I have problem to distinct values on column based on other column. The case study is:
Table: List
well | wbore | op|
------------------
wella|wbore_a|op_a|
wella|wbore_a|op_b|
wella|wbore_a|op_b|
wella|wbore_b|op_c|
wella|wbore_b|op_c|
wellb|wbore_g|op_t|
wellb|wbore_g|op_t|
wellb|wbore_h|op_k|
So, I want the output to be appear in different field/column like:
well | total_wbore | total_op
----------------------------
wella | 2 | 3
---------------------------
wellb | 2 | 2
the real study case come from different table but to simplify it I just assume this case happened in 1 table.
The sql query that I tried:
SELECT well.well_name, wellbore.wellbore_name, operation.operation_name, COUNT(*)
FROM well
INNER JOIN wellbore ON wellbore.well_uid = well.well_uid
INNER JOIN operation ON wellbore.well_uid = operation.well_uid
GROUP BY well.well_name,wellbore.wellbore_name
HAVING COUNT(*) > 1
But this query is to calculate the duplicate row which not meet the requirement. Anyone can help?