Ethereal-dev: Re: [Ethereal-dev] COL_INFO for andX commands

Note: This archive is from the project's previous web site, ethereal.com. This list is no longer active.

From: "Ronnie Sahlberg" <sahlberg@xxxxxxxxxxxxxxxx>
Date: Tue, 23 Jul 2002 07:28:21 +1000
Yes that would be really useful.
(however I dont like when you in the summary pane get multiple lines for one
single packet
but that is too late now since we have left behind the one packet one
dissected unit
with the reassembly and all)

In order to get this for TCP services I think we would need to first make
sure all
dissectors running ontop of TCP would return an int specifying the number of
bytes
eaten by the dissector so TCP would know how much was consumed
and then also move the loop-over-all-PDUs-in-one-tvb loop from the dissector
down
to the TCP layer so the TCP layer could add the new line to the summary
pane.

There might be some other hairy details to take care of for the summary pane
as well since
we would now start getting multiple lines for one frame.

Perhaps it would be easiest if we just allowed the COL_INFO field to become
multiline?

This may be something to place in the TODO line since it may be quite some
effort and may take some time
to implement.

----- Original Message -----
From: "Guy Harris"
Subject: Re: [Ethereal-dev] COL_INFO for andX commands


> On Tue, Jul 23, 2002 at 07:00:20AM +1000, Ronnie Sahlberg wrote:
> > It might be difficult to handle for the NBSS layer though since we dont
get
> > (and dont want to get?) information
> > to the SMB dissector of what is happening down in the NBSS dissector.
>
> It can find out if it's happening in the same frame by checking
> "pinfo->fd->num".  The trick is remembering the frame number in such a
> way that stale remembered data gets discarded.
>
> We might also want an option to display *multiple* lines in the summary
> pane for a single frame; that could come in handy if
>
> 1) you wanted to see the summary lines at *more than one
>    protocol layer* (e.g., TCP and some protocol above it);
>
> 2) you wanted a summary line for every higher-level PDU in a TCP
>    segment - or in some other lower-level protocol.