Question

I’m trying to get info for the next most recent entry based on the ID and location and add that to the query I am running. Not the second most recent entry in the table, only the previous entry based on the ID.

I’m not sure if this is possible or I just can’t figure out how to do it or even which direction I should go.

Here is how the previous programmer did it:

SELECT -- blah blah blah
       , (SELECT TOP 1 CAST(id AS VARCHAR) + '~' + CONVERT(VARCHAR,  creation_date,   120)
          FROM header hdr2 
          WHERE hdr2.id < hdr.id 
                AND hdr2.location = hdr.location 
          ORDER BY hdr2.id DESC) AS "prev_info"
FROM header hdr 
/*blah blah blah*/

I understand why he wanted to use only a single sub query and tried to put everything into that one column, but I’m trying to optimize this and I would like to do it in some other,easier and more efficient, way.

I feel like the answer is staring me in the face, but I just can’t reach out and grasp it. If I need to explain myself any more, please don’t hesitate to ask.

Answer 1

Consider using outer apply:

select  -- blah blah blah
,       prev_info.*
from    header hdr 
outer apply
        (
        select  top 1 id
        ,       creation_date
        from    header hdr2 
        where   hdr2.id < hdr.id 
                and hdr2.location = hdr.location 
        order by
                hdr2.id desc
        ) as prev_info