Re: [NSE] resolveall prerule, nmap.resolve(), nmap.address_family()

[David Fifield <david () bamsoftware com>]

On Wed, Aug 11, 2010 at 02:03:14AM -0500, Kris Katterjohn wrote:
> > I've merged the Script Pre-scanning and Post-scanning phases to Nmap
> > trunk. I hope that we'll see lot of scripts that make use of them.
>
> I've attached a patch and new prerule script for adding new targets to Nmap
> (target adding functionality is still not in trunk yet, but here's this anyway
> like my snmp-interfaces changes).  The script is called resolveall and just
> takes a host as it's only argument, resolveall.host.
>
> Several months ago there was discussion on Nmap scanning all addresses for a
> given host name, which is a very bad idea for default behavior.  As far as I
> know, nothing else came of this (no options, etc), so I decided to write this.
>
> I've also added general functions to the nmap library: resolve(host,af) takes
> a name and optional address family ("inet", etc) and returns a table with all
> of the addresses; address_family() returns Nmap's o.af() value as "inet" or
> "inet6" like resolve() expects.  If resolve() is not given an address family
> then it doesn't care and returns all addresses.

I think that resolve and address_family look good. Please commit them.
To document them, you just add a phony function in nselib/nmap.luadoc.

As for the script, it looks good to me but nobody has had much to say
about it. One suggestion I would make is to allow the script argument to
take an array of host names, not just a single one. It would be like
this:

--script-args='newtargets,resolveall.hosts={host1,host2}'

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