easyPCE

A better course evaluation tool.

Quick Start

Make sure you have Python 2.7, pip, and PostgreSQL, and npm installed. Then, run:

$ pip install -r requirements/local.txt
$ python manage.py migrate
$ npm install
$ npm run build

To run the server:

$ python manage.py runserver

Scraping the registrar is pretty simple. Make sure that you have RabbitMQ and you have migrated the site.

On a Mac with Homebrew set up:

$ brew install rabbitmq
$ brew services start rabbitmq
$ python manage.py migrate

Then, start the worker and assign it tasks:

$ celery multi start Jones -A easypce -l info
$ python manage.py scrape --all

If you intend to develop on the build, you may enjoy hot code reloading on the client side. To set this up, run:

$ npm run watch &
$ python manage.py runserver

Additionally, if you want scraping functionality to work, you'll need too install Redis and Celery for task scheduling. Then, you can run this:

$ celery worker -A easypce -l info
$ python manage.py runserver

You can also do lots of fun stuff with concurrency if you use Celery multi:

$ celery multi start num_workers -A easypce -l info

This project uses a mostly custom folder structure as an almagamation of various starting templates, adapted for our specific purposes.

Folder	Description
`api`	REST API-related code
`assets`	created on `npm run build`, bundle outputs for production use
`easypce`	The project folder; Django settings and routes
`requirements`	Python requirements, separated by environment
`static`	Static assets, including client-side code
`webpack`	Webpack configuration

celery worker -A easypce -l info: Run a celery worker for the project.
celery multi start num_workers -A easypce -l info: Run a worker cluster for the project.
python manage.py runserver: Run a development server.
python manage.py scrape [--all] [--meta] [--terms (terms)]: Run scraping tasks on workers.
npm run build: Builds the client-side bundle for production use.
npm run build-local: Like above, but for local use.
npm run clean: Remove all generated files.
npm run watch: Launch a server that watches and serves client-side files.