MySql选择动态行值作为列名

这是我给出的表格：

+---------+------------+-------------------+--------------------------+---------------+---------------+
| exec_id | project_id | flow_id           | job_id                   | start_time    | end_time      |
+---------+------------+-------------------+--------------------------+---------------+---------------+
|   10919 |         16 | my_flow_cleanup   | init                     | 1408480308611 | 1408480308686 |
|   10919 |         16 | my_flow_cleanup   | job_id_1                 | 1408480309212 | 1408480309426 |
|   10919 |         16 | my_flow_cleanup   | job_id_2                 | 1408480308721 | 1408480308776 |
|   10919 |         16 | my_flow_cleanup   | job_id_3                 | 1408480308827 | 1408480309171 |
+---------+------------+-------------------+--------------------------+---------------+---------------+

我想要实现这样的选择：

exec_id     init_start            init_end            job_id_1_start       job_id_1_end       job_id_2_start        job_id_2_end        job_id_3_start      job_id_3_end
10919       1408480308611         1408480308686       1408480309212        1408480309426      1408480308721         1408480308776       1408480308827       1408480309171

我花了两天多的时间,但收效甚微. This question对我有所帮助,但并非如此.如您所见,另一个问题涉及获取2列并将其转换为2行.但我必须采取3列并将它们转换为2行.

谁能帮我这个？它甚至可以在MySQL中使用吗？

编辑1

谢谢Khalid Junaid的答案,它解决了我的问题.我还要做一个修改.我必须通过start_time按升序对所选列进行排序.

例如.：

+---------+------------+-------------------+----------+---------------+---------------+
| exec_id | project_id | flow_id           | job_id   | start_time    | end_time      |
+---------+------------+-------------------+----------+---------------+---------------+
|   10919 |         16 | my_flow_cleanup   | init     | 10            | 15            |
|   10919 |         16 | my_flow_cleanup   | job_id_1 | 30            | 40            |
|   10919 |         16 | my_flow_cleanup   | job_id_2 | 40            | 50            |
|   10919 |         16 | my_flow_cleanup   | job_id_3 | 20            | 25            |
+---------+------------+-------------------+----------+---------------+---------------+

当前查询将返回：

exec_id     init_start            init_end            job_id_1_start       job_id_1_end       job_id_2_start        job_id_2_end        job_id_3_start      job_id_3_end
10919       10                    15                  30                   40                 40                    50                  20                  25

我需要：

exec_id     init_start            init_end            job_id_3_start       job_id_3_end       job_id_1_start        job_id_1_end        job_id_2_start      job_id_2_end
10919       10                    15                  20                   25                 30                    40                  40                  50

请注意,根据start_time,列的顺序现在已更改.

我尝试用临时表和视图来做这个,我觉得这很容易.不幸的是,我没有创建表/创建视图权限.有没有办法在没有临时表和视图的情况下实现这一目标？

解决方法:

从参考问题的使用group_concat的方法来看,你可以这样做,但是注意一件事,因为每个exec_id组你的工作id增加,然后group_concat方法不会是最佳的,因为它的默认长度为1024个字符要连接,对于动态列,这肯定会越过这个限制,但这个限制可以增加,如documentation所述

SET @sql = NULL;

SELECT GROUP_CONCAT(DISTINCT
  CONCAT('MAX(CASE WHEN job_id = ''',
         job_id, 
         ''' THEN start_time END) `',
         job_id,
         '_start`',
         ',',
         'MAX(CASE WHEN job_id = ''',
         job_id,
         ''' THEN end_time END) `',
         job_id,
         '_end`' 
         )

 )
  INTO @sql
  FROM t;

SET @sql = CONCAT('SELECT exec_id, ', @sql, ' 
                     FROM t 
                    GROUP BY exec_id');

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

Fiddle Demo

标签：sql,mysql,pivot-table
来源： https://codeday.me/bug/20191003/1847833.html