El Sábado, 29 de Marzo de 2008, M. Ranganathan escribió:

> That would be a very very bad idea because existing implementations
> are going to barf. 

It's curious you say that since the fact is that many of the today existing 
SIP implementations will fail if you write an horizontal TAB after the header 
name and before the ":". For example:

  Content-Length HTAB  :  25

But note that HTAB is allowed there:

  Content-Length  =   ( "Content-Length" / "l" ) HCOLON 1*DIGIT 
  HCOLON  =  *( SP / HTAB )   ":"   SWS 
  SP  =  %x20
  HTAB  =  %x09


This is, the ultra-permissive syntax of SIP makes most of the implementations 
fail when parsing a SIP message if it contains a non common but valid syntax 
(as in the example above).



> It would be best to invent a new protocol 
> entirely. I am looking for a catchy name... hmmm.... Lets call it
> "Jingle" shall we. It will be written in nice XML which you can also
> make entirely incomprehensible....

I don't like XML since it's not human readable (not easy to read).
Anyway XML wouldn't be the final solution since the same issue as now could 
occur if the XML values are so permissive as they are now.

Regards.


-- 
Iñaki Baz Castillo

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

Reply via email to