根据特定条件返回唯一行

Question

I need a query that will return distinct rows, but will not return any rows that have duplicate fruits where one of the rows is 'Approved'. Note: there will always be at most two fruits of the same name.

For example, the ideal result would be:

What I have so far is simply:

SELECT DISTINCT * FROM Food;

But this obviously still returns one of the rows that has the fruit 'Apple'

Answer 1

You can make use of some conditional aggregation for the desired result set like

select fruit, min(status)
from demo
group by fruit
having sum(status = 'Approved') = 0
and sum(status = 'Bought') > 0

or just use select fruit, 'Bought' as status

Demo

Answer 2

SELECT DISTINCT * FROM Food WHERE Fruit NOT IN (SELECT (UNIQUE Fruit) FROM Food WHERE Status = 'Approved');

试试这个查询。

Answer 3

I think you could solve this problem a different way perhaps? To get that same result you could add a WHERE clause, for instance:

SELECT DISTINCT * FROM Food WHERE Status = 'Bought';

You could also alter the data model so that a fruit's status is updated rather than adding another record for the same fruit (if that suits the application's purposes).

If you want to maintain the schema and get the same, I believe you can do: SELECT DISTINCT on (Fruit) Fruit, Status FROM Food ORDER BY Fruit, Status ASC;

Source