[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Scheme-reports] Scheme r7rs syntax described by ABNF



Alex -

I added the following TODO item to my local copy and also added one new item as a placeholder.

;;;
;;; TODO:
;;;   - Fix definition of peculiar-identifier and
;;;     pattern-peculiar-identifier not to include "+i", "-i", and
;;;     <infnan>.
;;;
;;;   - Review definitions to check if syntactically correct but
;;;     semantically incorrect numbers can be avoided (e.g. "#e0/0",
;;;     "#e+1/0", "#e-1/0", "#e"<infnan>).
;;;

I couldn't find any description for such issues in the r7rs draft (but maybe I need to look closer ?).

Or is it common practice to leave such items as rationals?

thanks,

Joe N.

On Jan 17, 2013, at 12:07 , Joseph Wayne Norton <norton@x> wrote:

> 
> Alex -
> 
> Without this extended syntax, the spec becomes more tedious (and error prone).
> 
> I'm aware of two ABNF parser libraries that support this extended syntax:
> 
> abnfgen  - http://www.quut.com/abnfgen/
> 
> abnfc - https://github.com/ubf/abnfc/tree/norton-upstream%2Bgenerators
> 
> I tested this draft of r7rs with the abnfc tool.
> 
> I'll add a TODO list at the top of this draft for the +i and +nan.0 symbols.
> 
> thanks,
> 
> Joe N.
> 
> 
> On Jan 17, 2013, at 11:33 AM, Alex Shinn <alexshinn@x> wrote:
> 
>> On Wed, Jan 16, 2013 at 11:24 PM, ノートン ジョーセフ ウェイ ン <norton@x> wrote:
>> Hello again.
>> 
>> I have an updated draft for the formal syntax of Scheme r7rs written in ABNF.  This draft is based on updates since r7rs-draft-8.pdf and includes all sections except for quasiquotations.  The datum and expression sections have undergone some quickcheck-style unit testing.  I'd appreciate any review and feedback.
>> 
>> Thanks for doing this.  The big advantage of the RFC format is
>> that it's machine readable - have you tried running this through
>> any ABNF parser libraries?
>> 
>> To use it practically you'd need to tie up some of the loose ends,
>> e.g. +i and +nan.0 should not parse as symbols.  This is possible
>> but a little tedious to spec out.
>> 
>> -- 
>> Alex
>> 
> 

_______________________________________________
Scheme-reports mailing list
Scheme-reports@x
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports