Simple, Sane Course Scheduling.
YACS is a web-based course schedule with an emphasis on usability. It is also flexible enough to work with other school course data.
Send questions and comments to @jeffhui or huij@rpi.edu.
If you're an RPI student and looking to use it, go to yacs.me
YACS is actively developed on Python 2.7. So go get that installed first. It's good python practice to put this project inside a virtualenv, but that's beyond the scope of this README.
You can download an archive from the top-right of the github page or clone the repo:
git clone git://github.com/jeffh/YACS.git
Which will download code into a YACS folder where you run this command.
YACS is built on top of Django 1.4rc1. Thus, it requires a database driver to run. Install the appropriate driver and its database, or just use the bundled SQLite.
-
YACS uses a lot of dependencies. It relies on pip to install them. Simply do:
pip install -r requirements/development.txt
Which will install all the dependencies YACS needs (minus the database driver). -
Edit
s.DATABASES
variable in theyacs/settings/development.py
file to your appropriate database settings. -
Run the following commands. When calling syncdb, you'll be ask to create a superuser, it is purely optional, only the debug-toolbar is visible for logged in super-users.
python manage.py syncdb python manage.py migrate
-
Next we need to get some data. Run these commands to import and setup the data for use. (These will take awhile).
python manage.py import_course_data # imports from RPI SIS python manage.py import_catalog_data # imports from RPI course catalog python manage.py create_section_cache # creates cache for generating schedules
-
Check it out by running the dev server
python manage.py runserver
and pointing your browser to http://localhost:8000/ and viola! -
(Optional) If you plan on editing the CSS. YACS uses SASS to generate the CSS. Once you installed it, use
make scss
to make sass automatically update the stylesheets as you change them. -
???
-
Profit?
Currently the project is laid out as follows:
- api: API application. Where all API related code is. Relies on courses and scheduler app.
- courses: The courses application. Contains the schema and manages the storage of course data. Also contains course-data-displaying views.
- courses_viz: An application that stores visualization of course data. (Part of Introduction to Visualization class)
- jslog: An application that can record basic JS data for debugging.
- lib: Contains library code that can, be potentially, separated into an independent project. To enforce this separability, this folder is added to the sys.path for absolute imports
- scheduler: This app handles course scheduling. Relies on the courses app for all the course data.
- yacs: Project files. Contains settings, root urls, templates, static files, etc.
- test_reports: Only appears when tests are executed. Used to see the test coverage.
- requirements: Contains various requirement files for PIP.
- Makefile: Used for running tests, cleaning python caches and deployment.
- manage.py: Django's CLI.
This project is still evolving. There are still issues to tackle. Go to the GitHub issues page to see them all.