nor is killRatioMustImprove respected either. I tested a drop of 5% coverage and it still passes the build. The unit tests on this project punch through to protected constructors and as such false-pass these issues, but the real code appears to be faulty in that the constructor always sets killRatioMustImprove to false and minimumKillRatio to zero. the subsequent setters set the public variables to the new values but its too late as the build condition arrays were already set to 0+false.
The commit that appears to have damaged this was in 1.0-18: Replace DataBoundConstructor with DataBoundSetter · jenkinsci/pitmutation-plugin@12635a9 · GitHub
note by doing this the constructor lost its parameters and as such the conditions are now built too early. If the unit tests were changed at this time to reference the public APIs then the fault would have been seen. Instead the constructor was marked as protected, so that the unit tests would continue to pass without changes.
nor is killRatioMustImprove respected either. I tested a drop of 5% coverage and it still passes the build. The unit tests on this project punch through to protected constructors and as such false-pass these issues, but the real code appears to be faulty in that the constructor always sets killRatioMustImprove to false and minimumKillRatio to zero. the subsequent setters set the public variables to the new values but its too late as the build condition arrays were already set to 0+false.
The commit that appears to have damaged this was in 1.0-18: Replace DataBoundConstructor with DataBoundSetter · jenkinsci/pitmutation-plugin@12635a9 · GitHub
note by doing this the constructor lost its parameters and as such the conditions are now built too early. If the unit tests were changed at this time to reference the public APIs then the fault would have been seen. Instead the constructor was marked as protected, so that the unit tests would continue to pass without changes.