Andre Kuhlenschmidt
Andre Kuhlenschmidt
I could go either way on allowing/disallowing this behavior. Allowing it, in general, would require a dynamic check of the non-existent runtime type information. Which means we would have to...
This seems like a fundamental problem with the way tuple coercions are currently implemented. Good to know! We will need to talk about this on Tuesday.
The type based casts implementation is a bug. The coercion is not a bug, but is unintuitive. You are starting with a Bool vector casting to a Dynamic vector and...
- I think a lot of our focus should be on having obvious unit testing in any new / modified development. - Any modification of a function needs to add...
Here is another program that we should make sure that the type-checker and runtime rejects. ```racket (ann (vector 1 (tuple 1 #f)) (Vect (Tuple Int))) ``` Here the contents of...
That sounds like an excellent way of fixing this.
@aryx Could I get you to unassign this issue from me?