如何选择两个不同的列？

Question

I want to be able to select two distinct from col1 and col2 ordered by id.

I'm struggling to do this because when I write the following SQL query...

SELECT DISTINCT col1, col2
FROM table
ORDER BY id

I can't ORDER BY id because it's not in the SELECT statement but if I put id in the SELECT statement it will take the DISTINCT id, col1 and col2. Which is basically the whole table as it is since the id column is unique.

How do I do this?

Answer 1

您可以使用聚合，并在order by子句中放置一个聚合函数：

select col1, col2 from mytable group by col1, col2 order by min(id) limit 10

Answer 2

This is one way to do it:

select A.col1, A.col2
from 
(select id, col1, col2
from Tablet
order by id) A
left join 
(select min(id) id2, col1, col2
from Tablet
GROUP BY COL1, COL2) B
on A.COL1 = B.COL1 AND A.COL2=b.COL2
where A.id = B.id2
LIMIT 4;

Here is the DEMO

Answer 3

If the task means "if there are duplicates by (col1, col2) then remove all such duplicates except one with the greatest id value", then

SELECT DISTINCT MAX(id) OVER (PARTITION BY col1, col2) id, col1, col2
FROM sourcetable
ORDER BY id DESC /* not sure in this point */
LIMIT 10

PS. Why "greatest"? because of

So it should eliminate row 3 and NOT row 4 as it is duplicated.

If the record with the least id value needed then replace MAX() to MIN().

PPS. It seems the real task is "Show last 10 original messages from chat history".