Mutation Testing with Stryker and Docker

Featured image

How do we know that we are writing quality code? Mostly we rely on unit tests or integration tests to answer that question. But then, the question repeats itself. How do we know that we are writing good tests? Let me introduce you Stryker, it allows you to test your tests with mutation testing. Stryker guys have a complete explanation and got mostly everything covered on their official site which you should visit.

Following an approach with Docker, I found it easier to implement in old projects since it is not necessary to upgrade libraries or care about the version of a testing framework. Just a few steps are needed.


We are going to replace Karma with Jest. Why should we do that? Well, this blog entry has a great explanation.


Ran all tests for this mutant.
Ran 0.80 tests per mutant on average.
File                   | % score | # killed | # timeout | # survived | # no cov | # error |
All files              |   46.67 |        4 |         3 |          8 |        0 |       0 |
 app.component.spec.ts |   46.67 |        4 |         3 |          8 |        0 |       0 |
15:53:38 (1) INFO DashboardReporter Dashboard report is not sent when not running on a build server
15:53:38 (1) INFO HtmlReporter Your report can be found at: file:///srv/reports/mutation/html/index.html
15:53:38 (1) INFO Stryker Done in 28 seconds.

And remember that:

Bugs, or mutants, are automatically inserted into your production code. Your tests are ran for each mutant. If your tests fail then the mutant is killed. If your tests passed, the mutant survived. The higher the percentage of mutants killed, the more effective your tests are. - Stryker Team.

Why is this awesome?

Mutation testing requires a lot of compute resources and it is time consuming. Delegating the process to a detached docker container makes the task easier since it required some dependencies as pre-requisite that might not be the best idea to have right now in your projects.

An example is hosted here and the dockerfile we used is here. If you have an idea about how to improve it, you are just about a Pull Request to cooperate.


To feel safe and make sure you don’t break anything during development, add tests to your project. Then, add mutation tests to verify the status of your tests and ensure their quality. If you don’t want to waste your time with dependencies and libraries, run the tests inside a docker container based on a custom image with everything you need.