Help Center> Data Lake Insight> SQL Syntax Reference (To Be Offline)> Spark SQL Syntax Reference (Unavailable Soon)> Built-in Functions> Aggregate Functions> count
Updated on 2023-10-25 GMT+08:00
View PDF

count

This function is used to return the number of records.

Syntax

count([distinct|all] <colname>)

Parameters

Table 1 Parameters

Parameter

Mandatory

Description

distinct or all

No

Determines whether duplicate records should be excluded during counting. all is used by default, indicating that all records will be included in the count.

If distinct is specified, only the number of unique values is counted.

colname

Yes

The column value can be of any type.

The value can be *, that is, count(*), indicating that the number of all rows is returned.

Return Values

The return value is of the BIGINT type.

If the value of colname is NULL, the row is not involved in calculation.

Example Code

  • Calculates the total number of records in the warehouse table. An example command is as follows:
    select count(*) from warehouse;

    The command output is as follows:

    _c0        
10
  • When used with group by, it groups all offerings by warehouse (warehouseId) and calculates the number of offerings in each warehouse (warehouseId). An example command is as follows:
    select warehouseId, count(*) from warehouse group by warehouseId;

    The command output is as follows:

    warehouseId _c1        
city1   6          
city2   5          
city3   6

    Example 3: Calculates the number of warehouses through distinct deduplication. An example command is as follows:

    select count(distinct warehouseId) from warehouse;

    The command output is as follows:

    _c0        
 3
Parent topic: Aggregate Functions