的MySQL如何使用變量的值的查詢在自己的子查詢

Question

我需要你的幫助：

我有一組爲用戶設置，遊戲和結果的關係表中，我試圖來計算每個用戶贏取百分比。

由於很多瀏覽我發現了以下方法：

SELECT Winner, (COUNT(Winner) * games.TotalPlayed) AS PercentWon 
FROM Results 
JOIN (
    SELECT 100/COUNT(*) AS TotalPlayed 
    FROM games 
    where Player_1 = 'John' OR Player_2 = 'John' 
) AS games 

where Winner == 'John' 
GROUP BY Winner 
ORDER BY PercentWon DESC; 

這完全適用於獲得約翰斯％，但我希望它掃描整個表的結果和打印值的每一個人。不過，我不允許使用：

SELECT Winner, (COUNT(Winner) * games.TotalPlayed) AS PercentWon 
FROM Results 
JOIN (
    SELECT 100/COUNT(*) AS TotalPlayed 
    FROM games 
    where Player_1 = Winner OR Player_2 = Winner 
) AS games 

where Winner != 'Draw' 
GROUP BY Winner 
ORDER BY PercentWon DESC; 

爲獲獎者沒有定義。 Winner在執行時如何獲得傳遞給子查詢的值？

Answer 1

我終於想出了一些能爲你工作的東西。這很棘手，因爲玩家可以在games表中的兩個字段之一。

SELECT player, (COALESCE(totalWon, 0)/count(*) * 100) as PercentWon 
FROM ((SELECT DISTINCT Player_1 as player FROM games) UNION (SELECT DISTINCT Player_2 as player FROM games)) players 
INNER JOIN games on (games.Player_1 = player or games.Player_2=player) 
LEFT JOIN 
(
    SELECT Winner, COUNT(Winner) as totalWon 
    FROM Results 
    WHERE Winner != 'Draw' 
    GROUP BY Winner 
) as winners 
ON winners.Winner = player 
GROUP BY player 
ORDER BY PercentWon DESC; 

的UNION在FROM子句中正在建設，我可以用它來加入遊戲用戶列表（找到每個玩家的遊戲總數）和獲獎者（找到每個玩家勝場數）。

COALESCE用於玩家沒有贏得任何遊戲的情況。由於我們正在連接表，這允許我們在沒有匹配時默認爲0而不是NULL。

這裏的一個基於sqlfiddle出了關於您的模式假設：

http://sqlfiddle.com/#!2/d762c/1/0