Unfortunately really busy week. I haven't had a chance to review yet. Will update when I get a chance. Thanks

George,

I will try and test this today. At a quick glance, it looks promising.

Thanks,
Chris

Andy and Jason,

Thanks for your solution. It works great on this data set, but unfortunately it doesn't work for all data sets. For example,

I'm starting to think...

Scott,

Very close, but I am seeing this:

Id 4 should not be null it should be 3. Thanks

Correct

Jason,

That looks like the answer I'm looking for, but I get an error converting varbinary to numeric. I will see if I can work that out.

Yes, I tried that, but it doesn't work. The group number changes base on an occurrence of 'S'. In the case of Site 1 seq 1 and 2 should be...

What about the changes below?

USE [TransactionalData]

GO

DECLARE @return_value int

declare @itm_cd varchar (30)

declare @del_doc_num varchar (30)

declare @ItemCode varchar (30)

declare @DelDoc varchar (30)

set @itm_cd = (select itm_cd from temp_sales_order_cbo where itm_cd...

Since you are returning 96% of the table rather than build the uid's you need, build the uid's you don't and exclude them. Maybe something like this:

;WITH excludesRef AS

(

...

I think you could do something like:

WHERE int_acct_num/10000000=1 AND int_acct_num=CAST(str_acct_num AS INT)

Are you using only the date part of the datetime in your dataset? Something like the following:

select distinct CONVERT(varchar(10), issueddate,101) as issueddate from table1

If you do, set the parameter type...