Sample video

Django_website_video.mp4

Overall diagram

Diagram's breakdown

Notes: This is the updated version, all the back-end jobs are compressed to the Huey worker, all of which will be executed by this worker when it recieves the POST request

Workflow:

1. Requests made by the users: When the users make the request to the webserver, it will be forward to the urls.py of Django framework
2. Mapping endpoint with the views function:
   • Views functions are Django's back-end functions, normally used to perform certain users-defined logic to response to the request coming from the users
   • Regarding the certain structure, the Views functions are mainly used to retrieve the parameters sent by the users, which in this case are the start_date, end_date, seller_id, and finally the launcher
3. View function perform according actions:
   • After receving the users' parameters, the view function will forward these paramaters to the Huey Task Queue Manager
     • The Huey Task Queue Manager is built to run asynchronously, meaning that this process will happen besides the main thread, allowing the views function to continue with its logic
     • Thanks to this feature, the view function is able to redirect the users back to the home page, otherwise, the page will continue loading until the task executed by the worker is done
   • After passing these params to the Huey Task Queue Manager, the views function can continue redirecting the users to the home page asynchronously
4. Huey Task Queue Manager puts the parameters with the assigned task to Redis Message Broker:
   • Normally, Redis is used as the cache database for the back-end, but in this case, it's used as the Message Broker, mostly used to store the queued tasks
   • Thanks to Redis, this will allow upcoming tasks not going to be missed, but rather stored on Redis and wait for idle workers (if any) to pick them up and execute the task
5. Huey Worker listens on Redis to pick up any upcoming tasks to execute:
   • Since we have the Huey Worker connected to the Redis server through the set up, it has been running in the back-end background
   • when the task arrived, it will be informed and pick up the task from Redis to execute