SQL problem [message #18839] |
Mon, 18 February 2002 06:25 |
Dave Adams
Messages: 1 Registered: February 2002
|
Junior Member |
|
|
Hi,
I wonder if you can help? I need to find a way to extract some information from a table:
DATE COUNT BILL TAX REF_FROM REF_TO
2/2/02 50 5237.40 247.02 101974033 101974082
2/2/02 55 5651.45 266.74 101974033 101974087
2/2/02 46 4351.23 102.12 101974992 101975102
The problem is the column REF_FROM can contain a unique value or there can be a number of duplicates.
If REF_FROM is unique I want the whole line, however if REF_FROM is not unique I only want the line where REF_TO is the largest number. Is this possible to do with an SQL script?
Thanks in advance,
Dave.
|
|
|
Re: SQL problem [message #18840 is a reply to message #18839] |
Mon, 18 February 2002 07:34 |
pratap kumar tripathy
Messages: 660 Registered: January 2002
|
Senior Member |
|
|
select * from dave
where
(ref_to >ref_from
and 0=(select count(*) from (select count(*) a,count(distinct ref_from) b from dave)
where a=b)
)
or 0!=(select count(*) from (select count(*) a,count(distinct ref_from) b from dave)
where a=b)
|
|
|