Window function (SQL)
In SQL, a window function or analytic function[1] is a function which uses values from one or multiple rows to return a value for each row. (This contrasts with an aggregate function, which returns a single value for multiple rows.) Window functions have an OVER clause; any function without an OVER clause is not a window function, but rather an aggregate or single-row (scalar) function.[2]
Example
[edit]As an example, here is a query which uses a window function to compare the salary of each employee with the average salary of their department (example from the PostgreSQL documentation):[3]
SELECT depname, empno, salary, avg(salary) OVER (PARTITION BY depname) FROM empsalary;
Output:
depname | empno | salary | avg ----------+-------+--------+---------------------- develop | 11 | 5200 | 5020.0000000000000000 develop | 7 | 4200 | 5020.0000000000000000 develop | 9 | 4500 | 5020.0000000000000000 develop | 8 | 6000 | 5020.0000000000000000 develop | 10 | 5200 | 5020.0000000000000000 personnel | 5 | 3500 | 3700.0000000000000000 personnel | 2 | 3900 | 3700.0000000000000000 sales | 3 | 4800 | 4866.6666666666666667 sales | 1 | 5000 | 4866.6666666666666667 sales | 4 | 4800 | 4866.6666666666666667 (10 rows)
The PARTITION BY
clause groups rows into partitions, and the function is applied to each partition separately. If the PARTITION BY
clause is omitted (such as with an empty OVER()
clause), then the entire result set is treated as a single partition.[4] For this query, the average salary reported would be the average taken over all rows.
Window functions are evaluated after aggregation (after the GROUP BY
clause and non-window aggregate functions, for example).[1]
Syntax
[edit]According to the PostgreSQL documentation, a window function has the syntax of one of the following:[4]
function_name ([expression [, expression ... ]]) OVER window_name function_name ([expression [, expression ... ]]) OVER ( window_definition ) function_name ( * ) OVER window_name function_name ( * ) OVER ( window_definition )
where window_definition
has syntax:
[ existing_window_name ] [ PARTITION BY expression [, ...] ] [ ORDER BY expression [ ASC | DESC | USING operator ] [ NULLS { FIRST | LAST } ] [, ...] ] [ frame_clause ]
frame_clause
has the syntax of one of the following:
{ RANGE | ROWS | GROUPS } frame_start [ frame_exclusion ] { RANGE | ROWS | GROUPS } BETWEEN frame_start AND frame_end [ frame_exclusion ]
frame_start
and frame_end
can be UNBOUNDED PRECEDING
, offset PRECEDING
, CURRENT ROW
, offset FOLLOWING
, or UNBOUNDED FOLLOWING
. frame_exclusion
can be EXCLUDE CURRENT ROW
, EXCLUDE GROUP
, EXCLUDE TIES
, or EXCLUDE NO OTHERS
.
expression
refers to any expression that does not contain a call to a window function.
Notation:
- Brackets [] indicate optional clauses
- Curly braces {} indicate a set of different possible options, with each option delimited by a vertical bar |
Example
[edit]Window functions allow access to data in the records right before and after the current record.[5][6][7][8] A window function defines a frame or window of rows with a given length around the current row, and performs a calculation across the set of data in the window.[9][10]
NAME | ------------ Aaron| <-- Preceding (unbounded) Andrew| Amelia| James| Jill| Johnny| <-- 1st preceding row Michael| <-- Current row Nick| <-- 1st following row Ophelia| Zach| <-- Following (unbounded)
In the above table, the next query extracts for each row the values of a window with one preceding and one following row:
SELECT LAG(name, 1) OVER(ORDER BY name) "prev", name, LEAD(name, 1) OVER(ORDER BY name) "next" FROM people ORDER BY name
The result query contains the following values:
| PREV | NAME | NEXT | |----------|----------|----------| | (null)| Aaron| Andrew| | Aaron| Andrew| Amelia| | Andrew| Amelia| James| | Amelia| James| Jill| | James| Jill| Johnny| | Jill| Johnny| Michael| | Johnny| Michael| Nick| | Michael| Nick| Ophelia| | Nick| Ophelia| Zach| | Ophelia| Zach| (null)|
History
[edit]Window functions were introduced to the SQL:2003 standard and had functionality expanded in later specifications.[11]
Support for particular database implementations was added as follows:
- PostgreSQL - version 8.4 in 2009.[12]
- MySQL - version 8 in 2018.[13][14]
- MariaDB - version 10.2 in 2016.[15]
See also
[edit]References
[edit]- ^ a b "Analytic function concepts in Standard SQL | BigQuery". Google Cloud. Retrieved 2021-03-23.
- ^ "Window Functions". sqlite.org. Retrieved 2021-03-23.
- ^ "3.5. Window Functions". PostgreSQL Documentation. 2021-02-11. Retrieved 2021-03-23.
- ^ a b "4.2. Value Expressions". PostgreSQL Documentation. 2021-02-11. Retrieved 2021-03-23.
- ^ Leis, Viktor; Kundhikanjana, Kan; Kemper, Alfons; Neumann, Thomas (June 2015). "Efficient Processing of Window Functions in Analytical SQL Queries". Proc. VLDB Endow. 8 (10): 1058–1069. doi:10.14778/2794367.2794375. ISSN 2150-8097.
- ^ Cao, Yu; Chan, Chee-Yong; Li, Jie; Tan, Kian-Lee (July 2012). "Optimization of Analytic Window Functions". Proc. VLDB Endow. 5 (11): 1244–1255. arXiv:1208.0086. doi:10.14778/2350229.2350243. ISSN 2150-8097.
- ^ "Probably the Coolest SQL Feature: Window Functions". Java, SQL and jOOQ. 2013-11-03. Retrieved 2017-09-26.
- ^ "Window Functions in SQL - Simple Talk". Simple Talk. 2013-10-31. Retrieved 2017-09-26.
- ^ "SQL Window Functions Introduction". Apache Drill.
- ^ "PostgreSQL: Documentation: Window Functions". www.postgresql.org. Retrieved 2020-04-04.
- ^ "Window Functions Overview". MariaDB KnowledgeBase. Retrieved 2021-03-23.
- ^ "PostgreSQL Release 8.4". www.postgresql.org. 24 July 2014. Retrieved 2024-03-10.
- ^ "MySQL :: What's New in MySQL 8.0? (Generally Available)". dev.mysql.com. Retrieved 2022-11-21.
- ^ "MySQL :: MySQL 8.0 Reference Manual :: 12.21.2 Window Function Concepts and Syntax". dev.mysql.com.
- ^ "MariaDB 10.2.0 Release Notes". mariadb.com. Retrieved 2024-03-10.