W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > October 2011

Re: ISSUE-112 (RDFParser and DataParser): RDFParser doesn't extend DataParser [RDF API]

From: Nathan <nathan@webr3.org>
Date: Tue, 25 Oct 2011 00:30:23 +0100
Message-ID: <4EA5F50F.7090002@webr3.org>
To: RDF Web Applications Working Group WG <public-rdfa-wg@w3.org>
RDF Web Applications Working Group Issue Tracker wrote:
> ISSUE-112 (RDFParser and DataParser): RDFParser doesn't extend DataParser [RDF API]
> http://www.w3.org/2010/02/rdfa/track/issues/112
> Raised by: Manu Sporny
> On product: RDF API
> This issue was raised by Christian Langanke:
> The RDFParser doesn't extend DataParser in the RDF API. This means that there is no way for the Data interface to get access to the triples in the graph. We should make the following changes to the RDF API:

all this is assuming that the latest draft of the RDF API is:

IIRC, the RDF API is completely detached from the RDF Interfaces and has 
no dependency on it, and doesn't mention the DataParser interface anywhere?

> 1. Ensure that there is a .graph member on the Data interface. We might as well expose the data since it's supposed to be there beneath the scenes.

The .parse method doesn't populate the RDF API in anyway, it simply 
takes a document and produces a sequence<Projection> for it, which it 
then passes on to a ParserCallback... which begs the question, just what 
do all the other methods of the Data interface operate on?

> 2. We should change RDFParser such that it extends DataParser.

Can't see how this would be possible, assuming that it's the DataParser 
from the RDFInterfaces that is being referred to - a toProject(Graph 
graph) method somewhere, or a series of interfaces to swap between 
Graphs and Projections would do the trick perhaps? (I'm sure we meant to 
do that!).


Received on Monday, 24 October 2011 23:31:35 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:19:53 UTC