Need to Pivot some data

  • Hi to all,

    I have this data:

    S1 S2 Crit

    1 3 9.0

    2 3 8.0

    3 3 7.0

    1 5 1.0

    2 5 2.0

    3 5 3.0

    and I need to make it:

    S2 Crit1 Crit2 Crit3

    3 9.0 8.0 7.0

    5 1.0 2.0 3.0

    Thank you for any help you can provide me.

  • If the number of values for s1 is fixed (in your example - 3) - then you can use the following

    create table #temp (s1 tinyint, s2 tinyint, crit float)

    insert into #temp select 1, 3, 9.0

    insert into #temp select 2, 3, 8.0

    insert into #temp select 3, 3, 7.0

    insert into #temp select 1, 5, 1.0

    insert into #temp select 2, 5, 2.0

    insert into #temp select 3, 5, 3.0

    select s2,

     (select crit from #temp t2 where t2.s2 = t1.s2 and t2.s1 = 1) as Crit1,

     (select crit from #temp t2 where t2.s2 = t1.s2 and t2.s1 = 2) as Crit2,

     (select crit from #temp t2 where t2.s2 = t1.s2 and t2.s1 = 3) as Crit3

    from #temp t1

    group by s2

    Cheers


    I feel the need - the need for speed

    CK Bhatia

  • This posted script from iecdba should do the trick...

    http://qa.sqlservercentral.com/scripts/viewscript.asp?scriptid=422

     

    -

  • CK Bhatia - If you feel the need for speed, you should use this query instead:

    select s2,

     sum(case when s1 = 1 then crit else 0 end) as Crit1,

     sum(case when s1 = 2 then crit else 0 end) as Crit2,

     sum(case when s1 = 3 then crit else 0 end) as Crit3

    from #temp t1

    group by s2

    Alberto - If you only need a static pivot, use the query above. If you need a dynamic piovt, these are the references I usually give:

    http://www.sqlteam.com/item.asp?ItemID=2955

    http://weblogs.sqlteam.com/jeffs/archive/2005/05/02/4842.aspx

    I've not compared these to the reference Jason posted, but all are generally the same idea.

    Ryan Randall

    Solutions are easy. Understanding the problem, now, that's the hard part.

Viewing 4 posts - 1 through 3 (of 3 total)

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