Re: N-Triples Updated 2001-07-08

Dave Beckett wrote:
> 
> I've updated the N-Triples document at:
> 
>    N-Triples
>    http://purl.org/net/dajobe/2001/06/ntriples/
> 
>    (RCS v1.4)

err... I'm surprised to see comments. Oh well... as
long as it's still line-oriented, that's fine.

But... please don't make the BNF informal by
way of external pointers like...

Nmtoken
            ::=
               See NT-Nmtoken in [XML10-2]
                                                              
 absoluteURI
            ::=
               An absolute URI-reference - see URI-reference in
               RFC 2396 - Uniform Resource Identifiers (URI):
               Generic Syntax

just define absoluteURI as [^<> ]+. (hmmm... the carat
raises the question of what characters are in the compliment...)
and nmtoken as [a-zA-Z0-9]+. And don't call it nmtoken.


Oh... I see this is called out as an issue:
 |Is Nmtoken an XML one or [A-Za-z0-9]+ as used in
 |    n-triples2kif.pl code and n3-simple.pl ? 

I don't see any reason to use XML's NMTOKEN construct.
It's unnecessarily hairy.


> following comments here and privately and made the following changes:
>   Reworked grammar to bring out line basis.
>   Remove vertical tab from ws.
>   Made eoln be: cr? lf
>     -- it can't really be only cr lf since that would make all
>        existing ntriples examples illegal
>   Added string escaping section
>   Removed URI encoding issue
>   Split references into history + references sections
>   Added section numbers, links
>   Updated unicode encoding issues
>   Lots of rewording.
> 
> There are still some unresolved issues at
>   http://purl.org/net/dajobe/2001/06/ntriples/#sec-issues
> that I'd appreciate feedback to resolve.
> 
> Dave

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/

Received on Monday, 9 July 2001 08:44:16 UTC