Question

我有一个包含以下结构的表

id      date      timestamp  licenseid   storeid    deviceid    value
1     2015-06-12   17:36:15   lic0001       1         0add      52
2     2015-06-12   17:36:15   lic0002       1         0add      54
3     2015-06-12   17:36:15   lic0003       1         0add      53
4     2015-06-12   17:36:21   lic0001       1         0add      54
5     2015-06-12   17:36:21   lic0002       1         0add      59
6     2015-06-12   17:36:21   lic0003       1         0add      62
7     2015-06-12   17:36:21   lic0004       1         0add      55
8     2015-06-12   17:36:15   lic0001       1         0bdd      53
9     2015-06-12   17:36:15   lic0002       1         0bdd      52
10    2015-06-12   17:36:15   lic0003       1         0bdd      52

我需要检查是否有任何许可证已停止报告值。我想到的方法是将time_stamps分成15分钟，然后检查是否缺少任何许可证的数据，持续时间更长，然后深潜。

需要帮助查询，可以从00:00开始每隔15分钟按时间戳分组，并显示该持续时间内的值计数。我已经尝试了以下查询，但时间戳没有显示正确的值。它只显示一行，时间键为0。

SELECT count(*),  round(unix_timestamp(time_stamp)/(15*60)) AS timekey
FROM     data
GROUP BY timekey;

PS：我已按照上述说明使用此帖子中的此查询，但查询为时间键返回0：Group mysql query by 15 min intervals

Answer 1

UNIX_TIMESTAMP()不适用于time参数。它适用于datetime。所以，一种方法是：

SELECT count(*), 
       round(unix_timestamp(addtime(date(0), time_stamp) )/(15*60)) AS timekey
FROM data
GROUP BY timekey;

编辑：

你总是可以使用蛮力方法：

SELECT COUNT(*), hour(timestamp) as hr, floor(minute(timestamp) / 15) as minute
FROM data
GROUP BY hr, minute;

Answer 2

我目前没有MySQL来测试我提出的解决方案，但您可以尝试使用CASE语句吗？

SELECT (CASE 
WHEN MINUTE(timestamp) BETWEEN 0 AND 15 THEN 1
WHEN MINUTE(timestamp) BETWEEN 16 AND 30 THEN 2
WHEN MINUTE(timestamp) BETWEEN 31 AND 45 THEN 3
ELSE 4 END) AS MINUTEGROUP

然后在MINUTEGROUP上分组

Answer 3

使用

create table foo (id int, date date, time_stamp time);

insert into foo (id, date, time_stamp) values (1, '2015-06-12', '17:26:15');
insert into foo (id, date, time_stamp) values (2, '2015-06-12', '17:26:15');
insert into foo (id, date, time_stamp) values (3, '2015-06-12', '17:26:15');
insert into foo (id, date, time_stamp) values (4, '2015-06-12', '17:36:21');
insert into foo (id, date, time_stamp) values (5, '2015-06-12', '17:36:21');
insert into foo (id, date, time_stamp) values (6, '2015-06-12', '17:36:21');
insert into foo (id, date, time_stamp) values (7, '2015-06-12', '18:36:21');
insert into foo (id, date, time_stamp) values (8, '2015-06-12', '18:36:15');
insert into foo (id, date, time_stamp) values (9, '2015-06-12', '18:36:15');
insert into foo (id, date, time_stamp) values (10, '2015-06-12', '18:36:15');

看起来像

select
    date,
    count(*),
    min(addtime(date, time_stamp)) min_time,
    max(addtime(date, time_stamp)) max_time,
    date_add(date, interval timestampdiff(second, date, addtime(date, time_stamp)) div 15 * 15 minute) start_time
from
    foo
group by
    date,
    timestampdiff(minute, date, addtime(date, time_stamp)) div 15

请参阅http://sqlfiddle.com/#!9/c5ced/37/0

SQL：组时间每15分钟一次

3 个答案: