ensure no abstractions leak through after lambda-lifting (for formulas with higher-order occurrences of quantifiers)
structure Nat =
struct
datatype nat = Suc of nat | Zero_nat;
val one_nat : nat = Suc Zero_nat;
fun plus_nat (Suc m) n = plus_nat m (Suc n)
| plus_nat Zero_nat n = n;
fun times_nat (Suc m) n = plus_nat n (times_nat m n)
| times_nat Zero_nat n = Zero_nat;
end; (*struct Nat*)
structure Codegen =
struct
fun fac (Nat.Suc n) = Nat.times_nat (Nat.Suc n) (fac n)
| fac Nat.Zero_nat = Nat.one_nat;
end; (*struct Codegen*)