每个值的 MySQL 行总数答案

【问题标题】：MySQL row totals per value每个值的 MySQL 行总数
【发布时间】：2020-09-30 09:11:23
【问题描述】：

MySQL表结构如下：

| id | company_id | project_id | status |
|  1 |          1 |          1 |  draft |
|  2 |          1 |          1 |   done |
|  3 |          1 |          1 |    wip |
|  4 |          1 |          2 |  draft |

我想在表上运行单个查询并为单个 company_id 获得类似的结果：

| project_id | draft | wip | done | total |
|          1 |     1 |   1 |    1 |     3 |
|          2 |     1 |   0 |    0 |     1 |

表格可能有 >.5M 行

【问题讨论】：

标签： mysql sql count sum pivot

【解决方案1】：

使用条件聚合：

select project_id, 
    sum(status = 'draft') draft,
    sum(status = 'wip') wip,
    sum(status = 'done') done,
    count(*) total
from mytable
group by project_id

【讨论】：

【解决方案2】：

您可以明确地选择所有 status 案例，然后动态旋转以获得所有 status 值，即使新插入的值不是当前值：

SET @sql = NULL;

SELECT GROUP_CONCAT(
             CONCAT(
                    'SUM( status =''', status, ''' ) AS ',status
                    )
       )
  INTO @sql
  FROM ( SELECT DISTINCT status
           FROM t 
       ) t;

SET @sql = CONCAT('SELECT company_id, project_id,',@sql,',COUNT(*) AS total'
                   ' FROM t
                    GROUP BY company_id, project_id'); 


PREPARE stmt FROM @sql;
EXECUTE stmt;
DEALLOCATE PREPARE stmt;

Demo

【讨论】：