Ethereal-dev: Re: [ethereal-dev] DIsplay of flags in packet pane

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 This list is no longer active.

From: John McDermott <jjm@xxxxxxxxxx>
Date: Wed, 12 Apr 2000 11:46:49 -0600

Ben Fowler wrote:

> >That doesn't necessarily mean they'll associate the letter you've chosen
> >with the field (especially if more than one field begins with the same
> >letter).  I *do* consider obviousness important, and would prefer not to
> >*have* to learn what the one-letter mnemonics mean before I can
> >understand a dissection.
> 
> That is a perfectly good point & makes you free to give less weight
> to my idea. My perspective is that I know the protocol (or I am prepared
> to learn it) and want a tool to give me the bits in a reasonable format. I am
> one of the 'they' in you paragraph.
> 

I know a lot of IP-family protocols pretty well.  The set/not-set is
fine for me as a one who knows the protocols.  A novice might need
something like
	Urgent: Set [Urgent data exists in this datagram]
but then we are discussing whether or not Ethereal is a teaching tool or
a tool for protocol *analysis*.  I prefer the latter even though I use
it in courses for teaching protocols.

I also would have a hard time associating letters with protocols unless
the letters were part of the spec (as is the case with C and Z bits in
some processors).


> >I consider the following:
> >
> >     Flags: 0x0012 (SYN, ACK)
> >         ..0. .... = Urgent: Not set
> >         ...1 .... = Acknowledgment: Set
> >         .... 0... = Push: Not set
> >         .... .0.. = Reset: Not set
> >         .... ..1. = Syn: Set
> >         .... ...0 = Fin: Not set
> >
> >to display a collection of 6 one-bit Boolean bitfields, so I *do*
> >consider you to be dealing with bitfields.

I agree.

> 
> You know the code better than I ever will. At the moment I feel
> that flags or booleans with are inherently on or off can with
> advantage be treated as a special case wheres bitfields
> are more like tiny its and should be treated as UINT or numeric.

I guess I'm misunderstanding.  The example above seems to contain
Boolean bitfields.  Some header values are 2-bit values (no pun
intended!) which can be interpreted as some sort of unsigned int as you
say.

>
--john

-- 
John McDermott jjm@xxxxxxxxxx
Writer and Computer Consultant
J-K International, Ltd.
+1 505/377-6293 - V
+1 505/377-6313 - F