Question

I have the following table:

Id  CategoryId  Code  Status
=================================
1   A100        0012  NULL
2   A100        0012  NULL
3   A100        0055  NULL
4   A100        0012  NULL
5   B201        1116  NULL
6   B201        1116  NULL
7   B201        1121  NULL
8   B201        1024  NULL

Logic: 1. For same CategoryID, code should be the same.
2. If multiple code exists, consider the one with lowest value.

Desired result:

Id  CategoryId  Code  Status
=================================
1   A100        0012  NULL
2   A100        0012  NULL
3   A100        0055  FAIL
4   A100        0012  NULL
5   B201        1116  FAIL
6   B201        1116  FAIL
7   B201        1121  FAIL
8   B201        1024  NULL

Thanks

Answer 1

Following your edit…

DECLARE @T TABLE
(
ID INT,
CategoryID CHAR(4),
Code CHAR(4),
Status CHAR(4) NULL
)
INSERT INTO @T (ID,CategoryID, Code)
SELECT 1,'A100',0012 UNION ALL SELECT 2,'A100',0012 UNION ALL
SELECT 3,'A100',0055 UNION ALL SELECT 4,'A100',0012 UNION ALL
SELECT 5,'B201',1116 UNION ALL SELECT 6,'B201',1116 UNION ALL
SELECT 7,'B201',1121 UNION ALL SELECT 8,'B201',1024;

WITH T AS
(
SELECT *, MIN(Code) OVER (PARTITION BY CategoryID ) AS MinCode
from @T
)
UPDATE T
SET Status = 'FAIL'
WHERE Code <> MinCode

SELECT *
FROM @T

Returns

ID          CategoryID Code Status
----------- ---------- ---- ------
1           A100       12   NULL
2           A100       12   NULL
3           A100       55   FAIL
4           A100       12   NULL
5           B201       1116 FAIL
6           B201       1116 FAIL
7           B201       1121 FAIL
8           B201       1024 NULL