Partitioning

Like aggregate functions, that may operate alone or in relation to a group, window functions may also operate on a group, which is called a “partition”.

Syntax

<window function>(...) OVER (PARTITION BY <expr> [, <expr> ...])

Aggregation over a group could produce more than one row, so the result set generated by a partition is joined with the main query using the same expression list as the partition.

Continuing the EMPLOYEE example, instead of getting the portion of each employee’s salary over the all-employees total, we would like to get the portion based on the employees in the same department:

select
    id,
    department,
    salary,
    salary / sum(salary) OVER (PARTITION BY department) portion
  from employee
  order by id;

Results

id  department  salary  portion
--  ----------  ------  ----------
1   R & D        10.00      0.3448
2   SALES        12.00      0.6000
3   SALES         8.00      0.4000
4   R & D         9.00      0.3103
5   R & D        10.00      0.3448

Ordering

The ORDER BY sub-clause can be used with or without partitions.The ORDER BY clause within OVER specifies the order in which the window function will process rows.This order does not have to be the same as the order rows appear in the output.

There is an important concept associated with window functions: for each row there is a set of rows in its partition called the window frame.By default, when specifying ORDER BY, the frame consists of all rows from the beginning of the partition to the current row and rows equal to the current ORDER BY expression.Without ORDER BY, the default frame consists of all rows in the partition.

As a result, for standard aggregate functions, the ORDER BY clause produces partial aggregation results as rows are processed.

Example

select
    id,
    salary,
    sum(salary) over (order by salary) cumul_salary
  from employee
  order by salary;

Results

id  salary  cumul_salary
--  ------  ------------
3     8.00          8.00
4     9.00         17.00
1    10.00         37.00
5    10.00         37.00
2    12.00         49.00

Then cumul_salary returns the partial/accumulated (or running) aggregation (of the SUM function).It may appear strange that 37.00 is repeated for the ids 1 and 5, but that is how it should work.The ORDER BY keys are grouped together, and the aggregation is computed once (but summing the two 10.00).To avoid this, you can add the ID field to the end of the ORDER BY clause.

It’s possible to use multiple windows with different orders, and ORDER BY parts like ASC/DESC and NULLS FIRST/LAST.

With a partition, ORDER BY works the same way, but at each partition boundary the aggregation is reset.

All aggregation functions can use ORDER BY, except for LIST().