Oracle PL/SQL Tutorial/Analytical Functions/Window Functions
Содержание
- 1 An Expanded Example of a Physical Window
- 2 Displaying which rows are used in the moving average calculations with two other analytical functions: FIRST_ VALUE and LAST_VALUE.
- 3 Moving average
- 4 Performing a Centered Average
- 5 Performing a Cumulative Sum
- 6 Performing a Moving Average
- 7 Row-ordering is done first and then the moving average
- 8 Use ROWS UNBOUNDED PRECEDING to implicitly indicate the end of the window is the current row
- 9 Use SUM for windowing
- 10 Use the COUNT aggregate analytical function to show how many rows are included in each window
- 11 Using the Window Functions
An Expanded Example of a Physical Window
To expand the moving average window, we can change the clause ROWS BETWEEN x PRECEDING AND y FOLLOWING to have different values for x and y.
<source lang="sql">
SQL> -- create demo table SQL> create table myTable(
2 id NUMBER(2), 3 value NUMBER(6,2) 4 ) 5 /
Table created. SQL> SQL> -- prepare data SQL> insert into myTable(ID, value)values (1,9); 1 row created. SQL> insert into myTable(ID, value)values (2,2.11); 1 row created. SQL> insert into myTable(ID, value)values (3,3.44); 1 row created. SQL> insert into myTable(ID, value)values (4,-4.21); 1 row created. SQL> insert into myTable(ID, value)values (5,10); 1 row created. SQL> insert into myTable(ID, value)values (6,3); 1 row created. SQL> insert into myTable(ID, value)values (7,-5.88); 1 row created. SQL> insert into myTable(ID, value)values (8,123.45); 1 row created. SQL> insert into myTable(ID, value)values (9,98.23); 1 row created. SQL> insert into myTable(ID, value)values (10,938.23); 1 row created. SQL> insert into myTable(ID, value)values (11,984.23); 1 row created. SQL> insert into myTable(ID, value)values (12,198.23); 1 row created. SQL> insert into myTable(ID, value)values (13,928.87); 1 row created. SQL> insert into myTable(ID, value)values (14,25.37); 1 row created. SQL> insert into myTable(ID, value)values (15,918.3); 1 row created. SQL> insert into myTable(ID, value)values (16,9.23); 1 row created. SQL> insert into myTable(ID, value)values (17,8.23); 1 row created. SQL> SQL> select * from myTable
2 /
ID VALUE
----------
1 9
2 2.11
3 3.44
4 -4.21
5 10
6 3
7 -5.88
8 123.45
9 98.23
10 938.23
11 984.23
12 198.23
13 928.87
14 25.37
15 918.3
16 9.23
17 8.23
17 rows selected. SQL> SQL> SQL> COLUMN ma FORMAT 99999.999 SQL> SELECT id, value,
2 AVG(value) OVER(ORDER BY id
3 ROWS BETWEEN 3 PRECEDING AND 1 FOLLOWING) ma
4 FROM myTable
5 ORDER BY id;
ID VALUE MA
---------- ----------
1 9 5.555
2 2.11 4.850
3 3.44 2.585
4 -4.21 4.068
5 10 2.868
6 3 1.270
7 -5.88 25.272
8 123.45 45.760
9 98.23 231.406
10 938.23 427.652
11 984.23 468.474
12 198.23 629.558
13 928.87 614.986
14 25.37 611.000
15 918.3 416.000
16 9.23 378.000
17 8.23 240.283
17 rows selected. SQL> SQL> SQL> SQL> SQL> -- clean the table SQL> drop table myTable
2 /
Table dropped.</source>
Displaying which rows are used in the moving average calculations with two other analytical functions: FIRST_ VALUE and LAST_VALUE.
These two functions tell us which rows are used in the calculation of the window function for each row.
<source lang="sql">
SQL> SQL> -- create demo table SQL> create table myTable(
2 id NUMBER(2), 3 value NUMBER(6,2) 4 ) 5 /
Table created. SQL> SQL> -- prepare data SQL> insert into myTable(ID, value)values (1,9); 1 row created. SQL> insert into myTable(ID, value)values (2,2.11); 1 row created. SQL> insert into myTable(ID, value)values (3,3.44); 1 row created. SQL> insert into myTable(ID, value)values (4,-4.21); 1 row created. SQL> insert into myTable(ID, value)values (5,10); 1 row created. SQL> insert into myTable(ID, value)values (6,3); 1 row created. SQL> insert into myTable(ID, value)values (7,-5.88); 1 row created. SQL> insert into myTable(ID, value)values (8,123.45); 1 row created. SQL> insert into myTable(ID, value)values (9,98.23); 1 row created. SQL> insert into myTable(ID, value)values (10,938.23); 1 row created. SQL> insert into myTable(ID, value)values (11,984.23); 1 row created. SQL> insert into myTable(ID, value)values (12,198.23); 1 row created. SQL> insert into myTable(ID, value)values (13,928.87); 1 row created. SQL> insert into myTable(ID, value)values (14,25.37); 1 row created. SQL> insert into myTable(ID, value)values (15,918.3); 1 row created. SQL> insert into myTable(ID, value)values (16,9.23); 1 row created. SQL> insert into myTable(ID, value)values (17,8.23); 1 row created. SQL> SQL> select * from myTable
2 /
ID VALUE
----------
1 9
2 2.11
3 3.44
4 -4.21
5 10
6 3
7 -5.88
8 123.45
9 98.23
10 938.23
11 984.23
12 198.23
13 928.87
14 25.37
15 918.3
16 9.23
17 8.23
17 rows selected. SQL> SQL> SQL> SQL> COLUMN first FORMAT 999999.99 SQL> COLUMN last LIKE first SQL> SELECT id, value,
2 AVG(value) OVER(ORDER BY id
3 ROWS BETWEEN 3 PRECEDING AND 1 FOLLOWING) ma,
4 FIRST_VALUE(value) OVER(ORDER BY id
5 ROWS BETWEEN 3 PRECEDING AND 1 FOLLOWING) first,
6 LAST_VALUE(value) OVER(ORDER BY id
7 ROWS BETWEEN 3 PRECEDING AND 1 FOLLOWING) last
8 FROM myTable
9 ORDER BY id;
ID VALUE MA FIRST LAST
---------- ---------- ---------- ----------
1 9 5.555 9.00 2.11
2 2.11 4.850 9.00 3.44
3 3.44 2.585 9.00 -4.21
4 -4.21 4.068 9.00 10.00
5 10 2.868 2.11 3.00
6 3 1.270 3.44 -5.88
7 -5.88 25.272 -4.21 123.45
8 123.45 45.760 10.00 98.23
9 98.23 231.406 3.00 938.23
10 938.23 427.652 -5.88 984.23
11 984.23 468.474 123.45 198.23
12 198.23 629.558 98.23 928.87
13 928.87 614.986 938.23 25.37
14 25.37 611.000 984.23 918.30
15 918.3 416.000 198.23 9.23
16 9.23 378.000 928.87 8.23
17 8.23 240.283 25.37 8.23
17 rows selected. SQL> SQL> SQL> -- clean the table SQL> drop table myTable
2 /
Table dropped. SQL> SQL></source>
Moving average
<source lang="sql">
SQL> SQL> -- create demo table SQL> create table Employee(
2 ID VARCHAR2(4 BYTE) NOT NULL, 3 First_Name VARCHAR2(10 BYTE), 4 Last_Name VARCHAR2(10 BYTE), 5 Start_Date DATE, 6 End_Date DATE, 7 Salary Number(8,2), 8 City VARCHAR2(10 BYTE), 9 Description VARCHAR2(15 BYTE) 10 ) 11 /
Table created. SQL> SQL> -- prepare data SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values ("01","Jason", "Martin", to_date("19960725","YYYYMMDD"), to_date("20060
725","YYYYMMDD"), 1234.56, "Toronto", "Programmer")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("02","Alison", "Mathews", to_date("19760321","YYYYMMDD"), to_date("19860
221","YYYYMMDD"), 6661.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("03","James", "Smith", to_date("19781212","YYYYMMDD"), to_date("19900
315","YYYYMMDD"), 6544.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("04","Celia", "Rice", to_date("19821024","YYYYMMDD"), to_date("19990
421","YYYYMMDD"), 2344.78, "Vancouver","Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("05","Robert", "Black", to_date("19840115","YYYYMMDD"), to_date("19980
808","YYYYMMDD"), 2334.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("06","Linda", "Green", to_date("19870730","YYYYMMDD"), to_date("19960
104","YYYYMMDD"), 4322.78,"New York", "Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("07","David", "Larry", to_date("19901231","YYYYMMDD"), to_date("19980
212","YYYYMMDD"), 7897.78,"New York", "Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("08","James", "Cat", to_date("19960917","YYYYMMDD"), to_date("20020
415","YYYYMMDD"), 1232.78,"Vancouver", "Tester")
3 /
1 row created. SQL> SQL> SQL> SQL> -- display data in the table SQL> select * from Employee
2 /
ID FIRST_NAME LAST_NAME START_DAT END_DATE SALARY CITY DESCRIPTION
---------- ---------- --------- --------- ---------- ---------- ---------------
01 Jason Martin 25-JUL-96 25-JUL-06 1234.56 Toronto Programmer 02 Alison Mathews 21-MAR-76 21-FEB-86 6661.78 Vancouver Tester 03 James Smith 12-DEC-78 15-MAR-90 6544.78 Vancouver Tester 04 Celia Rice 24-OCT-82 21-APR-99 2344.78 Vancouver Manager 05 Robert Black 15-JAN-84 08-AUG-98 2334.78 Vancouver Tester 06 Linda Green 30-JUL-87 04-JAN-96 4322.78 New York Tester 07 David Larry 31-DEC-90 12-FEB-98 7897.78 New York Manager 08 James Cat 17-SEP-96 15-APR-02 1232.78 Vancouver Tester 8 rows selected. SQL> SQL> SQL> SQL> SELECT id, salary,
2 AVG(salary) OVER(ORDER BY id 3 ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING) ma 4 FROM employee 5 ORDER BY id;
ID SALARY MA
---------- ----------
01 1234.56 3948.170 02 6661.78 4813.707 03 6544.78 5183.780 04 2344.78 3741.447 05 2334.78 3000.780 06 4322.78 4851.780 07 7897.78 4484.447 08 1232.78 4565.280 8 rows selected. SQL> SQL> SQL> SQL> -- clean the table SQL> drop table Employee
2 /
Table dropped. SQL></source>
Performing a Centered Average
<source lang="sql">
SQL> SQL> SQL> -- create demo table SQL> create table Employee(
2 ID VARCHAR2(4 BYTE) NOT NULL, 3 First_Name VARCHAR2(10 BYTE), 4 Last_Name VARCHAR2(10 BYTE), 5 Start_Date DATE, 6 End_Date DATE, 7 Salary Number(8,2), 8 City VARCHAR2(10 BYTE), 9 Description VARCHAR2(15 BYTE) 10 ) 11 /
Table created. SQL> SQL> -- prepare data SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values ("01","Jason", "Martin", to_date("19960725","YYYYMMDD"), to_date("20060725","YYYYMMDD"), 1234.56, "Toronto", "Programmer")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("02","Alison", "Mathews", to_date("19760321","YYYYMMDD"), to_date("19860221","YYYYMMDD"), 6661.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("03","James", "Smith", to_date("19781212","YYYYMMDD"), to_date("19900315","YYYYMMDD"), 6544.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("04","Celia", "Rice", to_date("19821024","YYYYMMDD"), to_date("19990421","YYYYMMDD"), 2344.78, "Vancouver","Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("05","Robert", "Black", to_date("19840115","YYYYMMDD"), to_date("19980808","YYYYMMDD"), 2334.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("06","Linda", "Green", to_date("19870730","YYYYMMDD"), to_date("19960104","YYYYMMDD"), 4322.78,"New York", "Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("07","David", "Larry", to_date("19901231","YYYYMMDD"), to_date("19980212","YYYYMMDD"), 7897.78,"New York", "Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("08","James", "Cat", to_date("19960917","YYYYMMDD"), to_date("20020415","YYYYMMDD"), 1232.78,"Vancouver", "Tester")
3 /
1 row created. SQL> SQL> SQL> SQL> -- display data in the table SQL> select * from Employee
2 /
ID FIRST_NAME LAST_NAME START_DAT END_DATE SALARY CITY DESCRIPTION
---------- ---------- --------- --------- ---------- ---------- ---------------
01 Jason Martin 25-JUL-96 25-JUL-06 1234.56 Toronto Programmer 02 Alison Mathews 21-MAR-76 21-FEB-86 6661.78 Vancouver Tester 03 James Smith 12-DEC-78 15-MAR-90 6544.78 Vancouver Tester 04 Celia Rice 24-OCT-82 21-APR-99 2344.78 Vancouver Manager 05 Robert Black 15-JAN-84 08-AUG-98 2334.78 Vancouver Tester 06 Linda Green 30-JUL-87 04-JAN-96 4322.78 New York Tester 07 David Larry 31-DEC-90 12-FEB-98 7897.78 New York Manager 08 James Cat 17-SEP-96 15-APR-02 1232.78 Vancouver Tester 8 rows selected. SQL> SQL> SQL> SQL> SELECT
2 city, SUM(salary) AS city_salary, 3 AVG(SUM(salary)) OVER 4 (ORDER BY city ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING) 5 AS moving_average 6 FROM employee 7 GROUP BY city 8 ORDER BY city;
CITY CITY_SALARY MOVING_AVERAGE
----------- --------------
New York 12220.56 6727.56 Toronto 1234.56 10858.0067 Vancouver 19118.9 10176.73 SQL> SQL> SQL> -- clean the table SQL> drop table Employee
2 /
Table dropped. SQL> SQL></source>
Performing a Cumulative Sum
<source lang="sql">
SQL> SQL> SQL> SQL> -- create demo table SQL> create table Employee(
2 ID VARCHAR2(4 BYTE) NOT NULL primary key, 3 First_Name VARCHAR2(10 BYTE), 4 Last_Name VARCHAR2(10 BYTE), 5 Start_Date DATE, 6 End_Date DATE, 7 Salary Number(8,2), 8 City VARCHAR2(10 BYTE), 9 Description VARCHAR2(15 BYTE) 10 ) 11 /
Table created. SQL> SQL> -- prepare data SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values ("01","Jason", "Martin", to_date("19960725","YYYYMMDD"), to_date("20060725","YYYYMMDD"), 1234.56, "Toronto", "Programmer")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("02","Alison", "Mathews", to_date("19760321","YYYYMMDD"), to_date("19860221","YYYYMMDD"), 6661.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("03","James", "Smith", to_date("19781212","YYYYMMDD"), to_date("19900315","YYYYMMDD"), 6544.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("04","Celia", "Rice", to_date("19821024","YYYYMMDD"), to_date("19990421","YYYYMMDD"), 2344.78, "Vancouver","Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("05","Robert", "Black", to_date("19840115","YYYYMMDD"), to_date("19980808","YYYYMMDD"), 2334.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("06","Linda", "Green", to_date("19870730","YYYYMMDD"), to_date("19960104","YYYYMMDD"), 4322.78,"New York", "Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("07","David", "Larry", to_date("19901231","YYYYMMDD"), to_date("19980212","YYYYMMDD"), 7897.78,"New York", "Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("08","James", "Cat", to_date("19960917","YYYYMMDD"), to_date("20020415","YYYYMMDD"), 1232.78,"Vancouver", "Tester")
3 /
1 row created. SQL> SQL> SQL> SQL> -- display data in the table SQL> select * from Employee
2 /
ID FIRST_NAME LAST_NAME START_DAT END_DATE SALARY CITY DESCRIPTION
-------------------- -------------------- --------- --------- ---------- ---------- ---------------
01 Jason Martin 25-JUL-96 25-JUL-06 1234.56 Toronto Programmer 02 Alison Mathews 21-MAR-76 21-FEB-86 6661.78 Vancouver Tester 03 James Smith 12-DEC-78 15-MAR-90 6544.78 Vancouver Tester 04 Celia Rice 24-OCT-82 21-APR-99 2344.78 Vancouver Manager 05 Robert Black 15-JAN-84 08-AUG-98 2334.78 Vancouver Tester 06 Linda Green 30-JUL-87 04-JAN-96 4322.78 New York Tester 07 David Larry 31-DEC-90 12-FEB-98 7897.78 New York Manager 08 James Cat 17-SEP-96 15-APR-02 1232.78 Vancouver Tester 8 rows selected. SQL> SQL> SQL> SQL> SELECT
2 city, SUM(salary) AS city_salary, 3 SUM(SUM(salary)) OVER 4 (ORDER BY city ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW) 5 AS cumulative_salary 6 FROM employee 7 GROUP BY city 8 ORDER BY city;
CITY CITY_SALARY CUMULATIVE_SALARY
----------- -----------------
New York 12220.56 12220.56 Toronto 1234.56 13455.12 Vancouver 19118.9 32574.02 3 rows selected. SQL> SQL> SQL> SQL> SQL> -- clean the table SQL> drop table Employee
2 /
Table dropped. SQL> SQL></source>
Performing a Moving Average
<source lang="sql">
SQL> SQL> SQL> SQL> -- create demo table SQL> create table Employee(
2 ID VARCHAR2(4 BYTE) NOT NULL primary key, 3 First_Name VARCHAR2(10 BYTE), 4 Last_Name VARCHAR2(10 BYTE), 5 Start_Date DATE, 6 End_Date DATE, 7 Salary Number(8,2), 8 City VARCHAR2(10 BYTE), 9 Description VARCHAR2(15 BYTE) 10 ) 11 /
Table created. SQL> SQL> -- prepare data SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values ("01","Jason", "Martin", to_date("19960725","YYYYMMDD"), to_date("20060725","YYYYMMDD"), 1234.56, "Toronto", "Programmer")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("02","Alison", "Mathews", to_date("19760321","YYYYMMDD"), to_date("19860221","YYYYMMDD"), 6661.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("03","James", "Smith", to_date("19781212","YYYYMMDD"), to_date("19900315","YYYYMMDD"), 6544.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("04","Celia", "Rice", to_date("19821024","YYYYMMDD"), to_date("19990421","YYYYMMDD"), 2344.78, "Vancouver","Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("05","Robert", "Black", to_date("19840115","YYYYMMDD"), to_date("19980808","YYYYMMDD"), 2334.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("06","Linda", "Green", to_date("19870730","YYYYMMDD"), to_date("19960104","YYYYMMDD"), 4322.78,"New York", "Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("07","David", "Larry", to_date("19901231","YYYYMMDD"), to_date("19980212","YYYYMMDD"), 7897.78,"New York", "Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("08","James", "Cat", to_date("19960917","YYYYMMDD"), to_date("20020415","YYYYMMDD"), 1232.78,"Vancouver", "Tester")
3 /
1 row created. SQL> SQL> SQL> SQL> -- display data in the table SQL> select * from Employee
2 /
ID FIRST_NAME LAST_NAME START_DAT END_DATE SALARY CITY DESCRIPTION
-------------------- -------------------- --------- --------- ---------- ---------- ---------------
01 Jason Martin 25-JUL-96 25-JUL-06 1234.56 Toronto Programmer 02 Alison Mathews 21-MAR-76 21-FEB-86 6661.78 Vancouver Tester 03 James Smith 12-DEC-78 15-MAR-90 6544.78 Vancouver Tester 04 Celia Rice 24-OCT-82 21-APR-99 2344.78 Vancouver Manager 05 Robert Black 15-JAN-84 08-AUG-98 2334.78 Vancouver Tester 06 Linda Green 30-JUL-87 04-JAN-96 4322.78 New York Tester 07 David Larry 31-DEC-90 12-FEB-98 7897.78 New York Manager 08 James Cat 17-SEP-96 15-APR-02 1232.78 Vancouver Tester 8 rows selected. SQL> SQL> SQL> SQL> SQL> SQL> SELECT
2 city, SUM(salary) AS city_salary, 3 AVG(SUM(salary)) OVER 4 (ORDER BY city ROWS BETWEEN 3 PRECEDING AND CURRENT ROW) 5 AS moving_average 6 FROM employee 7 GROUP BY city 8 ORDER BY city;
CITY CITY_SALARY MOVING_AVERAGE
----------- --------------
New York 12220.56 12220.56 Toronto 1234.56 6727.56 Vancouver 19118.9 10858.0067 3 rows selected. SQL> SQL> SQL> -- clean the table SQL> drop table Employee
2 /
Table dropped. SQL></source>
Row-ordering is done first and then the moving average
<source lang="sql">
SQL> SQL> SQL> SQL> -- create demo table SQL> create table Employee(
2 ID VARCHAR2(4 BYTE) NOT NULL, 3 First_Name VARCHAR2(10 BYTE), 4 Last_Name VARCHAR2(10 BYTE), 5 Start_Date DATE, 6 End_Date DATE, 7 Salary Number(8,2), 8 City VARCHAR2(10 BYTE), 9 Description VARCHAR2(15 BYTE) 10 ) 11 /
Table created. SQL> SQL> -- prepare data SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values ("01","Jason", "Martin", to_date("19960725","YYYYMMDD"), to_date("20060
725","YYYYMMDD"), 1234.56, "Toronto", "Programmer")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("02","Alison", "Mathews", to_date("19760321","YYYYMMDD"), to_date("19860
221","YYYYMMDD"), 6661.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("03","James", "Smith", to_date("19781212","YYYYMMDD"), to_date("19900
315","YYYYMMDD"), 6544.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("04","Celia", "Rice", to_date("19821024","YYYYMMDD"), to_date("19990
421","YYYYMMDD"), 2344.78, "Vancouver","Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("05","Robert", "Black", to_date("19840115","YYYYMMDD"), to_date("19980
808","YYYYMMDD"), 2334.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("06","Linda", "Green", to_date("19870730","YYYYMMDD"), to_date("19960
104","YYYYMMDD"), 4322.78,"New York", "Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("07","David", "Larry", to_date("19901231","YYYYMMDD"), to_date("19980
212","YYYYMMDD"), 7897.78,"New York", "Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("08","James", "Cat", to_date("19960917","YYYYMMDD"), to_date("20020
415","YYYYMMDD"), 1232.78,"Vancouver", "Tester")
3 /
1 row created. SQL> SQL> SQL> SQL> -- display data in the table SQL> select * from Employee
2 /
ID FIRST_NAME LAST_NAME START_DAT END_DATE SALARY CITY DESCRIPTION
---------- ---------- --------- --------- ---------- ---------- ---------------
01 Jason Martin 25-JUL-96 25-JUL-06 1234.56 Toronto Programmer 02 Alison Mathews 21-MAR-76 21-FEB-86 6661.78 Vancouver Tester 03 James Smith 12-DEC-78 15-MAR-90 6544.78 Vancouver Tester 04 Celia Rice 24-OCT-82 21-APR-99 2344.78 Vancouver Manager 05 Robert Black 15-JAN-84 08-AUG-98 2334.78 Vancouver Tester 06 Linda Green 30-JUL-87 04-JAN-96 4322.78 New York Tester 07 David Larry 31-DEC-90 12-FEB-98 7897.78 New York Manager 08 James Cat 17-SEP-96 15-APR-02 1232.78 Vancouver Tester 8 rows selected. SQL> SQL> SQL> SELECT id, salary,
2 AVG(salary) OVER(ORDER BY id 3 ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING) ma 4 FROM employee 5 ORDER BY id;
ID SALARY MA
---------- ----------
01 1234.56 3948.170 02 6661.78 4813.707 03 6544.78 5183.780 04 2344.78 3741.447 05 2334.78 3000.780 06 4322.78 4851.780 07 7897.78 4484.447 08 1232.78 4565.280 8 rows selected. SQL> SQL> SQL> SQL> -- clean the table SQL> drop table Employee
2 /
Table dropped. SQL> SQL></source>
Use ROWS UNBOUNDED PRECEDING to implicitly indicate the end of the window is the current row
<source lang="sql">
SQL> SQL> SQL> -- create demo table SQL> create table Employee(
2 ID VARCHAR2(4 BYTE) NOT NULL, 3 First_Name VARCHAR2(10 BYTE), 4 Last_Name VARCHAR2(10 BYTE), 5 Start_Date DATE, 6 End_Date DATE, 7 Salary Number(8,2), 8 City VARCHAR2(10 BYTE), 9 Description VARCHAR2(15 BYTE) 10 ) 11 /
Table created. SQL> SQL> -- prepare data SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values ("01","Jason", "Martin", to_date("19960725","YYYYMMDD"), to_date("20060725","YYYYMMDD"), 1234.56, "Toronto", "Programmer")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("02","Alison", "Mathews", to_date("19760321","YYYYMMDD"), to_date("19860221","YYYYMMDD"), 6661.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("03","James", "Smith", to_date("19781212","YYYYMMDD"), to_date("19900315","YYYYMMDD"), 6544.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("04","Celia", "Rice", to_date("19821024","YYYYMMDD"), to_date("19990421","YYYYMMDD"), 2344.78, "Vancouver","Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("05","Robert", "Black", to_date("19840115","YYYYMMDD"), to_date("19980808","YYYYMMDD"), 2334.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("06","Linda", "Green", to_date("19870730","YYYYMMDD"), to_date("19960104","YYYYMMDD"), 4322.78,"New York", "Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("07","David", "Larry", to_date("19901231","YYYYMMDD"), to_date("19980212","YYYYMMDD"), 7897.78,"New York", "Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date, Salary, City, Description)
2 values("08","James", "Cat", to_date("19960917","YYYYMMDD"), to_date("20020415","YYYYMMDD"), 1232.78,"Vancouver", "Tester")
3 /
1 row created. SQL> SQL> SQL> SQL> -- display data in the table SQL> select * from Employee
2 /
ID FIRST_NAME LAST_NAME START_DAT END_DATE SALARY CITY DESCRIPTION
---------- ---------- --------- --------- ---------- ---------- ---------------
01 Jason Martin 25-JUL-96 25-JUL-06 1234.56 Toronto Programmer 02 Alison Mathews 21-MAR-76 21-FEB-86 6661.78 Vancouver Tester 03 James Smith 12-DEC-78 15-MAR-90 6544.78 Vancouver Tester 04 Celia Rice 24-OCT-82 21-APR-99 2344.78 Vancouver Manager 05 Robert Black 15-JAN-84 08-AUG-98 2334.78 Vancouver Tester 06 Linda Green 30-JUL-87 04-JAN-96 4322.78 New York Tester 07 David Larry 31-DEC-90 12-FEB-98 7897.78 New York Manager 08 James Cat 17-SEP-96 15-APR-02 1232.78 Vancouver Tester 8 rows selected. SQL> SQL> SQL> SQL> SELECT
2 city, SUM(salary) AS city_salary, 3 SUM(SUM(salary)) OVER (ORDER BY city ROWS UNBOUNDED PRECEDING) AS 4 cumulative_salary 5 FROM employee 6 GROUP BY city 7 ORDER BY city;
CITY CITY_SALARY CUMULATIVE_SALARY
----------- -----------------
New York 12220.56 12220.56 Toronto 1234.56 13455.12 Vancouver 19118.9 32574.02 SQL> SQL> SQL> SQL> -- clean the table SQL> drop table Employee
2 /
Table dropped. SQL></source>
Use SUM for windowing
<source lang="sql">
SQL> SQL> SQL> SQL> -- create demo table SQL> create table Employee(
2 ID VARCHAR2(4 BYTE) NOT NULL, 3 First_Name VARCHAR2(10 BYTE), 4 Last_Name VARCHAR2(10 BYTE), 5 Start_Date DATE, 6 End_Date DATE, 7 Salary Number(8,2), 8 City VARCHAR2(10 BYTE), 9 Description VARCHAR2(15 BYTE) 10 ) 11 /
Table created. SQL> SQL> -- prepare data SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values ("01","Jason", "Martin", to_date("19960725","YYYYMMDD"), to_date("20060
725","YYYYMMDD"), 1234.56, "Toronto", "Programmer")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("02","Alison", "Mathews", to_date("19760321","YYYYMMDD"), to_date("19860
221","YYYYMMDD"), 6661.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("03","James", "Smith", to_date("19781212","YYYYMMDD"), to_date("19900
315","YYYYMMDD"), 6544.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("04","Celia", "Rice", to_date("19821024","YYYYMMDD"), to_date("19990
421","YYYYMMDD"), 2344.78, "Vancouver","Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("05","Robert", "Black", to_date("19840115","YYYYMMDD"), to_date("19980
808","YYYYMMDD"), 2334.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("06","Linda", "Green", to_date("19870730","YYYYMMDD"), to_date("19960
104","YYYYMMDD"), 4322.78,"New York", "Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("07","David", "Larry", to_date("19901231","YYYYMMDD"), to_date("19980
212","YYYYMMDD"), 7897.78,"New York", "Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("08","James", "Cat", to_date("19960917","YYYYMMDD"), to_date("20020
415","YYYYMMDD"), 1232.78,"Vancouver", "Tester")
3 /
1 row created. SQL> SQL> SQL> SQL> -- display data in the table SQL> select * from Employee
2 /
ID FIRST_NAME LAST_NAME START_DAT END_DATE SALARY CITY DESCRIPTION
---------- ---------- --------- --------- ---------- ---------- ---------------
01 Jason Martin 25-JUL-96 25-JUL-06 1234.56 Toronto Programmer 02 Alison Mathews 21-MAR-76 21-FEB-86 6661.78 Vancouver Tester 03 James Smith 12-DEC-78 15-MAR-90 6544.78 Vancouver Tester 04 Celia Rice 24-OCT-82 21-APR-99 2344.78 Vancouver Manager 05 Robert Black 15-JAN-84 08-AUG-98 2334.78 Vancouver Tester 06 Linda Green 30-JUL-87 04-JAN-96 4322.78 New York Tester 07 David Larry 31-DEC-90 12-FEB-98 7897.78 New York Manager 08 James Cat 17-SEP-96 15-APR-02 1232.78 Vancouver Tester 8 rows selected. SQL> SQL> SQL> SQL> COLUMN ma FORMAT 999999.9999 SQL> COLUMN sum LIKE ma SQL> COLUMN "sum/3" LIKE ma SQL> SELECT id, salary,
2 AVG(salary) OVER(ORDER BY id 3 ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING) ma, 4 SUM(salary) OVER(ORDER BY id 5 ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING) sum, 6 (SUM(salary) OVER(ORDER BY id 7 ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING))/3 "Sum/3" 8 FROM employee 9 ORDER BY id;
ID SALARY MA SUM Sum/3
---------- ------------ ------------ ------------
01 1234.56 3948.1700 7896.3400 2632.1133 02 6661.78 4813.7067 14441.1200 4813.7067 03 6544.78 5183.7800 15551.3400 5183.7800 04 2344.78 3741.4467 11224.3400 3741.4467 05 2334.78 3000.7800 9002.3400 3000.7800 06 4322.78 4851.7800 14555.3400 4851.7800 07 7897.78 4484.4467 13453.3400 4484.4467 08 1232.78 4565.2800 9130.5600 3043.5200 8 rows selected. SQL> SQL> SQL> -- clean the table SQL> drop table Employee
2 /
Table dropped. SQL> SQL></source>
Use the COUNT aggregate analytical function to show how many rows are included in each window
<source lang="sql">
SQL> SQL> SQL> -- create demo table SQL> create table Employee(
2 ID VARCHAR2(4 BYTE) NOT NULL, 3 First_Name VARCHAR2(10 BYTE), 4 Last_Name VARCHAR2(10 BYTE), 5 Start_Date DATE, 6 End_Date DATE, 7 Salary Number(8,2), 8 City VARCHAR2(10 BYTE), 9 Description VARCHAR2(15 BYTE) 10 ) 11 /
Table created. SQL> SQL> -- prepare data SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values ("01","Jason", "Martin", to_date("19960725","YYYYMMDD"), to_date("20060
725","YYYYMMDD"), 1234.56, "Toronto", "Programmer")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("02","Alison", "Mathews", to_date("19760321","YYYYMMDD"), to_date("19860
221","YYYYMMDD"), 6661.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("03","James", "Smith", to_date("19781212","YYYYMMDD"), to_date("19900
315","YYYYMMDD"), 6544.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("04","Celia", "Rice", to_date("19821024","YYYYMMDD"), to_date("19990
421","YYYYMMDD"), 2344.78, "Vancouver","Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("05","Robert", "Black", to_date("19840115","YYYYMMDD"), to_date("19980
808","YYYYMMDD"), 2334.78, "Vancouver","Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("06","Linda", "Green", to_date("19870730","YYYYMMDD"), to_date("19960
104","YYYYMMDD"), 4322.78,"New York", "Tester")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("07","David", "Larry", to_date("19901231","YYYYMMDD"), to_date("19980
212","YYYYMMDD"), 7897.78,"New York", "Manager")
3 /
1 row created. SQL> insert into Employee(ID, First_Name, Last_Name, Start_Date, End_Date,
Salary, City, Description)
2 values("08","James", "Cat", to_date("19960917","YYYYMMDD"), to_date("20020
415","YYYYMMDD"), 1232.78,"Vancouver", "Tester")
3 /
1 row created. SQL> SQL> SQL> SQL> -- display data in the table SQL> select * from Employee
2 /
ID FIRST_NAME LAST_NAME START_DAT END_DATE SALARY CITY DESCRIPTION
---------- ---------- --------- --------- ---------- ---------- ---------------
01 Jason Martin 25-JUL-96 25-JUL-06 1234.56 Toronto Programmer 02 Alison Mathews 21-MAR-76 21-FEB-86 6661.78 Vancouver Tester 03 James Smith 12-DEC-78 15-MAR-90 6544.78 Vancouver Tester 04 Celia Rice 24-OCT-82 21-APR-99 2344.78 Vancouver Manager 05 Robert Black 15-JAN-84 08-AUG-98 2334.78 Vancouver Tester 06 Linda Green 30-JUL-87 04-JAN-96 4322.78 New York Tester 07 David Larry 31-DEC-90 12-FEB-98 7897.78 New York Manager 08 James Cat 17-SEP-96 15-APR-02 1232.78 Vancouver Tester 8 rows selected. SQL> SQL> SQL> SQL> SELECT id, salary,
2 COUNT(salary) OVER(ORDER BY id 3 ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING) Howmanyrows 4 FROM employee 5 ORDER BY id;
ID SALARY HOWMANYROWS
---------- -----------
01 1234.56 2 02 6661.78 3 03 6544.78 3 04 2344.78 3 05 2334.78 3 06 4322.78 3 07 7897.78 3 08 1232.78 2 8 rows selected. SQL> SQL> -- clean the table SQL> drop table Employee
2 /
Table dropped. SQL> SQL> SQL></source>
Using the Window Functions
Window functions calculates things like
- cumulative sums,
- moving averages within a specified range of rows,
- a range of values, or
- an interval of time.
Window function can work with: SUM(), AVG(), MAX(), MIN(), COUNT(), VARIANCE(), and STDDEV().
Window function can also work with FIRST_VALUE() and LAST_VALUE(), which return the first and last values in a window.
16. 29. Window Functions 16. 29. 1. Using the Window Functions 16. 29. 2. <A href="/Tutorial/Oracle/0320__Analytical-Functions/PerformingaCumulativeSum.htm">Performing a Cumulative Sum</a> 16. 29. 3. <A href="/Tutorial/Oracle/0320__Analytical-Functions/UseROWSUNBOUNDEDPRECEDINGtoimplicitlyindicatetheendofthewindowisthecurrentrow.htm">Use ROWS UNBOUNDED PRECEDING to implicitly indicate the end of the window is the current row</a> 16. 29. 4. <A href="/Tutorial/Oracle/0320__Analytical-Functions/PerformingaMovingAverage.htm">Performing a Moving Average</a> 16. 29. 5. <A href="/Tutorial/Oracle/0320__Analytical-Functions/PerformingaCenteredAverage.htm">Performing a Centered Average</a> 16. 29. 6. <A href="/Tutorial/Oracle/0320__Analytical-Functions/Movingaverage.htm">Moving average</a> 16. 29. 7. <A href="/Tutorial/Oracle/0320__Analytical-Functions/Roworderingisdonefirstandthenthemovingaverage.htm">Row-ordering is done first and then the moving average</a> 16. 29. 8. <A href="/Tutorial/Oracle/0320__Analytical-Functions/UseSUMforwindowing.htm">Use SUM for windowing</a> 16. 29. 9. <A href="/Tutorial/Oracle/0320__Analytical-Functions/UsetheCOUNTaggregateanalyticalfunctiontoshowhowmanyrowsareincludedineachwindow.htm">Use the COUNT aggregate analytical function to show how many rows are included in each window</a> 16. 29. 10. <A href="/Tutorial/Oracle/0320__Analytical-Functions/AnExpandedExampleofaPhysicalWindow.htm">An Expanded Example of a Physical Window</a> 16. 29. 11. <A href="/Tutorial/Oracle/0320__Analytical-Functions/DisplayingwhichrowsareusedinthemovingaveragecalculationswithtwootheranalyticalfunctionsFIRSTVALUEandLASTVALUE.htm">Displaying which rows are used in the moving average calculations with two other analytical functions: FIRST_ VALUE and LAST_VALUE.</a>
