SQL挑戰 - 使用多個參考日期進行中斷

Question

考慮到下面的兩個表格，只顯示產品未處於離線狀態的活動時間。請記住，這是一小部分數據。活動表中有多個產品，離線表中有多個時間戳。

目標是在產品上線時輸出。因此，根據以下數據，我需要一個將返回的查詢：第3,6,7行 - 這些是產品聯機時的時間戳。

mysql> select * from activity; 
+------------+---------------------+ 
| product_id | activity_date  | 
+------------+---------------------+ 
|   1 | 2011-04-13 12:00:00 | 
|   1 | 2011-04-13 01:00:00 | 
|   1 | 2011-04-13 02:00:00 | 
|   1 | 2011-04-13 03:00:00 | 
|   1 | 2011-04-13 04:00:00 | 
|   1 | 2011-04-13 05:00:00 | 
|   1 | 2011-04-13 06:00:00 | 
+------------+---------------------+ 
7 rows in set (0.01 sec) 

mysql> select * from offline 
+------------+---------------------+---------------------+ 
| product_id | offline_start  | offline_end   | 
+------------+---------------------+---------------------+ 
|   1 | 2011-04-13 12:00:00 | 2011-04-13 01:00:00 | 
|   1 | 2011-04-13 03:00:00 | 2011-04-13 04:00:00 | 
+------------+---------------------+---------------------+ 
2 rows in set (0.01 sec) 

Answer 1

create table activity 
(
    product_id int, 
    activity_date datetime 
); 

create table [offline] 
(
    product_id int, 
    offline_start datetime, 
    offline_end datetime 
); 

insert into activity(product_id, activity_date) 
    values (1, '2011-04-13 12:00:00 AM') 
     ,(1, '2011-04-13 01:00:00 AM') 
     ,(1, '2011-04-13 02:00:00 AM') 
     ,(1, '2011-04-13 03:00:00 AM') 
     ,(1, '2011-04-13 04:00:00 AM') 
     ,(1, '2011-04-13 05:00:00 AM') 
     ,(1, '2011-04-13 06:00:00 AM'); 

insert into [offline](product_id, offline_start, offline_end) 
    values (1, '2011-04-13 12:00:00 AM', '2011-04-13 01:00:00 AM') 
     ,(1, '2011-04-13 03:00:00 AM', '2011-04-13 04:00:00 AM');   

select a.* 
from activity a 
     left join offline o 
      on a.product_id = o.product_id 
      and a.activity_date between o.offline_start and o.offline_end 
where o.product_id IS NULL 

不要有一個MySQL服務器設置在家裏測試，但如果AM不工作，然後12點爲00:00:00等活動表中的第一項應是

1, '2011-04-13 00:00:00' 

和離線表中的第一項應是

1, '2011-04-13 00:00:00', '2011-04-13 01:00:00' 

我的SQL服務器上測試這一點，這裏是輸出：

product_id activity_date 
1 2011-04-13 02:00:00.000 
1 2011-04-13 05:00:00.000 
1 2011-04-13 06:00:00.000