Good functional programming habbit?
in [Lisp]
|
Prev: Searching some references
Next: Comparing Lisp to Python, what you consider more important:speed or macros.
From: Andreas Eder on 29 Apr 2010 17:42 Hi Xah, >>>>> "Xah" == Xah Lee <xahlee(a)gmail.com> writes: Xah> using local variable is fine. In lisp, it's like this: Xah> (let (var1 var2 ...) Xah> (setq var1 val1) Xah> (setq var2 val2) Xah> ... Xah> body Xah> ) You can do this easier and more succintly in the following way: (let ((var1 val1) (var2 val2) ...) ... body) 'Andreas -- ceterum censeo redmondinem esse delendam.
From: Captain Obvious on 30 Apr 2010 03:34 DV> I also find particularly annoying when people assume that functional = DV> purely functional. That's the kind of vocabulary fuzziness that leads DV> to comments I've heard too often like: Usually people who mention functional programming want to juxtapose it to imperative programming. If your definition of functional programming language includes CL and JS, that's not juxtaposition at all. So your definition destroys meaning. Term "functional language" becomes unusable, but maybe if it pleases lispers so they know that they still use functional programming language it's ok? It doesn't need to be 100% pure functional, but I think it should have at least some support for immutable data structures to qualify as functional. DV> <Haskell guy>: DV> Lisp? Ah yeah, that's this weird imperative language with lots of DV> parens... Well, I guess I'm more like a Haskell guy on this, except that I'm programming mostly in CL. But I know a bit of Haskell, and I see that there is an abyss between them.
From: Tim Bradshaw on 30 Apr 2010 05:18 On 2010-04-29 16:42:18 +0100, RG said: > You know, Kenny, the arrogant Lisp weenie act would go over a lot better > if you actually knew what you're talking about. > > ? (defun foo () (values)) > FOO > ? (defun baz () (multiple-value-call (lambda (x) x) (foo))) > BAZ > ? (baz) >> Error: THis is not really any different to (defun foo (x y) (cons x y)) (foo 1)
From: RG on 30 Apr 2010 06:52 In article <hre79f$ul2$1(a)news.eternal-september.org>, Tim Bradshaw <tfb(a)tfeb.org> wrote: > On 2010-04-29 16:42:18 +0100, RG said: > > > You know, Kenny, the arrogant Lisp weenie act would go over a lot better > > if you actually knew what you're talking about. > > > > ? (defun foo () (values)) > > FOO > > ? (defun baz () (multiple-value-call (lambda (x) x) (foo))) > > BAZ > > ? (baz) > >> Error: > > THis is not really any different to > > (defun foo (x y) > (cons x y)) > > (foo 1) Good grief, I'm surrounded by idiots. Of course it's different. Here you're calling a function of two arguments with one argument. That has nothing to do with whether or not (values) "really" returns NIL as Kenny claimed. It doesn't. What does happen is that CL implicitly "expands" zero values to NIL in some contexts -- but not all. And in particular, not in the context of multiple-value-call. And this has nothing to do with any compile-time optimizations, as Kenny also incorrectly claimed. ? (defun zero-values () (values)) ZERO-VALUES ? (defun tim () (cons (zero-values) (zero-values))) TIM ? (defun kenny () (multiple-value-call 'cons (zero-values) (zero-values))) KENNY ? (tim) (NIL) ? (kenny) > Error: Too few arguments in call to #<Compiled-function CONS #x3000000D771F>: And if that doesn't convince you: ? (defun maybe-zero-values (x) (if x (values) 'kenny-is-a-smug-lisp-weenie)) MAYBE-ZERO-VALUES ? (defun kenny2 (x) (multiple-value-call 'cons (maybe-zero-values x) (maybe-zero-values x))) KENNY2 ? (kenny2 nil) (KENNY-IS-A-SMUG-LISP-WEENIE . KENNY-IS-A-SMUG-LISP-WEENIE) ? (kenny2 t) > Error: Too few arguments in call to #<Compiled-function CONS #x3000000D771F>: rg
From: RG on 30 Apr 2010 07:16
In article <4bda8817$0$280$14726298(a)news.sunsite.dk>, "Captain Obvious" <udodenko(a)users.sourceforge.net> wrote: > DV> I also find particularly annoying when people assume that functional = > DV> purely functional. That's the kind of vocabulary fuzziness that leads > DV> to comments I've heard too often like: > > Usually people who mention functional programming want to juxtapose it to > imperative programming. Usually people who mention "data" want to distinguish it from "code". That does not change the fact that code and data are fundamentally indistinguishable, nor does it change the fact that artificial constraints can be imposed to make them distinguishable, nor does that change the fact that imposing such artificial constraints is generally considered a bad idea here on C.L.L. What people usually want ought not be confused with actual facts. > If your definition of functional programming language includes CL and JS, > that's not juxtaposition at all. > So your definition destroys meaning. Term "functional language" becomes > unusable, but maybe if it pleases lispers so they know that they still use > functional programming language it's ok? > > It doesn't need to be 100% pure functional, but I think it should have at > least some support for immutable data structures to qualify as functional. Lisp can "provide some support for immutable data structures" by doing, e.g.: (shadow '(rplaca rplacd setf)) It doesn't seem to me that your definition is exactly a model of semantic clarity. rg |