Years ago I found a fantastic metric implemented by mixing cyclomatic complexity + code coverage. (crap4j) It feels like it might be targetting the same thing your tool it aiming at. And it generally yielded better code by trying to improve that metric. https://blog.ndepend.com/crap-metric-thing-tells-risk-code/
My team uses mostly Intellij for development. It would be very useful to have an extension for the IDE.
Support test file rating mode
mocha test files includes many describe, it,... statements nested in each other, should we rate those files differently...?
Track 'Testing' as a metric of Tech Debt
Working on a medium Angular 8 app where "shipping software" was prioritized over "shipping tested software." Most of the unit tests are either the bare-bones shell cranked out by the CLI or are skipped/commented out. E - No unit test file exists for that component. D - Only the CLI-generated spec.ts file exists. C through A - Each testable method (public with return type) has a test, each HTML element with an ID has a test, etc.
Support for React
Currently, the whole React component is seen is one excessively large function which always gets a bad rating. It'd be good to have the React component treated separately or at least have the individual functions within the React component assessed separately. Thanks.