This post is about an update that I’ve made to the Stryker Runner Visual Studio Code extension I built enabling it to work much better with monorepos. For more information about the extension please see this post.
What is Stryker?
The Stryker project is a collection of libraries for languages including C# , Scala and Javascript that are used to run mutation tests against your codebase.
What is mutation testing?
Mutation testing is the programmatic changing or rewriting of parts of your code (a mutant) and validating whether the existing unit tests fail. The idea is that they do fail (killing the mutant), that is the theory anyway if you want to kill 100% of mutants.
Unit testing frameworks (like Jest for Javascript) can tell you how much of your code is covered by unit tests but this really just means which code gets executed under the tests. Unit testing frameworks cannot tell you how much of the desired behaviour is adequately tested and this information is a part of what mutation testing attempts to solve. The Stryker docs have a brief intro on this as well.
What is VS Code?
Visual Studio Code (VS Code) is free and widely used IDE developed by Microsoft. VS Code has access to an official marketplace of extensions that are used to add additional functionality, integrate libraries like Stryker and even add additional language support. There are also unofficial marketplaces that can be used instead of, or as well as, the official one, e.g. the open source marketplace Open VSX.
Stryker Runner: what’s changed?
The Stryker Runner extension has been useful for engineers that I have introduced it to and for those who have discovered it through other channels. Recently though we have started working with monorepos and have found that the experience using the extension was not quite as good as when working on single project repositories.
What was the problem?
When attempting to run Stryker against a file or code selection in a monorepo Stryker was more often than not unable to locate a relevant node_modules
folder that it could symlink into the Stryker sandbox. This was not an issue if you choose to mutate in place.
When preparing a sandbox Stryker works its way up the file tree from the current working directory from where the stryker
command was run looking for node_modules
folder. Unfortunately, the extension had been built to execute commands from the root of VSCode Workspace that the target file belonged to. In a monorepo there may or may not be a relevant node_modules
folder in the root of the Workspace depending on how the dependencies are being manged.
How did we solve it?
I’ve recently released an 0.0.47
release of Stryker Runner that changes its behaviour so that it no longer uses the root of the VSCode Workspace as the working directory from which to run the stryker
command. Instead it locates the nearest ancestor package.json
file to the target on which Stryker is going to be run and uses that as the current working directory.
This change is intended to ensure that when the Stryker Runner extension is used and Stryker works its way up the file tree from the current working directory looking for a node_modules
folder it should always find the appropriate set of dependencies for the target.
The source for the Stryker Runner extension is available on Github.
The Stryker Runner extension can be downloaded from:
- As
.vsix
package directly from Github - The VS Code marketplace.
- The Open VSX Registry
Get in contact
If you have comments, questions or better ways to do anything that I have discussed in this post then please get in contact.