On Mon, Apr 10, 2000 at 04:37:25PM +0300, Santeri Paavolainen wrote:
> Two packets, and their display in the details window:
>
> Flags: 0x06
> .0.. = Don't fragment: Set
> ..0. = More fragments: Set
>
> Flags: 0x00
> .0.. = Don't fragment: Not set
> ..0. = More fragments: Not set
>
> Ahem... (quick diff between 0.8.4 and 0.8.6 packet.c and packet-ip.c
> didn't show me any obvious change in at least the bitfield manipulation
> routines and their use)
Gilbert, is this a result of the issue mentioned here?
Date: Fri, 07 Apr 2000 05:07:05 -0500
From: Gilbert Ramirez <gram@xxxxxxxxxx>
Subject: Re: [ethereal-dev] new release soon
To: Ben Fowler <wapdev@xxxxxxxxxxxx>
Cc: ethereal-dev@xxxxxxxx
On Thu, Apr 06, 2000 at 06:32:48AM +0100, Ben Fowler wrote:
> I am working on an 'issue' that I have that the FT_BOOLEAN type is
> not working as I expect (the bitmask parameter seems to be ignored),
> but I don't know if there is an ugly head here.
>
> Ben.
Yeah, that one wouldn't be an "ugly head issue", since that behavior
(bug? feature? :-) is old, not new.
In proto.c, proto_tree_set_uint() takes into account the bitmask.
proto_tree_set_boolean() does not.
The functions that then deal with numbers and bitmasks always
test hfinfo->bitshift, so if we were to set the bitmask in
proto_tree_set_boolean(), we might be lucky enough that it would just
automatically work. :)
--gilbert
