假设有一张表oldyang_bayern,取其中的10条数据作为参考:
    
  
现在需要合并此张表中姓名重复的数据,将符合条件的字段值,合并到重名数据中ID最高的那一条。
要合并的字段分为三种情况:
1. 常规字段(DEPARTMENT/JOBTITLE):取重名数据中ID最高的非空值,例如“施魏因施泰格”的DEPARTMENT字段合并后,应取值“组织部”;
2. 依据数值大小取值的字段(SALARY):取重名数据中的最大值;
3. 非常规字段(CITY/DITRICT/ADDRESS/ADDRESSRANK):这些字段都与地址相关,以ADDRESSRANK最高的那条数据为准(ADDRESSRANK由高到低分别为S|A|B|C|D),然后取此条数据中的CITY/DITRICT/ADDRESS值;如果ADDRESSRANK相同,取ID最高的数据。
我所给出的方法:
处理合并的存储过程 prc_oldyang_bayern
    create or replace procedure prc_oldyang_bayern(pname varchar2)
is
cursor cur is
 select * from oldyang_bayern
  where name=pname order by id desc; --NAME相同的数据按id降序排列
 
--申明记录表info_table,用于存放NAME相同的数据
 type info_table_type is table of oldyang_bayern%ROWTYPE index by binary_integer;
 info_table info_table_type;
 
 info_dest oldyang_bayern%rowtype;
 nCount number(2):=0; --计数变量
 vi number(2); --在更新地址字段时将会用于放入记录号
 
begin
 open cur;
  --通过循环计数的方式将重复的记录放入记录表
  loop
   nCount:= nCount+1;
   fetch cur into info_table(nCount);
   exit when cur%notfound;
  end loop;
 close cur;
 
 nCount := nCount - 1;
 
 --更新常规字段DEPARTMENT/JOBTITLE
 for i in 1..nCount-1 loop
  if i = 1 then
   info_dest.DEPARTMENT:= nvl(info_table(i).DEPARTMENT, info_table(i+1).DEPARTMENT);
  else
   info_dest.DEPARTMENT:= nvl(info_dest.DEPARTMENT, info_table(i+1).DEPARTMENT);
  end if;
 end loop;
 
 for i in 1..nCount-1 loop
  if i = 1 then
   info_dest.JOBTITLE:= nvl(info_table(i).JOBTITLE, info_table(i+1).JOBTITLE);
  else
   info_dest.JOBTITLE:= nvl(info_dest.JOBTITLE, info_table(i+1).JOBTITLE);
  end if;
 end loop;
 
 --更新SALARY字段
 for i in 1..nCount-1 loop
  if i = 1 then
   info_dest.SALARY:= greatest(nvl(info_table(i).SALARY,-1), nvl(info_table(i+1).SALARY,-1));
  else
   info_dest.SALARY:= greatest(nvl(info_dest.SALARY,-1), nvl(info_table(i+1).SALARY,-1));
  end if;
 end loop;
 info_dest.SALARY:= replace(info_dest.SALARY,-1); --处理当SALARY为空值时的状况
 
 --更新与ADDRESSRANK相关的字段
 --此处引用了一个自定义函数fun_get_ranknum
 --利用这个函数将地址等级S|A|B|C|D转换成对应的地址等级序号1|2|3|4|5(函数代码会在后面列出)
 --接着拼接地址等级序号+记录号,取拼接后最小值
 --此方法用于处理当ADDRESSRANK相同时,取ID最高的那条ADDRESSRANK相关记录
 --(注:在游标中记录已按ID降序排列)
 for i in 1..nCount-1 loop
  if i = 1 then
   info_dest.ADDRESSRANK:= least(fun_get_ranknum(info_table(i).ADDRESSRANK)||i, fun_get_ranknum(info_table(i+1).ADDRESSRANK)||i+1);
  else
   info_dest.ADDRESSRANK:= least(info_dest.ADDRESSRANK,  fun_get_ranknum(info_table(i+1).ADDRESSRANK)||i+1);
  end if;
 end loop;
 
 vi:= regexp_substr(info_dest.ADDRESSRANK, '\d$'); --使用正则表达式取出拼接值中的记录号
 info_dest.ADDRESSRANK:=info_table(vi).ADDRESSRANK;
 info_dest.CITY:=info_table(vi).CITY;
 info_dest.DISTRICT:=info_table(vi).DISTRICT;
 info_dest.ADDRESS:=info_table(vi).ADDRESS;
 
 --将筛选后的变量值更新到ID最高的记录当中
 update oldyang_bayern
   set DEPARTMENT=info_dest.DEPARTMENT,
     JOBTITLE=info_dest.JOBTITLE,
     SALARY=info_dest.SALARY,
     CITY=info_dest.CITY,
     DISTRICT=info_dest.DISTRICT,
     ADDRESS=info_dest.ADDRESS,
     ADDRESSRANK=info_dest.ADDRESSRANK
  where id = info_table(1).id;
 
 --删除重复的记录
 delete from oldyang_bayern where name=pname and id != info_table(1).id;
 
 commit;
end ;
  
  
存储过程中所用到的自定义函数 fun_get_ranknum
    create or replace function fun_get_ranknum(prank in varchar2)
return varchar2
is
 Result varchar(2);
 vcTemp varchar(2);
begin
 vcTemp:=prank;
 
 case
  when vcTemp = 'S' then
     vcTemp:= '1';
  when vcTemp = 'A' then
     vcTemp:= '2';
  when vcTemp = 'B' then
     vcTemp:= '3';
  when vcTemp = 'C' then
     vcTemp:= '4';
  when vcTemp = 'D' then
     vcTemp:= '5';
  when vcTemp is null then
     vcTemp:= '9';
 end case;
 
 Result := vcTemp;
 return(Result);
 
 exception
  when others then
   Result := '9';
  return result;
end ;
  
  
利用匿名块传递重复姓名到prc_oldyang_bayern,执行合并过程
    declare
cursor cur is select name from oldyang_bayern group by name having count(*)>1;
rec cur%rowtype;
begin
 open cur;
  loop
   fetch cur into rec;
     exit when cur%notfound;
   prc_oldyang_bayern(rec.name);
  end loop;
 close cur;
end;
  
  转自: http://www.oldyang.com/2009/06/23/using_plsql_remove_duplicates/


 
     
					 
					