改善，简化和加速我的sql的这些部分的最佳方法是什么？
-- do pivoting
drop temporary table if exists inverted;
create temporary table inverted as
select cast(s.name as char(30)) as name,
s.ename,
max(if(s.ts='00:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '0:30_it1',
max(if(s.ts='00:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '0:30_it2',
'#ffffff' as '0:30_color',
max(if(s.ts='01:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '1:30_it1',
max(if(s.ts='01:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '1:30_it2',
'#ffffff' as '1:30_color',
max(if(s.ts='02:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '2:30_it1',
max(if(s.ts='02:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '2:30_it2',
'#ffffff' as '2:30_color',
max(if(s.ts='03:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '3:30_it1',
max(if(s.ts='03:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '3:30_it2',
'#ffffff' as '3:30_color',
max(if(s.ts='04:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '4:30_it1',
max(if(s.ts='04:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '4:30_it2',
'#ffffff' as '4:30_color',
max(if(s.ts='05:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '5:30_it1',
max(if(s.ts='05:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '5:30_it2',
'#ffffff' as '5:30_color',
max(if(s.ts='06:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '6:30_it1',
max(if(s.ts='06:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '6:30_it2',
'#ffffff' as '6:30_color',
max(if(s.ts='07:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '7:30_it1',
max(if(s.ts='07:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '7:30_it2',
'#ffffff' as '7:30_color',
max(if(s.ts='08:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '8:30_it1',
max(if(s.ts='08:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '8:30_it2',
'#ffffff' as '8:30_color',
max(if(s.ts='09:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '9:30_it1',
max(if(s.ts='09:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '9:30_it2',
'#ffffff' as '9:30_color',
max(if(s.ts='10:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '10:30_it1',
max(if(s.ts='10:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '10:30_it2',
'#ffffff' as '10:30_color',
max(if(s.ts='11:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '11:30_it1',
max(if(s.ts='11:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '11:30_it2',
'#ffffff' as '11:30_color',
max(if(s.ts='12:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '12:30_it1',
max(if(s.ts='12:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '12:30_it2',
'#ffffff' as '12:30_color',
max(if(s.ts='13:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '13:30_it1',
max(if(s.ts='13:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '13:30_it2',
'#ffffff' as '13:30_color',
max(if(s.ts='14:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '14:30_it1',
max(if(s.ts='14:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '14:30_it2',
'#ffffff' as '14:30_color',
max(if(s.ts='15:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '15:30_it1',
max(if(s.ts='15:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '15:30_it2',
'#ffffff' as '15:30_color',
max(if(s.ts='16:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '16:30_it1',
max(if(s.ts='16:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '16:30_it2',
'#ffffff' as '16:30_color',
max(if(s.ts='17:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '17:30_it1',
max(if(s.ts='17:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '17:30_it2',
'#ffffff' as '17:30_color',
max(if(s.ts='18:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '18:30_it1',
max(if(s.ts='18:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '18:30_it2',
'#ffffff' as '18:30_color',
max(if(s.ts='19:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '19:30_it1',
max(if(s.ts='19:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '19:30_it2',
'#ffffff' as '19:30_color',
max(if(s.ts='20:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '20:30_it1',
max(if(s.ts='20:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '20:30_it2',
'#ffffff' as '20:30_color',
max(if(s.ts='21:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '21:30_it1',
max(if(s.ts='21:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '21:30_it2',
'#ffffff' as '21:30_color',
max(if(s.ts='22:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '22:30_it1',
max(if(s.ts='22:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '22:30_it2',
'#ffffff' as '22:30_color',
max(if(s.ts='23:30:00', cast(s.value_t1 as decimal(9,2)), null)) as '23:30_it1',
max(if(s.ts='23:30:00', cast(s.value_t2 as decimal(9,2)), null)) as '23:30_it2',
'#ffffff' as '23:30_color',
cast(null as time) as hour,
cast(null as decimal(9,2)) as value,
cast(null as char(17)) as series,
table_chart
from splitted_table s
group by s.table_chart, s.name, s.ename;
UPDATE inverted SET `0:30_color` = fn_rep_get_color_for_field(`0:30_it1`, `0:30_it2`, p_min, p_max), `1:30_color` = fn_rep_get_color_for_field(`1:30_it1`, `1:30_it2`, p_min, p_max),
`2:30_color` = fn_rep_get_color_for_field(`2:30_it1`, `2:30_it2`, p_min, p_max), `3:30_color` = fn_rep_get_color_for_field(`3:30_it1`, `3:30_it2`, p_min, p_max),
`4:30_color` = fn_rep_get_color_for_field(`4:30_it1`, `4:30_it2`, p_min, p_max), `5:30_color` = fn_rep_get_color_for_field(`5:30_it1`, `5:30_it2`, p_min, p_max),
`6:30_color` = fn_rep_get_color_for_field(`6:30_it1`, `6:30_it2`, p_min, p_max), `7:30_color` = fn_rep_get_color_for_field(`7:30_it1`, `7:30_it2`, p_min, p_max),
`8:30_color` = fn_rep_get_color_for_field(`8:30_it1`, `8:30_it2`, p_min, p_max), `9:30_color` = fn_rep_get_color_for_field(`9:30_it1`, `9:30_it2`, p_min, p_max),
`10:30_color` = fn_rep_get_color_for_field(`10:30_it1`, `10:30_it2`, p_min, p_max), `11:30_color` = fn_rep_get_color_for_field(`11:30_it1`, `11:30_it2`, p_min, p_max),
`12:30_color` = fn_rep_get_color_for_field(`12:30_it1`, `12:30_it2`, p_min, p_max), `13:30_color` = fn_rep_get_color_for_field(`13:30_it1`, `13:30_it2`, p_min, p_max),
`14:30_color` = fn_rep_get_color_for_field(`14:30_it1`, `14:30_it2`, p_min, p_max), `15:30_color` = fn_rep_get_color_for_field(`15:30_it1`, `15:30_it2`, p_min, p_max),
`16:30_color` = fn_rep_get_color_for_field(`16:30_it1`, `16:30_it2`, p_min, p_max), `17:30_color` = fn_rep_get_color_for_field(`17:30_it1`, `17:30_it2`, p_min, p_max),
`18:30_color` = fn_rep_get_color_for_field(`18:30_it1`, `18:30_it2`, p_min, p_max), `19:30_color` = fn_rep_get_color_for_field(`19:30_it1`, `19:30_it2`, p_min, p_max),
`20:30_color` = fn_rep_get_color_for_field(`20:30_it1`, `20:30_it2`, p_min, p_max), `21:30_color` = fn_rep_get_color_for_field(`21:30_it1`, `21:30_it2`, p_min, p_max),
