Working with Split Part in Postgres

04.04.2022

Intro

Postgresql split_part function allows us to split a string in to a number of substrings. This function help when formatting columns for a specific query. In this article, we will learn how to use Split Part in Postgresql.

The Syntax

The basic syntax of a Split Part is as follows:

SELECT split_part(string, delimiter, position)
FROM [table];

Getting Setup

We will be using docker in this article, but feel free to install your database locally instead. Once you have docker installed, create a new file called docker-compose.yml and add the following.

version: '3'
 
services:
  db:
    image: 'postgres:latest'
    ports:
      - 5432:5432
    environment:
      POSTGRES_USER: username
      POSTGRES_PASSWORD: password
      POSTGRES_DB: default_database
    volumes:
      - psqldata:/var/lib/postgresql

  phpmyadmin:
    image: phpmyadmin/phpmyadmin
    links:
      - db
    environment:
      PMA_HOST: db
      PMA_PORT: 3306
      PMA_ARBITRARY: 1
    restart: always
    ports:
      - 8081:80

volumes:
  psqldata:

Next, run docker-compose up.

Now, navigate to http://localhost:8081/ to access phpMyAdmin. Then log in with the username root and pass root_pass.

Click the SQL tab and you are ready to go.

Creating a DB

In this article, we will need some data to work with. If you don't understand these commands, don't worry, we will cover them in later articles.

We will be using the sample db provided here: https://dev.mysql.com/doc/sakila/en/. However, we will only enter what we need rather than import the whole db.

Next, let's create an film table. This is a slightly simplified version of the sakila database.

CREATE TABLE employees (
    emp_no      INT             NOT NULL,
    birth_date  DATE            NOT NULL,
    first_name  VARCHAR(14)     NOT NULL,
    last_name   VARCHAR(16)     NOT NULL,
    gender      VARCHAR(1),
    hire_date   DATE            NOT NULL,
    PRIMARY KEY (emp_no)
);

Now, let's enter a few rows

INSERT INTO employees VALUES (10001,'1953-09-02','Georgi','Facello','M','1986-06-26'),
(10002,'1964-06-02','Bezalel','Simmel','F','1985-11-21'),
(10003,'1959-12-03','Parto','Bamford','M','1986-08-28'),
(10004,'1954-05-01','Chirstian','Koblick','M','1986-12-01'),
(10005,'1955-01-21','Kyoichi','Maliniak','M','1989-09-12');

An Example

Let's try an example by splitting the birth date of the employees. To start, let's see how the split part function work.

select 
  split_part(birth_date::TEXT, '-', 1) as year
from employees

Here we force birth date to be a text type, we split on a hyphen, and we return the first result. We get the following.

year
1953
1964
1959
1954
1955

Now, let's do a slightly more complex example where we split each of the date parts into year, month, and day.

select 
	split_part(birth_date::TEXT, '-', 1) as year,
	split_part(birth_date::TEXT, '-', 2) as month,
	split_part(birth_date::TEXT, '-', 3) as date
from employees;

The only change between each column is the last parameter, the index of that split part. We have 1, 2, 3 which correspond to the date string split by the hyphen. You can think of the result from split_part as [1953, 09, 02].

The result is as follows.

year month date
1953 09 02
1964 06 02
1959 12 03
1954 05 01
1955 01 21