Dr. Arne Babenhauserheide writes:

> "Dr. Arne Babenhauserheide" <[email protected]> writes:
>
>> ' is explicitly defined as token, but so are other syntactic sugars
>> (` , @, .) and those are accepted in identifiers.
>>
>> So I think compatibility should go the other way round and have an opt
>> in for splitting up identifiers on '.
>
> This goes deeper: racket disallows any token in identifiers, but chicken
> accepts ` (but turns it into |a`b| on writing) while splitting on ' and
> failing on , (replaced by unquote).

Yes; that's how I read it when I wrote the reader, checked with Guile,
and so...

> The most robust way I found to try it out is '(a`b a'b a,b a,@b):
>
> Guile: (a`b a'b a,b a,@b)
> Racket/Chez: (a `b a 'b a ,b a ,@b)
> Chicken: (|a`b| a (quote b) a (unquote b) a (unquote-splicing b))
> sbcl: (A `B A 'B) ; fails on , and ,@
> emacs: (a `b a 'b a (\, b) a (\,@ b))
> kawa: (a (quasiquote b) a (quote b) a (unquote b) a (unquote-splicing b))
> mit: (a (quasiquote b) a (quote b) a (unquote b) a (unquote-splicing b))

Mes: (a`b a'b a,b a,@b)

;)

Greetings,
Janneke

-- 
Janneke Nieuwenhuizen <[email protected]>  | GNU LilyPond https://LilyPond.org
Freelance IT https://www.JoyOfSource.com | AvatarĀ® https://AvatarAcademy.com

Reply via email to