Ranking Query

Question

Ranking Query

cooljagadeesh

SSC Eights!

Points: 959
More actions
April 15, 2013 at 4:53 pm

#276461

Hi All,
i have a requirement to find out which employee pass three exams continuously
if you observe below query dataset
a passed three exams continuously
again b also passed
c and d failed
again e passed
finally it should display a,b,e rows only
declare @table table(id int ,name varchar(20),row int identity(1,1) )
insert into @table select 1,'a' union all
select 2,'a' union all
select 3,'a' union all
select 1,'b' union all
select 1,'b' union all
select 2,'b' union all
select 3,'b' union all
select 3,'c' union all
select 3,'c' union all
select 3,'c' union all
select 6,'d' union all
select 6,'d' union all
select 7,'d' union all
select 9,'e' union all
select 10,'e' union all
select 11,'e'
select * from @table
here i am trying to fetch by using ranking functions but not able to find out the query
please help any one
Regards,
Jagadeesh

Viewing 5 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic. Login to reply

Mansfield SSCrazy Points: 2396 More actions · Answer 1

You'll probably need a query that GROUP BY the name column. Combine this with a HAVING search condition that checks for MIN(), MAX() and COUNT(DISTINCT ) should get you what you want.

GROUP BY

HAVING

Alternatively, you can solve this problem by having two joins to the same table, but joining to the previous and next ID values, respectively.

Looking forward to see what query you come up with 🙂

Anjan@Sql SSCrazy Points: 2316 More actions · Answer 2

Select Name,COUNT(distinct id)

from @table

group by name

having COUNT(distinct id) =3

Pankaj067 Old Hand Points: 313 More actions · Answer 3

Hello cooljagadeesh,

You can implement RANK in your query like this

select *,

RANK() over (partition by name order by row) as rank

from @table

also for your filter purpose, the query is

select * from

(

select *,RANK() over (partition by name order by row) as rank from @table

) temp where rank =3

Thanks....:-)

ChrisM@Work SSC Guru Points: 186127 More actions · Answer 4

Pankaj067 (4/16/2013)
Hello cooljagadeesh,
You can implement RANK in your query like this
select *,
RANK() over (partition by name order by row) as rank
from @table
also for your filter purpose, the query is
select * from
(
select *,RANK() over (partition by name order by row) as rank from @table
) temp where rank =3
Thanks....:-)

Did you test it against the requirements?

declare @table table(id int ,name varchar(20),row int identity(1,1) )

insert into @table select 1,'a' union all

select 2,'a' union all

select 3,'a' union all

select 1,'b' union all

select 2,'b' union all

select 3,'b' union all

select 3,'c' union all

select 6,'d' union all

select 7,'d' union all

select 9,'e' union all

select 10,'e' union all

select 11,'e'

-------------------------------------------------

SELECT *,

rk = DENSE_RANK() over (partition by name order by ID)

FROM @table

--also for your filter purpose, the query is

SELECT *

FROM

(

SELECT *,

rk = DENSE_RANK() OVER (PARTITION BY name ORDER BY ID)

FROM @table

) temp

WHERE rk =3

😉

^{“Write the query the simplest way. If through testing it becomes clear that the performance is inadequate, consider alternative query forms.” - Gail Shaw}

For fast, accurate and documented assistance in answering your questions, please read this article.
Understanding and using APPLY, (I) and (II) Paul White
Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden