关于连接多个表的SQL查询的帮助
关于连接多个表的SQL查询的帮助
提问于 2011-08-12 22:03:27
首先，我将解释这种情况，我有一个表tbl_game，其结构就是这样的。此表包含游戏开始的时间和玩游戏的配对
| id | time | pair_id |
-----------+--------------+ ---------------
1 | 123123123 | 1 |
2 | 123168877 | 1 |我有另一个表tbl_throws，它保存每个玩家的分数。如果你想知道，this a basic dice rolling game
| id | game_id | player_id | score |
-----------+--------------+---------------+---------|
| 1 | 1 | 1 | 2 |
| 2 | 1 | 2 | 5 |
| 3 | 1 | 1 | 9 |
| 4 | 1 | 2 | 11 |
| 5 | 2 | 1 | 7 |
| 6 | 2 | 2 | 6 |现在，这里的id是投掷id，而不是游戏id。在这里，每个拥有player_id 1和2的玩家都掷了两次骰子，并在相同的游戏中获得了各自的得分，而在另一场比赛中只投了一次。
现在，使用这两个表，我需要创建一个记录集，即每个玩家在一场比赛中的总分
| game_id | game_time | player1_total | player2_total|
|------------+-----------+---------------+--------------|
| 1 | 123123123 | 11 | 16 |
| 2 | 123168877 | 7 | 6 |我尝试了很多胡言乱语的查询，但没有给出正确的结果？对此的正确查询是什么？
更新
因为，大多数答案都受到一个事实的约束，即player1id和player2id必须是已知的或固定的。
因此，我即将提供的信息可能有助于澄清混乱。还有另一个表，它保存了玩家的信息。tbl_pupil结构如下所示
| id | unique_id | name |
|---------+---------------+----------|
| 1 | 001 | some |
| 2 | 002 | another |这些玩家被统称为另一个表tbl_pair中的一对
| id | player1 | player2 |
|---------+---------------+----------|
| 1 | 1 | 2 |
So, now
select
g.id
g.time
p1.id as player1id
p1.name as player1name
t.score as player1score
p2.id as player2id
p2.name as player2name
t.score as player2score
FROM
tbl_game g,
inner join tbl_pair as pair on g.pair_id = pair.id
inner join tbl_pupil as p1 on p1.id = pair.player1
inner join tbl_pupil as p2 on p2.id = pair.player2
inner join tbl_throw as t on g.id = t.game_id这是我的初步查询，它在某种程度上带来了记录集
| id | time | player1id | player1name | player1score | player2id | player2name | player2score |
----------------------------------------------------------------------------------------------------------------------------
| 1 | 12 | 1 | some | 5 | 2 | another | 2 |
| 1 | 12 | 1 | some | 5 | 2 | another | 5 |
| 1 | 12 | 1 | some | 9 | 2 | another | 9 |
| 1 | 12 | 1 | some | 11 | 2 | another | 11 |顺便说一句，我现在只显示一个游戏id的结果。我没有保存足够的知识，无法将上面的记录分组到一个列中，在一列中使用player1单独的总和，而在另一列中使用playe2的单独总和。
回答 5
Stack Overflow用户
发布于 2011-08-12 22:08:17
试试这个：
SELECT
tbl_game.id AS game_id,
tbl_game.time AS game_time,
SUM(CASE WHEN player_id = 1 THEN score ELSE 0 END) AS player1_total,
SUM(CASE WHEN player_id = 2 THEN score ELSE 0 END) AS player2_total
FROM tbl_game JOIN tbl_thorws ON tbl_game.id = tbl_throws.game_id
GROUP BY tbl_game.idStack Overflow用户
发布于 2011-08-12 22:09:21
您需要内部连接这两个表，并汇总您的分数。为了进行基本的透视，我使用了一条CASE语句来按玩家聚合。
SELECT G.Id as Game_Id,
G.time as Game_Time,
SUM(CASE WHEN t.Player_id = 1 THEN t.score ELSE 0 END) as Player1_total,
SUM(CASE WHEN t.Player_id = 2 THEN t.score ELSE 0 END) as Player2_total
FROM tbl_game G
INNER JOIN tbl_throws T
ON g.id = t.game_id
GROUP BY g.ID, g.timeStack Overflow用户
发布于 2011-08-12 22:25:30
这与其他一些答案类似，但重要的是不依赖于播放器ID为1和2。
select
game_id = g.id,
game_time = g.time,
player1_total = SUM(case t.player_id when p.player1_id then t.score else 0 end),
player2_total = SUM(case t.player_id when p.player2_id then t.score else 0 end)
from
tbl_game g
join tbl_throws t on g.id = t.game_id
join ( --Get the player IDs for this game
select
game_id,
player1_id = MIN(player_id),
player2_id = MAX(player_id)
from
tbl_throws
group by game_id
) p
on p.game_id = t.game_id
group by
g.id, g.time只是为了好玩，我将上面的概括出来，允许更多的>2个玩家：
这两个CTE表只显示了我正在使用的测试数据
;WITH tbl_game as (
select ID = 1, time = 123123123, pair_id = 1
union select ID = 2, time = 123168877, pair_id = 1
),
tbl_throws as (
select id = 1, game_id = 1, player_id = 1, score = 2
union select id = 2, game_id = 1, player_id = 2, score = 5
union select id = 2, game_id = 1, player_id = 3, score = 5
union select id = 3, game_id = 1, player_id = 1, score = 9
union select id = 4, game_id = 1, player_id = 2, score = 11
union select id = 5, game_id = 2, player_id = 1, score = 7
union select id = 6, game_id = 2, player_id = 2, score = 6
)
select
game_id = g.id,
game_time = g.time,
player1_id = MAX(case x.player_no when 1 then t.player_id else 0 end),
player1_total = SUM(case x.player_no when 1 then t.score else 0 end),
player1_id = MAX(case x.player_no when 2 then t.player_id else 0 end),
player2_total = SUM(case x.player_no when 2 then t.score else 0 end),
player3_id = MAX(case x.player_no when 3 then t.player_id else 0 end),
player3_total = SUM(case x.player_no when 3 then t.score else 0 end),
player4_id = MAX(case x.player_no when 4 then t.player_id else 0 end),
player4_total = SUM(case x.player_no when 4 then t.score else 0 end)
/* Add more rows for the number of players permitted in a single game */
from
tbl_game g
join tbl_throws t on g.id = t.game_id
cross apply (
select player_no = COUNT(distinct player_id)
from tbl_throws sub
where sub.player_id <= t.player_id
and Sub.game_id = t.game_id
) x
group by
g.id, g.time页面原文内容由Stack Overflow提供。腾讯云小微IT领域专用引擎提供翻译支持
原文链接：
https://stackoverflow.com/questions/7041226
复制相关文章
相似问题
腾讯云开发者