`22:30_color` = fn_rep_get_color_for_field(`22:30_it1`, `22:30_it2`, p_min, p_max), `23:30_color` = fn_rep_get_color_for_field(`23:30_it1`, `23:30_it2`, p_min, p_max)
where SUBSTRING(table_chart, 2, 1) = 'p';
UPDATE inverted SET `0:30_color` = fn_rep_get_color_for_field(`0:30_it1`, `0:30_it2`, q_min, q_max), `1:30_color` = fn_rep_get_color_for_field(`1:30_it1`, `1:30_it2`, q_min, q_max),
`2:30_color` = fn_rep_get_color_for_field(`2:30_it1`, `2:30_it2`, q_min, q_max), `3:30_color` = fn_rep_get_color_for_field(`3:30_it1`, `3:30_it2`, q_min, q_max),
`4:30_color` = fn_rep_get_color_for_field(`4:30_it1`, `4:30_it2`, q_min, q_max), `5:30_color` = fn_rep_get_color_for_field(`5:30_it1`, `5:30_it2`, q_min, q_max),
`6:30_color` = fn_rep_get_color_for_field(`6:30_it1`, `6:30_it2`, q_min, q_max), `7:30_color` = fn_rep_get_color_for_field(`7:30_it1`, `7:30_it2`, q_min, q_max),
`8:30_color` = fn_rep_get_color_for_field(`8:30_it1`, `8:30_it2`, q_min, q_max), `9:30_color` = fn_rep_get_color_for_field(`9:30_it1`, `9:30_it2`, q_min, q_max),
`10:30_color` = fn_rep_get_color_for_field(`10:30_it1`, `10:30_it2`, q_min, q_max), `11:30_color` = fn_rep_get_color_for_field(`11:30_it1`, `11:30_it2`, q_min, q_max),
`12:30_color` = fn_rep_get_color_for_field(`12:30_it1`, `12:30_it2`, q_min, q_max), `13:30_color` = fn_rep_get_color_for_field(`13:30_it1`, `13:30_it2`, q_min, q_max),
`14:30_color` = fn_rep_get_color_for_field(`14:30_it1`, `14:30_it2`, q_min, q_max), `15:30_color` = fn_rep_get_color_for_field(`15:30_it1`, `15:30_it2`, q_min, q_max),
`16:30_color` = fn_rep_get_color_for_field(`16:30_it1`, `16:30_it2`, q_min, q_max), `17:30_color` = fn_rep_get_color_for_field(`17:30_it1`, `17:30_it2`, q_min, q_max),
`18:30_color` = fn_rep_get_color_for_field(`18:30_it1`, `18:30_it2`, q_min, q_max), `19:30_color` = fn_rep_get_color_for_field(`19:30_it1`, `19:30_it2`, q_min, q_max),
`20:30_color` = fn_rep_get_color_for_field(`20:30_it1`, `20:30_it2`, q_min, q_max), `21:30_color` = fn_rep_get_color_for_field(`21:30_it1`, `21:30_it2`, q_min, q_max),
`22:30_color` = fn_rep_get_color_for_field(`22:30_it1`, `22:30_it2`, q_min, q_max), `23:30_color` = fn_rep_get_color_for_field(`23:30_it1`, `23:30_it2`, q_min, q_max)
where SUBSTRING(table_chart, 2, 1) = 'q';
UPDATE inverted SET `0:30_color` = fn_rep_get_color_for_field(`0:30_it1`, `0:30_it2`, u_min, u_max), `1:30_color` = fn_rep_get_color_for_field(`1:30_it1`, `1:30_it2`, u_min, u_max),
`2:30_color` = fn_rep_get_color_for_field(`2:30_it1`, `2:30_it2`, u_min, u_max), `3:30_color` = fn_rep_get_color_for_field(`3:30_it1`, `3:30_it2`, u_min, u_max),
`4:30_color` = fn_rep_get_color_for_field(`4:30_it1`, `4:30_it2`, u_min, u_max), `5:30_color` = fn_rep_get_color_for_field(`5:30_it1`, `5:30_it2`, u_min, u_max),
`6:30_color` = fn_rep_get_color_for_field(`6:30_it1`, `6:30_it2`, u_min, u_max), `7:30_color` = fn_rep_get_color_for_field(`7:30_it1`, `7:30_it2`, u_min, u_max),
`8:30_color` = fn_rep_get_color_for_field(`8:30_it1`, `8:30_it2`, u_min, u_max), `9:30_color` = fn_rep_get_color_for_field(`9:30_it1`, `9:30_it2`, u_min, u_max),
`10:30_color` = fn_rep_get_color_for_field(`10:30_it1`, `10:30_it2`, u_min, u_max), `11:30_color` = fn_rep_get_color_for_field(`11:30_it1`, `11:30_it2`, u_min, u_max),
`12:30_color` = fn_rep_get_color_for_field(`12:30_it1`, `12:30_it2`, u_min, u_max), `13:30_color` = fn_rep_get_color_for_field(`13:30_it1`, `13:30_it2`, u_min, u_max),
`14:30_color` = fn_rep_get_color_for_field(`14:30_it1`, `14:30_it2`, u_min, u_max), `15:30_color` = fn_rep_get_color_for_field(`15:30_it1`, `15:30_it2`, u_min, u_max),
`16:30_color` = fn_rep_get_color_for_field(`16:30_it1`, `16:30_it2`, u_min, u_max), `17:30_color` = fn_rep_get_color_for_field(`17:30_it1`, `17:30_it2`, u_min, u_max),
`18:30_color` = fn_rep_get_color_for_field(`18:30_it1`, `18:30_it2`, u_min, u_max), `19:30_color` = fn_rep_get_color_for_field(`19:30_it1`, `19:30_it2`, u_min, u_max),
`20:30_color` = fn_rep_get_color_for_field(`20:30_it1`, `20:30_it2`, u_min, u_max), `21:30_color` = fn_rep_get_color_for_field(`21:30_it1`, `21:30_it2`, u_min, u_max),
`22:30_color` = fn_rep_get_color_for_field(`22:30_it1`, `22:30_it2`, u_min, u_max), `23:30_color` = fn_rep_get_color_for_field(`23:30_it1`, `23:30_it2`, u_min, u_max)
where SUBSTRING(table_chart, 2, 1) = 'u';
DROP TEMPORARY TABLE IF EXISTS deviation;
CREATE TEMPORARY TABLE deviation LIKE inverted;
insert into export_data.deviation (`name`,`0:30_it1`,`1:30_it1`,`2:30_it1`,`3:30_it1`,`4:30_it1`,`5:30_it1`,`6:30_it1`,`7:30_it1`,`8:30_it1`,`9:30_it1`,`10:30_it1`,`11:30_it1`,`12:30_it1`,`13:30_it1`,`14:30_it1`,`15:30_it1`,`16:30_it1`,`17:30_it1`,`18:30_it1`,`19:30_it1`,`20:30_it1`,`21:30_it1`,`22:30_it1`,`23:30_it1`,
`0:30_it2`,`1:30_it2`,`2:30_it2`,`3:30_it2`,`4:30_it2`,`5:30_it2`,`6:30_it2`,`7:30_it2`,`8:30_it2`,`9:30_it2`,`10:30_it2`,`11:30_it2`,`12:30_it2`,`13:30_it2`,`14:30_it2`,`15:30_it2`,`16:30_it2`,`17:30_it2`,`18:30_it2`,`19:30_it2`,`20:30_it2`,`21:30_it2`,`22:30_it2`,`23:30_it2`,`table_chart`)
SELECT
'table footer' as name,
SUM(abs(`0:30_it1`-`0:30_it2`)) as `0:30_it1`,
SUM(abs(`1:30_it1`-`1:30_it2`)) as `1:30_it1`,
SUM(abs(`2:30_it1`-`2:30_it2`)) as `2:30_it1`,
SUM(abs(`3:30_it1`-`3:30_it2`)) as `3:30_it1`,
SUM(abs(`4:30_it1`-`4:30_it2`)) as `4:30_it1`,
SUM(abs(`5:30_it1`-`5:30_it2`)) as `5:30_it1`,
SUM(abs(`6:30_it1`-`6:30_it2`)) as `6:30_it1`,
SUM(abs(`7:30_it1`-`7:30_it2`)) as `7:30_it1`,
SUM(abs(`8:30_it1`-`8:30_it2`)) as `8:30_it1`,
SUM(abs(`9:30_it1`-`9:30_it2`)) as `9:30_it1`,
SUM(abs(`10:30_it1`-`10:30_it2`)) as `10:30_it1`,
SUM(abs(`11:30_it1`-`11:30_it2`)) as `11:30_it1`,
SUM(abs(`12:30_it1`-`12:30_it2`)) as `12:30_it1`,
SUM(abs(`13:30_it1`-`13:30_it2`)) as `13:30_it1`,
SUM(abs(`14:30_it1`-`14:30_it2`)) as `14:30_it1`,
SUM(abs(`15:30_it1`-`15:30_it2`)) as `15:30_it1`,
SUM(abs(`16:30_it1`-`16:30_it2`)) as `16:30_it1`,
SUM(abs(`17:30_it1`-`17:30_it2`)) as `17:30_it1`,
SUM(abs(`18:30_it1`-`18:30_it2`)) as `18:30_it1`,
SUM(abs(`19:30_it1`-`19:30_it2`)) as `19:30_it1`,
SUM(abs(`20:30_it1`-`20:30_it2`)) as `20:30_it1`,
SUM(abs(`21:30_it1`-`21:30_it2`)) as `21:30_it1`,
SUM(abs(`22:30_it1`-`22:30_it2`)) as `22:30_it1`,
SUM(abs(`23:30_it1`-`23:30_it2`)) as `23:30_it1`,
MAX(abs(`0:30_it1`-`0:30_it2`)) as `0:30_it2`,
MAX(abs(`1:30_it1`-`1:30_it2`)) as `1:30_it2`,
MAX(abs(`2:30_it1`-`2:30_it2`)) as `2:30_it2`,
MAX(abs(`3:30_it1`-`3:30_it2`)) as `3:30_it2`,
MAX(abs(`4:30_it1`-`4:30_it2`)) as `4:30_it2`,
MAX(abs(`5:30_it1`-`5:30_it2`)) as `5:30_it2`,
MAX(abs(`6:30_it1`-`6:30_it2`)) as `6:30_it2`,
MAX(abs(`7:30_it1`-`7:30_it2`)) as `7:30_it2`,
MAX(abs(`8:30_it1`-`8:30_it2`)) as `8:30_it2`,
