SQL 查询连续n天登录的用户情况

以连续3天为例，使用工具：MySQL。

1.创建SQL表：

create table if not exists orde(id varchar(10),date datetime,orders varchar(10));
insert into orde values('1' , '2019/1/1',10 );
insert into orde values('1' , '2019/1/2',109 );
insert into orde values('1' , '2019/1/3',150 );
insert into orde values('1' , '2019/1/4',99);
insert into orde values('1' , '2019/1/5',145);
insert into orde values('1' , '2019/1/6',1455);
insert into orde values('1' , '2019/1/7',199);
insert into orde values('1' , '2019/1/8',188 );
insert into orde values('4' , '2019/1/1',10 );
insert into orde values('2' , '2019/1/2',109 );
insert into orde values('3' , '2019/1/3',150 );
insert into orde values('4' , '2019/1/4',99);
insert into orde values('5' , '2019/1/5',145);
insert into orde values('6' , '2019/1/6',1455);
insert into orde values('7' , '2019/1/7',199);
insert into orde values('8' , '2019/1/8',188 );
insert into orde values('9' , '2019/1/1',10 );
insert into orde values('9' , '2019/1/2',109 );
insert into orde values('9' , '2019/1/3',150 );
insert into orde values('9' , '2019/1/4',99);
insert into orde values('9' , '2019/1/6',145);
insert into orde values('9' , '2019/1/9',1455);
insert into orde values('9' , '2019/1/10',199);
insert into orde values('9' , '2019/1/13',188 );

查看数据表：

2.使用row_number() over()  排序函数计算每个id的排名，SQL如下：

select *,row_number() over(partition by id order by date ) 'rank'
from orde;

查看数据表:

 3.将date日期字段减去rank排名字段，SQL如下：

select *,DATE_SUB(a.date,interval a.rank day) 'date_sub'
from(
select *,row_number() over(partition by id order by date ) 'rank'
from orde
) a;

查看数据：

 

4.根据id和date分组并计算分组后的数量(count)、计算最早登录和最晚登录的时间，其中substr(b.date_sub,1,10) 为截取日期到年月日字段即可，SQL如下：

select b.id,min(date) 'start_time',max(date) 'end_time',count(*) 'date_count'
from(
select *,DATE_SUB(a.date,interval a.rank day) 'date_sub'
from(
select *,row_number() over(partition by id order by date ) 'rank'
from orde
) a
) b
group by substr(b.date_sub,1,10),id
having count(*) >= 3
;

 查看数据：

参考资料： 

(28条消息) SQL查询连续七天以上下单的用户_方便面袋袋的博客-CSDN博客_统计连续七天都有下单的https://blog.csdn.net/qq_43807789/article/details/99091753?spm=1001.2101.3001.6661.1&utm_medium=distribute.pc_relevant_t0.none-task-blog-2~default~CTRLIST~default-1.highlightwordscore&depth_1-utm_source=distribute.pc_relevant_t0.none-task-blog-2~default~CTRLIST~default-1.highlightwordscore