MySql查询关系数据库中的视图

Question

我有一个问题为视图构建MySQL查询。以下是我的关系数据库结构和一些示例数据。

travel 
| ID | NUM | 
| 01 | N01 | 
| 02 | N02 | 

travel_has_flight 
| travel_ID | flight_ID | 
| 01  |  01 | 
| 01  |  02 | 
| 02  |  03 | 

flight 
| ID | NUM | DEPARTURE_DATE | ARRIVAL_DATE | 
| 01 | A01 | 2010-10-10 12:50 | 2010-10-10 17:50 | 
| 02 | A02 | 2010-10-17 14:50 | 2010-10-17 14:50 | 
| 03 | A03 | 2010-10-22 12:50 | 2010-10-22 17:50 | 

预期输出：

从查询作为视图。

| travel_ID | travel_NUM | flight_NUM1 | flight_DEPARTURE_DATE | flight_NUM2 | flight_ARRIVAL_DATE | 
| 01  | N01  | A01  | 2010-10-10 12:50 |  A02  | 2010-10-17 14:50 | 
| 02  | N02  | A03  | 2010-10-10 12:50 |  A03  | 2010-10-17 14:50 | 

输出应该如下。旅行可以包含1班航班。在这种情况下，输出是来自此次航班的行程+日期的数据。如果旅行包含2班航班，则应返回旅行+第一班航班的出站日期和第二班航班的入境日期的数据。

我会很乐意为此提供任何帮助。

###更新###

我与SQL低于目前卡住了，问题是现在如何单独flight_NUM的加入...

SELECT t.ID, t.NUM, Min(f.DEPARTURE_DATE), MAX(f.ARRIVAL_DATE) 
FROM travel t 
INNER JOIN travel_has_flight tf ON tf.TRAVEL_ID = t.ID 
INNER JOIN flight f ON f.ID = tf.FLIGHT_ID 
GROUP BY t.NUM 

Answer 1

有办法少做它。我愿意做一个UNION 2所选择的，就像这样：

SELECT `travel`.`ID` as 'TRAVEL_ID',`travel`.`NUM` as 'TRAVEL_NUM', `flight`.`NUM` as 'FLIGHT_NUM1',`flight`.`DEPARTURE_DATE`as 'FLIGHT_DEPARTURE_DATE',`flight`.`NUM`as 'FLIGHT_NUM2',`flight`.`ARRIVAL_DATE`as 'FLIGHT_ARRIVAL_DATE' 
FROM `travel` 
INNER JOIN `travel_has_flight` on `travel`.`ID`=`travel_has_flight`.`travel_ID` 
INNER JOIN `flight` on `travel_has_flight`.`flight_ID`=`flight`.`ID` 
WHERE `travel`.ID in (SELECT travel_ID FROM `travel_has_flight` GROUP BY travel_ID HAVING count(`flight_ID`)=1) 
UNION 
SELECT T.TRAVEL_ID,T.TRAVEL_NUM,MIN(T.FLIGHT_NUM1),MIN(T.FLIGHT_DEPARTURE_DATE),MAX(T.FLIGHT_NUM2),MAX(T.FLIGHT_ARRIVAL_DATE) 
FROM 
(SELECT `travel`.`ID` as 'TRAVEL_ID',`travel`.`NUM` as 'TRAVEL_NUM', `flight`.`NUM` as 'FLIGHT_NUM1',`flight`.`DEPARTURE_DATE`as 'FLIGHT_DEPARTURE_DATE',`flight`.`NUM`as 'FLIGHT_NUM2',`flight`.`ARRIVAL_DATE`as 'FLIGHT_ARRIVAL_DATE' 
FROM `travel` 
INNER JOIN `travel_has_flight` on `travel`.`ID`=`travel_has_flight`.`travel_ID` 
INNER JOIN `flight` on `travel_has_flight`.`flight_ID`=`flight`.`ID` 
WHERE `travel`.ID in (SELECT travel_ID FROM `travel_has_flight` GROUP BY travel_ID HAVING count(`flight_ID`)>1)) as T 
GROUP BY T.TRAVEL_ID,T.TRAVEL_NUM 

Answer 2

你的表，

table travel (ID int, NUM varchar()) 
table leg (travel_ID int, flight_ID) 
table flight (ID int, NUM varchar(), DEPARTURE date, ARRIVAL date) 

加入旅游与腿，飞行加盟腿，

select <more below> 
from travel t 
<how> join leg l where t.ID = l.travel_ID 
<how> join flight f where l.flight_ID == f.ID 

现在，决定这些是内部连接还是外部连接？

即使在travel和travel_has_flight（腿）之间加入时没有匹配，您还想要结果吗？如果不是，则使用内连接，如果是，则使用外连接。 （对您的样本数据很重要）

即使在travel_has_flight（腿）和航班之间的联接没有匹配时，您还想要结果吗？如果不是，则使用内连接，如果是，则使用外连接。 （不要紧，你的样本数据）

你说使用内部连接，

select <more below> 
from travel t 
inner join leg l where t.ID = l.travel_ID 
inner join flight f where l.flight_ID == f.ID 

在您的样本数据，结果集是，

t. | t. | l.  | l.  | f. | f. | f.    | f. 
ID | NUM | travel_ID | flight_ID | ID | NUM | DEPARTURE  | ARRIVAL 
01 | N01 | 01  | 01  | 01 | A01 | 2010-10-10 12:50 | 2010-10-10 17:50 
01 | N01 | 01  | 02  | 02 | A02 | 2010-10-17 14:50 | 2010-10-17 14:50 
02 | N02 | 02  | 03  | 03 | A03 | 2010-10-22 12:50 | 2010-10-22 17:50 

但是，你说使用内部连接，所以您正在寻找1：1 您需要按日期排序的这些数据。假设理智数据（始发< =到达）

with (
select t.ID as tid, t.NUM as tnum, f.DEPARTURE as dep, f.ARRIVAL as arr 
from travel t 
inner join leg l where t.ID = l.travel_ID 
inner join flight f where l.flight_ID == f.ID 
group by t.NUM 
order by t.NUM, DEPARTURE 
) as parts, 
<query> 

现在，你将如何在两行合并在上面的“零件”表的连接部位？