On 3/3/20 7:06 AM, Arthur A. Gleckler wrote:
> On Tue, Mar 3, 2020 at 12:54 AM Shiro Kawai <xxxxxx@gmail.com <mailto:xxxxxx@gmail.com>> wrote:
>
>     At this moment, a string is a sequence of codepoints, and it's reasonable to expect a grapheme cluster is represented as a string.
>     I agree that it is good that the intention is noted, if it is indeed the original intention.   I think adding such note won't change the specification so it can be a post-finalization note?
>
>>             <http://per.bothner.com/>
>
>
>   Per, would this be okay?  If so, and someone can provide me with a patch or pull request, I"ll incorporate the change.

How about adding the sentence:

     If the mapper returns a string, the string is prepended to constructed string (without reversal).

and this example:

(string-unfold-right null? (lambda (x) (string  #\[ (car x) #\])) cdr '(#\a #\b #\c))
     => "[c][b][a]"
--
	--Per Bothner
xxxxxx@bothner.com   http://per.bothner.com/