Question

How could I achieve something like this??

DECLARE @TEMP_TABLE TABLE (DATA VARCHAR(MAX), SHADOW_ID INT)
DECLARE @TEMP_TABLE1 TABLE (ENT VARCHAR(MAX), RMR VARCHAR(MAX), DTM VARCHAR(MAX))

INSERT INTO @TEMP_TABLE1 (ENT) SELECT DATA from @TEMP_TABLE WHERE DATA LIKE 'ENT%'
INSERT INTO @TEMP_TABLE1 (RMR) SELECT DATA from @TEMP_TABLE WHERE DATA LIKE 'RMR%'
INSERT INTO @TEMP_TABLE1 (DTM) SELECT DATA from @TEMP_TABLE WHERE DATA LIKE 'DTM%'

without having any null values, because each statement populates 1 column and nulls into the other two.

Thanks!

Answer 1

It looks like you’re trying to create a pivot or cross-tab query.

insert into @TEMP_TABLE1 (ENT, RMR, DTM)
select a.ENT, a.RMR, a.DTM
from (
    select t.SHADOW_ID
        , (select DATA from @TEMP_TABLE where SHADOW_ID = t.SHADOW_ID and DATA like 'ENT%') as ENT
        , (select DATA from @TEMP_TABLE where SHADOW_ID = t.SHADOW_ID and DATA like 'RMR%') as RMR
        , (select DATA from @TEMP_TABLE where SHADOW_ID = t.SHADOW_ID and DATA like 'DTM%') as DTM
    from @TEMP_TABLE as t
    group by t.SHADOW_ID
) a

This is assuming that you have only one ENT, RMR and DTM value for each SHADOW_ID. If this is not the case, then I must agree with @JeremyHolovacs.

I would recommend putting SHADOW_ID into @TEMP_TABLE1 as well, so you could know where the data came from.