僅做記錄，原文：https://blog.csdn.net/scgaliguodong123_/article/details/60135385
視窗函數與分析函數
應用場景：
（1）用于分區排序
（2）動态Group By
（3）Top N
（4）累計計算
（5）層次查詢
視窗函數
FIRST_VALUE：取分組内排序後，截止到目前行，第一個值
LAST_VALUE： 取分組内排序後，截止到目前行，最後一個值
LEAD(col,n,DEFAULT) ：用于統計視窗内往下第n行值。第一個參數為列名，第二個參數為往下第n行（可選，預設為1），第三個參數為預設值（當往下第n行為NULL時候，取預設值，如不指定，則為NULL）
LAG(col,n,DEFAULT) ：與lead相反，用于統計視窗内往上第n行值。第一個參數為列名，第二個參數為往上第n行（可選，預設為1），第三個參數為預設值（當往上第n行為NULL時候，取預設值，如不指定，則為NULL）
OVER從句
1、使用标準的聚合函數COUNT、SUM、MIN、MAX、AVG
2、使用PARTITION BY語句，使用一個或者多個原始資料類型的列
3、使用PARTITION BY與ORDER BY語句，使用一個或者多個資料類型的分區或者排序列
4、使用視窗規範，視窗規範支援以下格式：
---------------------
?
1
2
3
(
ROWS
| RANGE)
BETWEEN
(UNBOUNDED | [num]) PRECEDING
AND
([num] PRECEDING |
CURRENT
ROW | (UNBOUNDED | [num]) FOLLOWING)
(
ROWS
| RANGE)
BETWEEN
CURRENT
ROW
AND
(
CURRENT
ROW | (UNBOUNDED | [num]) FOLLOWING)
(
ROWS
| RANGE)
BETWEEN
[num] FOLLOWING
AND
(UNBOUNDED | [num]) FOLLOWING
當ORDER BY後面缺少視窗從句條件，視窗規範預設是 RANGE BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW.
當ORDER BY和視窗從句都缺失, 視窗規範預設是 ROW BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING.
OVER從句支援以下函數， 但是并不支援和視窗一起使用它們。
Ranking函數: Rank, NTile, DenseRank, CumeDist, PercentRank.
Lead 和 Lag 函數.
分析函數
ROW_NUMBER() 從1開始，按照順序，生成分組内記錄的序列,比如，按照pv降序排列，生成分組内每天的pv名次,ROW_NUMBER()的應用場景非常多，再比如，擷取分組内排序第一的記錄;擷取一個session中的第一條refer等。
RANK() 生成資料項在分組中的排名，排名相等會在名次中留下空位
DENSE_RANK() 生成資料項在分組中的排名，排名相等會在名次中不會留下空位
CUME_DIST 小于等于目前值的行數/分組内總行數。比如，統計小于等于目前薪水的人數，所占總人數的比例
PERCENT_RANK 分組内目前行的RANK值-1/分組内總行數-1
NTILE(n) 用于将分組資料按照順序切分成n片，傳回目前切片值，如果切片不均勻，預設增加第一個切片的分布。NTILE不支援ROWS BETWEEN，比如 NTILE(2) OVER(PARTITION BY cookieid ORDER BY createtime ROWS BETWEEN 3 PRECEDING AND CURRENT ROW)。
Hive2.1.0及以後支援Distinct
在聚合函數（SUM, COUNT and AVG）中，支援distinct，但是在ORDER BY 或者 視窗限制不支援。
COUNT
(
DISTINCT
a) OVER (PARTITION
BY
c)
Hive 2.2.0中在使用ORDER BY和視窗限制時支援distinct
COUNT
(
DISTINCT
a) OVER (PARTITION
BY
c
ORDER
BY
d
ROWS
BETWEEN
1 PRECEDING
AND
1 FOLLOWING)
Hive2.1.0及以後支援在OVER從句中支援聚合函數
SELECT
rank() OVER (
ORDER
BY
sum
(b))
FROM
T
GROUP
BY
a;
測試資料集：
視窗函數詳細用法 4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
##
COUNT
、
SUM
、
MIN
、
MAX
、
AVG
select
user_id,
user_type,
sales,
--預設為從起點到目前行
sum
(sales) OVER(PARTITION
BY
user_type
ORDER
BY
sales
asc
)
AS
sales_1,
--從起點到目前行，結果與sales_1不同。
sum
(sales) OVER(PARTITION
BY
user_type
ORDER
BY
sales
asc
ROWS
BETWEEN
UNBOUNDED PRECEDING
AND
CURRENT
ROW)
AS
sales_2,
--目前行+往前3行
sum
(sales) OVER(PARTITION
BY
user_type
ORDER
BY
sales
asc
ROWS
BETWEEN
3 PRECEDING
AND
CURRENT
ROW)
AS
sales_3,
--目前行+往前3行+往後1行
sum
(sales) OVER(PARTITION
BY
user_type
ORDER
BY
sales
asc
ROWS
BETWEEN
3 PRECEDING
AND
1 FOLLOWING)
AS
sales_4,
--目前行+往後所有行
sum
(sales) OVER(PARTITION
BY
user_type
ORDER
BY
sales
asc
ROWS
BETWEEN
CURRENT
ROW
AND
UNBOUNDED FOLLOWING)
AS
sales_5,
--分組内所有行
SUM
(sales) OVER(PARTITION
BY
user_type)
AS
sales_6
from
order_detail
order
by
user_type,
sales,
user_id
+
----------+------------+--------+----------+----------+----------+----------+----------+----------+--+
| user_id | user_type | sales | sales_1 | sales_2 | sales_3 | sales_4 | sales_5 | sales_6 |
+
----------+------------+--------+----------+----------+----------+----------+----------+----------+--+
| liiu | new | 1 | 2 | 2 | 2 | 4 | 22 | 23 |
| qibaqiu | new | 1 | 2 | 1 | 1 | 2 | 23 | 23 |
| zhangsa | new | 2 | 4 | 4 | 4 | 7 | 21 | 23 |
| wanger | new | 3 | 7 | 7 | 7 | 12 | 19 | 23 |
| lilisi | new | 5 | 17 | 17 | 15 | 21 | 11 | 23 |
| qishili | new | 5 | 17 | 12 | 11 | 16 | 16 | 23 |
| wutong | new | 6 | 23 | 23 | 19 | 19 | 6 | 23 |
| lisi | old | 1 | 1 | 1 | 1 | 3 | 6 | 6 |
| wangshi | old | 2 | 3 | 3 | 3 | 6 | 5 | 6 |
| liwei | old | 3 | 6 | 6 | 6 | 6 | 3 | 6 |
+
----------+------------+--------+----------+----------+----------+----------+----------+----------+--+
注意:
結果和
ORDER
BY
相關,預設為升序
如果不指定
ROWS
BETWEEN
,預設為從起點到目前行;
如果不指定
ORDER
BY
，則将分組内所有值累加;
關鍵是了解
ROWS
BETWEEN
含義,也叫做WINDOW子句：
PRECEDING：往前
FOLLOWING：往後
CURRENT
ROW：目前行
UNBOUNDED：無界限（起點或終點）
UNBOUNDED PRECEDING：表示從前面的起點
UNBOUNDED FOLLOWING：表示到後面的終點
其他
COUNT
、
AVG
，
MIN
，
MAX
，和
SUM
用法一樣。
## first_value與last_value
select
user_id,
user_type,
ROW_NUMBER() OVER(PARTITION
BY
user_type
ORDER
BY
sales)
AS
row_num,
first_value(user_id) over (partition
by
user_type
order
by
sales
desc
)
as
max_sales_user,
first_value(user_id) over (partition
by
user_type
order
by
sales
asc
)
as
min_sales_user,
last_value(user_id) over (partition
by
user_type
order
by
sales
desc
)
as
curr_last_min_user,
last_value(user_id) over (partition
by
user_type
order
by
sales
asc
)
as
curr_last_max_user
from
order_detail;
+
----------+------------+----------+-----------------+-----------------+---------------------+---------------------+--+
| user_id | user_type | row_num | max_sales_user | min_sales_user | curr_last_min_user | curr_last_max_user |
+
----------+------------+----------+-----------------+-----------------+---------------------+---------------------+--+
| wutong | new | 7 | wutong | qibaqiu | wutong | wutong |
| lilisi | new | 6 | wutong | qibaqiu | qishili | lilisi |
| qishili | new | 5 | wutong | qibaqiu | qishili | lilisi |
| wanger | new | 4 | wutong | qibaqiu | wanger | wanger |
| zhangsa | new | 3 | wutong | qibaqiu | zhangsa | zhangsa |
| liiu | new | 2 | wutong | qibaqiu | qibaqiu | liiu |
| qibaqiu | new | 1 | wutong | qibaqiu | qibaqiu | liiu |
| liwei | old | 3 | liwei | lisi | liwei | liwei |
| wangshi | old | 2 | liwei | lisi | wangshi | wangshi |
| lisi | old | 1 | liwei | lisi | lisi | lisi |
+
----------+------------+----------+-----------------+-----------------+---------------------+---------------------+--+
## lead與lag
select
user_id,device_id,
lead(device_id) over (
order
by
sales)
as
default_after_one_line,
lag(device_id) over (
order
by
sales)
as
default_before_one_line,
lead(device_id,2) over (
order
by
sales)
as
after_two_line,
lag(device_id,2,
'abc'
) over (
order
by
sales)
as
before_two_line
from
order_detail;
+
----------+-------------+-------------------------+--------------------------+-----------------+------------------+--+
| user_id | device_id | default_after_one_line | default_before_one_line | after_two_line | before_two_line |
+
----------+-------------+-------------------------+--------------------------+-----------------+------------------+--+
| qibaqiu | fds | fdsfagwe |
NULL
| 543gfd | abc |
| liiu | fdsfagwe | 543gfd | fds | f332 | abc |
| lisi | 543gfd | f332 | fdsfagwe | dfsadsa323 | fds |
| wangshi | f332 | dfsadsa323 | 543gfd | hfd | fdsfagwe |
| zhangsa | dfsadsa323 | hfd | f332 | 65ghf | 543gfd |
| liwei | hfd | 65ghf | dfsadsa323 | fds | f332 |
| wanger | 65ghf | fds | hfd | dsfgg | dfsadsa323 |
| qishili | fds | dsfgg | 65ghf | 543gdfsd | hfd |
| lilisi | dsfgg | 543gdfsd | fds |
NULL
| 65ghf |
| wutong | 543gdfsd |
NULL
| dsfgg |
NULL
| fds |
+
----------+-------------+-------------------------+--------------------------+-----------------+------------------+--+
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
## RANK、ROW_NUMBER、DENSE_RANK
select
user_id,user_type,sales,
RANK() over (partition
by
user_type
order
by
sales
desc
)
as
r,
ROW_NUMBER() over (partition
by
user_type
order
by
sales
desc
)
as
rn,
DENSE_RANK() over (partition
by
user_type
order
by
sales
desc
)
as
dr
from
order_detail;
+
----------+------------+--------+----+-----+-----+--+
| user_id | user_type | sales | r | rn | dr |
+
----------+------------+--------+----+-----+-----+--+
| wutong | new | 6 | 1 | 1 | 1 |
| qishili | new | 5 | 2 | 2 | 2 |
| lilisi | new | 5 | 2 | 3 | 2 |
| wanger | new | 3 | 4 | 4 | 3 |
| zhangsa | new | 2 | 5 | 5 | 4 |
| qibaqiu | new | 1 | 6 | 6 | 5 |
| liiu | new | 1 | 6 | 7 | 5 |
| liwei | old | 3 | 1 | 1 | 1 |
| wangshi | old | 2 | 2 | 2 | 2 |
| lisi | old | 1 | 3 | 3 | 3 |
+
----------+------------+--------+----+-----+-----+--+
## NTILE
select
user_type,sales,
--分組内将資料分成2片
NTILE(2) OVER(PARTITION
BY
user_type
ORDER
BY
sales)
AS
nt2,
--分組内将資料分成3片
NTILE(3) OVER(PARTITION
BY
user_type
ORDER
BY
sales)
AS
nt3,
--分組内将資料分成4片
NTILE(4) OVER(PARTITION
BY
user_type
ORDER
BY
sales)
AS
nt4,
--将所有資料分成4片
NTILE(4) OVER(
ORDER
BY
sales)
AS
all_nt4
from
order_detail
order
by
user_type,
sales
+
------------+--------+------+------+------+----------+--+
| user_type | sales | nt2 | nt3 | nt4 | all_nt4 |
+
------------+--------+------+------+------+----------+--+
| new | 1 | 1 | 1 | 1 | 1 |
| new | 1 | 1 | 1 | 1 | 1 |
| new | 2 | 1 | 1 | 2 | 2 |
| new | 3 | 1 | 2 | 2 | 3 |
| new | 5 | 2 | 2 | 3 | 4 |
| new | 5 | 2 | 3 | 3 | 3 |
| new | 6 | 2 | 3 | 4 | 4 |
| old | 1 | 1 | 1 | 1 | 1 |
| old | 2 | 1 | 2 | 2 | 2 |
| old | 3 | 2 | 3 | 3 | 2 |
+
------------+--------+------+------+------+----------+--+
求取sale前20%的使用者ID
select
user_id
from
(
select
user_id,
NTILE(5) OVER(
ORDER
BY
sales
desc
)
AS
nt
from
order_detail
)A
where
nt=1;
## CUME_DIST、PERCENT_RANK
select
user_id,user_type,sales,
--沒有partition,所有資料均為1組
CUME_DIST() OVER(
ORDER
BY
sales)
AS
cd1,
--按照user_type進行分組
CUME_DIST() OVER(PARTITION
BY
user_type
ORDER
BY
sales)
AS
cd2
from
order_detail;
+
----------+------------+--------+------+----------------------+--+
| user_id | user_type | sales | cd1 | cd2 |
+
----------+------------+--------+------+----------------------+--+
| liiu | new | 1 | 0.3 | 0.2857142857142857 |
| qibaqiu | new | 1 | 0.3 | 0.2857142857142857 |
| zhangsa | new | 2 | 0.5 | 0.42857142857142855 |
| wanger | new | 3 | 0.7 | 0.5714285714285714 |
| lilisi | new | 5 | 0.9 | 0.8571428571428571 |
| qishili | new | 5 | 0.9 | 0.8571428571428571 |
| wutong | new | 6 | 1.0 | 1.0 |
| lisi | old | 1 | 0.3 | 0.3333333333333333 |
| wangshi | old | 2 | 0.5 | 0.6666666666666666 |
| liwei | old | 3 | 0.7 | 1.0 |
+
----------+------------+--------+------+----------------------+--+
select
user_type,sales
--分組内總行數
SUM
(1) OVER(PARTITION
BY
user_type)
AS
s,
--RANK值
RANK() OVER(
ORDER
BY
sales)
AS
r,
PERCENT_RANK() OVER(
ORDER
BY
sales)
AS
pr,
--分組内
PERCENT_RANK() OVER(PARTITION
BY
user_type
ORDER
BY
sales)
AS
prg
from
order_detail;
+
----+-----+---------------------+---------------------+--+
| s | r | pr | prg |
+
----+-----+---------------------+---------------------+--+
| 7 | 1 | 0.0 | 0.0 |
| 7 | 1 | 0.0 | 0.0 |
| 7 | 4 | 0.3333333333333333 | 0.3333333333333333 |
| 7 | 6 | 0.5555555555555556 | 0.5 |
| 7 | 8 | 0.7777777777777778 | 0.6666666666666666 |
| 7 | 8 | 0.7777777777777778 | 0.6666666666666666 |
| 7 | 10 | 1.0 | 1.0 |
| 3 | 1 | 0.0 | 0.0 |
| 3 | 4 | 0.3333333333333333 | 0.5 |
| 3 | 6 | 0.5555555555555556 | 1.0 |
+
----+-----+---------------------+---------------------+--+
![SQL優化SQL語句優化的目的[圖]](data:image/gif;base64,R0lGODlhAQABAIAAAP///wAAACwAAAAAAQABAAACAkQBADs=)