Merging multiple rows without using max group by

Question

Hi guys,
Is there any better way to convert
with
test as
select '001' as id, 'Apple' as name, 'A' as segment1,null as segment2,null as segment3 from dual union all
select '001' as id, 'Apple' as name,null as segment1, 'P' as segment2,null as segment3 from dual union all
select '001' as id, 'Apple' as name,null as segment1,null as segment2, 'L' as segment3 from dual
select * from test
to
select '001' as id, 'Apple' as name, 'A' as segment1,'P' as segment2,'L' as segment3 from dual
without using max and group by?
There are about 5 million records and using max group by is slowing it down by a lot.
It is needed to be done by SQL.
Thank You

Answer 1

with
test as
select '001' as id, 'Apple' as name, 'A' as segment1,null as segment2,null as segment3 from dual union all
select '001' as id, 'Apple' as name,null as segment1, 'P' as segment2,null as segment3 from dual union all
select '001' as id, 'Apple' as name,null as segment1,null as segment2, 'L' as segment3 from dual
select distinct id, name
, last_value(segment1 ignore nulls) over (partition by id, name order by segment1) as seg1
, last_value(segment2 ignore nulls) over (partition by id, name order by segment2) as seg2
, last_value(segment3 ignore nulls) over (partition by id, name order by segment3) as seg3
from test
;No max or group by, but not necessarily any faster. If you post your actual query, some sample data with expected results, and the additional information requested in the following posts, then someone may be able to provide better suggestions.
When your query takes too long
When your query takes too long ...
How to Post a SQL statement tuning request
HOW TO: Post a SQL statement tuning request - template posting
Regards,
Bob