On Fri, Dec 12, 2008 at 4:13 PM, Scott Lawrence
<[email protected]> wrote:
>
> On Fri, 2008-12-12 at 15:59 -0500, M. Ranganathan wrote:
>> On Fri, Dec 12, 2008 at 3:19 PM, Arjun Nair <[email protected]> wrote:
>> > Dale Worley wrote:
>> >> Looking at 3261 and 2543, I think the point is that in 2543, both from-
>> >> and to-tags are optional, but people rapidly learned that the UAS needs
>> >> to provide a to-tag to allow forking to be handled correctly.
>> >>
>> >> So I think the "compatibility with 2543" point is that the from-tag may
>> >> be missing, but that to-tag processing will be as we expect from 3261.
>> >>
>> >>> if we don't find a from tag in a request, can we just compare the
>> >>> whole from field instead?
>> >>
>> >> That seems to be the correct thing to do.
>> >>
>> >
>> > Right, I understand.. I will implement it as:
>> >
>> > if [ first_cseq == second_cesq ]
>> > then
>> >   if [ first_to_tag == second_to_tag
>> >        || first_to_tag.isNull         # <-- Support for matching dialog 
>> > forming
>> >        || second_to_tag.isNull ]      # <-- requests with the correct 
>> > dialog
>> >   then
>> >
>> >      if [ first_from_tag.isNull
>> >           && second_from_tag.isNull ]
>> >      then
>> >         if [ RFC 2543 : COMPARE THE ENTIRE FROM FIELD ]
>> >         then
>> >            DIALOG_MATCH = TRUE;
>> >         fi
>> >      else if [ first_from_tag == second_from_tag ]
>> >      then
>> >         DIALOG_MATCH = TRUE;
>> >      fi
>> >
>> >   fi
>> > fi
>> >
>> >
>>
>> I think it ought to be fine to reject requests without a From: tag.
>
> Be Liberal In What You Accept.
>
> What would it break to allow requests without a From tag?
>
>

I believe it would cause problems for forked subscriptions but I will
admit, I have a hard time coming up with an example.


-- 
M. Ranganathan
_______________________________________________
sipx-dev mailing list
[email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-dev
Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-dev

Reply via email to