From: "Attila Sipos" <[EMAIL PROTECTED]>

   Looking on an ethernet snooper, the beginning of
   the display-name is:
   22 cc e1

   Now obviously 22 is the open quote ( " ).

   Next is cc.

   Now, to me, cc is UTF8-NONASCII so next comes 1 UTF8-CONT.
   But no, the next byte is e1 which is not a UTF8-CONT.

   Is "22 cc e1", the beginning of an illegal display name?

   Or where am I going wrong?

That way of looking at it is correct.  (That is, the provided
display-name is malformed.)

An alternative approach (with the same effect) is to consider that SIP
message headers are sequences of characters in the Unicode character
set, represented as octets using UTF-8.  (Bodies can have other
encodings, IIRC.)  But in UTF-8, an octet in the range C0-DF is always
followed by an octet in the range 80-BF.  Since this rule is violated
by the octets you report above, the octets do not represent *any*
characters via UTF-8, and the purported display-name is an invalid
encoding.

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

Reply via email to