Question : sql query

below is the code I am getting duplicate values. I want to change this query so the id's in first section and not used again in the second section where we do the union
--first section
select
a.*
from mytable a
inner join (select ID,srno from mytable b
      where srno >= (select max(srno) from mytable c
            where b.ID = c.ID and date1 < @start_date )
      and (case when date1 < @start_date and status = 'start' then 'start' end )= 'start'
      ) b on a.ID = b.ID
where not exists(select ID from mytable b where a.ID = b.ID
and (status = 'end' or date1>@end_date)) and
b.srno = a.srno - 1
when we are doing the union
the below the section should select only the id's not used in the section1
--section2
select
a.*
from mytable a
inner join (
      select ID , max(srno) srno from mytable f where exists(select ID from mytable e
            where status = 'start' and f.ID = e.ID and e.date1 >= @start_date
            and e.date1 < dateadd(d,1,@end_date))
group by ID
      ) b on a.ID = b.ID
where not exists(select ID from mytable b where a.ID = b.ID
and (status = 'end' or date1>@end_date)) and
b.srno = a.srno
order by a.ID
end

please help

Code Snippet:

         
           
             1:
2:
3:
4:
5:
6:
7:
8:
9:
10:
11:
12:
13:
14:
15:
16:
17:
18:
19:
20:
21:
22:
23:
24:
25:
26:
27:
28:
29:

           
           
             select 
a.*
from mytable a
inner join (select ID,srno from mytable b
	where srno >= (select max(srno) from mytable c
		where b.ID = c.ID and date1 < @start_date )
	and (case when date1 < @start_date and status = 'start' then 'start' end )= 'start'
	) b on a.ID = b.ID 
where not exists(select ID from mytable b where a.ID = b.ID 
and (status = 'end' or date1>@end_date)) and
 b.srno = a.srno - 1

union 


select 
a.*
from mytable a
inner join (
	select ID , max(srno) srno from mytable f where exists(select ID from mytable e
		where status = 'start' and f.ID = e.ID and e.date1 >= @start_date 
		and e.date1 < dateadd(d,1,@end_date))
group by ID
	) b on a.ID = b.ID 
where not exists(select ID from mytable b where a.ID = b.ID 
and (status = 'end' or date1>@end_date)) and
 b.srno = a.srno 
order by a.ID
end

           
         

Open in New Window

Select All

Answer : sql query

try like this.

                       
         
           
             1:
2:
3:
4:
5:
6:
7:
8:
9:
10:
11:
12:
13:
14:
15:
16:
17:
18:
19:
20:
21:
22:
23:

           
           
             ;with cte_1 as (
select a.*
from mytable a
inner join (select ID,srno from mytable b
        where srno >= (select max(srno) from mytable c
                where b.ID = c.ID and date1 < @start_date )
        and (case when date1 < @start_date and status = 'start' then 'start' end )= 'start'
        ) b on a.ID = b.ID 
where not exists(select ID from mytable b where a.ID = b.ID 
and (status = 'end' or date1>@end_date)) and
 b.srno = a.srno - 1),
cte_2 as (
select 
a.*
from mytable a
inner join (
        select ID , max(srno) srno from mytable f where exists(select ID from mytable e
                where status = 'start' and f.ID = e.ID and e.date1 >= @start_date 
                and e.date1 < dateadd(d,1,@end_date))
group by ID
        ) b on a.ID = b.ID )
select * from cte_1 union
select * from cte_2 c2 where not exists (select 1 from cte_1 c1 where c1.id = c2.id)

           
         

Open in New Window

Select All