对于一个体育比赛，人们可以两人一组对抗另一个由两人组成的队，我想输出一个特定球员面对的每个对手的列表，以及特定球员和该对手之间比赛的赢、平和输的次数。
例如，如果结果是：
玩家1和玩家2 2-1玩家3和玩家4
玩家3和玩家1 1-0玩家2和玩家4
玩家2和玩家4 4-2玩家1和玩家3
玩家1和玩家4 0-0玩家3和玩家2
我希望玩家1的输出是：
播放器2，W1 D1 L1
播放器3，W1 D1 L0
玩家4，W2 D0 L1
事实证明，要做到这一点非常困难。
我的数据被组织成两个表格，一个是所有球员（球员）的表格，一个是得分（得分）的表格。
玩家有PlayerID和PlayerName两个字段。
分数有Player1、Player2、Player3、Player4和Score字段，其中每个球员字段都可以包含任何PlayerID（因此，根据行的不同，任何给定的ID都可以出现在不同的列中），分数的存储方式是，如果小于21，则Player1和Player2将失败，如果大于21，则获胜。
这个查询有什么问题？很明显它返回了0。
SELECT opponent as opp, COUNT(result='win') as c1, COUNT(result='draw') as c2, COUNT(result='loss') as c3
FROM (SELECT
CASE
WHEN '$player' IN(Players1.PlayerName, Players2.PlayerName) THEN Players3.PlayerName
WHEN '$player' IN(Players3.PlayerName, Players4.PlayerName) THEN Players1.PlayerName
ELSE 'NA'
END as opponent,
CASE
WHEN '$player' IN(Players1.PlayerName, Players2.PlayerName)
THEN
CASE
WHEN Scores.Score > 21 THEN 'win'
ELSE 'NA'
END as result,
CASE
WHEN Scores.Score < 21 THEN 'loss'
ELSE 'NA'
END as result,
WHEN '$player' IN(Players3.PlayerName, Players4.PlayerName)
THEN
CASE
WHEN Scores.Score < 21 THEN 'win'
ELSE 'NA'
END as result,
CASE
WHEN Scores.Score > 21 THEN 'loss'
ELSE 'NA'
END as result,
WHEN Scores.Score = 21 THEN 'draw'
ELSE 'NA'
END as result,
FROM Scores
INNER JOIN Players Players1 ON Scores.Player1=Players1.PlayerID
INNER JOIN Players Players2 ON Scores.Player2=Players2.PlayerID
INNER JOIN Players Players3 ON Scores.Player3=Players3.PlayerID
INNER JOIN Players Players4 ON Scores.Player4=Players4.PlayerID
WHERE Players1.PlayerName = '$player' OR Players2.PlayerName = '$player' OR Players3.PlayerName = '$player' OR Players4.PlayerName = '$player'
UNION ALL
SELECT
CASE
WHEN '$player' IN(Players1.PlayerName, Players2.PlayerName) THEN Players4.PlayerName
WHEN '$player' IN(Players3.PlayerName, Players4.PlayerName) THEN Players2.PlayerName
ELSE 'NA'
END as opponent,
CASE
WHEN '$player' IN(Players1.PlayerName, Players2.PlayerName) AND Scores.Score > 21 THEN 'win'
WHEN '$player' IN(Players1.PlayerName, Players2.PlayerName) AND Scores.Score < 21 THEN 'loss'
WHEN '$player' IN(Players3.PlayerName, Players4.PlayerName) AND Scores.Score < 21 THEN 'win'
WHEN '$player' IN(Players3.PlayerName, Players4.PlayerName) AND Scores.Score > 21 THEN 'loss'
WHEN Scores.Score = 21 THEN 'draw'
ELSE 'NA'
END as result,
FROM Scores
INNER JOIN Players Players1 ON Scores.Player1=Players1.PlayerID
INNER JOIN Players Players2 ON Scores.Player2=Players2.PlayerID
INNER JOIN Players Players3 ON Scores.Player3=Players3.PlayerID
INNER JOIN Players Players4 ON Scores.Player4=Players4.PlayerID
WHERE Players1.PlayerName = '$player' OR Players2.PlayerName = '$player' OR Players3.PlayerName = '$player' OR Players4.PlayerName = '$player'
) AS SUBQUERY
GROUP BY opp
最佳答案
这并不漂亮，但是由于MySQL缺少Pivot函数，您可以使用一个内部查询来分析结果，该查询可以分解谁玩的以及结果是什么。然后外部查询可以聚合这些结果。下面这些对我很管用，不过，同样，也不好看：
select players.name as playerName,
opponents.name as opponentName,
SUM(CASE WHEN result = 'W'THEN 1 ELSE 0 END) AS win,
SUM(CASE WHEN result = 'L'THEN 1 ELSE 0 END) AS lose,
SUM(CASE WHEN result = 'D' THEN 1 ELSE 0 END) AS draw
from (
select player1 as player, player3 as opponent, score, CASE WHEN score > 21 THEN 'W' WHEN score < 21 THEN 'L' ELSE 'D' END as result
from scores
union all
select player1 as player, player4 as opponent, score, CASE WHEN score > 21 THEN 'W' WHEN score < 21 THEN 'L' ELSE 'D' END as result
from scores
union all
select player2 as player, player3 as opponent, score, CASE WHEN score > 21 THEN 'W' WHEN score < 21 THEN 'L' ELSE 'D' END as result
from scores
union all
select player2 as player, player4 as opponent, score, CASE WHEN score > 21 THEN 'W' WHEN score < 21 THEN 'L' ELSE 'D' END as result
from scores
-- now reverse result because opponents are being slotted into the player position
union all
select player3 as player, player1 as opponent, score, CASE WHEN score > 21 THEN 'L' WHEN score < 21 THEN 'W' ELSE 'D' END as result
from scores
union all
select player4 as player, player1 as opponent, score, CASE WHEN score > 21 THEN 'L' WHEN score < 21 THEN 'W' ELSE 'D' END as result
from scores
union all
select player3 as player, player2 as opponent, score, CASE WHEN score > 21 THEN 'L' WHEN score < 21 THEN 'W' ELSE 'D' END as result
from scores
union all
select player4 as player, player2 as opponent, score, CASE WHEN score > 21 THEN 'L' WHEN score < 21 THEN 'W' ELSE 'D' END as result
from scores
) resultsByPlayer
inner join players on players.playerId = resultsByPlayer.player
inner join players as opponents on opponents.playerId = resultsByPlayer.opponent
group by players.playerId, opponents.playerId