Re: update to nmap-mac-prefixes

[Daniel Miller <bonsaiviking () gmail com>]

On Wed, Jun 17, 2015 at 10:15 AM, Robin Wood <robin@digi.ninja> wrote:

> On 17 June 2015 at 16:12, Jacek Wielemborek <d33tah () gmail com> wrote:
> > W dniu 17.06.2015 o 16:48, Robin Wood pisze:
> > > I can understand script dependancies changing and being a pain to
> > > maintain but what about datafiles like this one? How often does the
> > > format of these change?
> > >
> > > Robin
> >
> > I second the question - I wouldn't expect the datafiles to rely on
> > implementation changes as much as NSE. Maybe instead of nmap-update, an
> > NSE binding for pulling the datafiles would be feasible? I mean
> > something like:
> >
> > nmap --script=update-datafiles
>
> Also, if the script pulled down the raw data then it would parse it
> into the file format it required so whatever version of the script a
> user has the file created will be appropriate for them.
>
> A flaw with this is if the provider changes the file format, the best
> way around this would be to have some very solid validity checking and
> keep a backup of the previous datafile in case a problem is found.
This approach could work for nmap-mac-prefixes, nmap-rpc, and
nmap-protocols, but nmap-services contains a port frequency column that is
generated from private Nmap scan data. I suppose we could separate that
frequency data out from the names and port numbers in the services file,
but that would require rewriting a lot of the parsing routines within Nmap
itself.

See https://github.com/nmap/nmap/issues/162 and
https://github.com/nmap/nmap/issues/163 for more files like this and their
sources (please add any I have missed!).

Dan

_______________________________________________
Sent through the dev mailing list
https://nmap.org/mailman/listinfo/dev
Archived at http://seclists.org/nmap-dev/