Nmap Development mailing list archives

Re: --lua-exec packaging


From: David Fifield <david () bamsoftware com>
Date: Fri, 6 Sep 2013 19:01:21 -0700

On Thu, Sep 05, 2013 at 12:16:00AM +0200, Jacek Wielemborek wrote:
2013/9/5 David Fifield <david () bamsoftware com>:
On Thu, Sep 05, 2013 at 12:04:39AM +0200, Jacek Wielemborek wrote:
The way I see it, Ncat --lua-exec scripts could install on Unix to
$PREFIX/share/ncat/scripts. This leaves us some room for future
$PREFIX/share/ncat/filters ($PREFIX is usually /usr or /usr/local). On
Windows (though it's definitely not my district, so feel free to
correct me), I'd suggest %ProgramFiles%\Nmap\Ncat\Scripts (and later,
filters).

I agree on $datadir/ncat/scripts for the installation directory.

I don't think we should put everything that is in the ncat/scripts
directory in the source code there. Some of them are demos that are
better worked into documentation. Some of them are trivial and not
likely to be ever used.

I think the only scripts I would install, of those existing, are
httpd.lua and whatismyip.lua.

I believe that echo.lua, discard.lua and echo.lua could be useful as
well - definitely way less than httpd.lua, still - remember that it's
not easy to achieve this kind of functionality on Windows and this way
you could do it not only easily, but also in a portable manner.

I think that those scripts are better as tutorial examples. I can't
picture anyone actually using them even if they are installed by
default. Can you picture someone using Nmap for their job or whatever,
saying to themself, "I need an echo server right now," and next
wondering if there is an echo Lua script installed with Ncat by default?

What do you think about the search path concept?

I'd prefer to avoid that kind of magic until we see a need for it. It
may seem really trivial, but still searching paths is more likely to
have bugs than not. If someone writes to say, "I love your script
xxx.lua, but I hate typing /usr/local/share/ncat/scripts/xxx.lua every
time," then let's consider it more.

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


Current thread: