mysql – SQL连接,用于将子表中的值与子表值连接为单行中的不同列

2023年4月18日 152次阅读

我有一个survey_datas表包含这样的数据

survey_data_id  | title
1               | Paul 
3               | Anna 
4               | Alan

另一个表project_playlist_indexes包含这样的数据

survey_id  |survey_data_id  | favorite
1          | 1              | 22.10
2          | 1              | 24.00
3          | 3              | 12.00

我想将survey_datas表与project_playlist_indexes表一起使用,以便project_playlist_indexes表中包含的与survey_datas表相同的survey_data_id的值应该是最喜欢的time1,最喜欢的时间2,…最喜欢的时间n,我想得到的结果表是像这样

survey_data_id  |title | favorite_time1 | favorite_time2
            1   | paul | 22.10          |24.00
            3   | anna | 12.00          | null
            4   | alan | null           | null

目前我正在使用该查询

SELECT s.*,GROUP_CONCAT(pi.favorite) ,pi.*
FROM survey_datas s
LEFT JOIN  project_playlist_indexes pi 
ON pi.survey_data_id = s.survey_data_id  
GROUP BY pi.survey_data_id

但最喜欢的值是进入一个字段,我希望它在不同的列中.我怎样才能做到这一点

最佳答案 您可以通过执行动态SQL查询来完成此操作.我所做的是,首先根据survey_data_id列给出一个行号.然后通过survey_data_id选择每个行编号项作为每个列组.不知道代码的效率如何.

询问

set @query = null;
select
  group_concat(distinct
    concat(
      'max(case when `rn` = ',
      `rn`,
      ' then `favorite` end) as `favorite', `rn` , '`'
    )
  ) into @query
from (
  select `survey_id`, `survey_data_id`, `favorite`, (
    case `survey_data_id` when @curA 
    then @curRow := @curRow + 1 
    else @curRow := 1 and @curA := `survey_data_id` end 
  ) as `rn`
  from `project_playlist_indexes` t, 
  (select @curRow := 0, @curA := '') r 
  order by `survey_data_id`, `survey_data_id`
) t;

set @query = concat('select t2.`survey_data_id`, t2.`title`,', 
                @query,
              ' from (select `survey_id`, `survey_data_id`, `favorite`, (
              case `survey_data_id` when @curA 
              then @curRow := @curRow + 1 
              else @curRow := 1 and @curA := `survey_data_id` end 
              ) as `rn`
              from `project_playlist_indexes` t, 
              (select @curRow := 0, @curA := '''') r 
              order by `survey_data_id`, `survey_data_id`) t1
              right join `survey_datas` t2
              on t1.survey_data_id = t2.`survey_data_id`
              group by t1.`survey_data_id`
              order by t2.`survey_data_id`;'
     );

prepare stmt from @query;
execute stmt;
deallocate prepare stmt;

Find a demo here

点赞