Re: Is Named RDF List (Collection) valid?

Hi Rui,

On 18/04/2024 13:24, Rui Zhao wrote:
> Hi,
> 
> We are using Turtle as a data format for our development, and uses RDF 
> Lists on it. However, we see some weird behaviours over the tools we use.
> 
> In short, can I have a *Named* RDF List (Collection)?

Yes, if you mean can the list cells have URI subjects and not blank nodes.

It won't print in a pretty form. The Turtle syntax for lists assumes 
blank nodes and exactly one each of rdf:first, rdf:rest per list 
element. It may look messy if the data deviates from that.

> For example, is 
> the following data valid?

As Turtle, there is no DOT at the end but otherwise, yes.

> @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>.
> @prefix : <http://example.org#>.
>  
> :node1 
>   :has :node2.
>  
> :node2
>   a rdf:List;
>   rdf:first [a :Wow];
>   rdf:rest rdf:nil
> 
> We have found that this is rejected by rdflib (JS), and 
> https://www.easyrdf.org/converter also showed unexpected output with 
> this data (e.g. if outputting in N3 syntax).

What did you get?

rdf:nil may be printed as () as in:

"""
:node2  rdf:type   rdf:List;
         rdf:first  [ rdf:type  :Wow ];
         rdf:rest   () .
"""

which is correct (its the same triples) in the sense that "()" is a 
syntax for rdf:nil.

> 
> ||
> 
> I have consulted the RDF spec https://www.w3.org/TR/rdf-schema/#ch_list, 
> which did not seem to say anything about this.
> 
> 
> Best regards,
> 
> Rui
>

Received on Thursday, 18 April 2024 15:02:35 UTC