Kangaroux / imgtweets

Image browser for Twitter. Easily view the images a user has posted to their timeline. React + DRF. https://imgtweets.com/

imgtweets - Twitter Image Viewer

Check it out here!

The goal of this project is to scrape images from Twitter and provide a website for viewing those images.

Twitter is a great platform for sharing but terrible for searching. It can also be difficult to find something in a sea of tweets.

Local Dev Setup

This project requires the following:

• Python 3.10
• Node 14 (and yarn)

Docker is used for production but isn't necessary for development.

Frontend

The frontend uses create-react-app. It runs a development server and serves the frontend assets at http://localhost:3000. This is the URL you should visit to view the site.

cd frontend
yarn
yarn start

Backend

The backend uses Django and DRF. It runs a development server at http://localhost:8000 and a browsable API at http://localhost:8000/api/.

pip install -r requirements.txt
cd server
./manage.py runserver

Production

The app is intended to be ran in Docker for production.

Production assets are currently built outside of Docker. You will need to follow the dev setup steps first.

# Update .env with your settings
cp .env.example .env

# Build production assets and collect everything to dist/
./tasks.sh build

# Start everything and serve the site on port 80
docker-compose up --build

Scraping Process

The scrape process is relatively simple.

1. Provide the scraper a username
2. The scraper fetches tweets from the user's timeline
3. Fetched tweets are inspected for media, i.e. images and videos
4. Media URLs are collected and added to a database for caching

A REST API provides access to the database for easy searching.