Build multi-tenant RAG with Neon's database-per-user model — no nosy neighbors, max isolation, minimal costs

PostgreSQL jsonb_object_agg() Function

Summary: in this tutorial, you will learn how to use the PostgreSQL jsonb_object_agg() function to aggregate key/value pairs into a JSON object.

Introduction to the PostgreSQL jsonb_object_agg() function

The PostgreSQL jsonb_object_agg() function is an aggregate function that allows you to collect key/value pairs into a JSON object.

The jsonb_object_agg() can be useful when you want to aggregate data from multiple rows into a single JSON object or construct complex JSON output.

Here’s the syntax of the jsonb_object_agg() function:

jsonb_object_agg(key, value)

In this syntax:

  • key represents the key for the JSON object. The key must not be null.
  • value represents the value for the corresponding key.

The jsonb_object_agg() returns a JSON object that consists of key/value pairs.

PostgreSQL jsonb_object_agg() function examples

Let’s explore some examples of using the PostgreSQL jsonb_object_agg() function.

1) Basic PostgreSQL jsonb_object_agg() function example

First, create a table called departments:

CREATE TABLE departments(
    id SERIAL PRIMARY KEY,
    department_name VARCHAR(255) NOT NULL
);

Second, insert some rows into the departments table:

INSERT INTO departments(department_name)
VALUES
   ('Sales'),
   ('Marketing')
RETURNING *;

Third, use the jsonb_object_agg() function to create an object whose key is the department name and value is the id:

SELECT
  jsonb_object_agg(department_name, id) departments
FROM
  departments;

Output:

departments
------------------------------
 {"Sales": 1, "Marketing": 2}
(1 row)

2) Using the jsonb_object_agg() function with GROUP BY clause

First, create a new table called employees:

CREATE TABLE employees (
    id SERIAL PRIMARY KEY,
    name VARCHAR(100) NOT NULL,
    salary DECIMAL(10, 2),
    department_id INT NOT NULL,
    FOREIGN KEY(department_id)
       REFERENCES departments(id) ON DELETE CASCADE
);

Second, insert some rows into the employees table:

INSERT INTO employees (name, salary, department_id)
VALUES
  ('John Doe', 60000, 1),
  ('Jane Smith', 65000, 1),
  ('Alice Johnson', 55000, 2),
  ('Bob Williams', 70000, 2),
  ('Alex Miller', NULL , 2)
RETURNING *;

Output:

id |     name      |  salary  | department_id
----+---------------+----------+---------------
  1 | John Doe      | 60000.00 |             1
  2 | Jane Smith    | 65000.00 |             1
  3 | Alice Johnson | 55000.00 |             2
  4 | Bob Williams  | 70000.00 |             2
  5 | Alex Miller   |     null |             2
(5 rows)

Third, use the jsonb_object_agg() function to get the department name and a JSON object that contains employee details of the department including employee name and salary:

SELECT
  department_name,
  jsonb_pretty(
    jsonb_object_agg(e.name, e.salary)
  ) AS employee_details
FROM
  departments d
  INNER JOIN employees e ON e.department_id = d.id
GROUP BY
  department_name;

Output:

department_name |       employee_details
-----------------+-------------------------------
 Marketing       | {                            +
                 |     "Alex Miller": null,     +
                 |     "Bob Williams": 70000.00,+
                 |     "Alice Johnson": 55000.00+
                 | }
 Sales           | {                            +
                 |     "John Doe": 60000.00,    +
                 |     "Jane Smith": 65000.00   +
                 | }
(2 rows)

Note that we use the jsonb_pretty() function to format JSON.

Alex Miller has not had a salary yet so his salary is null. The jsonb_object_agg() also collects the null into the JSON object.

To skip nulls, you can use the jsonb_object_agg_strict() function as follows:

SELECT
  department_name,
  jsonb_pretty(
    jsonb_object_agg_strict(e.name, e.salary)
  ) AS employee_details
FROM
  departments d
  INNER JOIN employees e ON e.department_id = d.id
GROUP BY
  department_name;

Output:

department_name |       employee_details
-----------------+-------------------------------
 Marketing       | {                            +
                 |     "Bob Williams": 70000.00,+
                 |     "Alice Johnson": 55000.00+
                 | }
 Sales           | {                            +
                 |     "John Doe": 60000.00,    +
                 |     "Jane Smith": 65000.00   +
                 | }
(2 rows)

The jsonb_object_agg_strict() function works like the jsonb_object_agg() function except that it skips null values.

Summary

  • Use the jsonb_object_agg() function to aggregate key/value pairs into a JSON object.
  • Use the jsonb_object_agg() function to aggregate key/value pairs into a JSON object and skip null values.

Last updated on

Was this page helpful?