1.rownum和rowid都是伪列,但两者的根本是不同的,rownum是根据sql查询出来的结果给每行分配一个
逻辑编号,不同的sql也就会导致rownum不同,但是rowid是物理结构上的,在每条记录Insert到数据库时,
就会有一个唯一的物理记录。rowid可以说是屋里存在的,表示记录在表空间中的一个唯一位置ID,在DB
中唯一。只要记录没被搬动过,rowid是不变的。rowid相对于表来说相当于一般列。
rownum常用语分页查询,rowid则常用语去除重复记录。
2.rownum是对结果集加的一个伪列,即先查到结果集之后再加上去的一个列(先有结果集)。简单的说,rownum
是对符合条件结果的序列号。总是从1开始排起,所以选出的结果中不能没有1而有其他大于1的值。
也可以理解为rownum是oracle数据库从数据文件或缓冲区中读取数据的顺序。它取得第一条记录则rownum为1,
第2条为2。如果使用>,>=,=,between...and这些条件,因为从缓冲区或数据文件中得到的第一条记录的rownum
为1,则被删除,接着取下条,可是rownum依旧为1,有被删除,便没有了数据。
case1:
SQL> select * from emp where rownum between 2 and 5; EMPNO ENAME JOB MGR HIREDATE SAL COMM DEPTNO ----- ---------- --------- ----- ----------- --------- --------- ------
case2:
SQL> select * from emp where rownum<>3; EMPNO ENAME JOB MGR HIREDATE SAL COMM DEPTNO ----- ---------- --------- ----- ----------- --------- --------- ------ 7369 SMITH CLERK 7902 1980/12/17 800.00 20 7499 ALLEN SALESMAN 7698 1981/2/20 1600.00 300.00 30
case3:
SQL> select * from emp where rownum>0; EMPNO ENAME JOB MGR HIREDATE SAL COMM DEPTNO ----- ---------- --------- ----- ----------- --------- --------- ------ 7369 SMITH CLERK 7902 1980/12/17 800.00 20 7499 ALLEN SALESMAN 7698 1981/2/20 1600.00 300.00 30 7521 WARD SALESMAN 7698 1981/2/22 1250.00 500.00 30 7566 JONES MANAGER 7839 1981/4/2 2975.00 20 7654 MARTIN SALESMAN 7698 1981/9/28 1250.00 1400.00 30 7698 BLAKE MANAGER 7839 1981/5/1 2850.00 30 7782 CLARK MANAGER 7839 1981/6/9 2450.00 10 7788 SCOTT ANALYST 7566 1987/4/19 3000.00 20 7839 KING PRESIDENT 1981/11/17 5000.00 10 7844 TURNER SALESMAN 7698 1981/9/8 1500.00 0.00 30 7876 ADAMS CLERK 7788 1987/5/23 1100.00 20 7900 JAMES CLERK 7698 1981/12/3 950.00 30 7902 FORD ANALYST 7566 1981/12/3 3000.00 20 7934 MILLER CLERK 7782 1982/1/23 1300.00 10 14 rows selected
case4:
SQL> select * from emp where rownum between 1 and 3; EMPNO ENAME JOB MGR HIREDATE SAL COMM DEPTNO ----- ---------- --------- ----- ----------- --------- --------- ------ 7369 SMITH CLERK 7902 1980/12/17 800.00 20 7499 ALLEN SALESMAN 7698 1981/2/20 1600.00 300.00 30 7521 WARD SALESMAN 7698 1981/2/22 1250.00 500.00 30
3.rownum和row_numbe() over:
使用rownum进行排序的时候是先对结果集加入伪列rownum然后再进行排序,而row_number()在包含
排序从句后是先排序再计算行号码
case:
order by:
SQL> select a.*,rownum from emp a order by a.ename; EMPNO ENAME JOB MGR HIREDATE SAL COMM DEPTNO ROWNUM ----- ---------- --------- ----- ----------- --------- --------- ------ ---------- 7876 ADAMS CLERK 7788 1987/5/23 1100.00 20 11 7499 ALLEN SALESMAN 7698 1981/2/20 1600.00 300.00 30 2 7698 BLAKE MANAGER 7839 1981/5/1 2850.00 30 6 7782 CLARK MANAGER 7839 1981/6/9 2450.00 10 7 7902 FORD ANALYST 7566 1981/12/3 3000.00 20 13 7900 JAMES CLERK 7698 1981/12/3 950.00 30 12 7566 JONES MANAGER 7839 1981/4/2 2975.00 20 4 7839 KING PRESIDENT 1981/11/17 5000.00 10 9 7654 MARTIN SALESMAN 7698 1981/9/28 1250.00 1400.00 30 5 7934 MILLER CLERK 7782 1982/1/23 1300.00 10 14 7788 SCOTT ANALYST 7566 1987/4/19 3000.00 20 8 7369 SMITH CLERK 7902 1980/12/17 800.00 20 1 7844 TURNER SALESMAN 7698 1981/9/8 1500.00 0.00 30 10 7521 WARD SALESMAN 7698 1981/2/22 1250.00 500.00 30 3 14 rows selected
