Writing tests for code that represents legislation is a bit fiddly.

Ideally we’d have perfect test cases but as noted in Overview there are concrete examples of a candidate rejecting their nomination, pushing back the publishing of the SoPN papers.

This project has two sets of tests:

  • Unit tests, with single-specified examples
  • Approval tests, with test data sourced from parsing historic SoPNs (provided by Democracy Club, who maintain a database of candidates and elections)