Lets create an employee table with id, name and salary columns and load some data,

Create a dummy table

CREATE TABLE public.employee (
    id int8 NOT NULL,
    name varchar(120) NOT NULL,
    salary int8 NOT NULL,
    CONSTRAINT emp_pk PRIMARY KEY (id)
);

Generate dummy data

Below sql query generates 1 Million employee records,

WITH salary_list AS (
    SELECT '{1000, 2000, 5000}'::INT[] salary
)
INSERT INTO public.employee
(id, name, salary)
SELECT n, 'Employee ' || n as name, salary[1 + mod(n, array_length(salary, 1))]
FROM salary_list, generate_series(1, 1000000) as n

Lets try to decouple the query,

salary_list

WITH salary_list AS (
    SELECT '{1000, 2000, 5000}'::INT[] salary
)

This is a common table expression, which returns an integer array of salaries. we can use this salary column within this query.

generate_series

generate_series is a postgres function which takes a min and max number and generates values, an example below

select n from generate_series(1, 10) as n

'Employee ' || n as name generates employee names with their id as suffix.

salary[1 + mod(n, array_length(salary, 1))] this picks up salary based on n value returned from generate series.

So the final output that we will get is below,

id	name	salary
1	Employee 1	2000
2	Employee 2	5000
3	Employee 3	1000
4	Employee 4	2000
5	Employee 5	5000

Hope you liked this trick to import random data into postgres. please follow this blog to get more postgres tips.