Data Mining using Python project repository

We recommend using virtualenv when setting up the application.

To install required modules, simply type:

python setup.py install

On some computers this command fails at first, because numpy is not installed correctly with setuptools. To get arround this issue install numpy directly with pip as pip install numpy==1.9.1 and afterwards execute setup.py install.

For optimization reasons, the course information is fetched from a MongoDB database, which is an requirement for running the application. In order to import courses into MongoDB from the xml, run the command:

python course_import.py

To configure the application, it is possible to either add an app.cfg file that contains the relevant configurations (see app.cfg.example for an example configuration file) or environment variables (see cv_kickstarter_config.py for environment variables used).

The app defaults to using MongoDB through localhost unless a MONGO_URL environment variable is given.

To run the web server after the setup, execute:

python webapp.py

or to run with gunicorn, execute:

gunicorn webapp:app --log-file=-

If the virtual env is activated for the first time, it might need to deactivated and activated again to use gunicorn.

The CV can also be exported in a json format through the CLI by using the command:

jsoncv s123456 secret

where 's123456' should be your student id and 'secret' be your password to CampusNet.

If the virtual env is activated for the first time, it might need to deactivated and activated again to use command line script.

The project uses py.test for testing, so to run tests, execute:

py.test

The project is also set up with tox and is tested against python 2.7, 3.4, pypy and pypy3. To run tox locally, type:

python setup.py test