Re: [apps-discuss] Link rel uniqueness

On Oct 21, 2012, at 5:52 AM, Mark Nottingham wrote:

> Hi Jan,
> 
> What's the specific requirement you're referring to?

Section 5.3.:

"The relation type of a link is conveyed in the "rel" parameter's
   value.  The "rel" parameter MUST NOT appear more than once in a given
   link-value; occurrences after the first MUST be ignored by parsers."

I was unsure what the link-value refers to but from what you write below I understand that it refers to the combination of all header values of the same name.

That would then mean that parsers, when seeing

>> Link: </a>; rel="describedBy"
>> Link: </b>; rel="describedBy"

would, per 5.3., have to produce only

>> Link: </a>; rel="describedBy"

(ignoring the Link to /b)

Correct?

P.S. I find that a bit counter intuitive. Given that further links might occur elsewhere in the entity.

Jan

> 
> Semantically, in ANY HTTP header defined using the "," list production, the following are equivalent:
> 
> Foo: 1
> Foo: 2
> 
> and
> 
> Foo: 1, 2
> 
> Regards,
> 
> 
> On 21/10/2012, at 9:33 AM, Jan Algermissen <jan.algermissen@nordsc.com> wrote:
> 
>> Hi,
>> 
>> I am seeking a clarification ragarding link rel uniqueness in http://tools.ietf.org/html/rfc5988
>> 
>> AFAIU the required uniqueness of rel is 'per header' and thus, the following HTTP response headers in a single response would be valid according to RFC 5988:
>> 
>> 
>> Link: </a>; rel="describedBy"
>> Link: </b>; rel="describedBy"
>> 
>> Correct?
>> 
>> Jan
>> _______________________________________________
>> apps-discuss mailing list
>> apps-discuss@ietf.org
>> https://www.ietf.org/mailman/listinfo/apps-discuss
> 
> --
> Mark Nottingham   http://www.mnot.net/
> 
> 
> 