where new_db=0;
1-16将所有订单的数量减去“正常”状态+“即将超时”状态的订单数量之和(即count_will+count_normal)
set count_invalid=count_already-(count_will+count_normal);
1-17输出“已经超时”状态的订单数量
select count_already;
1-18结束
END //
1-19将结束符修改为“;”
DELIMITER ;
**********书山有路,学海无涯,无数个孤独的夜晚,需要一点小小的成就感!**********
网友评论