[Guy Harris <guy@xxxxxxxxxx>]

On Tue, Feb 19, 2002 at 10:56:52AM +0100, andreas.sikkema@xxxxxxxxxxx wrote:
> > I'm not sure the RTP and RTCP dissectors are an issue - see how
> > "packet-rtsp.c" does it.  It just gets handles for the RTP and RTCP
> > dissectors, and uses "conversation_set_dissector()".
> 
> Can you do this from inside a plugin?

Not currently, but we can add the relevant routine to the plugin API
vector ("conversation_set_dissector()" - "find_dissector()",
"find_conversation()', and "conversation_new()" are already there).

> > TPKT is both a dissector (for port 102, as per RFC 1006) and a set of
> > service routines for other dissectors to use.
> 
> I know, I wrote the first implementation, but can I get to the service
> routines from inside a plugin?

You won't need to with my proposed scheme - the TPKT dissector would
decide whether it sees a TPKT header or not, and would only call the
subdissector's "check" routine if it sees one (so the subdissector
wouldn't need to call the "is this TPKT?" routine), and would also put
the TPKT header into the protocol tree and set up the tvbuff for the
"dissect" routine (so the subdissector wouldn't need to call the
"dissect a TPKT header" routine).

> > seem to show TPKT headers in a TCP segment by themselves, the heuristic
> > TPKT dissector should probably treat as the beginning of a TPKT PDU a
> > TCP segment that either
> 
> That is implementation dependant.

Dependent on what?