Sequence Requirement Question

[Claude Warren <cl...@xenei.com>]

background:

I am working on a jena security package and have a question about Seq objects.

The security system allows restriction on triples.  This leads to a
potential case where the an element in the list can not be returned to
the user.   In this case should the sequence be returned with holes in
it or should the numbers in the list be re-sequenced.

Example

x RDF.li(1) A
x RDF.li(2) B <--- this one the user may not see
x RDF.li(3) C

Should the system return
x RDF.li(1) A
x RDF.li(3) C

or
x RDF.li(1) A
x RDF.li(2) C

Opinions?

--Claude

-- 
I like: Like Like - The likeliest place on the web
Identity: https://www.identify.nu/user.php?claude@xenei.com
LinkedIn: http://www.linkedin.com/in/claudewarren

Re: Sequence Requirement Question

[Dave Reynolds <da...@gmail.com>]

On 29/11/12 22:12, Claude Warren wrote:
> background:
>
> I am working on a jena security package and have a question about Seq objects.
>
> The security system allows restriction on triples.  This leads to a
> potential case where the an element in the list can not be returned to
> the user.   In this case should the sequence be returned with holes in
> it or should the numbers in the list be re-sequenced.
>
> Example
>
> x RDF.li(1) A
> x RDF.li(2) B <--- this one the user may not see
> x RDF.li(3) C
>
> Should the system return
> x RDF.li(1) A
> x RDF.li(3) C
>
> or
> x RDF.li(1) A
> x RDF.li(2) C
>
> Opinions?

The semantics explicitly say that gaps are allowed in SEQ so I wouldn't 
bother renumbering unless that gives away important information that 
something has been restricted.

http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#Containers

Dave