I've just noticed (after sending previous email...), R6RS has errata
for the test-200:

    Fx4 must be less than or equal to the difference between fx3 and fx2.

So it's only negative count I want clarification.

Cheers,

On 16 May 2017 at 22:11, Takashi Kato <ktakashi19@gmail.com> wrote:
> Hi,
>
> Test-200 and test-202 seem violating R6RS fxrotate-bit-field. R6RS says:
>
>     Fx2, fx3, and fx4 must be non-negative and less than (fixnum-width).
>     Fx2 must be less than or equal to fx3. Fx4 must be less than the
>     difference between fx3 and fx2.
>
> However, the test cases are:
>
> #| test-200 |# (fxbit-field-rotate #b110 1 1 2)
> ;; -> (fxrotate-bit-field 6 1 2 1)
> ;; thus fx4 is equal to the difference between fx3 and fx2
>
> #| test-202 |# (fxbit-field-rotate #b0111 -1 1 4)
> ;; -> (fxrotate-bit-field 7 1 4 -1)
> ;; thus fx4 is negative
>
> Most of the implementations allow test-200 except Mosh and Sagittarius
> (not sure about Guile, Vicare and IronScheme) and pnly Ypsilon allows
> test-202 (not sure about Guile, Vicare and IronScheme)
>
> Are these cases invalid on this SRFI? Or does it have to be allowed?
>
> Cheers,
>
> --
> _/_/
> Takashi Kato
> E-mail: ktakashi19@gmail.com

--
_/_/
Takashi Kato
E-mail: ktakashi19@gmail.com