You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have created a instrument_classes file with the classes that i want to be mutated, but at the end, all mutants listed in mutatnts.log file belong to the class modified by the fix. In version 51 its org.apache.commons.lang.BooleanUtils.
Im runing the following comand for the checkout of version 51f: defects4j mutation -i classes.src
This seems to be caching issue. For efficiency reasons, the classes are not re-mutated after the first execution of defects4j mutation, which is useful when running multiple test suites against the same set of mutants. The cache of mutated classes, however, needs to be updated if setting the "-i" flag changes the set of classes that should be mutated. This is apparently not the case and clearly a bug.
As a workaround, the "-i" option is working properly on a fresh working copy (i.e., defects4j checkout followed by defects4j mutation -i ...).
I have created a instrument_classes file with the classes that i want to be mutated, but at the end, all mutants listed in mutatnts.log file belong to the class modified by the fix. In version 51 its org.apache.commons.lang.BooleanUtils.
Im runing the following comand for the checkout of version 51f:
defects4j mutation -i classes.src
classes.src contains the following:
It results in 391 generated mutants:
Which is the same result as running
defects4j mutation
without the "-i" parameterThe text was updated successfully, but these errors were encountered: