From: Guy Harris [mailto:guy@xxxxxxxxxx]
Sent: Tuesday, October 09, 2001 3:42 PM
> "follow_read_stream()" appears to consider the "client" to be the
> host/port from which the first packet in the stream is sent (which isn't
> necessarily the case, of course, if you've caught the traffic in
> mid-stream...).
>
> However, "follow_stream_cb()", I suspect, considers the "client" to be
> the host/port from which the *currently selected packet* is sent. That
> controls which of the last two two items in the option menu is
> considered the "client" direction and which is considered the "server"
> direction.
>
> "follow_read_stream()"s choice is more likely to be correct than is
> "follow_stream_cb()"s choice, so it shouldn't be changed.
>
> Perhaps, instead, we need to have the callbacks for the option menu not
> choose whether to show "client->server" or "server->client", but choose
> one or the other address/port pairs as the source to show.
I think the easy patch would be to add client address & port fields
to the follow_info_t struct. The follow_stream_cb could populate the
field and the follow_read_stream would use the fields instead of
guessing based upon the first packet in the save file.
The titles client/server in the code don't represent the actual
environment, but we don't show the names client/server in the GUI.
The naming problem in the code isn't important at this point and
should be added to the 'gotta clean this when I get time' list.
Jeff Foster
jfoste@xxxxxxxxxxxx
***
The information in this e-mail is confidential and intended solely for the
individual or entity to whom it is addressed. If you have received this
e-mail in error please notify the sender by return e-mail, delete this
e-mail, and refrain from any disclosure or action based on the information.
****
