Improvements to property-based testing

Lead Research Organisation: Imperial College London
Department Name: Dept of Computing


focus on
evaluating and further developing these ideas.
The primary questions I wish to answer are:
How effective at finding real bugs is property-based testing?
What factors influence that effectiveness?
How can we make it more effective?
Why Hypothesis?
Hypothesis would make a good vehicle for this project for a number of reasons. Principally:
1. It is a mature project where the main engineering aspects have already been settled,
allowing me to focus almost exclusively on the research aspects.
2. There is a large and growing body of open source code already tested using it. This
provides an excellent source of empirical data for evaluating changes for their effect
on the tests that people write in practice, rather than focusing on artificial examples.
3. The public Hypothesis data generation API is declarative, with little control over the
distribution. This provides a great deal of freedom to experiment with the engine.
These combine to make Hypothesis an excellent testbed for the practical implementation
and evaluation of research on this topic.

Research area: testing and verification


10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/R512540/1 01/10/2017 30/09/2021
2286962 Studentship EP/R512540/1 01/10/2017 31/03/2021 David Ritchie Maciver