JSON

PostgreSQL supports JSON functions and operators which gives flexibility when storing data inside a database column.

PostgreSQL supports two types of JSON columns: JSON and JSONB.

The recommended type is JSONB for almost all cases. When you use the JSONB format, the data is parsed when it's put into the database so it's faster when querying and also it can be indexed.

Steps#

Create a table with a JSON column.#

create table books (
  id serial primary key,
  title text,
  author text,
  metadata jsonb
);

Insert data into the table.#

Data
SQL
UI
JS

id	title	author	metadata
1	The Poky Little Puppy	Janette Sebring Lowrey	{"ages":[3,6],"price":5.95,"description":"Puppy is slower than other, bigger animals."}
2	The Tale of Peter Rabbit	Beatrix Potter	{"ages":[2,5],"price":4.49,"description":"Rabbit eats some vegetables."}
3	Tootle	Gertrude Crampton	{"ages":[2,5],"price":3.99,"description":"Little toy train has big dreams."}
4	Green Eggs and Ham	Dr. Seuss	{"ages":[4,8],"price":7.49,"description":"Sam has changing food preferences and eats unusually colored food."}
5	Harry Potter and the Goblet of Fire	J.K. Rowling	{"ages":[10,99],"price":24.95,"description":"Fourth year of school starts, big drama ensues."}

View the data.#

SQL
JS
Data

select *
from books;

id	title	author	metadata
1	The Poky Little Puppy	Janette Sebring Lowrey	{"ages":[3,6],"price":5.95,"description":"Puppy is slower than other, bigger animals."}
2	The Tale of Peter Rabbit	Beatrix Potter	{"ages":[2,5],"price":4.49,"description":"Rabbit eats some vegetables."}
3	Tootle	Gertrude Crampton	{"ages":[2,5],"price":3.99,"description":"Little toy train has big dreams."}
4	Green Eggs and Ham	Dr. Seuss	{"ages":[4,8],"price":7.49,"description":"Sam has changing food preferences and eats unusually colored food."}
5	Harry Potter and the Goblet of Fire	J.K. Rowling	{"ages":[10,99],"price":24.95,"description":"Fourth year of school starts, big drama ensues."}

NOTICE: The data as it appears here will have the JSONB fields in a different order than when we inserted them. As we said earlier about the differences between the JSON and JSONB fields:

When you use the JSONB format, the data is parsed when it's put into the database...

Query the `JSONB` data.#

Select title, description, price, and age range for each book.#

SQL
JS
Results

select
  title,
  metadata -> 'description' AS description,
  metadata -> 'price' as price,
  metadata -> 'ages' -> 0 as low_age,
  metadata -> 'ages' -> 1 as high_age
from
  books;

title	description	price	low_age	high_age
The Poky Little Puppy	Puppy is slower than other, bigger animals.	5.95	3	6
The Tale of Peter Rabbit	Rabbit eats some vegetables.	4.49	2	5
Tootle	Little toy train has big dreams.	3.99	2	5
Green Eggs and Ham	Sam has changing food preferences and eats unusually colored food.	7.49	4	8
Harry Potter and the Goblet of Fire	Fourth year of school starts, big drama ensues.	24.95	10	99

Data Types#

One last thing to note: the -> operator returns JSONB data. If you want TEXT/STRING data returned, you need to use the ->> operator.

metadata -> 'description' (this returns a JSON object)
metadata ->> 'description' (this returns STRING/TEXT data)