> On Aug 17, 2022, at 12:44 PM, Carsten Bormann <[email protected]> wrote:
> 
> On 2022-08-17, at 16:54, Russ Housley <[email protected]> wrote:
>> 
>>        Generic_Headers /= (
> 
> This is unfortunately not doing the right thing, as it tries to turn generic 
> headers into a type choice with a second alternative.
> 
> The way Generic_Headers is defined in 8152 does not lend itself to expanding 
> it.
> 
> (TBD10 and TBD11 also don’t quite act like generic headers, because you may 
> need to include them in crit.)
> 
> The extension point actually is in the second line of the group (8152, 
> Section 3):
> 
>   header_map = {
>       Generic_Headers,
>       * label => values   ;;; extension point
>   }
> 
> No socket or similar construct is provided here to make the link to 
> extensions explicit, so unless we want to retrofit our own socket here, I’d 
> just say:
> 
>  CountersignatureV2_header = (
>        TBD10 => COSE_Countersignature / [+COSE_Countersignature]
>  )
> 
>  Countersignature0V2_header = (
>        TBD11 => COSE_Countersignature0
>  )
> 
> 
> Grüße, Carsten


Carsten:

Thanks for catching this.  I have incorporated your suggestion into my edit 
buffer.

Russ

_______________________________________________
COSE mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/cose

Reply via email to