我正在尝试使用查询来获取特定输出。这是我的要求：
我有两张桌子，其中一张是生物识别机指纹打孔细节的摘录：
我有两个名为 raw_test 和日历
的表格mysql> desc raw_test;
+---------------+----------+------+-----+---------+-------+
| Field | Type | Null | Key | Default | Extra |
+---------------+----------+------+-----+---------+-------+
| emp_id | int(4) | YES | | NULL | |
| time_loginout | datetime | YES | | NULL | |
+---------------+----------+------+-----+---------+-------+
2 rows in set (0.00 sec)
mysql> desc calender;
+---------------+----------+------+-----+---------+-------+
| Field | Type | Null | Key | Default | Extra |
+---------------+----------+------+-----+---------+-------+
| time_loginout | datetime | YES | | NULL | |
+---------------+----------+------+-----+---------+-------+
1 row in set (0.00 sec)
raw_test表包含以下数据：
mysql> select * from raw_test;
+--------+---------------------+
| emp_id | time_loginout |
+--------+---------------------+
| 111 | 2014-01-01 09:40:00 |
| 111 | 2014-01-01 16:00:00 |
| 222 | 2014-01-01 09:45:00 |
| 222 | 2014-01-01 17:04:00 |
| 333 | 2014-01-01 09:10:00 |
| 333 | 2014-01-01 17:10:00 |
| 444 | 2014-01-01 09:32:00 |
| 444 | 2014-01-01 17:40:00 |
| 555 | 2014-01-01 09:40:00 |
| 555 | 2014-01-01 17:50:00 |
| 111 | 2014-01-02 09:21:00 |
| 111 | 2014-01-02 17:20:00 |
| 222 | 2014-01-02 09:11:00 |
| 222 | 2014-01-02 17:30:00 |
| 333 | 2014-01-02 09:10:00 |
| 333 | 2014-01-02 17:10:00 |
| 444 | 2014-01-02 09:18:00 |
| 444 | 2014-01-02 16:54:00 |
| 555 | 2014-01-02 09:14:00 |
| 555 | 2014-01-02 16:40:00 |
| 111 | 2014-01-03 09:13:00 |
| 111 | 2014-01-03 16:30:00 |
| 222 | 2014-01-03 09:12:00 |
| 222 | 2014-01-03 17:40:00 |
| 333 | 2014-01-03 09:11:00 |
| 333 | 2014-01-03 17:10:00 |
| 444 | 2014-01-03 09:10:00 |
| 444 | 2014-01-03 17:22:00 |
| 555 | 2014-01-03 09:18:00 |
| 555 | 2014-01-03 17:21:00 |
| 111 | 2014-01-04 09:20:00 |
| 111 | 2014-01-04 17:12:00 |
| 222 | 2014-01-04 09:30:00 |
| 222 | 2014-01-04 17:11:00 |
| 333 | 2014-01-04 09:43:00 |
| 333 | 2014-01-04 17:42:00 |
| 444 | 2014-01-04 09:32:00 |
| 444 | 2014-01-04 17:23:00 |
| 555 | 2014-01-04 09:21:00 |
| 555 | 2014-01-04 17:54:00 |
+--------+---------------------+
40 rows in set (0.00 sec)
和日历保存数据如下：
mysql> select * from calender;
+---------------------+
| time_loginout |
+---------------------+
| 2014-01-01 09:40:00 |
| 2014-01-01 16:00:00 |
| 2014-01-01 09:45:00 |
| 2014-01-01 17:04:00 |
| 2014-01-01 09:10:00 |
| 2014-01-01 17:10:00 |
| 2014-01-01 09:32:00 |
| 2014-01-01 17:40:00 |
| 2014-01-01 09:40:00 |
| 2014-01-01 17:50:00 |
| 2014-01-02 09:21:00 |
| 2014-01-02 17:20:00 |
| 2014-01-02 09:11:00 |
| 2014-01-02 17:30:00 |
| 2014-01-02 09:10:00 |
| 2014-01-02 17:10:00 |
| 2014-01-02 09:18:00 |
| 2014-01-02 16:54:00 |
| 2014-01-02 09:14:00 |
| 2014-01-02 16:40:00 |
| 2014-01-03 09:13:00 |
| 2014-01-03 16:30:00 |
| 2014-01-03 09:12:00 |
| 2014-01-03 17:40:00 |
| 2014-01-03 09:11:00 |
| 2014-01-03 17:10:00 |
| 2014-01-03 09:10:00 |
| 2014-01-03 17:22:00 |
| 2014-01-03 09:18:00 |
| 2014-01-03 17:21:00 |
| 2014-01-04 09:20:00 |
| 2014-01-04 17:12:00 |
| 2014-01-04 09:30:00 |
| 2014-01-04 17:11:00 |
| 2014-01-04 09:43:00 |
| 2014-01-04 17:42:00 |
| 2014-01-04 09:32:00 |
| 2014-01-04 17:23:00 |
| 2014-01-04 09:21:00 |
| 2014-01-04 17:54:00 |
+---------------------+
40 rows in set (0.00 sec
）
此处日历表是 raw_test 表的摘录： 即，我已按如下方式创建了日历表：
create table calender as (select time_loginout from raw_test);
如果我执行以下查询，则无法获得预期的输出：
select r.emp_id,
max(case
when c.time_loginout >= '2014-01-01 09:00:00' and c.time_loginout <= '2014-01-01 09:15:00' then 'P'
when c.time_loginout > '2014-01-01 09:15:00' and c.time_loginout <= '2014-01-01 09:30:00' then 'L'
when c.time_loginout > '2014-01-01 09:30:00' and c.time_loginout <= '2014-01-01 10:00:00' then '1H'
when c.time_loginout > '2014-01-01 10:00:00' and c.time_loginout <= '2014-01-01 11:00:00' then '2H'
when c.time_loginout > '2014-01-01 11:00:00' and c.time_loginout <= '2014-01-01 12:00:00' then '3H'
when c.time_loginout > '2014-01-01 12:00:00' and c.time_loginout <= '2014-01-01 13:30:00' then 'HD'
when c.time_loginout > '2014-01-01 13:30:00' and c.time_loginout <= '2014-01-01 14:00:00' then 'HD'
when c.time_loginout > '2014-01-01 14:00:00' and c.time_loginout <= '2014-01-01 17:00:00' then 'FD'
else 'EO' end)'01',
max(case
when c.time_loginout >= '2014-01-02 09:00:00' and c.time_loginout <= '2014-01-02 09:15:00' then 'P'
when c.time_loginout > '2014-01-02 09:15:00' and c.time_loginout <= '2014-01-02 09:30:00' then 'L'
when c.time_loginout > '2014-01-02 09:30:00' and c.time_loginout <= '2014-01-02 10:00:00' then '1H'
when c.time_loginout > '2014-01-02 10:00:00' and c.time_loginout <= '2014-01-02 11:00:00' then '2H'
when c.time_loginout > '2014-01-02 11:00:00' and c.time_loginout <= '2014-01-02 12:00:00' then '3H'
when c.time_loginout > '2014-01-02 12:00:00' and c.time_loginout <= '2014-01-02 13:30:00' then 'HD'
when c.time_loginout > '2014-01-02 13:30:00' and c.time_loginout <= '2014-01-02 14:00:00' then 'HD'
when c.time_loginout > '2014-01-02 14:00:00' and c.time_loginout <= '2014-01-02 17:00:00' then 'FD'
else 'EO' end)'02',
max(case
when c.time_loginout >= '2014-01-03 09:00:00' and c.time_loginout <= '2014-01-03 09:15:00' then 'P'
when c.time_loginout > '2014-01-03 09:15:00' and c.time_loginout <= '2014-01-03 09:30:00' then 'L'
when c.time_loginout > '2014-01-03 09:30:00' and c.time_loginout <= '2014-01-03 10:00:00' then '1H'
when c.time_loginout > '2014-01-03 10:00:00' and c.time_loginout <= '2014-01-03 11:00:00' then '2H'
when c.time_loginout > '2014-01-03 11:00:00' and c.time_loginout <= '2014-01-03 12:00:00' then '3H'
when c.time_loginout > '2014-01-03 12:00:00' and c.time_loginout <= '2014-01-03 13:30:00' then 'HD'
when c.time_loginout > '2014-01-03 13:30:00' and c.time_loginout <= '2014-01-03 14:00:00' then 'HD'
when c.time_loginout > '2014-01-03 14:00:00' and c.time_loginout <= '2014-01-03 17:00:00' then 'FD'
else 'EO' end)'03',
max(case
when c.time_loginout >= '2014-01-04 09:00:00' and c.time_loginout <= '2014-01-04 09:15:00' then 'P'
when c.time_loginout > '2014-01-04 09:15:00' and c.time_loginout <= '2014-01-04 09:30:00' then 'L'
when c.time_loginout > '2014-01-04 09:30:00' and c.time_loginout <= '2014-01-04 10:00:00' then '1H'
when c.time_loginout > '2014-01-04 10:00:00' and c.time_loginout <= '2014-01-04 11:00:00' then '2H'
when c.time_loginout > '2014-01-04 11:00:00' and c.time_loginout <= '2014-01-04 12:00:00' then '3H'
when c.time_loginout > '2014-01-04 12:00:00' and c.time_loginout <= '2014-01-04 13:30:00' then 'HD'
when c.time_loginout > '2014-01-04 13:30:00' and c.time_loginout <= '2014-01-04 14:00:00' then 'HD'
when c.time_loginout > '2014-01-04 14:00:00' and c.time_loginout <= '2014-01-04 17:00:00' then 'FD'
else 'EO' end)'04'
from raw_test r left join calender c on r.time_loginout=c.time_loginout group by r.emp_id;
以上查询的输出：
+--------+------+------+------+------+
| emp_id | 01 | 02 | 03 | 04 |
+--------+------+------+------+------+
| 111 | FD | L | P | L |
| 222 | EO | P | P | L |
| 333 | P | P | P | EO |
| 444 | EO | L | P | EO |
| 555 | EO | P | L | L |
+--------+------+------+------+------+
