【ORACLE&DB2】中如何删除一个表中重复数据，仅留其中一行
首先看一下ORACLE中的情况：
delete from test a where rowid < (select max(rowid) from test where col1=a.col1)；

但是DB2下由于没有序列rowid,所以不能用如上方法删除，只能实现一个存储过程，示例如下：
test表中只有一个列，其他可以仿照如下实现
CREATE PROCEDURE Delete_same_items (
   IN p_a INTEGER)
LANGUAGE SQL
SPECIFIC Delete_same_items
MODIFIES SQL DATA 
BEGIN ATOMIC
    DECLARE SQLSTATE CHAR(5);
    DECLARE v_a INTEGER;
    DECLARE c_emp CURSOR WITH HOLD FOR
        SELECT a
        FROM   test where a=p_a
        FOR UPDATE OF a; 
     DECLARE EXIT HANDLER FOR  NOT FOUND
     SET p_a = NULL;   
    OPEN c_emp;
    FETCH FROM c_emp INTO v_a;   
    WHILE ( SQLSTATE = '00000' ) DO
       FETCH FROM c_emp INTO v_a;      
       DELETE from test where  current of c_emp;
    END WHILE;   
    CLOSE c_emp;
END