Re: [External] : Re: How to make M-x TAB not work on (interactive) declaration?

[Andreas Eder]

On Mo 16 Jan 2023 at 18:38, Jean Louis <bugs@gnu.support> wrote:

> * Anders Munch <ajm@flonidan.dk> [2023-01-16 13:58]:
>> Jean Louis wrote:
>> > I actually expect function to tell me wrong number of arguments or no 
>> > arguments, as I find it safer for programming that way.
>> 
>> It makes sense for 
>>       (+)
>> to be an error, if it's written like that.
>> But if the code goes
>>      (apply #'+ a-list-of-numbers)
>> then a-list-of-numbers being empty is normal and expected, and 0 is 
>> invariably the desired result.
>> You can't make the former an error without the latter becoming an error as 
>> well, so it isn't.
>
> (apply #'+ nil) that is what you mean why is it usable?
>
> than this is disaster:
>
> (apply #'* nil) ➜ 1
>
> Do you wish to say that `apply' function is reason for this below?
>
> (*) ➜ 1 and 
> (+) ➜ 0 and
> (-) ➜ 0
>
> Maybe you know some reference to Lisp beginning why is it so?

See http://www.lispworks.com/documentation/HyperSpec/Body/f_pl.htm for
#'+ and http://www.lispworks.com/documentation/HyperSpec/Body/f_st.htm
for #'*.

'Andreas