El Jueves, 12 de Marzo de 2009, Hadriel Kaplan escribió:
> > -----Original Message-----
> > From: [email protected] [mailto:sip-
> > [email protected]] On Behalf Of Iñaki Baz
> > Castillo
> >
> > About it I would really understand why TEL2SIP conversion is *so*
> > complex. Why
> > TEL parameters are appended to the SIP userinfo part instead of being
> > converted into SIP URI parameters?
> > Any useful reason for it?
>
> In theory the idea is that a parameter is an attribute of the thing, so a
> user parameter is an attribute of the user part, while the URI parameter is
> an attribute of the whole URI.  So originally it made sense that the TEL
> URI parameters be SIP user part ones, because they were attributes of the
> number itself, not of a SIP URI.
>
> There was also some concern about namespace collision of parameter names,
> if the two types of URI's just happened to define the same param name with
> different meaning. (though the odds of that are nill)


Thanks, let me one question more:

When converting to SIP URI, RFC 3261 doesn't mandate that the TEL URI number 
is "cleaned" (visual separators removed). This is confusing:

tel1 = tel:+123-456
tel2 = tel:(+1)-23-45-6

Both TEL URI are equal since the number is compared after removing visual 
separators, but when converting to SIP:

sip1 = sip:[email protected];user=phone
sip1 = sip:(+1)[email protected];user=phone

both SIP URI are different since userinfo doesn't match. Or perhaps userinfo 
should be compared by removing visual separators when the SIP URI contains 
a "user=phone" parameter?

Thanks.

PD: This is too much complex, really a pain. I suspect it is really unusable 
and nobody will implement it "properly".


-- 
Iñaki Baz Castillo

_______________________________________________
Sip-implementors mailing list
[email protected]
https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors

Reply via email to