Retrieve the source

$ git clone https://github.com/antoviaque/languagelearning.git

You will need a version of distribute >= 0.6.28. If you use https://github.com/brainsik/virtualenv-burrito you can upgrade with this command:

$ ~/.venvburrito/bin/virtualenv-burrito upgrade

Install the dependencies:

$ cd languagelearning
$ mkvirtualenv languagelearning
$ pip install -r requirements.txt

Download the corpus data for the nltk library:

$ make update-nltk

Edit the configuration file:

$ cp local_settings.py.dist local_settings.py
$ gvim local_settings.py

To set the GOOGLE_API_KEY variable:

• Go to https://code.google.com/apis/console
• Get the API key at the bottom of the page
• Activate Google Translate in "Services"
• Setup billing

To set the BING_API_KEY variable:

• Go to http://www.bing.com/developers/
• Click on "Search API"
• Click on "Sign up" for 5000 requests (free plan)

To run the development server (loading static files directly from the /static directory and avoid having to reload the server during development):

$ ./manage.py runserver

Then go to http://localhost:8000

To compile static assets, run

$ make build

This will compile and uglify all assets into static/build. You will then need to set up a proxy to serve all assets at static from static/build. The proxy is responsible for gziping all content if you want to have it compressed.

phantomjs is used for automated headless browser testing. To run these tests, you'll need to make sure you've got phantomjs. On Mac, that would be:

$ brew install phantomjs

On debian/ubuntu, you will need to install phantomjs from source to have the latest version (the repository currently contains 1.6.0). Download the source from http://phantomjs.org/download.html and create a symlink to the bin/phantomjs binary from your path.

Then, in the root project directory:

$ npm install mocha-phantomjs mocha

Finally, in the root project directory (while the test server is running on port 8000):

$ make test-client

To run all tests:

$ ./manage.py test

To run only the application tests (excluding Django & external packages):

$ ./manage.py test languagelearning

To run a specific test:

$ ./manage.py test languagelearning.ASampleTestClass.test_sample_test