oracle拆分函數，將字符串拆分紅多行多字段表數據

--建立OBJECT類型,包含三個字段
CREATE OR REPLACE TYPE OBJ_V_TABLE AS OBJECT
(
  ID VARCHAR2(50),
  NAME VARCHAR2(50),
  DESCP VARCHAR2(50)
);


--建立一個表類型 
CREATE OR REPLACE TYPE V_TABLE IS TABLE OF OBJ_V_TABLE;


--建立拆分函數,分隔符每三個對應三個表字段

CREATE OR REPLACE FUNCTION SPLIT(P_STR       VARCHAR2,
                                 P_DELIMITER VARCHAR2 DEFAULT ',')
  RETURN V_TABLE

 IS

  RS V_TABLE := V_TABLE();

  L_STR VARCHAR2(4000) := '';
  L_LEN NUMBER := 0;
  ID    VARCHAR2(50);
  NAME  VARCHAR2(50);
  DESCP VARCHAR2(50);

BEGIN

  L_STR := P_STR;
  L_LEN := LENGTH(P_DELIMITER);

  WHILE  LENGTH(L_STR) > 0 LOOP
    RS.EXTEND;
    IF INSTR(L_STR, P_DELIMITER,1,3) > 0 THEN
      ID    := SUBSTR(L_STR, 1, INSTR(L_STR, P_DELIMITER) - 1);
      L_STR := SUBSTR(L_STR, INSTR(L_STR, P_DELIMITER) + L_LEN);
      NAME  := SUBSTR(L_STR, 1, INSTR(L_STR, P_DELIMITER) - 1);
      L_STR := SUBSTR(L_STR, INSTR(L_STR, P_DELIMITER) + L_LEN);
      DESCP := SUBSTR(L_STR, 1, INSTR(L_STR, P_DELIMITER) - 1);
      L_STR := SUBSTR(L_STR, INSTR(L_STR, P_DELIMITER) + L_LEN);
      RS(RS.COUNT) := OBJ_V_TABLE(ID, NAME, DESCP);
    ELSIF INSTR(L_STR, P_DELIMITER,1,2) > 0 THEN
           ID    := SUBSTR(L_STR, 1, INSTR(L_STR, P_DELIMITER) - 1);
           L_STR := SUBSTR(L_STR, INSTR(L_STR, P_DELIMITER) + L_LEN);
           NAME  := SUBSTR(L_STR, 1, INSTR(L_STR, P_DELIMITER) - 1);
           L_STR := SUBSTR(L_STR, INSTR(L_STR, P_DELIMITER) + L_LEN);
           DESCP := L_STR;
           RS(RS.COUNT) := OBJ_V_TABLE(ID, NAME, DESCP);
           EXIT;
     ELSIF INSTR(L_STR, P_DELIMITER,1,1) > 0 THEN
           ID    := SUBSTR(L_STR, 1, INSTR(L_STR, P_DELIMITER) - 1);
           NAME  := SUBSTR(L_STR, INSTR(L_STR, P_DELIMITER) + L_LEN);
           RS(RS.COUNT) := OBJ_V_TABLE(ID, NAME, '');
           EXIT;
     ELSE
        RS(RS.COUNT) := OBJ_V_TABLE(L_STR, '', '');
       EXIT;
     END  IF ;
  END LOOP ;

  RETURN  RS;

END ;

--運行 

SELECT * FROM TABLE(SPLIT('a,張山,1,b,李四,2,c,王五,3,5,六七'));

運行結果: