Last year I shipped a few untyped ES6 JavaScript node.js services with 100% code coverage, 100% statement coverage, 100% branch coverage
-
-
Interesting. How would scope work? If you only mutate locally and then leave the function, does the taint follow?
-
No idea, but it’d be fun to find out… I haven’t really thought this through very deeply, I just like the idea :)
End of conversation
New conversation -
-
-
And I still want to make it happen, assuming I can find the time.
-
My mutation testing tool for Ruby code runs some axioms for the direction of the mutations but only for value computations, I'm curious if these map well to types.
- 2 more replies
New conversation -
-
-
We did this! "How Good are Your Types?" at Mutation'17. The results weren't great, but it was in a relatively type-impoverished setting (MutPy). Would be very interesting to try in a richer setting like Idris... cc:
@_rahulgopinath http://web.engr.oregonstate.edu/~walkiner/papers/mutation17-how-good-are-your-types.pdf …Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
It would be fun to mutate types to find more general signatures i.e. Int -> Int to a -> a.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
And for crazy dynlangs like OP used here, mutation testing is IMO the only option to get any kind of "at least something" correctness.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.