时间范围之间的过滤和组结果

Question

1. timesetup表（Unnormalized），此表用于过滤结果的时间条件

   | time_id | period_from | period_to  | session1_from | session1_to | session2_from | session2_to | session3_from | session3_to | session4_from | session4_to | session5_from | session5_to |
   |---------|-------------|------------|---------------|-------------|---------------|-------------|---------------|-------------|---------------|-------------|---------------|-------------|
   | 1       | 10/09/2015  | 11/09/2015 | 04:00:00      | 05:00:00    | 12:00:00      | 13:00:00    | 15:00:00      | 16:00:00    | 18:00:00      | 18:35:00    | 19:00:00      | 20:00:00    |
   | 2       | 12/09/2015  | 13/09/2015 | 04:10:00      | 05:10:00    | 12:10:00      | 13:10:00    | 15:10:00      | 16:10:00    | 18:10:00      | 18:45:00    | 19:10:00      | 20:10:00    |
   |------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|

规范化版本：

dateperiode表：

| period_id | date_from  | date_to    |
|-----------|------------|------------|
| 1         | 10/09/2015 | 11/09/2015 |
| 2         | 12/09/2015 | 13/09/2015 |
|-------------------------------------|

sessionrange表：

| sessionrange_id | session   | session_from | session_to  |
|-----------------|-----------|--------------|-------------|
| 1               | 1         | 04:00:00     | 05:00:00    |
| 2               | 2         | 12:00:00     | 13:00:00    |
| 3               | 3         | 15:00:00     | 16:00:00    |
| 4               | 4         | 18:00:00     | 18:35:00    |
| 5               | 5         | 19:00:00     | 20:00:00    |
| 6               | 1         | 04:10:00     | 05:10:00    |
| 7               | 2         | 12:10:00     | 13:10:00    |
| 8               | 3         | 15:10:00     | 16:10:00    |
| 9               | 4         | 18:10:00     | 18:45:00    |
| 10              | 5         | 19:10:00     | 20:10:00    |
|-----------------|-----------|--------------|-------------|

timsetup表：

| period_id | sessionrange_id |
|-----------|-----------------|
| 1         | 1               |
| 1         | 2               |
| 1         | 3               |
| 1         | 4               |
| 1         | 5               |
| 2         | 6               |
| 2         | 7               |
| 2         | 8               |
| 2         | 9               |
| 2         | 10              |
|-----------|-----------------|

2. checktime表，此表包含指纹扫描仪（tapping_time字段）的数据

   |userid  | tapping_time         |
   |--------|----------------------|
   |234     | 10/09/2015 04:20:04  |
   |234     | 10/09/2015 04:20:06  |
   |234     | 10/09/2015 12:15:35  |
   |234     | 10/09/2015 15:31:11  |
   |234     | 10/09/2015 18:19:10  |
   |234     | 10/09/2015 18:19:15  |
   |234     | 10/09/2015 19:37:53  |
   |234     | 11/09/2015 04:38:42  |
   |234     | 11/09/2015 04:38:47  |
   |234     | 11/09/2015 12:21:27  |
   |234     | 11/09/2015 15:45:30  |
   |234     | 11/09/2015 15:45:37  |
   |234     | 11/09/2015 18:27:15  |
   |234     | 11/09/2015 19:55:08  |
   |234     | 11/09/2015 19:55:12  |
   |234     | 12/09/2015 04:45:10  |
   |234     | 12/09/2015 04:45:13  |
   |234     | 12/09/2015 13:12:55  |
   |234     | 12/09/2015 16:35:08  |
   |234     | 12/09/2015 18:49:10  |
   |234     | 12/09/2015 20:20:57  |
   |234     | 13/09/2015 05:11:56  |
   |234     | 13/09/2015 05:12:05  |
   |234     | 13/09/2015 12:45:13  |
   |234     | 13/09/2015 15:47:25  |
   |234     | 13/09/2015 18:31:27  |
   |234     | 13/09/2015 18:31:30  |
   |234     | 13/09/2015 20:01:18  |
   |-------------------------------|

我正在尝试创建一个Access查询，将日期/时间分组到timesetup表中的句点。根据tapping_time表中的会话条件，timesetup每个会话应该只有一个结果（即使用户一次点击多次）。此外，用户在会话之外的时间（session_from和session_to之间的时间）字段应该永远不会计入或出现在结果表中。

期望的结果：

| userid | date       | tapping_on | session |
|--------|------------|------------|---------|
| 234    | 10/09/2015 | 04:20:04   | 1       |
| 234    | 10/09/2015 | 12:15:35   | 2       |
| 234    | 10/09/2015 | 15:31:11   | 3       |
| 234    | 10/09/2015 | 18:19:10   | 4       |
| 234    | 10/09/2015 | 19:37:53   | 5       |
| 234    | 11/09/2015 | 04:38:42   | 1       |
| 234    | 11/09/2015 | 12:21:27   | 2       |
| 234    | 11/09/2015 | 15:45:30   | 3       |
| 234    | 11/09/2015 | 18:27:15   | 4       |
| 234    | 11/09/2015 | 19:55:08   | 5       |
| 234    | 12/09/2015 | 04:45:10   | 1       |
| 234    | 12/09/2015 | 16:35:08   | 3       |
| 234    | 12/09/2015 | 18:49:10   | 4       |
| 234    | 13/09/2015 | 12:45:13   | 2       |
| 234    | 13/09/2015 | 15:47:25   | 3       |
| 234    | 13/09/2015 | 18:31:27   | 4       |
| 234    | 13/09/2015 | 20:01:18   | 5       |
|--------------------------------------------|

Answer 1

您可能在规范化方面有点过分，因为dateperiode和sessionrange之间的关系看起来像是一对多关系，因此不需要联结表。

有了以下数据，我相信以下查询应该可以正常工作，但遗憾的是我在准备好时没有一个像样的测试设置：

SELECT userid, tDatepart As [date], Min(t.tapping_time) As tapping_on, session FROM 
    (
        SELECT session_from, session_to, date_from, date_to, session
        FROM timsetup i
        INNER JOIN dateperiode d ON i.period_id = d.period_id
        INNER JOIN sessionrange q ON t.sessionrange_id = q.sessionrange_id
    ) As s
INNER JOIN 
    (SELECT Int(tapping_time) As tDatepart, CDate(tapping_time - Int(tapping_time)) As tTimepart, userid, tapping_time
    FROM checktime) t
ON (t.tDatePart >= s.date_from AND t.tDatePart <= s.date_to AND t.tTimePart >= s.session_from AND t.tTimePart <= s.session_to)
GROUP BY userid, tDatepart, session