Larry Clapp wrote:
> Every so often you get people on cll that say "I can never seem to
> understand LET [or COND or DO or what-have-you] -- I understand the
> various parts, but I can never remember where to put the parentheses!"
That's a good indication that something is broken.
> [...]
> Below, is J a variable to be bound or a value to be assigned?
>
> (LET (a b c d e f g h i j k l m n o p)
> code)
>
> You don't know, you have to count!
Take a look at
(LET (a b c d e f g h i j k l m n o p)
...)
-or-
(LET (a b
c d
e f
g h
i j
k l
m n
o p)
...)
-or-
(LET (a b c d e f g h
i j k l m n o p)
...)
et cetera. Fairly obvious what's a variable name and value, isn't it?
> That sucks. So add another set of sublists.:
No. Write a macro utility that works around the design flaw. Isn't it
the Lisp way to let the computer do the work for you? :-)
Here is an example from Graham's Common Lisp (p.83, "5.2 Context").
;; old style
(let ((x 7)
(y 2))
(format t "Number")
(+ x y))
;; new style
(let (x 7
y 2)
(format t "Number")
(+ x y))
Here's the macro code:
;; un-flatten a list like (a 1 b 2 c 3) into ((a 1) (b 2) (c 3))
(defun slet->let (seq)
(if (first seq)
`((,(first seq) ,(second seq)) ,@(slet->let (nthcdr 2 seq)))
nil))
;; simple let
(defmacro slet (assoclist &body body)
`(let ,(slet->let assoclist)
,@body))
;; simple let*
(defmacro slet* (assoclist &body body)
`(let* ,(slet->let assoclist)
,@body))
I think there are easier ways to do this though. :)
> "Where to put the parentheses" falls out just as naturally from the
> basic structure of COND. COND consists of a list of alternatives.
>
> [...]
>
> So
>
> (COND (choice1
> code1a code1b code1c) ; implicit PROGN
> (choice2 code2a)
> (choice3 code3a code3b))
Could have been written as:
(COND choice1 (code1a code1b code1c)
choice2 code2a
choice3 (code3a code3b))
Another example from Graham's Common Lisp (p.86, "our-member").
;; old style
(defun our-member (obj lst)
(cond ((atom lst) nil)
((eql (car lst) obj) lst)
(t (our-member obj (cdr lst)))))
;; new style
(defun our-member (obj lst)
(cond (atom lst) nil
(eql (car lst) obj) lst
t (our-member obj (cdr lst))))
I can't provide a macro utility for that yet. ;)
(Even McCarthy stated in [1] that "The unexpected appearance of an
interpreter tended to freeze the form of the language, and some of the
decisions made rather lightheartedly for the ``Recursive functions ...''
paper later proved unfortunate. These included the COND notation for
conditional expressions which leads to an unnecessary depth of
parentheses.")
> So anyway, hope this provides some fertilizer (hopefully not in the
> "this is crap!" sense :) for Lisp Gardeners.
Braces are Lisp's strength. Like everything else, they can be abused.
Just adding some gasoline and a spark to that fertilizer. :-)
[1] http://www-formal.stanford.edu/jmc/history/lisp/node3.html
--
chris c. eineke
_______________________________________________
Gardeners mailing list
[email protected]
http://www.lispniks.com/mailman/listinfo/gardeners
