Re: [Wireshark-dev] dissect_sip_tcp_heur()

2013-08-12 Thread Hadriel Kaplan
On Aug 12, 2013, at 6:22 AM, Martin Kaiser wrote: >> I'm not really familiar with SIP over TCP (in all my use cases, SIP is >> over UDP) but your proposal makes sense. > > Thanks Pascal, > > that explains why nobody spotted this so far. I think it's more because it's only the heuristic SIP/TC

Re: [Wireshark-dev] dissect_sip_tcp_heur()

2013-08-12 Thread Martin Kaiser
> I'm not really familiar with SIP over TCP (in all my use cases, SIP is > over UDP) but your proposal makes sense. Thanks Pascal, that explains why nobody spotted this so far. I committed the change in r51313 and I'll schedule it for backporting to 1.10 and 1.8 if nobody complains. Best regard

Re: [Wireshark-dev] dissect_sip_tcp_heur()

2013-08-12 Thread Pascal Quantin
2013/8/10 Martin Kaiser > I came across this while browsing through the coverity defects. > > gboolean first = TRUE; is never changed > > Does the following change make sense to those who are familiar with SIP? > > diff --git a/epan/dissectors/packet-sip.c b/epan/dissectors/packet-sip.c > index f