Re: Portreasons Changes

[David Fifield <david () bamsoftware com>]

On Fri, May 27, 2011 at 09:55:52PM -0700, David Fifield wrote:
> On Mon, May 23, 2011 at 02:46:28PM -0700, Colin L. Rice wrote:
> > Hello,
> >
> > I wrote a small patch to redo the port-reasons mapping system from ICMP
> > code and type to error type to output strings.
> >
> > Instead of using icmp_mod_code and icmp_mod_type which was confusing
> > there is now a single function icmp_to_reason(type,code) which
> > translates ICMP codes and types to reason codes. This has the additional
> > benefit of allowing us to not worry about overlapping ICMP types and
> > codes in the modular system.
> >
> > Instead of using to arrays to translate the reason codes to plural and
> > singular strings we now use a map combined with a reason_string class.
> > This allows us to add more items to the reason_string in the future such
> > as a detailed description as to how we got this as well as avoiding
> > synchronization issues.
>
> Colin, I'm getting some strange output with this change:
>
> $ sudo ./nmap -PE google.com -sn --reason
>
> Starting Nmap 5.51SVN ( http://nmap.org ) at 2011-05-27 21:54 PDT
> Nmap scan report for google.com (74.125.224.114)
> Host is up, received unknown-responses (0.014s latency).
>
> I expect echo-reply (singular). Do you get the same?

I added handling for echo reply to icmp_to_reason, and swapped the order
of singular and plural definitions, which were backwards. Would you and
make sure that all the reasons that were previously calculated by adding
ER_ICMPTYPE_MOD and ER_ICMPCODE_MOD are handled properly?

David Fifield
_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://seclists.org/nmap-dev/