我对SQL不太熟练,所以希望有人可以帮助我。
我的表格中有一个date_of_post列,如下所示(示例)2015-08-31 11:00:00
。
我使用INTERVAL 1 DAY
获取过去24小时。然而,它看起来比过去24小时还要多。这是我用来获取数据的查询
SELECT DATE_ADD(date(t.date_of_post),
INTERVAL hour(t.date_of_post) HOUR) AS dateTime,
count(*) as entries
FROM `soc_stat` t
WHERE `main_tag` = 'morgenmad'
AND t.date_of_post > DATE_SUB(CURDATE(), INTERVAL 1 DAY)
GROUP BY date(t.date_of_post), hour(t.date_of_post)
它返回以下内容:
2015-08-31 11:00:00 = 11
2015-08-31 12:00:00 = 2
2015-08-31 13:00:00 = 3
2015-08-31 14:00:00 = 3
2015-08-31 15:00:00 = 1
2015-08-31 16:00:00 = 3
2015-08-31 17:00:00 = 2
2015-08-31 19:00:00 = 1
2015-09-01 04:00:00 = 1
2015-09-01 05:00:00 = 3
2015-09-01 06:00:00 = 9
2015-09-01 07:00:00 = 33
2015-09-01 08:00:00 = 38
2015-09-01 09:00:00 = 29
2015-09-01 10:00:00 = 13
2015-09-01 11:00:00 = 12
2015-09-01 12:00:00 = 6
2015-09-01 13:00:00 = 5
我不明白11:00:00
和12:00:00
中存在13:00:00
,2015-08-31
和2015-09-01
的原因。它不应该只返回过去24小时吗?
答案 0 :(得分:1)
.table-header-rotated th.rotate-45 {
height: 80px;
width: 40px;
min-width: 40px;
max-width: 40px;
position: relative;
vertical-align: bottom;
padding: 0;
font-size: 12px;
line-height: 0.8;
white-space: nowrap;
}
返回“今天的开头”。将其替换为CURDATE()
答案 1 :(得分:1)
视觉效果可能有所帮助。如果你使用别名坚持使用它们。当您使用聚合函数(如count)时,将按所有非聚合列进行分组。
for me it is 2015-09-01 08:47:00
create table soc_stat
( id int auto_increment primary key,
main_tag varchar(20) not null,
date_of_post datetime not null
);
truncate table soc_stat;
insert soc_stat (main_tag,date_of_post) values ('morgenmad','2015-09-02 11:00:00');
insert soc_stat (main_tag,date_of_post) values ('morgenmad','2015-09-01 11:00:00');
insert soc_stat (main_tag,date_of_post) values ('morgenmad','2015-09-01 09:00:00');
insert soc_stat (main_tag,date_of_post) values ('morgenmad','2015-09-01 08:00:00');
insert soc_stat (main_tag,date_of_post) values ('morgenmad','2015-09-01 07:00:00');
insert soc_stat (main_tag,date_of_post) values ('morgenmad','2015-08-31 09:00:00');
insert soc_stat (main_tag,date_of_post) values ('morgenmad','2015-08-31 08:00:00');
insert soc_stat (main_tag,date_of_post) values ('morgenmad','2015-08-31 07:00:00');
SELECT date(t.date_of_post) dt, hour(t.date_of_post) hr,count(*) as entries
FROM `soc_stat` t
WHERE t.`main_tag` = 'morgenmad'
AND t.date_of_post between DATE_SUB(now(), INTERVAL 1 DAY) and now()
GROUP BY dt,hr
order by t.date_of_post desc;
+------------+------+---------+
| dt | hr | entries |
+------------+------+---------+
| 2015-09-01 | 8 | 1 |
| 2015-09-01 | 7 | 1 |
| 2015-08-31 | 9 | 1 |
+------------+------+---------+