On Feb 21, 2014, at 6:36 PM, Guy Harris <[email protected]> wrote:

> On Feb 21, 2014, at 12:08 PM, Hadriel Kaplan <[email protected]> 
> wrote:
> 
>> Also, FT_IPv4 and FT_IPv6 are frequently in duplicate fields.  Should they 
>> be/not-be?  Display filter input/verification might have issues with it, but 
>> it seems logical to have generic "foo.src"/"foo.dst"/etc. fields of both 
>> types.
> 
> The one place where we're doing that with ".src" and ".dst" is in the PGM 
> dissector; in, for example, a Source Path Message, there's a field specifying 
> the Address Family Indicator (AFI) for the source address and another 
> specifying the address, which could be IPv4, IPv6, or, in theory, a number of 
> other types.

And it's also done in: h245.network, h248.address, mih.mihf_id, 
openflow_v4.oxm.value, openflow_v5.oxm.value, pflog.saddr, 
rsip.parameter.address, rsvp.notify_request.notify_node_address_ipv4, 
sap.originating_source, sflow_245.nexthop, and a bunch in pim.

I don't know if it actually works properly as display filters, however, in a 
capture/file of mixed address families.

-hadriel

___________________________________________________________________________
Sent via:    Wireshark-dev mailing list <[email protected]>
Archives:    http://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
             mailto:[email protected]?subject=unsubscribe

Reply via email to