Oracle 分析函数的使用(zt)

2/11/2005 09:43 FPOracle 分析函数使用介绍




分析函数是oracle816引入的一个全新的概念,为我们分析数据提供了一种简单高效的处理方式.在分析函数出现以前,我们必须使用自联查询,子查询或者内联视图,甚至复杂的存储过程实现的语句,现在只要一条简单的sql语句就可以实现了,而且在执行效率方面也有相当大的提高.下面我将针对分析函数做一些具体的说明.基础数据

除本文内容外，你还可参考：
ROLLUP与CUBE http://xsb.itpub.net/post/419/29159
分析函数参考手册：http://xsb.itpub.net/post/419/33028

今天我主要给大家介绍一下以下几个函数的使用方法
1. 自动汇总函数rollup,cube,
2. rank 函数, rank,dense_rank,row_number
3. lag,lead函数
4. sum,avg,的移动增加,移动平均数
5. ratio_to_report报表处理函数
6. first,last取基数的分析函数

Code: 
06:34:23 SQL> select * from t;

BILL_MONTH AREA_CODE NET_TYPE LOCAL_FARE
--------------- ---------- ---------- --------------
200405 5761 G 7393344.04
200405 5761 J 5667089.85
200405 5762 G 6315075.96
200405 5762 J 6328716.15
200405 5763 G 8861742.59
200405 5763 J 7788036.32
200405 5764 G 6028670.45
200405 5764 J 6459121.49
200405 5765 G 13156065.77
200405 5765 J 11901671.70
200406 5761 G 7614587.96
200406 5761 J 5704343.05
200406 5762 G 6556992.60
200406 5762 J 6238068.05
200406 5763 G 9130055.46
200406 5763 J 7990460.25
200406 5764 G 6387706.01
200406 5764 J 6907481.66
200406 5765 G 13562968.81
200406 5765 J 12495492.50
200407 5761 G 7987050.65
200407 5761 J 5723215.28
200407 5762 G 6833096.68
200407 5762 J 6391201.44
200407 5763 G 9410815.91
200407 5763 J 8076677.41
200407 5764 G 6456433.23
200407 5764 J 6987660.53
200407 5765 G 14000101.20
200407 5765 J 12301780.20
200408 5761 G 8085170.84
200408 5761 J 6050611.37
200408 5762 G 6854584.22
200408 5762 J 6521884.50
200408 5763 G 9468707.65
200408 5763 J 8460049.43
200408 5764 G 6587559.23

BILL_MONTH AREA_CODE NET_TYPE LOCAL_FARE
--------------- ---------- ---------- --------------
200408 5764 J 7342135.86
200408 5765 G 14450586.63
200408 5765 J 12680052.38

40 rows selected.

Elapsed: 00:00:00.00




1. 使用rollup函数的介绍
Quote:
下面是直接使用普通sql语句求出各地区的汇总数据的例子
06:41:36 SQL> set autot on
06:43:36 SQL> select area_code,sum(local_fare) local_fare
06:43:50 2 from t
06:43:51 3 group by area_code
06:43:57 4 union all
06:44:00 5 select '合计' area_code,sum(local_fare) local_fare
06:44:06 6 from t
06:44:08 7 /

AREA_CODE LOCAL_FARE
---------- --------------
5761 54225413.04
5762 52039619.60
5763 69186545.02
5764 53156768.46
5765 104548719.19
合计 333157065.31

6 rows selected.

Elapsed: 00:00:00.03

Execution Plan
----------------------------------------------------------
0 SELECT STATEMENT Optimizer=ALL_ROWS (Cost=7 Card=1310 Bytes=
24884)

1 0 UNION-ALL
2 1 SORT (GROUP BY) (Cost=5 Card=1309 Bytes=24871)
3 2 TABLE ACCESS (FULL) OF 'T' (Cost=2 Card=1309 Bytes=248
71)

4 1 SORT (AGGREGATE)
5 4 TABLE ACCESS (FULL) OF 'T' (Cost=2 Card=1309 Bytes=170
17)





Statistics
----------------------------------------------------------
0 recursive calls
0 db block gets
6 consistent gets
0 physical reads
0 redo size
561 bytes sent via SQL*Net to client
503 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
6 rows processed


下面是使用分析函数rollup得出的汇总数据的例子
06:44:09 SQL> select nvl(area_code,'合计') area_code,sum(local_fare) local_fare
06:45:26 2 from t
06:45:30 3 group by rollup(nvl(area_code,'合计'))
06:45:50 4 /

AREA_CODE LOCAL_FARE
---------- --------------
5761 54225413.04
5762 52039619.60
5763 69186545.02
5764 53156768.46
5765 104548719.19
333157065.31

6 rows selected.

Elapsed: 00:00:00.00

Execution Plan
----------------------------------------------------------
0 SELECT STATEMENT Optimizer=ALL_ROWS (Cost=5 Card=1309 Bytes=
24871)

1 0 SORT (GROUP BY ROLLUP) (Cost=5 Card=1309 Bytes=24871)
2 1 TABLE ACCESS (FULL) OF 'T' (Cost=2 Card=1309 Bytes=24871
)





Statistics
----------------------------------------------------------
0 recursive calls
0 db block gets
4 consistent gets
0 physical reads
0 redo size
557 bytes sent via SQL*Net to client
503 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
6 rows processed


从上面的例子我们不难看出使用rollup函数,系统的sql语句更加简单,耗用的资源更少,从6个consistent gets降到4个consistent gets,如果基表很大的话,结果就可想而知了.




1. 使用cube函数的介绍
Quote:
为了介绍cube函数我们再来看看另外一个使用rollup的例子
06:53:00 SQL> select area_code,bill_month,sum(local_fare) local_fare
06:53:37 2 from t
06:53:38 3 group by rollup(area_code,bill_month)
06:53:49 4 /

AREA_CODE BILL_MONTH LOCAL_FARE
---------- --------------- --------------
5761 200405 13060433.89
5761 200406 13318931.01
5761 200407 13710265.93
5761 200408 14135782.21
5761 54225413.04
5762 200405 12643792.11
5762 200406 12795060.65
5762 200407 13224298.12
5762 200408 13376468.72
5762 52039619.60
5763 200405 16649778.91
5763 200406 17120515.71
5763 200407 17487493.32
5763 200408 17928757.08
5763 69186545.02
5764 200405 12487791.94
5764 200406 13295187.67
5764 200407 13444093.76
5764 200408 13929695.09
5764 53156768.46
5765 200405 25057737.47
5765 200406 26058461.31
5765 200407 26301881.40
5765 200408 27130639.01
5765 104548719.19
333157065.31

26 rows selected.

Elapsed: 00:00:00.00系统只是根据rollup的第一个参数area_code对结果集的数据做了汇总处理,而没有对bill_month做汇总分析处理,cube函数就是为了这个而设计的.
下面,让我们看看使用cube函数的结果

06:58:02 SQL> select area_code,bill_month,sum(local_fare) local_fare
06:58:30 2 from t
06:58:32 3 group by cube(area_code,bill_month)
06:58:42 4 order by area_code,bill_month nulls last
06:58:57 5 /

AREA_CODE BILL_MONTH LOCAL_FARE
---------- --------------- --------------
5761 200405 13060.43
5761 200406 13318.93
5761 200407 13710.27
5761 200408 14135.78
5761 54225.41
5762 200405 12643.79
5762 200406 12795.06
5762 200407 13224.30
5762 200408 13376.47
5762 52039.62
5763 200405 16649.78
5763 200406 17120.52
5763 200407 17487.49
5763 200408 17928.76
5763 69186.54
5764 200405 12487.79
5764 200406 13295.19
5764 200407 13444.09
5764 200408 13929.69
5764 53156.77
5765 200405 25057.74
5765 200406 26058.46
5765 200407 26301.88
5765 200408 27130.64
5765 104548.72200405 79899.53
200406 82588.15
200407 84168.03
200408 86501.34
333157.05

30 rows selected.

Elapsed: 00:00:00.01

可以看到,在cube函数的输出结果比使用rollup多出了几行统计数据.这就是cube函数根据bill_month做的汇总统计结果


1 rollup 和 cube函数的再深入
Quote:
从上面的结果中我们很容易发现,每个统计数据所对应的行都会出现null,
我们如何来区分到底是根据那个字段做的汇总呢,
这时候,oracle的grouping函数就粉墨登场了.
如果当前的汇总记录是利用该字段得出的,grouping函数就会返回1,否则返回0

1 select decode(grouping(area_code),1,'all area',to_char(area_code)) area_code,
2 decode(grouping(bill_month),1,'all month',bill_month) bill_month,
3 sum(local_fare) local_fare
4 from t
5 group by cube(area_code,bill_month)
6* order by area_code,bill_month nulls last
07:07:29 SQL> /

AREA_CODE BILL_MONTH LOCAL_FARE
---------- --------------- --------------
5761 200405 13060.43
5761 200406 13318.93
5761 200407 13710.27
5761 200408 14135.78
5761 all month 54225.41
5762 200405 12643.79
5762 200406 12795.06
5762 200407 13224.30
5762 200408 13376.47
5762 all month 52039.62
5763 200405 16649.78
5763 200406 17120.52
5763 200407 17487.49
5763 200408 17928.76
5763 all month 69186.54
5764 200405 12487.79
5764 200406 13295.19
5764 200407 13444.09
5764 200408 13929.69
5764 all month 53156.77
5765 200405 25057.74
5765 200406 26058.46
5765 200407 26301.88
5765 200408 27130.64
5765 all month 104548.72all area 200405 79899.53
all area 200406 82588.15
all area 200407 84168.03
all area 200408 86501.34
all area all month 333157.05
30 rows selected.

Elapsed: 00:00:00.01
07:07:31 SQL>
可以看到,所有的空值现在都根据grouping函数做出了很好的区分,这样利用rollup,cube和grouping函数,我们做数据统计的时候就可以轻松很多了.





2. rank函数的介绍介绍完rollup和cube函数的使用,下面我们来看看rank系列函数的使用方法.

问题2.我想查出这几个月份中各个地区的总话费的排名.
Quote:
为了将rank,dense_rank,row_number函数的差别显示出来,我们对已有的基础数据做一些修改,将5763的数据改成与5761的数据相同.
1 update t t1 set local_fare = (
2 select local_fare from t t2
3 where t1.bill_month = t2.bill_month
4 and t1.net_type = t2.net_type
5 and t2.area_code = '5761'
6* ) where area_code = '5763'
07:19:18 SQL> /

8 rows updated.

Elapsed: 00:00:00.01

我们先使用rank函数来计算各个地区的话费排名.
07:34:19 SQL> select area_code,sum(local_fare) local_fare,
07:35:25 2 rank() over (order by sum(local_fare) desc) fare_rank
07:35:44 3 from t
07:35:45 4 group by area_codee
07:35:50 5
07:35:52 SQL> select area_code,sum(local_fare) local_fare,
07:36:02 2 rank() over (order by sum(local_fare) desc) fare_rank
07:36:20 3 from t
07:36:21 4 group by area_code
07:36:25 5 /

AREA_CODE LOCAL_FARE FARE_RANK
---------- -------------- ----------
5765 104548.72 1
5761 54225.41 2
5763 54225.41 2
5764 53156.77 4
5762 52039.62 5

Elapsed: 00:00:00.01我们可以看到红色标注的地方出现了,跳位,排名3没有出现
下面我们再看看dense_rank查询的结果.

07:36:26 SQL> select area_code,sum(local_fare) local_fare,
07:39:16 2 dense_rank() over (order by sum(local_fare) desc ) fare_rank
07:39:39 3 from t
07:39:42 4 group by area_code
07:39:46 5 /

AREA_CODE LOCAL_FARE FARE_RANK
---------- -------------- ----------
5765 104548.72 1
5761 54225.41 2
5763 54225.41 2
5764 53156.77 3 这是这里出现了第三名
5762 52039.62 4

Elapsed: 00:00:00.00
在这个例子中,出现了一个第三名,这就是rank和dense_rank的差别,
rank如果出现两个相同的数据,那么后面的数据就会直接跳过这个排名,而dense_rank则不会,
差别更大的是,row_number哪怕是两个数据完全相同,排名也会不一样,这个特性在我们想找出对应没个条件的唯一记录的时候又很大用处

1 select area_code,sum(local_fare) local_fare,
2 row_number() over (order by sum(local_fare) desc ) fare_rank
3 from t
4* group by area_code
07:44:50 SQL> /

AREA_CODE LOCAL_FARE FARE_RANK
---------- -------------- ----------
5765 104548.72 1
5761 54225.41 2
5763 54225.41 3
5764 53156.77 4
5762 52039.62 5在row_nubmer函数中,我们发现,哪怕sum(local_fare)完全相同,我们还是得到了不一样排名,我们可以利用这个特性剔除数据库中的重复记录.

这个帖子中的几个例子是为了说明这三个函数的基本用法的. 下个帖子我们将详细介绍他们的一些用法.


2. rank函数的介绍a. 取出数据库中最后入网的n个用户
select user_id,tele_num,user_name,user_status,create_date
from (
select user_id,tele_num,user_name,user_status,create_date,
rank() over (order by create_date desc) add_rank
from user_info
)
where add_rank <= :n;

b.根据object_name删除数据库中的重复记录
create table t as select obj#,name from sys.obj$;
再insert into t1 select * from t1 数次.
delete from t1 where rowid in (
select row_id from (
select rowid row_id,row_number() over (partition by obj# order by rowid ) rn
) where rn <> 1
);

c. 取出各地区的话费收入在各个月份排名.
SQL> select bill_month,area_code,sum(local_fare) local_fare,
2 rank() over (partition by bill_month order by sum(local_fare) desc) area_rank
3 from t
4 group by bill_month,area_code
5 /

BILL_MONTH AREA_CODE LOCAL_FARE AREA_RANK
--------------- --------------- -------------- ----------
200405 5765 25057.74 1
200405 5761 13060.43 2
200405 5763 13060.43 2
200405 5762 12643.79 4
200405 5764 12487.79 5
200406 5765 26058.46 1
200406 5761 13318.93 2
200406 5763 13318.93 2
200406 5764 13295.19 4
200406 5762 12795.06 5
200407 5765 26301.88 1
200407 5761 13710.27 2
200407 5763 13710.27 2
200407 5764 13444.09 4
200407 5762 13224.30 5
200408 5765 27130.64 1
200408 5761 14135.78 2
200408 5763 14135.78 2
200408 5764 13929.69 4
200408 5762 13376.47 5

20 rows selected.
SQL>3. lag和lead函数介绍取出每个月的上个月和下个月的话费总额
1 select area_code,bill_month, local_fare cur_local_fare,
2 lag(local_fare,2,0) over (partition by area_code order by bill_month ) pre_local_fare,
3 lag(local_fare,1,0) over (partition by area_code order by bill_month ) last_local_fare,
4 lead(local_fare,1,0) over (partition by area_code order by bill_month ) next_local_fare,
5 lead(local_fare,2,0) over (partition by area_code order by bill_month ) post_local_fare
6 from (
7 select area_code,bill_month,sum(local_fare) local_fare
8 from t
9 group by area_code,bill_month
10* )
SQL> /
AREA_CODE BILL_MONTH CUR_LOCAL_FARE PRE_LOCAL_FARE LAST_LOCAL_FARE NEXT_LOCAL_FARE POST_LOCAL_FARE
--------- ---------- -------------- -------------- --------------- --------------- ---------------
5761 200405 13060.433 0 0 13318.93 13710.265
5761 200406 13318.93 0 13060.433 13710.265 14135.781
5761 200407 13710.265 13060.433 13318.93 14135.781 0
5761 200408 14135.781 13318.93 13710.265 0 0
5762 200405 12643.791 0 0 12795.06 13224.297
5762 200406 12795.06 0 12643.791 13224.297 13376.468
5762 200407 13224.297 12643.791 12795.06 13376.468 0
5762 200408 13376.468 12795.06 13224.297 0 0
5763 200405 13060.433 0 0 13318.93 13710.265
5763 200406 13318.93 0 13060.433 13710.265 14135.781
5763 200407 13710.265 13060.433 13318.93 14135.781 0
5763 200408 14135.781 13318.93 13710.265 0 0
5764 200405 12487.791 0 0 13295.187 13444.093
5764 200406 13295.187 0 12487.791 13444.093 13929.694
5764 200407 13444.093 12487.791 13295.187 13929.694 0
5764 200408 13929.694 13295.187 13444.093 0 0
5765 200405 25057.736 0 0 26058.46 26301.881
5765 200406 26058.46 0 25057.736 26301.881 27130.638
5765 200407 26301.881 25057.736 26058.46 27130.638 0
5765 200408 27130.638 26058.46 26301.881 0 0
20 rows selected.

利用lag和lead函数,我们可以在同一行中显示前n行的数据,也可以显示后n行的数据.4. sum,avg,max,min移动计算数据介绍计算出各个连续3个月的通话费用的平均数
1 select area_code,bill_month, local_fare,
2 sum(local_fare)
3 over ( partition by area_code
4 order by to_number(bill_month)
5 range between 1 preceding and 1 following ) "3month_sum",
6 avg(local_fare)
7 over ( partition by area_code
8 order by to_number(bill_month)
9 range between 1 preceding and 1 following ) "3month_avg",
10 max(local_fare)
11 over ( partition by area_code
12 order by to_number(bill_month)
13 range between 1 preceding and 1 following ) "3month_max",
14 min(local_fare)
15 over ( partition by area_code
16 order by to_number(bill_month)
17 range between 1 preceding and 1 following ) "3month_min"
18 from (
19 select area_code,bill_month,sum(local_fare) local_fare
20 from t
21 group by area_code,bill_month
22* )
SQL> /

AREA_CODE BILL_MONTH LOCAL_FARE 3month_sum 3month_avg 3month_max 3month_min
--------- ---------- ---------------- ---------- ---------- ---------- ----------
5761 200405 13060.433 26379.363 13189.6815 13318.93 13060.433
5761 200406 13318.930 40089.628 13363.2093 13710.265 13060.433
5761 200407 13710.265 41164.976 13721.6587 14135.781 13318.9340089.628 = 13060.433 + 13318.930 + 13710.265
13363.2093 = (13060.433 + 13318.930 + 13710.265) / 3
13710.265 = max(13060.433 + 13318.930 + 13710.265)
13060.433 = min(13060.433 + 13318.930 + 13710.265)
5761 200408 14135.781 27846.046 13923.023 14135.781 13710.265
5762 200405 12643.791 25438.851 12719.4255 12795.06 12643.791
5762 200406 12795.060 38663.148 12887.716 13224.297 12643.791
5762 200407 13224.297 39395.825 13131.9417 13376.468 12795.06
5762 200408 13376.468 26600.765 13300.3825 13376.468 13224.297
5763 200405 13060.433 26379.363 13189.6815 13318.93 13060.433
5763 200406 13318.930 40089.628 13363.2093 13710.265 13060.433
5763 200407 13710.265 41164.976 13721.6587 14135.781 13318.93
5763 200408 14135.781 27846.046 13923.023 14135.781 13710.265
5764 200405 12487.791 25782.978 12891.489 13295.187 12487.791
5764 200406 13295.187 39227.071 13075.6903 13444.093 12487.791
5764 200407 13444.093 40668.974 13556.3247 13929.694 13295.187
5764 200408 13929.694 27373.787 13686.8935 13929.694 13444.093
5765 200405 25057.736 51116.196 25558.098 26058.46 25057.736
5765 200406 26058.460 77418.077 25806.0257 26301.881 25057.736
5765 200407 26301.881 79490.979 26496.993 27130.638 26058.46
5765 200408 27130.638 53432.519 26716.2595 27130.638 26301.881

20 rows selected.5. ratio_to_report函数的介绍
Quote:
1 select bill_month,area_code,sum(local_fare) local_fare,
2 ratio_to_report(sum(local_fare)) over
3 ( partition by bill_month ) area_pct
4 from t
5* group by bill_month,area_code
SQL> break on bill_month skip 1
SQL> compute sum of local_fare on bill_month
SQL> compute sum of area_pct on bill_month
SQL> /

BILL_MONTH AREA_CODE LOCAL_FARE AREA_PCT
---------- --------- ---------------- ----------
200405 5761 13060.433 .171149279
5762 12643.791 .165689431
5763 13060.433 .171149279
5764 12487.791 .163645143
5765 25057.736 .328366866
********** ---------------- ----------
sum 76310.184 1

200406 5761 13318.930 .169050772
5762 12795.060 .162401542
5763 13318.930 .169050772
5764 13295.187 .168749414
5765 26058.460 .330747499
********** ---------------- ----------
sum 78786.567 1

200407 5761 13710.265 .170545197
5762 13224.297 .164500127
5763 13710.265 .170545197
5764 13444.093 .167234221
5765 26301.881 .327175257
********** ---------------- ----------
sum 80390.801 1

200408 5761 14135.781 .170911147
5762 13376.468 .161730539
5763 14135.781 .170911147
5764 13929.694 .168419416
5765 27130.638 .328027751
********** ---------------- ----------
sum 82708.362 1


20 rows selected.



6 first,last函数使用介绍
Quote:
取出每月通话费最高和最低的两个用户.
1 select bill_month,area_code,sum(local_fare) local_fare,
2 first_value(area_code)
3 over (order by sum(local_fare) desc
4 rows unbounded preceding) firstval,
5 first_value(area_code)
6 over (order by sum(local_fare) asc
7 rows unbounded preceding) lastval
8 from t
9 group by bill_month,area_code
10* order by bill_month
SQL> /

BILL_MONTH AREA_CODE LOCAL_FARE FIRSTVAL LASTVAL
---------- --------- ---------------- --------------- ---------------
200405 5764 12487.791 5765 5764
200405 5762 12643.791 5765 5764
200405 5761 13060.433 5765 5764
200405 5765 25057.736 5765 5764
200405 5763 13060.433 5765 5764
200406 5762 12795.060 5765 5764
200406 5763 13318.930 5765 5764
200406 5764 13295.187 5765 5764
200406 5765 26058.460 5765 5764
200406 5761 13318.930 5765 5764
200407 5762 13224.297 5765 5764
200407 5765 26301.881 5765 5764
200407 5761 13710.265 5765 5764
200407 5763 13710.265 5765 5764
200407 5764 13444.093 5765 5764
200408 5762 13376.468 5765 5764
200408 5764 13929.694 5765 5764
200408 5761 14135.781 5765 5764
200408 5765 27130.638 5765 5764
200408 5763 14135.781 5765 5764

20 rows selected.


ref: http://sunsonbaby.cnblogs.com/archive/2004/09/29/47876.aspx