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
Currently when a test throws a runtime exception, we catch that and return a test failure. (Good.)
But we don't shrink the input.
This should be doable: instead of
try {
do {
input = fuzz()
test result = runTest(input)
} while test result is not failure and attempt count limit is not reached
if test result is failure {
test result = shrink(test result)
}
return test result
} catch {
return failure
}
we could do:
do {
input = fuzz()
try {
test result = runTest(input)
} catch {
test result = failure
}
} while test result is not failure and attempt count limit is not reached
if test result is failure {
test result = shrink(test result)
}
return test result
so that crash failures behave the same way as other failures, with "Given $INPUT:" reported, that input properly shrunk, etc.
The text was updated successfully, but these errors were encountered:
Currently when a test throws a runtime exception, we catch that and return a test failure. (Good.)
But we don't shrink the input.
This should be doable: instead of
we could do:
so that crash failures behave the same way as other failures, with
"Given $INPUT:"
reported, that input properly shrunk, etc.The text was updated successfully, but these errors were encountered: