How to Use MQTT in The Django Project

MQTT is a lightweight IoT messaging protocol based on publish/subscribe model, which can provide real-time reliable messaging services for connected devices with very little code and bandwidth. It is widely used in industries such as IoT, mobile Internet, smart hardware, Internet of vehicles, and power and energy.

Django is an open-source Web framework and one of the more popular Python Web frameworks. This article mainly introduces how to connect, subscribe, unsubscribe, and send and receive messages between MQTT clients and MQTT brokers in the Django project.

We will write a simple MQTT client using the paho-MQTT client library. paho-mqtt is a widely used MQTT client library in Python that provides client support for MQTT 5.0, 3.1.1, and 3.1 on Python 2.7 and 3.x.

Project Initialization

This project uses Python 3.8 for development testing, and the reader can confirm the version of Python with the following command.

$ python3 --version
Python 3.8.2

Install Django and paho-mqtt using Pip.

pip3 install django
pip3 install paho-mqtt

Create a Django project.

django-admin startproject mqtt-test

The directory structure after creation is as follows.

├── manage.py
└── mqtt_test
  ├── __init__.py
  ├── asgi.py
  ├── settings.py
  ├── urls.py
  ├── views.py
  └── wsgi.py

Using Paho-MQTT

This article will use free public MQTT Broker provided by EMQ. The service is created based on MQTT Cloud service - EMQX Cloud. The server access information is as follows:

• Broker: broker.emqx.io
• TCP Port: 1883
• Websocket Port: 8083

Import Paho-MQTT

import paho.mqtt.client as mqtt

Writing Connection Callback

Successful or failed MQTT connections can be handled in this callback function, and this example will subscribe to the django/mqtt topic after a successful connection.

def on_connect(mqtt_client, userdata, flags, rc):
   if rc == 0:
       print('Connected successfully')
       mqtt_client.subscribe('django/mqtt')
   else:
       print('Bad connection. Code:', rc)

Writing Message Callback

This function will print the messages received by the django/mqtt topic.

def on_message(mqtt_client, userdata, msg):
   print(f'Received message on topic: {msg.topic} with payload: {msg.payload}')

Adding Django Configuration Items

Add configuration items for the MQTT broker in settings.py. Readers who have questions about the following configuration items and MQTT-related concepts mentioned in this article can check out the blog The Easiest Guide to Getting Started with MQTT.

This example uses anonymous authentication, so the username and password are set to empty.

MQTT_SERVER = 'broker.emqx.io'
MQTT_PORT = 1883
MQTT_KEEPALIVE = 60
MQTT_USER = ''
MQTT_PASSWORD = ''

Configuring the MQTT Client

client = mqtt.Client()
client.on_connect = on_connect
client.on_message = on_message
client.username_pw_set(settings.MQTT_USER, settings.MQTT_PASSWORD)
client.connect(
    host=settings.MQTT_SERVER,
    port=settings.MQTT_PORT,
    keepalive=settings.MQTT_KEEPALIVE
)

Creating a Message Publishing API

We create a simple POST API to implement MQTT message publishing.

In actual applications, the API code may require more complex business logic processing.

Add the following code in views.py.

import json
from django.http import JsonResponse
from mqtt_test.mqtt import client as mqtt_client


def publish_message(request):
    request_data = json.loads(request.body)
    rc, mid = mqtt_client.publish(request_data['topic'], request_data['msg'])
    return JsonResponse({'code': rc})

Add the following code in urls.py.

from django.urls import path
from . import views

urlpatterns = [
    path('publish', views.publish_message, name='publish'),
]

Start the MQTT Client

Add the following code in __init__.py.

from . import mqtt
mqtt.client.loop_start()

At this point, we have finished writing all the code, and the full code can be found on GitHub.

Finally, execute the following command to run the Django project.

python3 manage.py runserver

When the Django application starts, the MQTT client will connect to the MQTT Broker and subscribe to the topic django/mqtt.

Testing

Next, we will use an open-source cross-platform MQTT client - MQTT X, to test connection, subscription, and publishing.