Open oflatt opened 3 months ago
Will this be too radical? Will a warning / opt-in feature be more appropriate? For instance, in eggcc, we have templates for generating query atoms like (= e (Add e1 e2))
and now all of a sudden we need to condition on whether e,e1,e2
are used in the body, which is a burden to users.
Actually, reopening this issue since this can be part of a (set-option strict)
similar to #325
If you write a rule with an unused pattern variable, we should throw an error:
If you want to ignore something, use an underscore