分割函数和

CREATE OR REPLACE TYPE str_split IS TABLE OF VARCHAR2 (4000);
CREATE OR REPLACE FUNCTION splitstr(p_string IN VARCHAR2, p_delimiter IN VARCHAR2)
    RETURN str_split 
    PIPELINED
AS
    v_length   NUMBER := LENGTH(p_string);
    v_start    NUMBER := 1;
    v_index    NUMBER;
BEGIN
    WHILE(v_start <= v_length)
    LOOP
        v_index := INSTR(p_string, p_delimiter, v_start);
 
        IF v_index = 0
        THEN
            PIPE ROW(SUBSTR(p_string, v_start));
            v_start := v_length + 1;
        ELSE
            PIPE ROW(SUBSTR(p_string, v_start, v_index - v_start));
            v_start := v_index + 1;
        END IF;
    END LOOP;
 
    RETURN;
END splitstr;
 
 
--使用方法--
select * from table(splitstr(‘16021-16022‘,‘-‘));
 
--执行出的结果
 
COLUMN_VALUE
16021
16022

CREATE OR REPLACE FUNCTION  fn_GetDead (v_deadId varchar2)
return varchar2
is
  
v_deadString varchar2(100);
v_deadString_all VARCHAR2(4000);
cursor c_deadIdSpilt  --设置一个分割字符串游标--
 is
select *  from table(splitstr(‘‘||v_deadId||‘‘,‘-‘));
c_row c_deadIdSpilt%rowtype;
  
begin  
    
 if  instr(v_deadId,‘-‘) > 0 then  ---判读变量中是否含有"-"字符，如果有，则循环遍历这个字符，取出亡人姓名
  open c_deadIdSpilt;              --打开游标--
  loop
    fetch c_deadIdSpilt into c_row;
    EXIT WHEN c_deadIdSpilt%NOTFOUND;
      SELECT wrxm into v_deadString FROM wrxxb WHERE id=C_ROW.COLUMN_VALUE;
        
       v_deadString_all:=v_deadString_all||‘-‘||v_deadString;       
       
  end loop;
  
 close c_deadIdSpilt;
 
  return ltrim(v_deadString_all,‘-‘);
 else
     
  SELECT wrxm INTO v_deadString FROM wrxxb WHERE id=‘‘||v_deadId||‘‘;
  
  return v_deadString;
 end if;
  
--异常处理--
exception
  
  when others then dbms_output.put_line(‘error‘);
  return -1;
end fn_GetDead;
/
 --调用--
SELECT  fn_GetDead(‘16021-16022‘) FROM dual;
 
--我执行出的结果-- 
FN_GETDEAD(‘16021-16022‘)
马瑞祥-赵桂芝