oracle之分析函数解析及其应用场景

ORACLE 分析函数FIRST_VALUE,LAST_VALUE用法
sum over
avg over
first_value over
last_value over
...聚合函数结合over就是分析函数
备注LAST_VALUE一般这样操作:
【ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING / ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW】
3、列出每月的订单总额以及截至到当前月的订单总额
SQL> select month,
  2         sum(tot_sales) month_sales,
  3         sum(sum(tot_sales)) over(order by month
  4           rows between unbounded preceding and current row) current_total_sales
  5    from orders
  6   group by month;

     MONTH MONTH_SALES CURRENT_TOTAL_SALES
---------- ----------- -------------------
         1      610697              610697
         2      428676             1039373
         3      637031             1676404
         4      541146             2217550
         5      592935             2810485
         6      501485             3311970
         7      606914             3918884
         8      460520             4379404
         9      392898             4772302
        10      510117             5282419
        11      532889             5815308
        12      492458             6307766

        
SQL> select month,
  2         sum(tot_sales) month_sales,
  3         sum(sum(tot_sales)) over(order by month
  4         rows between unbounded preceding and current row) current_total_sales,
  5         sum(sum(tot_sales)) over(order by month
  6         rows between unbounded preceding and unbounded following) total_sales
  7    from orders
  8   group by month;

     MONTH MONTH_SALES CURRENT_TOTAL_SALES TOTAL_SALES
---------- ----------- ------------------- -----------
         1      610697              610697     6307766
         2      428676             1039373     6307766
         3      637031             1676404     6307766
         4      541146             2217550     6307766
         5      592935             2810485     6307766
         6      501485             3311970     6307766
         7      606914             3918884     6307766
         8      460520             4379404     6307766
         9      392898             4772302     6307766
        10      510117             5282419     6307766
        11      532889             5815308     6307766
        12      492458             6307766     6307766
4、统计当天销售额和五天内的评价销售额
 select trunc(order_dt) day,
             sum(sale_price) daily_sales,
             avg(sum(sale_price)) over (order by trunc(order_dt)
                      range between interval '2' day preceding
                                     and interval '2' day following) five_day_avg
   from cust_order
 where sale_price is not null
     and order_dt between to_date('01-jul-2001','dd-mon-yyyy')
     and to_date('31-jul-2001','dd-mon-yyyy')
为了对指定范围进行统计,Oracle使用关键字range、interval来指定一个范围。上面的例子告诉Oracle查找当前日期的前2天,后2天范围内的记录,并统计其销售平均值



5、Oracle提供了2个额外的函数:first_value、last_value,用于在窗口记录集中查找第一条记录和最后一条记录。假设我们的报表需要显示当前月、上一个月、后一个月的销售情况,以及每3个月的销售平均值,这两个函数就可以派上用场了
select month,
             first_value(sum(tot_sales)) over (order by month
                                    rows between 1 preceding and 1 following) prev_month,
 
             sum(tot_sales) monthly_sales,
 
             last_value(sum(tot_sales)) over (order by month
                                  rows between 1 preceding and 1 following) next_month,
 
             avg(sum(tot_sales)) over (order by month
                                 rows between 1 preceding and 1 following) rolling_avg
    from orders
 where year = 2001
      and region_id = 6
  group by month
 order by month;
首先我们来看:rows between 1 preceding and 1 following告诉Oracle在当前记录的前一条、后一条范围内查找并统计,而first_value和last_value在这3条记录中至分别找出第一条、第三条记录,这样我们就轻松地得到相邻三个月的销售记录及平均值了!


6、我们知道了如何利用窗口函数来显示相邻的记录,现在假如我们想每次显示当月的销售额和上个月的销售额,应该怎么做呢?

从第五点的介绍我们可以知道,利用first_value(sum(tot_sales) over (order by month rows between 1 preceding and 0 following))就可以做到了,其实Oracle还有一个更简单的方式让我们来比较2条记录,它就是lag函数。

leg函数类似于preceding和following

子句,它能够通过和当前记录的相对位置而被应用,在比较同一个相邻的记录集内两条相邻记录的时候特别有用。

select  month,            
          sum(tot_sales) monthly_sales,
          lag(sum(tot_sales), 1) over (order by month) prev_month_sales
   from orders
 where year = 2001
      and region_id = 6
  group by month
 order by month;
 
 *************************************************
 1、over函数的写法:

  over(partition by class order by sroce) 按照sroce排序进行累计,order by是个默认的开窗函数,按照class分区。

  2、开窗的窗口范围:

  over(order by sroce range between 5 preceding and 5 following):窗口范围为当前行数据幅度减5加5后的范围内的。

  over(order by sroce rows between 5 preceding and 5 following):窗口范围为当前行前后各移动5行。

  3、与over()函数结合的函数的介绍

  (1)、查询每个班的第一名的成绩:如下 

1 SELECT * FROM (select t.name,t.class,t.sroce,rank() over(partition by t.class order by t.sroce desc) mm from T2_TEMP t) where mm = 1;
 *************************************************

分析函数应用场景:https://blog.csdn.net/WuLex/article/details/82796991
原文地址:https://www.cnblogs.com/lhl-shubiao/p/10418870.html