주 콘텐츠로 건너뛰기

검색

Welcome to Tibero GTS!

프로시저를 이용한 동적 PIVOT 수행 방법

문서유형ㅣ기술정보

분야ㅣ관리/환경설정

적용제품버전ㅣTibero 6

문서번호ㅣTADTI079

 

개요

데이터 PIVOT을 수행하는 방법은 아래와 같이 세 가지 방법이 있습니다.
 
  1. SUM(DECODE()) 구문을 이용한 피벗
  2. 프로시저를 이용한 동적 피벗
  3. PIVOT 구문을 이용한 피벗
 
이 문서에서는 프로시저(procedure)를 이용하여 동적으로 PIVOT을 수행하는 방법을 설명합니다.
동적 PIVOT 방식은 컬럼 값이 자주 변경되는 환경에서 유용합니다.
참고
테스트 환경은 Tibero 6 (DB 6.0 FS06_CS_1703) 버전 환경을 사용했습니다.
 

방법

 

테스트 테이블 생성

CREATE TABLE TEST( 
		YEARDATA VARCHAR(4), 
		DATA1    VARCHAR(6), 
		DATA2    VARCHAR(6)
);

 

데이터 입력

INSERT INTO TEST VALUES ('1990','100000','999999');
INSERT INTO TEST VALUES ('1991','100001','999998');
INSERT INTO TEST VALUES ('1992','100002','999997');
INSERT INTO TEST VALUES ('1993','100003','999996');
INSERT INTO TEST VALUES ('1994','100004','999995');
INSERT INTO TEST VALUES ('1995','100005','999994');
INSERT INTO TEST VALUES ('1996','100006','999993');
INSERT INTO TEST VALUES ('1997','100007','999992');
INSERT INTO TEST VALUES ('1998','100008','999991');
INSERT INTO TEST VALUES ('1999','100009','999990');
INSERT INTO TEST VALUES ('2000','100010','999989');
INSERT INTO TEST VALUES ('2001','100011','999988'); 
COMMIT;

SELECT * FROM TEST;

YEARDATA DATA1  DATA2 
-------- ------ ------
1990     100000 999999
1991     100001 999998
1992     100002 999997
1993     100003 999996
1994     100004 999995
1995     100005 999994
1996     100006 999993
1997     100007 999992
1998     100008 999991
1999     100009 999990
2000     100010 999989
2001     100011 999988 

12 rows selected.

 

프로시저를 이용한 동적 PIVOT 수행 방법

동적 PIVOT 방식은 컬럼 값이 자주 변경되는 환경에서 유용합니다. (반면에 정적방식으로 수행시 테이블의 row값이 증가하는 경우에는 작성이 어렵습니다. )
CREATE OR REPLACE PROCEDURE SP_PIVOT(OUT_CURSOR OUT SYS_REFCURSOR)
IS
	SQL_STMT VARCHAR2(30000);
	V_ROWIDX NUMBER(3) := 0;
	
BEGIN

	SQL_STMT := 'SELECT ';
	FOR X IN
		(
	SELECT  YEARDATA
	FROM  TEST
	ORDER BY YEARDATA
)
LOOP
	IF V_ROWIDX = 0 THEN
	V_ROWIDX := V_ROWIDX + 1;
	SQL_STMT := SQL_STMT||' MAX(DECODE(YEARDATA,'||X.YEARDATA||',NVL(DATA1,0),'''')) AS 
	"'||X.YEARDATA||'"';
ELSE
			SQL_STMT := SQL_STMT||', MAX(DECODE(YEARDATA,'||X.YEARDATA||',NVL(DATA1,0),'''')) 
			AS "'||X.YEARDATA||'"';
	END IF; 
END LOOP;
SQL_STMT := SQL_STMT||' FROM TEST';
SQL_STMT := SQL_STMT||' UNION ALL SELECT ';
V_ROWIDX := 0;
FOR Y IN
	(
	SELECT  YEARDATA
	FROM  TEST
	ORDER BY YEARDATA
)
LOOP
	IF V_ROWIDX = 0 THEN
	V_ROWIDX := V_ROWIDX + 1;
SQL_STMT := SQL_STMT||' MAX(DECODE(YEARDATA,'||Y.YEARDATA||',NVL(DATA2,0),'''')) AS "'||Y.YEARDATA||'"';
ELSE
		SQL_STMT := SQL_STMT||', MAX(DECODE(YEARDATA,'||Y.YEARDATA||',NVL(DATA2,0),'''')) AS "'||Y.YEARDATA||'"';
		END IF;
	END LOOP;
	SQL_STMT := SQL_STMT||' FROM TEST;';

	OPEN OUT_CURSOR FOR(SQL_STMT); 
	DBMS_OUTPUT.PUT_LINE(SQL_STMT);
END;
/

Procedure 'SP_PIVOT' created.

