Log Service:Window functions

Summary

Aggregate functions calculate the single result for a group of rows, and window functions calculate the result for each row in a group. A window function has three elements: partition, order, and frame. For more information, see Window Function Concepts and Syntax.

function over (
    [partition by partition_expression]
    [order by order_expression]
    [frame]
)

• Partition: The partition element is defined by the PARTITION BY clause. The PARTITION BY clause separates rows into partitions. If you do not specify the PARTITION BY clause, all rows are treated as a single partition.
• Order: The order element is defined by the ORDER BY clause. The ORDER BY clause sorts rows in all partitions.
  Note If you use the ORDER BY clause to sort rows on fields that have the same value, the order of these rows is non-deterministic. You can include additional fields in the ORDER BY clause to obtain the expected order of these rows. Example: order by request_time, request_method.
• Frame: The frame element is defined by the FRAME clause. The FRAME clause specifies a subset of each partition. A frame further refines the rows in each partition. You cannot specify the FRAME clause for ranking functions. Syntax of the FRAME clause: { rows | range} { frame_start | frame_between }. Example: range between unbounded preceding and unbounded following. For more information, see Window Function Frame Specification.

Window functions

Aggregate functions

You can use all aggregate functions as window functions. For more information about aggregate functions, see Aggregate function. The following example shows how to use the sum function as a window function.

Syntax

sum() over (
    [partition by partition_expression]
    [order by order_expression]
    [frame]
)

Parameters

Return value type

The double data type.

Examples

Calculate the percentage of each employee salary in each department.

• Query statement

  * |
  SELECT
    department,
    staff_name,
    salary,
    round ( salary * 1.0 / sum(salary) over(partition by department), 3) AS salary_percentage 

• Query result

cume_dist function

Calculates the cumulative distribution of each value in a partition. The result is obtained by using division. The numerator is the number of rows whose field values are smaller or equal to the field value of the specified row. The specified row is also counted. The denominator is the total number of rows in the partition. The calculation is based on the order of the rows in the partition. Value range: (0,1].

Syntax

cume_dist() over (
    [partition by partition_expression]
    [order by order_expression]
)

Parameters

Return value type

The double data type.

Examples

Calculate the cumulative distribution of the size of each object in an OSS bucket named bucket00788.

• Query statement

  bucket=bucket00788 |
  select
    object,
    object_size,
    cume_dist() over (
      partition by object
      order by
        object_size
    ) as cume_dist
  from  oss-log-store

• Query result

dense_rank function

Calculates the rank of each value in a partition. Rows that have the same field value are assigned the same rank. The ranks are consecutive. For example, if two rows have the same rank of 1, the rank of the next row is 2.

Syntax

dense_rank() over (
    [partition by partition_expression]
    [order by order_expression]
)

Parameters

Return value type

The bigint data type.

Examples

Calculate the rank of each employee salary in each department.

• Query statement

  * |
  select
    department,
    staff_name,
    salary,
    dense_rank() over(
      partition by department
      order by
        salary desc
    ) as salary_rank
  order by
    department,
    salary_rank

• Query result

ntile function

Divides the rows in each partition into a number of groups. The number of groups is specified by the N parameter.

Syntax

ntile(n) over (
    [partition by partition_expression]
    [order by order_expression]
)

Parameters

Return value type

The bigint data type.

Examples

Divide the rows in each partition into three groups.

• Query statement

  object=245-da918c.model |
  select
    object,
    object_size,
    ntile(3) over (
      partition by object
      order by
        object_size
    ) as ntile
  from  oss-log-store

• Query result

percent_rank function

Calculates the percentage ranking of each row in a partition. The calculation formula is (rank - 1)/(total_rows - 1). In the formula, rank represents the rank of the current row, and total_rows represents the total number of rows in a partition.

Syntax

percent_rank() over (
    [partition by partition_expression]
    [order by order_expression]
)

Parameters

Return value type

The double data type.

Examples

Calculate the percentage ranking of the size of each OSS object.

• Query statement

  object=245-da918c3e2dd9dc9cb4d9283b%2F555e2441b6a4c7f094099a6dba8e7a5f.model|
  select
    object,
    object_size,
   percent_rank() over (
      partition by object
      order by
        object_size
    ) as ntile
  FROM  oss-log-store

• Query result

rank function

Calculates the rank of each row in a partition. Rows that have the same field value are assigned the same rank. The ranks are not consecutive. For example, if two rows have the same rank of 1, the rank of the next row is 3.

Syntax

rank() over (
    [partition by partition_expression]
    [order by order_expression]
)

Parameters

Return value type

The bigint data type.

Examples

Calculate the rank of each employee salary in each department.

• Query statement

  * |
  select
    department,
    staff_name,
    salary,
    rank() over(
      partition by department
      order by
        salary desc
    ) as salary_rank
  order by
    department,
    salary_rank

• Query result

row_number function

Calculate the rank of each row in a partition. The ranks are unique, and start from 1.

Syntax

row_number() over (
    [partition by partition_expression]
    [order by order_expression]
)

Parameters

Return value type

The bigint data type.

Examples

Calculate the rank of each employee on the salary in each department.

• Query statement

  * |
  select
    department,
    staff_name,
    salary,
    row_number() over(
      partition by department
      order by
        salary desc
    ) as salary_rank
  order by
    department,
    salary_rank

• Query result

first_value function

Returns the value of the specified field in the first row of each partition.

Syntax

first_value(x) over (
    [partition by partition_expression]
    [order by order_expression]
    [frame]
)

Parameters

Return value type

The data type is the same as the data type of the x parameter.

Examples

Return the minimum size of each object in the specified OSS bucket.

• Query statement

  bucket :bucket90 |
  select
    object,
    object_size,
    first_value(object_size) over (
      partition by object
      order by
        object_size 
       range between unbounded preceding and unbounded following
    ) as first_value
  from  oss-log-store

• Query result

last_value function

Returns the value of the specified field in the last row of each partition.

Syntax

last_value(x) over (
    [partition by partition_expression]
    [order by order_expression]
    [frame]
)

Parameters

Return value type

The data type is the same as the data type of the x parameter.

Examples

Return the maximum size of each object in the specified OSS bucket.

• Query statement

  bucket :bucket90 |
  select
    object,
    object_size,
    last_value(object_size) over (
      partition by object
      order by
        object_size 
       range between unbounded preceding and unbounded following
    ) as last_value
  from  oss-log-store

• Query result

lag function

Returns the value of the specified field in the row that is at the specified offset before the current row in each partition. The offset is specified by the offset parameter.

Syntax

lag(x,offset,defaut_value) over (
    [partition by partition_expression]
    [order by order_expression]
    [frame]
)

Parameters

Return value type

The data type is the same as the data type of the x parameter.

Examples

Count the daily unique visitors (UVs) to your website and calculates the percentage of the increase in UVs over the previous day.

• Query statement

  * |
  select
    day,
    UV,
    UV * 1.0 /(lag(UV, 1, 0) over()) as diff_percentage
  from  (
      select
        approx_distinct(client_ip) as UV,
        date_trunc('day', __time__) as day
      from  log
      group by
        day
      order by
        day asc
    )

• Query result

lead function

Returns the value of the specified field in the row that is at the specified offset after the current row in each partition. The offset is specified by the offset parameter.

Syntax

lead(x,offset,defaut_value) over (
    [partition by partition_expression]
    [order by order_expression]
    [frame]
)

Parameters

Return value type

The data type is the same as the data type of the x parameter.

Examples

Count the hourly unique visitors (UVs) to your website on 2021-08-26 and calculates the difference in percentage between UVs of two consecutive hours.

• Query statement

  * |
  select
    time,
    UV,
    UV * 1.0 /(lead(UV, 1, 0) over()) as diff_percentage
  from  (
      select
        approx_distinct(client_ip) as uv,
        date_trunc('hour', __time__) as time
      from    log
      group by
        time
      order by
        time asc
    )

• Query result

nth_value function

Returns the value of the specified field in the row that is at the specified offset from the beginning of each partition. The offset is specified by the offset parameter.

Syntax

nth_value(x,offset) over (
    [partition by partition_expression]
    [order by order_expression]
    [frame]
)

Parameters

Return value type

The data type is the same as the data type of the x parameter.

Examples

Return the employees whose salary is the second highest in each department.

• Query statement

  * |
  select
    department,
    staff_name,
    salary,
    nth_value(staff_name, 2) over(
      partition by department
      order by
        salary desc
        range between unbounded preceding and unbounded following
    ) as second_highest_salary from log

• Query result