ANNOUNCEMENT: Live Wireshark University & Allegro Packets online APAC Wireshark Training Session
April 17th, 2024 | 14:30-16:00 SGT (UTC+8) | Online

Wireshark-dev: Re: [Wireshark-dev] questions about conversations

From: "Anders Broman \(AL/EAB\)" <anders.broman@xxxxxxxxxxxx>
Date: Wed, 16 May 2007 12:03:07 +0200
Hi,
If there is stuff that should only be done once you can use:
if((!pinfo->fd->flags.visited)
See packet-uma.c

Regards
Anders 

-----Original Message-----
From: wireshark-dev-bounces@xxxxxxxxxxxxx
[mailto:wireshark-dev-bounces@xxxxxxxxxxxxx] On Behalf Of Tomasz Noinski
Sent: den 16 maj 2007 11:57
To: wireshark-dev@xxxxxxxxxxxxx
Subject: [Wireshark-dev] questions about conversations

Hi,


I'm writing my own dissector and I've been quite successful, but lately
I wanted to add some dynamic analysis (based on info from other packets)
and I stumbled into some problems.

All I know about conversations is from README.developer.
My dissector works well if it's called once for each frame, in correct
order. The problems start when I click on a frame in Wireshark window -
then my dissector is called for this frame again, _twice_.

I guess it's the correct behaviour that the dissector is called several
times, so my main question is:
What is the suggested way to make sure the  dissection result is the
same in the first dissector call and the later ones (when I click)?


Also:

Is the conversation data somehow saved for each frame when it changes or
is there always exactly one such data structure for one conversation in
the program?

Is there an easy way to tell my dissector is being run the first time
for a specific frame (i.e. just after capture, not after filtering or
after clicking on a packet)?

Why is the dissector called twice with every click on a frame in the
Wireshark window - is this the desired behaviour?


Thanks in advance,

Noix
_______________________________________________
Wireshark-dev mailing list
Wireshark-dev@xxxxxxxxxxxxx
http://www.wireshark.org/mailman/listinfo/wireshark-dev