The project is written in Python and requires Python 2.7 to be installed. It uses a PostgreSQL background so PostgreSQL 9.4.2 or higher must be installed.

This project has the following dependencies:

$ pip install flask pyparsing sqlalchemy pycurl slate sklearn nltk pandas numpy scipy psycopg2

For the scraper:

$ pip install beautifulsoup

Rename src/config.py.sample to src/config.py and update the variables inside accordingly. PAPER_DIR is a directory to download the exam papers to.

Import the database dump in data/dumps/exam_papers.sql using the psql tool:

$ psql -f data/dumps/exam_papers-29112015.sql

• pycurl requires libcurl to be installed.
  • On debian: sudo apt-get install libcurl4-openssl-dev
• pandas requires python-dev.l
• nltk requires running nltk.download() to download it's files. Type d and download the stopwords dataset.
• slate is broken with the latest version of it's dependency, PDFMiner. Fix it by running sudo pip install --upgrade --ignore-installed slate==0.3 pdfminer==20110515.

Once you have all the dependencies installed and database running, it's simple a matter of starting the server and visiting http://localhost:5000/.

$ pwd
/downloads/ct422-project
$ cd ..
$ python -m project.src.web.api
 * Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)

Note: To start the server, you must run the python command from the parent directory of the repository.