Hana 和Oracle拆分字符串

前面分享了mysql的写法,这里分享Hana内存数据库和Oracle数据库的写法.原理都是一样的。

HANA内存数据库写法:
CREATE PROCEDURE SPLIT_TEST(IN   V_ROID DECIMAL,
	                          TEXT NVARCHAR(1000))
AS 
BEGIN
    BEGIN 
        DELETE FROM ROLE_FUNCLIST WHERE ROLE_ID=:V_ROID;
    END;
 BEGIN
      DECLARE _ITEMS NVARCHAR(1000) ARRAY;
      DECLARE _TEXT NVARCHAR(1000);
      DECLARE _INDEX INTEGER;
      _TEXT := :TEXT;
      _INDEX := 1;
          WHILE LOCATE(:_TEXT,',') > 0 DO
             _ITEMS[:_INDEX] := SUBSTR_BEFORE(:_TEXT,',');
             INSERT INTO ROLE_FUNCLIST VALUES(:V_ROID,:_ITEMS[:_INDEX]);
            _TEXT := SUBSTR_AFTER(:_TEXT,',');
            _INDEX := :_INDEX + 1;
          END WHILE;
            INSERT INTO ROLE_FUNCLIST VALUES(:V_ROID,:_TEXT);
          _ITEMS[:_INDEX] := :_TEXT;
          RST = UNNEST(:_ITEMS) AS ("ITEMS"); 
          SELECT * FROM :RST;
 END; 
END;

CREATE OR REPLACE PROCEDURE SP_STR(V_STR_1 IN VARCHAR2)
AS
V_START NUMBER :=1;
V_LOCATE NUMBER;
V_STR_BEFORE VARCHAR2(4000);
V_STR VARCHAR2(100):=V_STR_1;
BEGIN
     WHILE INSTR(V_STR,',')>0 LOOP
     V_LOCATE :=INSTR(V_STR,',');
     V_STR_BEFORE :=SUBSTR(V_STR,V_START,V_LOCATE-V_START);
     INSERT INTO T(NAME) VALUES(V_STR_BEFORE);
     V_STR :=SUBSTR(V_STR,V_LOCATE+1);
     END LOOP;
     INSERT INTO T(NAME) VALUES(V_STR);
commit;  
END;