Ethereal-dev: Re: [Ethereal-dev] IrDA dissector plugin

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 This list is no longer active.

Date Index Thread Index Other Months All Mailing Lists
Date Prev Date Next Thread Prev Thread Next
From: Jean Tourrilhes <jt@xxxxxxxxxxxxxxxxxx>
Date: Wed, 7 Jan 2004 11:52:26 -0800
On Wed, Jan 07, 2004 at 11:43:45AM -0800, Guy Harris wrote:
> 
> On Jan 7, 2004, at 9:28 AM, Jean Tourrilhes wrote:
> 
> >	As far as I can see, the IrDA protocol really uses 0x2F as the
> >XID command. This is easy to verify, as XID is used for basic
> >discovery. The log below is my Palm-Pilot discovering my Linux box :
> 
> I.e., that's your Palm Pilot sending the XID command?

	Correct (check the last line). Linux does the same.

> Sigh.  I don't know whether that was just somebody putting a "0" where 
> there should have been a "1" in the table of UI command values, or a 
> deliberate choice on the part of the people developing the IrDA spec, 
> but it looks as if "dissect_xdlc_control()" would have to be changed in 
> order to handle IrDA's non-HDLC code for XID commands.
> 
> Probably the right thing to do is to have it take a "const value_string 
> *" argument that points to a value_string table for mapping UI commands 
> to short names; if it's null, the HDLC table is used, otherwise that 
> table is used.  (The long names are used in fields, and the fields are 
> supplied by the dissector calling "dissect_xdlc_control()", so that's 
> not an issue.)
> 
> I'll look at doing that.

	Well, it might be a case where the API/semantic of the
function would become so complex that the benefit of merging the code
would not pan out, because you would not actually reduce the
maintainance overhead.
	Something to ponder... But you know better.

	Jean