Description

A web app that follows the general structure of a photo-sharing based social media platform, with sentiment analysis based classification of comments, which get highlighted as red if they are positive, blue if they are negative and similar shades in between.

Final view can be seen at : https://project-karma-dk.herokuapp.com/

Initial Steps

• Create a virtual environment using:

python -m venv venv\

• In the environment install dependencies from requirements.txt

pip install -r requirements.txt

• The vader_lexicon might need to be installed seperately for local development

Firebase :

Get config files for firebase and use them in

• app/static/js/firebase_init.js
• app/main/utils/firebase_init.py
• app/main/utils/project-firebase.json

Reference structure has been kept here to make setting up easier. Note : The .py file imports the .json file as well Refer official documentation for firebase_sdk for python and using firebase on web.

• Create a migration using the following commands :

• python manage.py db init

• python manage.py db migrate

• python manage.py db upgrade

Important steps while deploying

• In the migrations/ folder, put an empty .keep file in versions/ if initally there is no migration as the folder would then not be created
• Change secret key in config.py

TODO

• Update access rules to firestorage to allow only authenticated reads (Updated this, still not sure)
• Add friends feature in relational db
• Make the UI better, as some images are displayed smaller due to rigid flexbox rules
• Update Navbar text to be on the right side (d-flex solution tried-> Failed)

NAVIGATION

• Regsiter at /register, this takes you to the login page
• At login page enter email id and password
• You arrive at upload_files/, just upload an image, this is where a user would make a new post
• Visiting "/all" would get images/posts, new posts get updated on the home page (atleast one is visible always)