PrairieLearn

PrairieLearn is an online problem-driven learning system. It consists of a server component that presents questions and other data via an API, and a webapp that interfaces with the user and communicates with the server.

Building

1. Install the pre-requisites:

• Node.js
• npm
• MongoDB

On OS X these can be installed with MacPorts or Homebrew.

2. Next clone the latest code:

    $ git clone https://github.com/PrairieLearn/PrairieLearn.git
   

3. Install the backend libraries:

    $ cd PrairieLearn/backend
    $ npm install
   

4. (Optional) Render question TeX image files for figures

    $ cd PrairieLearn/backend
    $ make
   

Running the server

1. Run the database:

    $ mongod --dbpath ~/db
   

2. Run the server:

    $ cd PrairieLearn/backend
    $ node server
   

3. In a web-browswer view the PrairieLearn/frontend/index.html file:

    $ open file://...<path>.../PrairieLearn/frontend/index.html
   

To test the database, you can access it directly:

$ mongo
> show dbs
> use data
> show collections
> db.users.find()
> db.users.save({uid: "user1@illinois.edu"})

To test the server, you can access it from the commandline:

$ curl http://localhost:3000/questions
$ curl http://localhost:3000/questions/scalarAdd

$ curl -H "X-Auth-UID: user1@illinois.edu" -H "X-Auth-Name: User Name" -H "X-Auth-Date: 2013-08-17T09:44:18Z" -H "X-Auth-Signature: 3d38a7acba63047cf8bcf29f9691c68a2cae30e3ae5057ef1ea4616d2060a4be" http://localhost:3000/users
$ curl -H "X-Auth-UID: user1@illinois.edu" -H "X-Auth-Name: User Name" -H "X-Auth-Date: 2013-08-17T09:44:18Z" -H "X-Auth-Signature: 3d38a7acba63047cf8bcf29f9691c68a2cae30e3ae5057ef1ea4616d2060a4be" http://localhost:3000/users/user1@illinois.edu

$ curl -H "X-Auth-UID: user1@illinois.edu" -H "X-Auth-Name: User Name" -H "X-Auth-Date: 2013-08-17T09:44:18Z" -H "X-Auth-Signature: 3d38a7acba63047cf8bcf29f9691c68a2cae30e3ae5057ef1ea4616d2060a4be" -H "Accept: application/json" -H "Content-type: application/json" -X POST -d '{"uid": "user1@illinois.edu", "qid": "scalarAdd"}' http://localhost:3000/qInstances

$ curl http://localhost:3000/questions/scalarAdd/1/question.html
$ curl http://localhost:3000/questions/scalarAdd/1/client.js
$ curl http://localhost:3000/questions/scalarAdd/1/params

$ curl -H "X-Auth-UID: user1@illinois.edu" -H "X-Auth-Name: User Name" -H "X-Auth-Date: 2013-08-17T09:44:18Z" -H "X-Auth-Signature: 3d38a7acba63047cf8bcf29f9691c68a2cae30e3ae5057ef1ea4616d2060a4be" -H "Accept: application/json" -H "Content-type: application/json" -X POST -d '{"uid": "user1@illinois.edu", "qid": "scalarAdd", "vid": "1", "qiid": "qi32", "submittedAnswer": {"c": "43"}}' http://localhost:3000/submissions

Useful curl options:

--trace
--trace-time
-w, --write-out
-v, --verbose

Accessing the client

To access the client we can either run a webserver locally or acess the files directly from disk.

Accessing files directly from disk

Simply open frontend/index.html in a web browser. Chrome will give cross-origin errors when loading non-javascript from disk, so it must be run with the --allow-file-access-from-files commandline argument. Under OS X this can be done with:

open '/Applications/Google Chrome.app' --new --args --allow-file-access-from-files

On Windows this can be done by adding the --allow-file-access-from-files flag to a special shortcut (edit it with right-click and properties -> target).

To disable cross-origin errors in Firefox go to about:config and set security.fileuri.strict_origin_policy to false.

Run a local webserver

Any webserver that can serve up the frontend directory tree can be used to access the client. One particularly easy way to do this is to run python -m SimpleHTTPServer in the frontend directory, and then point the webbrowser to http://localhost:8000.

Restore of the database

When the DB is stopped:

mongorestore --dbpath ~/db ~/path-to-dump

Miscellaneous Notes

Run grunt to check code

$ sudo npm install -g grunt-cli
$ grunt

Use marked to process markdown

$ sudo npm install -g marked
$ marked --gfm README.md > README.html

Tools and libraries

• Node.js
• npm
• MongoDB
• Express
• jQuery
• Underscore.js
• Backbone.js
• Bootstrap
• Grunt
• async
• JSHint
• RequireJS
• tween.js
• Sylvester
• MathJax
• Rivets.js

Deploying to a Vagrant virtual machine

Install:

• VirtualBox: http://www.virtualbox.org/
• Vagrant: http://www.vagrantup.com/

In the PrairieLearn directory run:

vagrant up            # download and boot the VM and install packages

To run the PrairieLearn server do:

vagrant ssh           # login to the VM
cd /vagrant/backend   # change to the PrairieLearn backend server directory
npm install           # install libary packages (one time only)
make                  # compile latex in figures (only after changing figure latex)
grunt                 # check code for syntax and style (optional)
node server           # run the PrairieLearn server

When the server is running, access PrairieLearn in a web browser on the host at the URL http://localhost:8080/

To stop the server and logout of the VM:

ctrl-c                # stop the server
logout                # logout of the VM (can also use ctrl-d)

Other useful vagrant commands on the host:

vagrant halt          # shutdown the VM
vagrant destroy       # delete the VM completely
vagrant up            # start the VM again (after halt or destroy)
vagrant suspend       # suspend the VM
vagrant resume        # resume a suspended VM