What do you mean? Nothing in that post is related to Lisp parenthesis syntax. It is a post about expressing a custom mathematical syntax in valid Haskell.(Which is obviously a clever game and not a meaningful endorsement of the syntax, since statement syntax is not the point of McCarthy's paper.)

 I mean that the Haskell version is not anywhere close to being type safe. If you change almost all of the [] to () you can achieve a great deal of type safety at no extra cost. Having said that, there's still a lot of low-hanging type safety fruit remaining! (I also took the liberty of removing Lambda, which was redundant). diff (y, x) = atom (y) → (eq (y, x) → ONE \$ ZERO) \$ eq (car (y), PLUS) → cons (PLUS, maplist (cdr (y), \z -> diff (car (z), x))) \$ eq (car (y), TIMES) → cons (PLUS, maplist (cdr (y), \z -> cons (TIMES, maplist (cdr (y), \w -> z /= w → car (w) \$ diff (car (w), x))))) \$ error ("Unexpected term" ++ show y) mctest1 = diff (List [TIMES, X, List [PLUS, X, A], Y], X) data SExpr = ZERO | ONE | PLUS | TIMES | X | A | Y | List [SExpr] deriving (Eq, Show) car (List t) = head t car _ = error "car of non-pair" cdr (List t) = List (tail t) cdr _ = error "cdr of non-pair" cons (h, List t) = List (h : t) cons _ = error "cons with non-list" -- our lists aren't pairs atom (List _) = False atom _ = True eq (x, y) = x == y maplist (List x, f) = List (ml x) where ml [] = [] ml x@(_:t) = f (List x) : ml t infixl 1 → (→) :: Bool -> a -> a -> a (→) True x _ = x (→) False _ y = y

Search: