Re: [Sip] Support for Multipart/MIME

[Cullen Jennings <[e-mail]>]

I like the draft - this looks like exactly they type of think I was  
hoping someone would write. Thank you,  Cullen

On May 27, 2007, at 12:22 PM, Gonzalo Camarillo wrote:

> Hi Paul,
>
> I have added a few guidelines about this in Section 5 of the new  
> version of the draft:
>
> http://users.piuha.net/gonzalo/temp/draft-camarillo-sip-body- 
> handling-01.txt
>
> With those guidelines and the behavior defined in other sections,  
> we should be OK. Let me know if you agree or want to add further  
> clarifications or behavior.
>
> Thanks,
>
> Gonzalo
>
>
> Paul Kyzivat wrote:
>> Revising one point:
>> Paul Kyzivat wrote:
>>> - What "handling" value should be used for multipart bodies and  
>>> the contained body parts?
>>>
>>> If a body part is referenced by a cid: url in a header (or in  
>>> another body part I suppose), then the handling of the body part  
>>> should be required.
>> On reflection, this is tricky.
>> If the header containing the reference is optional (as most are),  
>> then recipients that don't understand the header will have no need  
>> for the referenced part. If it were required and they don't  
>> support the type, then they would fail the request. This is bad.
>> So if processing of the header is optional, then the referenced  
>> body part should have optional handling.
>> But then there could be a problem if the recipient does understand  
>> and process the header, but doesn't understand the body part, and  
>> so because it is optional doesn't process it. In that case, the  
>> cid reference will be unsatisfied. This could also result in an  
>> inappropriate error.
>> I think the solution here is that if a header contains a cid url,  
>> and if that url cannot be resolved to a body part that is  
>> supported, then the UA should treat it the same as if the header  
>> itself was not understood - typically ignoring it unless there is  
>> a Require in effect that demands failing the request.
>> An alternative is to require that Content-ID be remembered even  
>> for parts that are not understood, so that references can be  
>> resolved even if they can't be processed. This would allow more  
>> precise errors to be generated. This wouldn't be so hard at an  
>> outer level, but it might be troublesome in conjunction with nesting.
>>     Paul
>
>
>
> _______________________________________________
> Sip mailing list  https://www1.ietf.org/mailman/listinfo/sip
> This list is for NEW development of the core SIP Protocol
> Use [e-mail] for questions on current sip
> Use [e-mail] for new developments on the application of sip


_______________________________________________
Sip mailing list  https://www1.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use [e-mail] for questions on current sip
Use [e-mail] for new developments on the application of sip