Hi Ian,

Sorry for the late reply.

Ian Price <ianprice90@googlemail.com> skribis:

> ludo@gnu.org (Ludovic Courtès) writes:
>
>> Hi Göran,
>>
>> Sorry for the delay.
>>
>> Göran Weinholt <goran@weinholt.se> skribis:
>>
>>> the case-lambda form is specified in r6rs-lib as accepting any number of
>>> clauses, including zero. So this should not give an error:
>>
>> My interpretation of the ‘case-lambda’ implementation on p. 15 of
>> r6rs-lib.pdf is that ‘case-lambda-help’ raises an assertion violation
>> when ‘case-lambda’ is called with zero clauses.
> The case-lambda-help macro is expanded from within
> (lambda args
>   (let ((n (length args)))
>     (case-lambda-help args n
>       (fmls b1 b2 ...) ...)))
>
> So, the full expansion is
> (lambda args
>   (let ((n (length args)))
>     (assertion-violation #f "unexpected number of arguments")))
>
> and thus a procedure that always returns an assertion violation.

Indeed, thanks for the correction (I was thinking of
‘assertion-violation’ as a compile-time assertion.)

So, here’s a tentative patch for review: