Re: [nmap-svn] r18902 - nmap

[David Fifield <david () bamsoftware com>]

On Sat, Jul 17, 2010 at 08:04:16AM -0400, Patrick Donnelly wrote:
> On Sat, Jul 17, 2010 at 7:59 AM,  <commit-mailer () insecure org> wrote:
> > Author: batrick
> > Date: Sat Jul 17 04:59:04 2010
> > New Revision: 18902
> >
> > Log:
> > Clean up code. Fix port.version.service_fp not getting set properly.
> > Added field service_dtype (which was replacing service_fp) to version table.
> >
> >
> > Modified:
> >   nmap/nse_nmaplib.cc
>
> In addition to cleaning up this code there appears to be an error
> where the port.version.service_fp was being replaced with what I think
> was supposed to be a service_dtype field:
>
> -  SCRIPT_ENGINE_PUSHSTRING_NOTNULL(sd->service_fp, "service_fp");
> -
> -  switch(sd->dtype) {
> -    case(SERVICE_DETECTION_TABLE):
> -      SCRIPT_ENGINE_PUSHSTRING_NOTNULL("table", "service_fp");
> -      break;
> -    case(SERVICE_DETECTION_PROBED):
> -      SCRIPT_ENGINE_PUSHSTRING_NOTNULL("probed", "service_fp");
> -      break;
> -    default:
> -      fatal("%s: In: %s:%i This should never happen.",
> -        SCRIPT_ENGINE, __FILE__, __LINE__);
> -      break;
> -  }
>
> As you can see, the first set to the "service_fp" field is always
> replaced by this sd->dtype value. I went ahead and added a
> "service_dtype" field instead. I will document this change assuming no
> one has a problem with it.

That looks fine. I was going to suggest calling it just "dtype" instead
of "service_dtype", because the other fields match their C struct names.
But on reflection I think that doesn't matter so much. How about calling
it "method", which matches the attribute in XML output?

<service name="X11" method="table" conf="3"/>

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