Re: [apps-discuss] URI templates in Link header fields

[Sean Leonard <[e-mail]>]

Overall it's a good idea, so +1 to the effort (to add URI template 
support to HTTP Links).

I give +½ to the consolidation comment. It would be good to consolidate, 
but I do not know if consolidating to one spec would put excessive 
burdens on diverse implementations that only need selected features 
(i.e., to support all features). Lack of knowledge of this space, that's 
all.

Regards,

Sean

On 8/30/2016 5:33 AM, Graham Klyne wrote:
> Hi!
>
> I've wanted something like this in the past, and I *thought* I had 
> some code that accepted a template in a custom header field.  But on 
> closer examination, I see that I passed a template in the body of a 
> POST submission.
>
> It's something I could see myself using if specified.
>
> And +1 to Erik Wilde's comment: "it would be great to consolidate all 
> of this so that tooling and implementations can rely on a robust spec".
>
> #g
> -- 
>
>
> On 29/08/2016 09:42, Julian Reschke wrote:
>> Hi there.
>>
>> RFC 5988, defining the HTTP Link header field, predates RFC 6570, 
>> defining URI
>> templates.
>>
>> In the meantime, I heard of a few cases where people were trying to use
>> templates in Link header fields (an obvious thing to do), and came up 
>> with hacks
>> to do so with the existing Link header field.
>>
>> I'm tempted to work on a small spec that defines an experimental new 
>> header
>> field, similar to Link but allowing templates.
>>
>> Is anybody aware of existing code trying to use templates in Link 
>> header fields?
>> If so, please report over here (or in doubt send private email).
>>
>> Best regards, Julian