Re: [Gzz] ``canon3_file_format``: A canonical, N3-based file format

[Benja Fallenstein]

Antti-Juhani Kaijanaho wrote:
> On 20030401T214550+0200, Benja Fallenstein wrote:
> > The ``NEWLINE`` token may be any of CR, LF, and CRLF.
>
> Any particular reason for not allowing the other Unicode
> newline-function characters (such as the "canonical" Unicode newline,
> U+2028 LINE SEPARATOR)?  It does make sense not to generate them in CVS
> context, but is this format exclusively for CVS?

Done.

> > In contexts where the specific form used matters,
> > the newline character is LF. (In particular, when computing
> > a content hash-- e.g., when creating a Canon3 Storm block.)
>
> Why not LINE SEPARATOR?

As said in the PEG, this would be incompatible with N3. But we're 
unfortunately not 100% compatible with N3 anyway, because it doesn't 
allow using local names for anonymous nodes. :-/ So we could do this also.

What do people think?

- Benja