Plugin does not have to be csharp specific

[rouke-broersma]

Hi! I am a maintainer of stryker-net and I just came across this. Awesome!

I notice that you market/name the plugin as being stryker-net specific but our json schema is actually a shared schema between a lot of different mutation testing tools. StrykerJS, Stryker4S, infectionphp.. etc and even pitest I believe have support for this schema!

I also see that you implemented your own file parser for reading the json but we actually have a Scala (should be compatible with Java I think?) package that already implements this that you could use: https://github.com/stryker-mutator/mutation-testing-elements/tree/master/packages/metrics-scala @hugo-vrijswijk could you perhaps comment on how usable the package is in a Java project?

[mstfymrtc]

Hi @rouke-broersma, I get that StrykerJS, Stryker4S.. etc. all use the same schema but the point is there are some limitations in Generic Issues that you've mentioned in the documentation.

Also, we've just tried to make the plugin compatible with both JS, C# and Scala but there are some problems with Stryker.NET. The JSON report which is produced by Stryker.NET contains a field which is mutatorName but that field is not an enum. For instance, StrykerJS uses enums for this field such as AssignmentOperator, ArithmeticOperator etc. Because Stryker.NET does not use specific mutator names, we can't create a rule for each mutator in Sonarqube.

If we could make mutatorName field in Stryker.NET enum just like the StrykerJS, we could create a single sonar plugin for JS,C# and Scala. What do you think?

[rouke-broersma]

Hi @mstfymrtc we have a proposal open for shared mutator name terminology here: stryker-mutator/mutation-testing-elements#1458. You could add your feedback there.

That said, from the point of the json schema the mutator name is not an enum. That strykerjs choses to implement this as an enum internally is not relevant. Mutator names could be different between all mutation testing frameworks until we have a common terminology. It will most likely never be the same for all mutation testing frameworks. This is why we explicitly don't enforce mutator names in the schema.

I also don't really see how the mutator name not being an enum matters for the rules, because all mutants are treated equally, I assume?

[mstfymrtc]

Thanks for the fast response @rouke-broersma :)

• It matters because if they would be enum, we can allow users to activate/disactivate mutator rules like that.

• It's also possible to define examples/description like below for each mutator.

In the current implementation of the plugin, we have just 2 different rules like below. (Because we can't categorize mutatorName)

[rouke-broersma]

Well the mutant names in stryker.net are static, they should not change unless we have a breaking change. They're just not placed in an enum data structure.