Closed emersion closed 6 years ago
I suspect Nat_num.natural predates Nat_big_num, which subsumes it. I don't think Natnum.natural is used anywhere. Hence, the right thing to do is to just remove type natural (and the natural* functions) from Nat_num, which will eliminate all Big_int references from Nat_num.
I suspect Nat_num.natural predates Nat_big_num, which subsumes it. I don't think Natnum.natural is used anywhere. Hence, the right thing to do is to just remove type natural (and the natural* functions) from Nat_num, which will eliminate all Big_int references from Nat_num.
Thanks for the review. Just done that, works, all tests pass.
This fixes the compilation on OCaml 4.06.
I've no idea why
nat_num.ml{i,}
were using directlyBig_int
over ourBig_int_impl
wrappers. There's a chance it always picked the ocaml-num implementation.Also
run-ocaml-tests.sh
was requiringnums.cma
- no idea why.All the tests pass with this patch.
There's some noise because of trailing whitespace chars, but w/e.