Re: RDFa parser implementation issues

Hi Jan,

Partial answer: you will find datetime and time processing steps in the
HTML+RDFa spec at http://www.w3.org/TR/html-rdfa/

Steph.


On Fri, Feb 21, 2014 at 11:10 AM, Jan Wielemaker <J.Wielemaker@vu.nl> wrote:

> Hi,
>
> I am trying to implement the RDFa spec [1] for SWI-Prolog's RDF toolkit.
> Running the XHTML5 test suite, I'm now at:
>
>     Processed 137 tests, 14 failed (89.78% success)
>
> I do not understand the remaining 14 though.  Here they are:
>
> #0066: I think this should be a blank node using this production:
>        7.5.5.2: "otherwise, if @typeof is present, then new subject
>        is set to be a newly created bnode;"  What rule did I miss?
> #0134: The docs only speak about reserved words in the 1.0 --> 1.1
>        changenotes. There are no reserved words in 1.1.  What makes
>        this happen?  Are there implicitely defined terms?
> #0188: 7.5.13, skip_element=true: should also copy vocab from local
> context.
> #0198: Fragile test due to underdefined serialization of the XMLLiteral.
> #0206: Where the the specification of the default context? Seems to
>        require at least rdf, xsd and owl.  What else?
> #0272: Where is the spec what should happen to @datetime?
> #0273: Idem
> #0274: Idem
> #0275: <time> is special?  I cannot find that in the document.
> #0276: Idem
> #0277: Idem
> #0281: As #0272
> #0282: Idem
> #0287: Idem
> #0312: Why is rel="nofollow" not processed as a term relative to the
>        default vocabulary?
> #0319: Seems a bit fragile.  This happens after writing turtle and
>        reading it back with an @base.  Possibly a negative test that
>        the triple relative to the base of the RDFa document is not
>        generated would have been better.
>
>         Thanks --- Jan
>
> P.s.    Are implementation reports still appreciated?  If so, I'll
>         generate one.
>
> [1] http://www.w3.org/TR/2013/REC-rdfa-core-20130822/
>
>


-- 
Steph.