Build multi-tenant RAG with Neon's database-per-user model — no nosy neighbors, max isolation, minimal costs
PostgreSQL Tutorial/JSON Functions/jsonb_path_query_array

PostgreSQL jsonb_path_query_array() Function

Summary: in this tutorial, you will learn how to use the PostgreSQL jsonb_path_query_array() function to query JSONB data using a JSON path and return matched elements as a JSON array.

Introduction to PostgreSQL jsonb_path_query_array() function

The jsonb_path_query_array() function allows you to query JSONB data using a JSON path expression.

Here’s the basic syntax of the jsonb_path_query_array() function:

jsonb_path_query_array(jsonb_data, json_path)

In this syntax:

  • First, specify the jsonb_data that you want to query.
  • Second, provide a json_path that you want to match elements within the jsonb_data.

The jsonb_path_query_array() function returns the matched elements as a JSON array.

If the function does not find any matched element, it returns an empty array.

If either argument is NULL, the function returns NULL.

PostgreSQL jsonb_path_query_array() function example

Let’s explore some examples of using the jsonb_path_query_array() function

1) Basic PostgreSQL jsonb_path_query_array() function example

The following example uses the jsonb_path_query_array() function to get the employee names as an array:

SELECT
  jsonb_path_query_array(
    '{"employees": [{"name": "Alice", "age": 25}, {"name": "Bob", "age": 30}]}',
    '$.employees[*].name'
  );

Output:

jsonb_path_query_array
------------------------
 ["Alice", "Bob"]
(1 row)

In this example, the JSON path expression $.employees[*].name locates the value of the name key of all elements in the employees array.

2) Using jsonb_path_query_array() function with table data

First, create a new table called employees:

CREATE TABLE employees (
    id SERIAL PRIMARY KEY,
    data JSONB
);

In the employees table, the data column has the type of JSONB.

Second, insert some rows into the employees table:

INSERT INTO employees (data) VALUES
    ('{"name": "Alice", "age": 30, "pets": [{"type": "cat", "name": "Fluffy"}, {"type": "dog", "name": "Buddy"}]}'),
    ('{"name": "Bob", "age": 35, "pets": [{"type": "dog", "name": "Max"}]}'),
    ('{"name": "Charlie", "age": 40, "pets": [{"type": "rabbit", "name": "Snowball"}]}')
RETURNING *;

Output:

id |                                                    data

----+-------------------------------------------------------------------------------------------------------------
  1 | {"age": 30, "name": "Alice", "pets": [{"name": "Fluffy", "type": "cat"}, {"name": "Buddy", "type": "dog"}]}
  2 | {"age": 35, "name": "Bob", "pets": [{"name": "Max", "type": "dog"}]}
  3 | {"age": 40, "name": "Charlie", "pets": [{"name": "Snowball", "type": "rabbit"}]}
(3 rows)

Third, use the jsonb_path_query_array() function to retrieve the pet names of employees as a JSON array:

SELECT jsonb_path_query_array(data, '$.pets[*].name') AS employee_pet_names
FROM employees;

Output:

employee_pet_names
---------------------
 ["Fluffy", "Buddy"]
 ["Max"]
 ["Snowball"]
(3 rows)

3) Handling missing paths

If the specified path doesn’t exist in the JSONB data, the jsonb_path_query_array() function returns an empty array. For example:

SELECT jsonb_path_query_array(data, '$.address')
FROM employees;

Output:

jsonb_path_query_array
------------------------
 []
 []
 []
(3 rows)

In this example, the employee object doesn’t have an address key, so the result is an empty array.

Summary

  • Use the jsonb_path_query_array() function to query JSONB data using a JSON path and return matched elements as a JSON array.

Last updated on

Was this page helpful?