From: Paul Kyzivat <[EMAIL PROTECTED]> Consider the following:
From: PaulKyzivat <sip:[EMAIL PROTECTED]> From: "PaulKyzivat" <sip:[EMAIL PROTECTED]> I can certainly imagine that a stack might receive either of these, and convert into some internal form that is identical for both. Then when generating a new request it might re-encode it using a different form than it received. For instance it might always use the quoted form. Or it might always use the token form if it could be represented that way, and only use the quoted form if it had to. Its not clear if that would be valid behavior or not. It's not even clear that there is a *definition* of valid behavior. There is a set of rules for comparing URIs, so one can assert that a particular change leaves a URI equivalent to the original. But I don't think there are any rules for name-addr's, or for any other header usages. IMHO the *effect* of any value should be specified only in terms of the "un-quoted" value, and this should be a blanket policy. But that's not stated anywhere in SIP. However, it's also not denied anywhere, and such an effect is stated for some particular circumstances (e.g., components of SIP URIs). So the situation is unclear, and it probably leads to more confustion and interoperability failure than it should. Dale _______________________________________________ Sip-implementors mailing list [email protected] https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors
