I checked the answer here but didn't solve my problem because I even adding the else
still throwing the same error:
ERROR: column table1.column11 must appear in the GROUP BY clause or be used in an aggregate function
I would like to remove table1.column11
from SELECT
because is affecting the GROUP BY
. Instead of each ENUM
type be a column in a single line this query below is returning a table with a ENUM
column filled in different lines.
Query example:
SELECT table1.column11, -- enum column
table2.id,
(case when table1.column11 = 'enum_value_1' then SUM(table1.column12) end) as enum_value_1,
(case when table1.column11 = 'enum_value_2' then SUM(table1.column12) end) as enum_value_2,
--- ...
(case when table1.column11 = 'enum_value_9' then SUM(table1.column19) end) as enum_value_9,
FROM table1
LEFT JOIN table2 ON table2.id = table1.table2_id
WHERE table1.column13 IS NULL
GROUP BY table2.id, table1.column11
ORDER BY table2.id ASC
How is now:
table1.column11 | table2.id | enum_value_1 | enum_value_2 | ... | enum_value_9
================|===========|==============|==============|=====|=============
enum_value_1 | 1 | 10 | [null] | ... | [null]
enum_value_2 | 1 | [null] | 20 | ... | [null]
... | ... | ... | ... | ... | ...
enum_value_9 | 1 | [null] | [null] | ... | 15
How I would like to have:
table2.id | enum_value_1 | enum_value_2 | ... | enum_value_9
==========|==============|==============|=====|=============
1 | 10 | 20 | ... | 15
CodePudding user response:
The example in the question is difficult to reproduce, but I think my suggestions will be helpful. Remove table1.column11
from the select list and use the column only inside the aggregates.
So instead of:
SELECT table1.column11, -- enum column
table2.id,
(case when table1.column11 = 'enum_value_1' then SUM(table1.column12) end) as enum_value_1,
use:
SELECT
table2.id,
sum(case when table1.column11 = 'enum_value_1' then table1.column12 end) as enum_value_1,