R

React Dj - A open source Blog Management System

Publish anywhere , Manage here.

The problem React Dj - A open source Blog Management System solves

We started sharing our journey of building scalable and maintainable web applications using Django and React on different sites like Medium, Reddit, Quora, Facebook Groups etc. just two weeks ago. We and the readers faces lot of issues following and managing those articles , beacause each time we share some article we have to give lot of urls like refer this and that , on the other end readers were unable to track those articles and their progress. One of challenging thing for readers , how to connect with us? .There are lot of platforms but no one provides the Support Chat Bot kind of things . So we built a Open Source Blog Management System , any one can use with giving us some credits , which supports multiple features like User Registration through Email Verification Code , Secured JWT Based Login and API Authentication , Mark as Read for the posts you completed and provide rating for the posts.
Our favourite feature Zulip Chat Bot Support in which you can interact with zulip bot and ask your queries like What is my progress? , Top rated blogs ? , Job opportunities ( It will recommend the openings based on your reading habits ).
That's it , We will be working hard to launch this as a product.

Challenges we ran into

Some of the challenges we ran into while building the project were:-

  1. Configuration of Django-server to run React bundles: We solved this problem by writing webpack configurations for production and webpack-django-loader to serve the static bundles.
  2. Tracking progress (reading blogs) for various users.
  3. For added security and reduce spamming of "Mark as Read" feature, we implemented E-Mail OTP system on all important actions. (Register, Mark as Read/Assign Rating) etc. Sweet-Alerts were used to capture OTP and display messages.
  4. Employing Natural Language Processing (Semantic Analysis) to comprehend user commands on zulip-bot. The commands were not hard-coded and thus their meaning needed to be extracted and associated callback methods fired to deliver a response.
  5. Deployment of Zulip-bot on Zulipchat, subscribing to streams, listen on message by a user and perform response. (call_on_each_message functionality).

Technologies used

Discussion