Django and Langflow Integration

This project demonstrates how to set up a Django project that interacts with a Langflow instance locally. It includes WebSocket connections handled by Django Channels and the deployment of the application using Docker.

Prerequisites

• Python 3.x
• Django
• Docker
• Langflow

Setup and Running Locally

1. Create a Django Project

First, create a new Django project by running:

django-admin startproject project_name

2. Start the Django Backend on Localhost

Navigate to your project directory and start the development server:

python manage.py runserver

3. Create a New App for WebSocket Connections

Within your Django project, create a new app to handle WebSocket connections:

python manage.py startapp chat

4. Spin Up Langflow Locally

Run the Langflow service locally by executing:

langflow % python -m langflow run

5. Notes

If you encounter issues with WebSocket connections, refer to this helpful Stack Overflow discussion.

Docker Setup and Deployment

6. Build the Docker Images

Navigate to the appropriate directories and build the Docker images for the Django backend and Langflow service:

docker build -t your-docker-username/django-backend:latest .
docker build -t your-docker-username/langflow-service:latest .

7. Push the Docker Images to a Registry

Push the updated images to your Docker registry:

docker push your-docker-username/django-backend:latest
docker push your-docker-username/langflow-service:latest

8. Deploy the Stack Using Docker Compose

Deploy the stack using Docker Compose:

docker stack deploy -c docker-compose.yml mystack

Communicate with server

Get new tokens for existing user

In the body include {"username":"your username", "password":"your password"}

http://localhost:8000/api/token/

In the body include {"refresh":"token"}

http://localhost:8000/api/token/refresh/

Start react app

In the react-chat folder

npm start