Re: Handling deep list recursion

Heime <[email protected]> writes:

>> 
>> >
>> > 1. If someone does not provide a recursive solution to your
>> > question, then you might consider writing an iterative
>> > solution using ‘dolist’ or ‘while’ or ‘do’.
>> 
>> I could use the `while' approach, because there is no binding
>> overhead.
>> 
>> >    (info "(elisp) Iteration")
>> >
>> > 2. Besides ‘cl-every’, you could write a solution using
>> > ‘cl-find-if’ or ‘cl-find-if-not’ or ‘cl-notany’.
>> 
>> cl-notany looks the most appropriate because it exits immediately
>> on first non-consp cell, matching the recursive early-return logic.
>
> Is this equivalent to (cl-every #'consp object)?  Does cl-notany
> return immediately as soon as an entry is not a consp?

I leave it to the Emacs maintainers to answer whether
cl-every and cl-notany short-circuit.  Both almost certainly
do.

>
>       (cl-notany (lambda (x) (not (consp x))) object)

More simply: (cl-notany #'atom object)

Examples:
   
;; A valid association list and two invalid alists:
(setq
 alist1     '((a . b) (c . d) ("e" "f") (gee . 8))
 not-alist1 '((a . b) (c . d) ("e" "f") nil (gee . 8))
 not-alist2 '(4.5 (a . b) (c . d) ("e" "f") (gee . 8)))

;; Using 'cl-notany':
(cl-notany #'atom alist1)
;;=> t

(cl-notany #'atom not-alist1)
;;=> nil

(cl-notany #'atom not-alist2)
;;=> nil

;; Using 'cl-every'
(cl-every #'consp alist1)
;;=> t

(cl-every #'consp not-alist1)
;;=> nil

(cl-every #'consp not-alist2)
;;=> nil

-- 
The lyf so short, the craft so long to lerne.
- Geoffrey Chaucer, The Parliament of Birds.