MAX(abs(`9:30_it1`-`9:30_it2`)) as `9:30_it2`,
MAX(abs(`10:30_it1`-`10:30_it2`)) as `10:30_it2`,
MAX(abs(`11:30_it1`-`11:30_it2`)) as `11:30_it2`,
MAX(abs(`12:30_it1`-`12:30_it2`)) as `12:30_it2`,
MAX(abs(`13:30_it1`-`13:30_it2`)) as `13:30_it2`,
MAX(abs(`14:30_it1`-`14:30_it2`)) as `14:30_it2`,
MAX(abs(`15:30_it1`-`15:30_it2`)) as `15:30_it2`,
MAX(abs(`16:30_it1`-`16:30_it2`)) as `16:30_it2`,
MAX(abs(`17:30_it1`-`17:30_it2`)) as `17:30_it2`,
MAX(abs(`18:30_it1`-`18:30_it2`)) as `18:30_it2`,
MAX(abs(`19:30_it1`-`19:30_it2`)) as `19:30_it2`,
MAX(abs(`20:30_it1`-`20:30_it2`)) as `20:30_it2`,
MAX(abs(`21:30_it1`-`21:30_it2`)) as `21:30_it2`,
MAX(abs(`22:30_it1`-`22:30_it2`)) as `22:30_it2`,
MAX(abs(`23:30_it1`-`23:30_it2`)) as `23:30_it2`,
CONCAT('s',SUBSTRING(table_chart,2)) as table_chart
from export_data.inverted
group by table_chart;
目前需要花费太多时间，我认为可能会有一些比一次又一次重复代码更简单的方法，但我不确定如何。
答案 0 :(得分：2)
非常奇怪的查询，但看起来你正在根据半小时的增量对T1和T2值进行数据透视表样式查询。我会暂停枢轴，直到预先聚合的唯一行之后。由于您的“ts”列似乎代表“时间段”，并且半小时增量没有变化，因此我会将您的第一个查询简化为。
drop temporary table if exists inverted;
create temporary table inverted as
select
s.table_chart,
cast(s.name as char(30)) as name,
s.ename,
s.ts,
MAX( CAST( s.value_t1 as decimal(9,2)) as t1Val,
MAX( CAST( s.value_t2 as decimal(9,2)) as t2Val,
'#ffffff' as tsColor
from
splitted_table s
group by
s.table_chart,
s.name,
s.ename,
s.ts
现在，每个图表，名称，ename可能有24个时隙行，但已经在自己的行中预先聚合。您的下一个UPDATE可以针对通用行进行简化。我删除了你的（小时，数值，系列）列，因为它们在最终插入之前没有被使用，所以我在那里添加了它们。
UPDATE inverted
SET tsColor = fn_rep_get_color_for_field( t1Val, t2Val, p_min, p_max)
where SUBSTRING(table_chart, 2, 1) = 'p';
UPDATE inverted
SET tsColor = fn_rep_get_color_for_field(t1Val, t2Val, q_min, q_max)
where SUBSTRING(table_chart, 2, 1) = 'q';
UPDATE inverted
SET tsColor = fn_rep_get_color_for_field( t1Val, t2Val, u_min, u_max)
where SUBSTRING(table_chart, 2, 1) = 'u';
现在，你可以转动。由于您的第一个查询是按表格和ename分组，每个T1或T2值获得MAX（）值，每个时间段/表格/ eName只会有一条记录。
DROP TEMPORARY TABLE IF EXISTS deviation;
CREATE TEMPORARY TABLE deviation as
select
'table footer' as name,
SUM( if( i.ts='00:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '0:30_it1',
SUM( if( i.ts='01:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '1:30_it1',
SUM( if( i.ts='02:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '2:30_it1',
SUM( if( i.ts='03:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '3:30_it1',
SUM( if( i.ts='04:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '4:30_it1',
SUM( if( i.ts='05:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '5:30_it1',
SUM( if( i.ts='06:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '6:30_it1',
SUM( if( i.ts='07:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '7:30_it1',
SUM( if( i.ts='08:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '8:30_it1',
SUM( if( i.ts='09:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '9:30_it1',
SUM( if( i.ts='10:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '10:30_it1',
SUM( if( i.ts='11:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '11:30_it1',
SUM( if( i.ts='12:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '12:30_it1',
SUM( if( i.ts='13:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '13:30_it1',
SUM( if( i.ts='14:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '14:30_it1',
SUM( if( i.ts='15:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '15:30_it1',
SUM( if( i.ts='16:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '16:30_it1',
SUM( if( i.ts='17:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '17:30_it1',
SUM( if( i.ts='18:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '18:30_it1',
SUM( if( i.ts='19:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '19:30_it1',
SUM( if( i.ts='20:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '20:30_it1',
SUM( if( i.ts='21:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '21:30_it1',
SUM( if( i.ts='22:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '22:30_it1',
SUM( if( i.ts='23:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '23:30_it1',
MAX( if( i.ts='00:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '0:30_it2',
MAX( if( i.ts='01:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '1:30_it2',
MAX( if( i.ts='02:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '2:30_it2',
MAX( if( i.ts='03:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '3:30_it2',
MAX( if( i.ts='04:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '4:30_it2',
MAX( if( i.ts='05:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '5:30_it2',
MAX( if( i.ts='06:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '6:30_it2',
MAX( if( i.ts='07:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '7:30_it2',
MAX( if( i.ts='08:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '8:30_it2',
MAX( if( i.ts='09:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '9:30_it2',
MAX( if( i.ts='10:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '10:30_it2',
MAX( if( i.ts='11:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '11:30_it2',
MAX( if( i.ts='12:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '12:30_it2',
MAX( if( i.ts='13:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '13:30_it2',
MAX( if( i.ts='14:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '14:30_it2',
MAX( if( i.ts='15:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '15:30_it2',
MAX( if( i.ts='16:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '16:30_it2',
MAX( if( i.ts='17:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '17:30_it2',
MAX( if( i.ts='18:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '18:30_it2',
MAX( if( i.ts='19:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '19:30_it2',
MAX( if( i.ts='20:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '20:30_it2',
MAX( if( i.ts='21:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '21:30_it2',
MAX( if( i.ts='22:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '22:30_it2',
MAX( if( i.ts='23:30:00', ABS( i.t1Val - i.t2Val ), 0 )) as '23:30_it2',
cast(null as time) as hour,
cast(null as decimal(9,2)) as value,
cast(null as char(17)) as series,
CONCAT('s',SUBSTRING(i.table_chart,2)) as table_chart
from
export_data.inverted
group by
table_chart;
现在，为您的COLOR。既然你永远不会查询那些，它们甚至不会成为你的结果集的一部分，并且可能存在基于“p”，“q”或“u”值的不同值，你想要对另一个进行预测...... MIN，MAX或者一个特别基于“你”的胜利？或者“p”？您必须为...之类的内容添加IF（）或case / when。
您的表结构将具有基于您的原始数据透视查询结构和CREATE LIKE条件的列，但由于您从未插入这些列，因此它们始终为空。所以要额外花费数小时，数值，系列列。
MAX( if( i.ts='00:30:00',
case when SUBSTR( i.table_chart, 2, 1 ) = 'p' then i.tsColor
when SUBSTR( i.table_chart, 2, 1 ) = 'q' then i.tsColor
ELSE i.tsColor end,
null )) as '0:30:00_Color',
etc for rest of colors.
希望这有助于您从另一个角度看待可能的解决方案。