原數據：
id code name value
----------- ---------- --------
1 2014000 A 10
2 2014000 B 9
3 2014001 C 100
4 2014002 D 4
5 2014002 E 5
6 2014003 F 9
7 2014003 G 10
期望結果：
code name1 value1 name2 value2
----------- --------- --------- --------- ---------
2014000 A 10 B 9
2014001 C 100 null null
2014002 D 4 E 5
2014003 F 9 G 10
注意：同1個ID最多只2條記錄，會出現相同value。
CREATE TABLE # (ID INT, code INT, NAME CHAR(1), VALUE VARCHAR(10))
INSERT INTO # VALUES
(1,2014000,'A','10'),
(2,2014000,'B','9'),
(3,2014001,'C','100'),
(4,2014002,'D','4'),
(5,2014002,'E','5'),
(6,2014003,'F','9'),
(7,2014003,'G','10')
SELECT
code,MIN(name) AS name1,MIN(CAST(REPLACE(value,' ','') AS INT)) AS value1,
CASE COUNT(ID) WHEN 1 THEN NULL ELSE MAX(name) END AS name2,
CASE COUNT(ID) WHEN 1 THEN NULL ELSE MAX(CAST(REPLACE(value,' ','') AS INT)) END AS value2
FROM
#
GROUP BY
code
