Re: Some hints for nmap scripts

[jah <jah () zadkiel plus com>]

On 02/12/2008 17:44, Patrick Donnelly wrote:
> On Tue, Dec 2, 2008 at 9:00 AM, jah <jah () zadkiel plus com> wrote:
>   
> > On 02/12/2008 15:41, Patrick Donnelly wrote:
> >     
> > > On Tue, Dec 2, 2008 at 7:49 AM, jah <jah () zadkiel plus com> wrote:
> > >
> > >       
> > > > The only script currently using this stuff is one named promiscuous.nse
> > > > in Nmap versions up to 4.76 (now named sniffer-detect.nse) (as an aside,
> > > > this script has a commented-out section between --[[ and ]]-- which is
> > > > not the correct way [3] - it should be --[[ and --]], but for some
> > > > reason the script still works as expected and considers the improperly
> > > > commented section as a properly commented section - v. odd.)
> > > >
> > > >         
> > > This is actually a "trick" Roberto mentions in his book, PiL, in [1].
> > > The comments in sniffer-detect.nse are correct.
> > >
> > > [1] http://www.lua.org/pil/1.3.html
> > >       
> > I think the trick you refer to is to uncomment a previously commented
> > block by adding a third hyphen ---[[ --]]
> >     
>
> This is correct. I misread your earlier post taking to mean that --[[
> ]] commenting should be --[[ --]]. The --[[ ]]-- works because the
> final two hyphens are interpreted as a single line comment (which of
> course is empty as there is nothing between the hyphens and the
> newline).
I expected that improperly closing a multiline comment in this way would
mean that the comment would extend to the end of the file, but it
clearly doesn't.
]] is all that's required to close such a comment.

It all starts to get rather messy if a commented block happens to
contain a multiline string  foo=[[  ]]  which results in a script
compilation error: nesting of [[...]] is deprecated near '['

Not a big deal of course, but worth noting.

jah

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