On Wed, 2008-12-17 at 00:10 +0100, Iñaki Baz Castillo wrote: > Hi, I'm trying a softphone which sends a MESSAGE with: > > -------- > Content-Type: text/plain > Content-Encoding: UTF-8 > Content-Length: 5 > > HELLO > --------
Content-Encoding is used to specify a process that is applied to the "real" body's octets to produce the octets that are actually the body of the message. The default value is "identity", which means no change, and "gzip" is defined. But I don't think there's any real use for it in SIP. It makes sense in HTTP when returning large documents. The character set that is used to encode the body is (or would be) specified by the "charset" parameter of the Content-Type header -- see RFC 3261 section 20.15 and RFC 4021 section 2.2.5. But I would be surprised if any implementation handles anything other than UTF-8, except maybe for MESSAGE. I would be surprised if many implementations can even handle the charset parameter, though it's defined in 3261. Dale _______________________________________________ Sip-implementors mailing list [email protected] https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors
