Re: Correction: no *relativeURI* in HTTP in RDF

Hi Christophe

Christophe Strobbe schrieb:

> I sent the previous message with the wrong subject :-(
> 
> ==
> 
> As some of you know, the Test Case Description Language 2.0, 
> <http://bentoweb.org/refs/TCDL2.0.html>, borrows some language from HTTP 
> Vocabulary in RDF to point to test files: 
> <http://bentoweb.org/refs/TCDL2.0.html#edef-file>.
> HTTP in RDF defines the request URI properties http:aboluteURI, 
> http:abs_path and http:authority
> (<http://www.w3.org/TR/HTTP-in-RDF/#requestURI>) but not relative URI.
> However, TCDL 2.0 uses relative URIs to point from the metadata to the 
> test files (so the whole suite of test case can be moved to somewhere 
> else without changing the links to the test files) and is forced to 
> abuse http:absoluteURI for this in the absence of http:relativeURI.
> It would be nice to have http:relativeURI in HTTP in RDF, although I 
> understand why only the other sub-properties were introduced (the URIs 
> in the EARL report need to be correct, independent of the location where 
> you put the report, right?).

No, HTTP-in-RDF does not rely on EARL, it may be the other way round. 
HTTP-in-RDF is just an RDF representation of the vocabulary defined in 
HTTP 1.1. And in section 5.1.2 of HTTP 1.1 we see that

Request-URI    = "*" | absoluteURI | abs_path | authority

HTTP 1.1 does not permit relative URIs here. So there's no need for 
HTTP-in-RDF to provide a relativeURI property.

Having said that, feel free to create your own relativeURI as a 
sub-property of requestURI, just like absoluteURI, abs_path and 
authority are sub-properties of requestURI.

-- 
Johannes Koch
BIKA Web Compliance Center - Fraunhofer FIT
Schloss Birlinghoven, D-53757 Sankt Augustin, Germany
Phone: +49-2241-142628    Fax: +49-2241-142065

Received on Monday, 19 May 2008 21:12:19 UTC