row_number() over(order by )
SQL> select a.*,row_number() over(order by a.ename) from emp a; EMPNO ENAME JOB MGR HIREDATE SAL COMM DEPTNO ROW_NUMBER()OVER(ORDERBYA.ENAM ----- ---------- --------- ----- ----------- --------- --------- ------ ------------------------------ 7876 ADAMS CLERK 7788 1987/5/23 1100.00 20 1 7499 ALLEN SALESMAN 7698 1981/2/20 1600.00 300.00 30 2 7698 BLAKE MANAGER 7839 1981/5/1 2850.00 30 3 7782 CLARK MANAGER 7839 1981/6/9 2450.00 10 4 7902 FORD ANALYST 7566 1981/12/3 3000.00 20 5 7900 JAMES CLERK 7698 1981/12/3 950.00 30 6 7566 JONES MANAGER 7839 1981/4/2 2975.00 20 7 7839 KING PRESIDENT 1981/11/17 5000.00 10 8 7654 MARTIN SALESMAN 7698 1981/9/28 1250.00 1400.00 30 9 7934 MILLER CLERK 7782 1982/1/23 1300.00 10 10 7788 SCOTT ANALYST 7566 1987/4/19 3000.00 20 11 7369 SMITH CLERK 7902 1980/12/17 800.00 20 12 7844 TURNER SALESMAN 7698 1981/9/8 1500.00 0.00 30 13 7521 WARD SALESMAN 7698 1981/2/22 1250.00 500.00 30 14 14 rows selected
4.
case1:在最外层控制分页的最小值和最大值
SELECT * FROM (SELECT A.*, ROWNUM RN FROM GAS_CA.A_CARD_METER_PAY A WHERE A.PURCH_GAS_TIME BETWEEN TO_DATE('2015/1/1', 'YYYY/MM/DD') AND TO_DATE('2016/1/1', 'YYYY/MM/DD') AND A.ORG_NO LIKE '1009%') WHERE RN BETWEEN 10 AND 20;
case2:
SELECT * FROM (SELECT ROWNUM RN, A.* FROM GAS_CA.A_CARD_METER_PAY A WHERE A.PURCH_GAS_TIME BETWEEN TO_DATE('2015/1/1', 'YYYY/MM/DD') AND TO_DATE('2016/1/1', 'YYYY/MM/DD') AND A.ORG_NO LIKE '1009%' AND ROWNUM <= 20) WHERE RN >= 10;
对于case1和case2,绝大多数情况下,case2要比第一个查询效率高得多。
因为在CBO优化模式下,oracle可以将外层的查询条件推到内层查询中,以提高内层查询的执行效率。对于第二份语句,
第二层的查询条件where rownum<=40就可以被oracle推入到内层查询中,这样oracle查询的结果一旦查过rownum限制的条件,
就终止查询将结果返回。而第一个语句,oracle无法将第三层的条件推到最内层(即使推到最内层也没有意义,因为不知道RN代表什么)
。
case3:有order by 的写法(较高效率)
SELECT T1.*, RN FROM (SELECT ROWNUM RN, T.* FROM (SELECT * FROM GAS_CA.A_CARD_METER_PAY A WHERE A.PURCH_GAS_TIME BETWEEN TO_DATE('2015/1/1', 'YYYY/MM/DD') AND TO_DATE('2016/1/1', 'YYYY/MM/DD') AND A.ORG_NO LIKE '1009%' ORDER BY A.PURCH_GAS_TIME) T WHERE ROWNUM <= 20) T1 WHERE RN >= 10;
case4:将查询条件放置最外层
SELECT * FROM (SELECT T.*, ROWNUM RN FROM (SELECT * FROM GAS_CA.A_CARD_METER_PAY A WHERE A.PURCH_GAS_TIME BETWEEN TO_DATE('2015/1/1', 'YYYY/MM/DD') AND TO_DATE('2016/1/1', 'YYYY/MM/DD') AND A.ORG_NO LIKE '1009%' ORDER BY A.PURCH_GAS_TIME) T) T1 WHERE T1.RN BETWEEN 10 AND 20;