日期:2014-05-17 浏览次数:21139 次
create table testa
(
    sname    varchar2(30),
    stime    varchar2(20)
);
create table testb as select * from testa;
create or replace trigger tri_testb
  after insert on testa  
  for each row
declare
  -- local variables here
begin
    if inserting then
        merge into testb a
        using (select :new.sname as sname, :new.stime as stime from dual) b
           on (a.sname = b.sname)
         when matched then
        update set a.stime = b.stime
         when not matched then
        insert (a.sname, a.stime)
        values (b.sname, b.stime);
    end if;
end tri_testb;
/
select * from testb;
insert into testa values('张三', '20110912');
insert into testa values('张三', '20110913');
insert into testa values('张三', '20110914');
insert into testa values('李四', '20110912');
insert into testa values('张三', '20110913');
commit;
select * from testb;
insert into testa values('张三', '20110915');
commit;
select * from testb;
------解决方案--------------------
建立一个针对A表insert的触发器,如果表为a、b,字段为name,time
CREATE OR REPLACE TRIGGER TR_AUTO
  after INSERT ON 表A
  FOR EACH ROW
DECLARE
  NEXTID NUMBER;
BEGIN
   SELECT SEQ_ADV_IDX.NEXTVAL INTO NEXTID FROM DUAL;
   update b set b.time= :NEW.time where b.name=:NEW.name
END;
------解决方案--------------------
楼上都已经些出来了,我就不再写了
这个其实就是触发器的基本功能啊:
A表插入后 (after insert)触发触发器,
触发器的功能有两步:1,查询B表(查看B表中是否已经有该数据了);2,根据查询结果作更新或插入操作(1的返回结果为true,则更新B表;否则插入B表)
ps:A表中新插入的数据可以用(:NEW.xxxx)取得