W3C home > Mailing lists > Public > spec-prod@w3.org > April to June 2010

Re: W3C EBNF catalog?

From: Dan Connolly <connolly@w3.org>
Date: Thu, 29 Apr 2010 12:56:22 -0500
To: Dominique Hazael-Massieux <dom@w3.org>
Cc: spec-prod <spec-prod@w3.org>
Message-ID: <1272563782.4466.20905.camel@pav.lan>
On Thu, 2010-04-29 at 18:18 +0200, Dominique Hazael-Massieux wrote:
> Hi,
> 
> As I was having a cursory look at the definition of the RIF language in
> EBNF [1], and wanted to quickly check the validity of the grammar using
> existing tools such as abnfgen [2] or bnf_check, the following (somewhat
> unstructured) thoughts occurred to me:
> • for better or for worse, XML defined its own EBNF syntax [4] that
> differs from the ISO standard [5]; given that a number of W3C specs keep
> using that specific syntax, and given that there are more tools that
> recognize the ISO syntax, it would be nice to have a converter from the
> W3C syntax into the ISO one — does anyone know if that exists?

I have python code to read "W3C syntax" (i.e. the one in the XML spec)
and write it out in turtle...

  http://www.w3.org/2000/10/swap/grammar/ebnf2turtle.py

it should be straightforward to adapt it to write out the ISO syntax...
but... why?

Is there a network of tools and practice around the ISO syntax?

I also have javascript code to read IETF ABNF syntax
a la that abnfgen tool. Hmm... 
http://bitbucket.org/DanC/urlp/src/tip/abnf.js
(also in http://jay.w3.org/~connolly/projects/urlp/ )


Some other relevant pointers...

a notation3 grammar in XML formal grammar notation / EBNF
http://lists.w3.org/Archives/Public/public-cwm-talk/2006AprJun/0032.html

bnf2turtle -- write a turtle version of an EBNF grammar
by connolly on Fri, 2006-02-10
http://dig.csail.mit.edu/breadcrumbs/node/85

We need a EBNF spec Bjoern Hoehrmann (Monday, 9 January)
http://lists.w3.org/Archives/Public/spec-prod/2006JanMar/thread.html#msg0


> • RIF refers to a number of EBNF defined in other W3C specs (in
> particular XML Namespaces and SPARQL); I wonder if having a catalog of
> the existing EBNF defined in various specifications wouldn't facilitate
> both the re-usage of these EBNF and their testing — right now, to create
> the full RIF EBNF grammar requires a lot of easy-to-get wrong copy &
> pasting, that an EBNF catalog could presumably facilitate; is this a
> goofy idea? has anyone heard of a tool allowing to manage/catalog EBNF
> productions?

Having the RIF BNF as one file that you don't have to copy/paste
from is a very good idea.

Trying to collect the BNFs from all W3C specs... ugh...
resist the urge to centralize! I suppose the QA matrix has
been valuable, but... is it maintained any more?

People ask for lists of DTDs... I don't really see the appeal.
Why not lists of .png images in specs while we're at it?

> 
> Dom
> 
> 1.
> http://www.w3.org/TR/2009/CR-rif-dtb-20091001/#sec-shortcuts-constants
> 2. http://www.quut.com/abnfgen/
> 3. http://www.icosaedro.it/bnf_chk/index.html
> 4. http://www.w3.org/TR/REC-xml/#sec-notation
> 5. http://www.iso.org/iso/catalogue_detail.htm?csnumber=26153
> 
> 
> 
> 


-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
gpg D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E
Received on Thursday, 29 April 2010 17:56:24 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 10 March 2012 06:19:17 GMT