Re: Report on SPARQL Grammar implementation experience using ANTLR.

Dan Connolly wrote:
> On Mar 9, 2005, at 12:17 PM, Seaborne, Andy wrote:
> 
>> Thompson, Bryan B. wrote:
>>
>>> Hello,
>>> I wanted to report on our implementation experience for the SPARQL
>>> grammar using the ANTLR parser generator based on the Working Draft
>>> of the SPARQL query language[1][2].
>>
>>
>> Excellent news!
> 
> 
> yes, quite...
> 
>> Hmm - the productions come fairly directly from a javacc grammar
> 
> 
> I'd like a link to that from the editor's draft, Andy. I dunno if it's 
> worth
> keeping when we publish or not... but I hate the idea of not giving
> the readers all the bits we use to produce the draft.

The grammar in rq23 will be master one - the javacc is the development one 
because it is executable and hence checkable against the test suite.  The 
one in rq23 is mechaniscally produced but manually cleaned for presentation.

The downside of the javacc one is that all the detail in it that doesn't end 
up in rq23 is the java code that is specific to implementation.

The grammar going into Eric's yacker is the rq23 one.

> 
>> I'm most of the way through changing the grammar to reflect 0227  If 
>> the javacc master would be of help, it is in the ARQ module of CVS in 
>> the Jena repsoitory on SourceForge, directory /Grammar.
> 
> 
> Care to spell that in URI syntax? ;-)

:-)
Repository/module:
http://cvs.sourceforge.net/viewcvs.py/jena/ARQ/

Grammar:
http://cvs.sourceforge.net/viewcvs.py/jena/ARQ/Grammar/sparql.jj?view=markup

Beware that the web view of CVS trails the live repository - SF are in the 
process of upgrades to remove this problem.

I just checked (21:50Z) and it looks to be slightly lagging (qname 
productions aren't right).  You can get a live version via anonymous CVS.

	Andy

Received on Wednesday, 9 March 2005 21:55:16 UTC