Lisp

55 readers

3 users here now

founded 1 year ago

MODERATORS

rglullis

defmacro

variable binding (alien.top)

submitted 1 year ago by InevitableCurrent310@alien.top to c/lisp

5 comments fedilink hide all child comments

I'm learning common lisp and the subtleties of scoping really troubles me. For example, running this piece of ("double-let") code

(let ((x 10))
  (funcall
    (let((x 20))
      (lambda () x ))))

gives me 20 in Clisp and 10 in SBCL. I'm not even sure this is a dynamic/lexical scoping issue, because a typical example that demonstrates scoping would concern special variables defined with defvar/defparameter, like

(defvar *x* 10)
(defun foo ()
  (let ((*x* 20))
    (bar)))
(defun bar ()
  (print *x*))
(foo)

a dynamic-binding language is supposed to give 10, and lexical-binding one 20. (BTW I tested this with Clisp and SBCL, both gave me 10.)

So what is causing different behavior with the "double-let" example? Does SBCL not create an closure with the lambda function?

I would appreciate it if someone would explain, or point out references.

you are viewing a single comment's thread
view the rest of the comments

[–] Patrick-Poitras@alien.top 1 points 1 year ago (1 children)

My understanding was that the first example should give 20 on SBCL. I tried it and it gave me 20.

On your second example, on SBCL, I got 20 for foo and 10 for bar.

* (defvar *x* 10)
*X*
* (defun foo ()
     (let ((*x* 20))
         (bar)))
; in: DEFUN FOO
;     (BAR)
; 
; caught STYLE-WARNING:
;   undefined function: COMMON-LISP-USER::BAR
; 
; compilation unit finished
;   Undefined function:
;     BAR
;   caught 1 STYLE-WARNING condition
FOO
* (defun bar ()
     (print *x*))
BAR
* (foo)

20 
20
* (bar)

10 
10

Not sure if that's helpful in any way. Is there some context we're missing?

[–] Rockola_HEL@alien.top 1 points 1 year ago

SBCL 2.2.9.debian gives 20 for the 1st example, as expected. OP might want to double-check that result.