Re: [NSE] ASN ipOps and whois

[jah <jah () zadkiel plus com>]

On 06/09/2008 04:54, David Fifield wrote:
> This looks really good. I have checked in the new ipOps.lua and ASN.nse,
> and the patch to dns.lua. I modified ASN.nse and whois.nse to use the
> library instead of duplicating the functions. It appears to work okay
> but I'd like you to check my work. I was surprised once when a function
> I deleted out of the *** UTILITY FUNCTIONS *** section wasn't in ipOps
> (get_prefix_length). I left it duplicated in both scripts.
>   
Nice!
I don't think get_prefix_length would be very useful as part of the
ipOps library as it's really only any use in the context of IP address
assignments/BGP where the prefix length of a range can be determined by
comparing the host bits.
> I found two functions in ipOps.nse that don't appear to be used
> anywhere: todword and get_parts_as_number. Am I right that they are
> unused, or did I miss something? I guess they were in the old ipOps, but
> if we don't use them let's get rid of them. Especially as todword
> doesn't support IPv6 addresses.  
get_parts_as_number was used in the previous isPrivate. Nothing is
apparently using either that or todword.  I kept their functionality
(modifying get_parts_as_number for IPv6) because they're documented
functions and also because people might be using them in unknown scripts.
> Why did you have compare_ip take two addresses and an operator? Maybe
> there's a good reason for it. I would have expected the function to
> return negative, zero, or positive like strcmp.  
I did it this way because both compare_ip and expand_ip are based on
perl functions in Net::IP and the code was straightforward.  compare_ip
is a bit cumbersome, but it works well in conditional statements.  Do
you think strcmp-like return values would suit most people better? 
Could do it both ways and choose your output based on whether the
operator is given...
> If you scan a whole netblock with ASN.nse you get a ton of the same
> answers. Is there a way to make it say "See the result for" like
> whois.nse does?  
This can be done although it's not completely straightforward and will
require some serious testing.  It's a good idea though and I'll get on it.
> Anyway, please check that I integrated everything correctly. This has
> been a lot of work over a few months.
It's all good, no problems.

Regards,

jah


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