5 rows in set (0.00 sec)
实际上我正在寻找每个emp_id可以是以下之一：
P - present
L - late
EO - eary out
1H - 1 hour late
2H - 2 hour late
3H - 3 hour late
HD - half day late
FD - full day late
但是，emp_id 111在9:40之前打了一拳，并显示了FD，它应该显示为1H。
请在这方面帮助我。 感谢你。
答案 0 :(得分：0)
如果您正在寻找入住时间，您只希望最小值不是最大值。此外，我认为没有必要你的第二个表日历。在这个例子中，它似乎没有做任何事情。试试这个：
select emp_id,
MIN(case
when time_loginout >= '2014-01-01 09:00:00' and time_loginout <= '2014-01-01 09:15:00' then 'P'
when time_loginout > '2014-01-01 09:15:00' and time_loginout <= '2014-01-01 09:30:00' then 'L'
when time_loginout > '2014-01-01 09:30:00' and time_loginout <= '2014-01-01 10:00:00' then '1H'
when time_loginout > '2014-01-01 10:00:00' and time_loginout <= '2014-01-01 11:00:00' then '2H'
when time_loginout > '2014-01-01 11:00:00' and time_loginout <= '2014-01-01 12:00:00' then '3H'
when time_loginout > '2014-01-01 12:00:00' and time_loginout <= '2014-01-01 13:30:00' then 'HD'
when time_loginout > '2014-01-01 13:30:00' and time_loginout <= '2014-01-01 14:00:00' then 'HD'
when time_loginout > '2014-01-01 14:00:00' and time_loginout <= '2014-01-01 17:00:00' then 'FD'
else 'EO' end)'01',
MIN(case
when time_loginout >= '2014-01-02 09:00:00' and time_loginout <= '2014-01-02 09:15:00' then 'P'
when time_loginout > '2014-01-02 09:15:00' and time_loginout <= '2014-01-02 09:30:00' then 'L'
when time_loginout > '2014-01-02 09:30:00' and time_loginout <= '2014-01-02 10:00:00' then '1H'
when time_loginout > '2014-01-02 10:00:00' and time_loginout <= '2014-01-02 11:00:00' then '2H'
when time_loginout > '2014-01-02 11:00:00' and time_loginout <= '2014-01-02 12:00:00' then '3H'
when time_loginout > '2014-01-02 12:00:00' and time_loginout <= '2014-01-02 13:30:00' then 'HD'
when time_loginout > '2014-01-02 13:30:00' and time_loginout <= '2014-01-02 14:00:00' then 'HD'
when time_loginout > '2014-01-02 14:00:00' and time_loginout <= '2014-01-02 17:00:00' then 'FD'
else 'EO' end)'02',
MIN(case
when time_loginout >= '2014-01-03 09:00:00' and time_loginout <= '2014-01-03 09:15:00' then 'P'
when time_loginout > '2014-01-03 09:15:00' and time_loginout <= '2014-01-03 09:30:00' then 'L'
when time_loginout > '2014-01-03 09:30:00' and time_loginout <= '2014-01-03 10:00:00' then '1H'
when time_loginout > '2014-01-03 10:00:00' and time_loginout <= '2014-01-03 11:00:00' then '2H'
when time_loginout > '2014-01-03 11:00:00' and time_loginout <= '2014-01-03 12:00:00' then '3H'
when time_loginout > '2014-01-03 12:00:00' and time_loginout <= '2014-01-03 13:30:00' then 'HD'
when time_loginout > '2014-01-03 13:30:00' and time_loginout <= '2014-01-03 14:00:00' then 'HD'
when time_loginout > '2014-01-03 14:00:00' and time_loginout <= '2014-01-03 17:00:00' then 'FD'
else 'EO' end)'03',
MIN(case
when time_loginout >= '2014-01-04 09:00:00' and time_loginout <= '2014-01-04 09:15:00' then 'P'
when time_loginout > '2014-01-04 09:15:00' and time_loginout <= '2014-01-04 09:30:00' then 'L'
when time_loginout > '2014-01-04 09:30:00' and time_loginout <= '2014-01-04 10:00:00' then '1H'
when time_loginout > '2014-01-04 10:00:00' and time_loginout <= '2014-01-04 11:00:00' then '2H'
when time_loginout > '2014-01-04 11:00:00' and time_loginout <= '2014-01-04 12:00:00' then '3H'
when time_loginout > '2014-01-04 12:00:00' and time_loginout <= '2014-01-04 13:30:00' then 'HD'
when time_loginout > '2014-01-04 13:30:00' and time_loginout <= '2014-01-04 14:00:00' then 'HD'
when time_loginout > '2014-01-04 14:00:00' and time_loginout <= '2014-01-04 17:00:00' then 'FD'
else 'EO' end)'04'
from raw_test group by emp_id;