Oracle PostgreSQL (ROW_NUMBER() OVER)分析函数

浅浅的花香味﹌ 2022-08-23 11:59 237阅读 0赞

**一、定义**

语法：ROW\_NUMBER() OVER(PARTITION BY COLUMN ORDER BY COLUMN)

简单的说row\_number()从1开始，为每一条分组记录返回一个数字，这里的ROW\_NUMBER() OVER (ORDER BY xlh DESC) 是先把xlh列降序，再为降序以后的每条xlh记录返回一个序号。   
示例：   
xlh           row\_num   
1700              1   
1500              2   
1085              3   
710                4

**二、语法**

**row\_number() OVER (PARTITION BY COL1 ORDER BY COL2)** 表示根据COL1分组，在分组内部根据 COL2排序，而此函数计算的值就表示每组内部排序后的顺序编号（组内连续的唯一的)

**三、例子**

初始化数据

create table employee (empid int ,deptid int ,salary decimal(10,2))  
insert into employee values(1,10,5500.00)  
insert into employee values(2,10,4500.00)  
insert into employee values(3,20,1900.00)  
insert into employee values(4,20,4800.00)  
insert into employee values(5,40,6500.00)  
insert into employee values(6,40,14500.00)  
insert into employee values(7,40,44500.00)  
insert into employee values(8,50,6500.00)  
insert into employee values(9,50,7500.00)

数据显示为

empid       deptid      salary  
\----------- ----------- ---------------------------------------  
1           10          5500.00  
2           10          4500.00  
3           20          1900.00  
4           20          4800.00  
5           40          6500.00  
6           40          14500.00  
7           40          44500.00  
8           50          6500.00  
9           50          7500.00

需求：根据部门分组，显示每个部门的工资等级

预期结果：

empid       deptid      salary                                  rank  
\----------- ----------- --------------------------------------- --------------------  
1           10          5500.00                                 1  
2           10          4500.00                                 2  
4           20          4800.00                                 1  
3           20          1900.00                                 2  
7           40          44500.00                               1  
6           40          14500.00                               2  
5           40          6500.00                                 3  
9           50          7500.00                                 1  
8           50          6500.00                                 2

SQL脚本：

SELECT \*, Row\_Number() OVER (partition by deptid ORDER BY salary desc) rank FROM employee

**四、Row\_Number() OVER 去重详解**  
 SELECT \*  
          FROM (SELECT K.USRID,  
                ROW\_NUMBER() OVER(PARTITION BY USRID ORDER BY RECVTIME DESC) ROW\_NUM  
                  FROM STAGE.TS\_A\_TRADECOMPFEE K  
                 )  
         WHERE ROW\_NUM = 1;

如果这张表中有两条除RECVTIME外完全一样的记录，而现在我们只需要一条，那么就可以用这种方法去重，通过ROW\_NUMBER() OVER函数通过RECVTIME排序，下边令ROW\_NUM = 1就可以找到一条记录了，这样可以去冗余。

**注意：在使用over等开窗函数时，over里头的分组及排序的执行晚于“where，group by，order by”的执行。**

如下代码：

1.  select   
2.  ROW\_NUMBER() over(partition by customerID  order by insDT) as rows,  
3.  customerID,totalPrice, DID  
4.  from OP\_Order where insDT**>**'2011-07-22'

以上代码是先执行where子句，执行完后，再给每一条记录进行编号。