MySQL分组方式-限制到第一个X记录

我在网络上做了一个很好的观察，似乎很多人都有类似的问题，但我似乎找不到一个解决我问题的明确方法

为了提供一些背景信息，我正在运行一个体育网站，并拥有一个数据库，其中包含了过去的成绩——跨越数千条记录

下面是一些示例数据-表名：matches

| match_id | manager_id | outcome |
| -------- | ---------- | ------- |
|        1 |          1 |       W |
|        2 |          1 |       D |
|        3 |          1 |       D |
|        4 |          2 |       L |
|        5 |          2 |       L |
|        6 |          2 |       W |
|        7 |          2 |       D |
|        8 |          2 |       L |
|        9 |          2 |       L |
|       10 |          3 |       W |
|       11 |          3 |       W |
|       12 |          3 |       W |
|       13 |          3 |       W |
|       14 |          3 |       D |
|       15 |          3 |       D |
|       16 |          4 |       L |
|       17 |          4 |       L |
|       18 |          4 |       D |
|       19 |          5 |       W |
|       20 |          5 |       W |

我想做的是计算每个经理的匹配数，以及结果——这是我通过使用此查询获得的结果

SELECT
`manager_id`,
COUNT(*) AS `played`,
SUM(`outcome` = 'W') AS `won`,
SUM(`outcome` = 'D') AS `drawn`,
SUM(`outcome` = 'L') AS `lost`,
ROUND((SUM(`outcome` = 'W')/COUNT(*)) * 100, 0) AS `win_percentage`
FROM
`matches`
GROUP BY `manager_id`

查询结果

| manager_id | played | won | drawn | lost | win_percentage |
| ---------- | ------ | --- | ----- | ---- | -------------- |
|          1 |      3 |   1 |     2 |    0 |             33 |
|          2 |      6 |   1 |     1 |    4 |             16 |
|          3 |      6 |   4 |     2 |    0 |             66 |
|          4 |      3 |   0 |     1 |    2 |              0 |
|          5 |      2 |   2 |     0 |    0 |            100 |

这一切都很好，而且做起来相对简单

然而，我想要实现的是根据每个管理者id的前X个记录数找出完全相同的数据

例如，假设我需要每个管理器的前两个匹配项的上述数据。我应该得出如下结果

| manager_id | played | won | drawn | lost | win_percentage |
| ---------- | ------ | --- | ----- | ---- | -------------- |
|          1 |      2 |   1 |     1 |    0 |             50 |
|          2 |      2 |   0 |     0 |    2 |              0 |
|          3 |      2 |   2 |     0 |    0 |            100 |
|          4 |      2 |   0 |     0 |    2 |              0 |
|          5 |      2 |   2 |     0 |    0 |            100 |

以上只是一个例子，因为在现实中，我们将搜索每组的前50或100个匹配项

---

非常感谢您对我的支持。

因为MySql中没有窗口函数，所以您可以使用中描述的变量

---

你首先如何定义？顺序是什么？按匹配id的顺序，按经理id分组。基本上，每个经理id的前X个匹配id记录数。

set @howmany := 2;

SELECT
  manager_id,
  COUNT(*) AS played,
  SUM(outcome = 'W') AS won,
  SUM(outcome = 'D') AS drawn,
  SUM(outcome = 'L') AS lost,
  ROUND((SUM(outcome = 'W')/COUNT(*)) * 100, 0) AS win_percentage
FROM
  (SELECT 
     match_id, 
     manager_id, 
     outcome, 
    CASE WHEN manager_id != @manager THEN @row := 1 
      ELSE @row := @row + 1 END as rownum,
    CASE WHEN manager_id != @manager THEN @manager := manager_id 
      ELSE @manager END as _
  FROM (SELECT * from matches ORDER BY manager_id, match_id) temp1
  JOIN (SELECT @manager := 0) temp2
  JOIN (SELECT @row := 0) temp3
  ) temp
WHERE temp.rownum <= @howmany
GROUP BY manager_id;