There is a script that can scrape the CSE department of UCSD for courses, their names, number of unites, descriptions, and requirements. After scraping, a database is created. The scraping is implemented in the file scrape.pl and can be run as a script from the command line:

$ swipl -s builddb.pl

This creates a "database" file, db.pl, used by the backend.

At the moment, the backend exposes one predicate, courses/1. It can be used to get a list of all courses. The result is a list with one element per course. Each element is a term course/5, with arguments:

1. Course ID, an atom
2. Course Title, an SWI7-style string
3. Units, a term (see below)
4. Text description, an SWI7-style string
5. Prerequisites, an atom (for now)

Units are represented as one of three terms:

• exactly(Units): a course is worth Units units (an integer)
• from_to(From, To): a course is worth anything from From to To units
• one_of(List): a course is worth one of the integer values in List

The user should decide how to represent these for a human reader.