SL Baur <steve(a)xemacs.org> writes:
> No. Even if we accept integers, this is a kludge liable to
change,
> and should be (un)documented as such. No correct code should ever
> rely on `string' accepting integers.
I mean it should have the kind of warning `concat' has.
Oh. Well, concat has that warning because it existed long before
XEmacs/Mule. I don't see the gain by actually *introducing* broken
behavior to a new function and then adding a warning not to use it.
[...] Oh, the comment in `concat' is wrong:
(concat 65 65)
=> "6565"
This is the hysterical behaviour, yes. What comment are you referring
to?
Well, make `string' consistent.
(string 65)
=> "65"
Yuck. I don't think we want to do anything like this. FSFmacs
doesn't do it.
(string 65)
=> "A"
--
Hrvoje Niksic <hniksic(a)srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
A radioactive cat has eighteen half-lives.