Re: [MMUSIC] 10 BUNDLE questions

[Justin Uberti <[e-mail]>]

On Mon, Mar 18, 2013 at 2:05 PM, Harald Alvestrand <[e-mail]>wrote:

>  On 03/18/2013 04:59 PM, Justin Uberti wrote:
>
> You're looking at the handling of a=mid, not a=group.
>
>  Section 9.2 indicates the handling of a=group, which seems to indicate a
> subset is allowed:
>
>     A SIP entity that receives an offer that contains an "a=group" line
>    with semantics that are understood MUST return an answer that
>    contains an "a=group" line with the same semantics.  The
>    identification-tags contained in this "a=group" line MUST be the same
>    as those received in the offer, or a subset of them (zero
>    identification-tags is a valid subset).  When the identification-tags
>    in the answer are a subset, the "group" value to be used in the
>    session MUST be the one present in the answer.
>
>
> Aha, I misinterpreted the phrase "BUNDLE mids in the answer doesn't match".
>
> This is, in some sense, a subset of the question "is unbundling possible".
> I don't have a strong opinion on this.
>
> The key issue here is whether the remote side can reject a BUNDLE. IOW, if
caller has A and B bundled, and adds m= line C (BUNDLEd), does the caller
need to allocate a new port, so that it can handle the case where the
callee decides only A and B should be BUNDLEd? Or can the caller offer C on
the existing A+B BUNDLEd port?

I think life would be simpler if we enforced the rule that the callee could
only accept C as BUNDLEd, or reject C entirely. IOW, any difference in the
a=group answer (other than removal of rejected mids) is fatal.