From: Jeroen Scheerder (js@cwi.nl)
Date: Tue Apr 11 2000 - 06:05:10 CDT
Jonathan Grobe (08-04-2000 18:24 -0500) [Special Characters]:
> [From news.software.nntp:]
>
> Subject: Re: [ANNOUNCE] Diablo 2.0 BETA01 Available
> From: curt@kcwc.com (Curt Welch)
> Date: 08 Apr 2000 23:11:51 GMT
[snip]
> Looking at 822, 1036, son-of-1036, USEFOR, and DRUMS, I conclude that
> in all of those, the above ID is invalid because of the [ and ].
>
> But all the standards are also inconsistent. What a mess. And even the
> new USEFOR seems to have problems. It starts by droping one of the " in
> the double quote convention (a typo no doubt). Then makes reference
> to strict-qtext strict-quoted-pair which I can't find defined anywhere.
[snip]
Being a parsing theory researcher (and technology developer) myself, a
question springs to mind.
Wouldn't it make sense to take the syntax definition for the described
message format and generate a parser (or a parse table) from that
syntax definition?
If we did, obvious mistakes and omissions will be apparent immediately;
and a simple suite of sample messages could serve to validate the
soundness of the defined syntax.
If nobody's interested anyway, there's no point; but if actually having
a proof-of-concept parser that raises confidence in the
well-definedness and soundness the Usenet message format syntax
definition does sound like a good idea to some others, I could try to
find the time to generate the parser. And we could start a little
syntax-validation project from there. Idea?