Question

I’ve got the query:

select distinct index, name, code, amount as amount_min 
from sales s 
where (amount = 
(
    select mindex(s1.amount) 
    from sales s1 
    where  s1.name = s.name
)) 
order by s.name

Query works very well, but I would like to modify them so that when the column index, code, amount in several rows are identical, only one show, the first row. The rest (although different code columns) should be ignored. Can someone help me?

Answer 1

Since I don’t know if it’s oracle, sql server, sybase ase, etc., you way to do it is like this:

Step 1: insert the data in a temporary table with an identity column

select distinct counter = identity(10), index, name, code, amount as amount_min
into #t1
from sales s 
where (amount = 
(
    select mindex(s1.amount) 
    from sales s1 
    where  s1.name = s.name
)) 
order by s.name

Step 2: make the needed select

select index, name, code, amount_min from #t1
where counter in (select min(counter) from #t1 group by index, name, amount_min)

Step 3: drop the temporary table

drop table #t1