프로시저를 수행하여 쿼리를 동적으로 생성합니다. (YEARDATA 증가시에도 동적으로 쿼리 구문 작성 가능합니다.)
SQL> SET SERVEROUTPUT ON
SQL> var out_cursor refcursor
SQL> exec SP_PIVOT(:out_cursor);
SELECT  MAX(DECODE(YEARDATA,1990,NVL(DATA1,0),'')) AS "1990", 
MAX(DECODE(YEARDATA,1991,NVL(DATA1,0),'')) AS "1991", 
MAX(DECODE(YEARDATA,1992,NVL(DATA1,0),'')) AS "1992", 
MAX(DECODE(YEARDATA,1993,NVL(DATA1,0),'')) AS "1993", 
MAX(DECODE(YEARDATA,1994,NVL(DATA1,0),'')) AS "1994", 
MAX(DECODE(YEARDATA,1995,NVL(DATA1,0),'')) AS "1995", 
MAX(DECODE(YEARDATA,1996,NVL(DATA1,0),'')) AS "1996", 
MAX(DECODE(YEARDATA,1997,NVL(DATA1,0),'')) AS "1997", 
MAX(DECODE(YEARDATA,1998,NVL(DATA1,0),'')) AS "1998", 
MAX(DECODE(YEARDATA,1999,NVL(DATA1,0),'')) AS "1999", 
MAX(DECODE(YEARDATA,2000,NVL(DATA1,0),'')) AS "2000", 
MAX(DECODE(YEARDATA,2001,NVL(DATA1,0),'')) AS "2001" FROM TEST UNION ALL 
SELECT  MAX(DECODE(YEARDATA,1990,NVL(DATA2,0),'')) AS "1990", 
MAX(DECODE(YEARDATA,1991,NVL(DATA2,0),'')) AS "1991", 
MAX(DECODE(YEARDATA,1992,NVL(DATA2,0),'')) AS "1992", 
MAX(DECODE(YEARDATA,1993,NVL(DATA2,0),'')) AS "1993", 
MAX(DECODE(YEARDATA,1994,NVL(DATA2,0),'')) AS "1994", 
MAX(DECODE(YEARDATA,1995,NVL(DATA2,0),'')) AS "1995", 
MAX(DECODE(YEARDATA,1996,NVL(DATA2,0),'')) AS "1996", 
MAX(DECODE(YEARDATA,1997,NVL(DATA2,0),'')) AS "1997", 
MAX(DECODE(YEARDATA,1998,NVL(DATA2,0),'')) AS "1998", 
MAX(DECODE(YEARDATA,1999,NVL(DATA2,0),'')) AS "1999", 
MAX(DECODE(YEARDATA,2000,NVL(DATA2,0),'')) AS "2000", 
MAX(DECODE(YEARDATA,2001,NVL(DATA2,0),'')) AS "2001" FROM TEST;
 
쿼리 수행 결과는 다음과 같습니다. 
SQL> SELECT  MAX(DECODE(YEARDATA,1990,NVL(DATA1,0),'')) AS "1990", 
MAX(DECODE(YEARDATA,1991,NVL(DATA1,0),'')) AS "1991", 
MAX(DECODE(YEARDATA,1992,NVL(DATA1,0),'')) AS "1992", 
MAX(DECODE(YEARDATA,1993,NVL(DATA1,0),'')) AS "1993", 
MAX(DECODE(YEARDATA,1994,NVL(DATA1,0),'')) AS "1994", 
MAX(DECODE(YEARDATA,1995,NVL(DATA1,0),'')) AS "1995", 
MAX(DECODE(YEARDATA,1996,NVL(DATA1,0),'')) AS "1996", 
MAX(DECODE(YEARDATA,1997,NVL(DATA1,0),'')) AS "1997", 
MAX(DECODE(YEARDATA,1998,NVL(DATA1,0),'')) AS "1998", 
MAX(DECODE(YEARDATA,1999,NVL(DATA1,0),'')) AS "1999", 
MAX(DECODE(YEARDATA,2000,NVL(DATA1,0),'')) AS "2000", 
MAX(DECODE(YEARDATA,2001,NVL(DATA1,0),'')) AS "2001" FROM TEST UNION ALL 
SELECT  MAX(DECODE(YEARDATA,1990,NVL(DATA2,0),'')) AS "1990", 
MAX(DECODE(YEARDATA,1991,NVL(DATA2,0),'')) AS "1991", 
MAX(DECODE(YEARDATA,1992,NVL(DATA2,0),'')) AS "1992", 
MAX(DECODE(YEARDATA,1993,NVL(DATA2,0),'')) AS "1993", 
MAX(DECODE(YEARDATA,1994,NVL(DATA2,0),'')) AS "1994", 
MAX(DECODE(YEARDATA,1995,NVL(DATA2,0),'')) AS "1995", 
MAX(DECODE(YEARDATA,1996,NVL(DATA2,0),'')) AS "1996", 
MAX(DECODE(YEARDATA,1997,NVL(DATA2,0),'')) AS "1997", 
MAX(DECODE(YEARDATA,1998,NVL(DATA2,0),'')) AS "1998", 
MAX(DECODE(YEARDATA,1999,NVL(DATA2,0),'')) AS "1999", 
MAX(DECODE(YEARDATA,2000,NVL(DATA2,0),'')) AS "2000", 
MAX(DECODE(YEARDATA,2001,NVL(DATA2,0),'')) AS "2001" FROM TEST;

1990   1991   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001
------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------
100000 100001 100002 100003 100004 100005 100006 100007 100008 100009 100010 100011 
999999 999998 999997 999996 999995 999994 999993 999992 999991 999990 999989 999988 

2 rows selected.

 

 

 

관련 대상