Re: "\r\r" in script output

[David Fifield <david () bamsoftware com>]

On Mon, Sep 08, 2008 at 01:35:47PM -0600, David Fifield wrote:
> On Sun, Sep 07, 2008 at 02:18:33AM -0600, David Fifield wrote:
> > On Sat, Sep 06, 2008 at 01:30:19PM -0700, Fyodor wrote:
> > > I'm happy to report that David's setup.py fix allowed me to create a
> > > working Nmap Windows installer.  I tested Zenmap, and it worked
> > > against scanme.nmap.org.  But then I tried against a machine on my
> > > locan network and it crashed :(.  Here is the command:
> > >
> > > nmap -T Aggressive -A flog.yuma.net
> > >
> > > Zenmap reported:
> > >
> > > --
> > > "Gtk-ERROR **: Char offset 64 is off the end of the line
> > > aborting...
> > > [OK]
> > > --
> >
> > We figured this bug out eventually. It had to do with Zenmap breaking
> > lines differently than PyGTK was in the syntax highlighting module.
> >
> > The bug was uncovered because some script output has \r\r sequences in
> > it. (PyGTK was breaking lines between each \r, and Zenmap wasn't.) We
> > fixed the bug but the question arises: Where are the \r\r sequences
> > coming from?
>
> It might be in this code in formatScriptOutput in output.cc:
>
> #ifdef WIN32
>         const char* sep = "\r\n";
> #else
>         const char* sep = "\n";
> #endif
>         std::string line_prfx = "|  ";
>
>         char* token = strtok(c_output, sep);
>
>         result += line_prfx + sr.get_id() + ": ";
>
>         while(token != NULL) {
>                 if(line > 0)
>                         result += line_prfx;
>                 result += std::string(token) + sep;
>                 token = strtok(NULL, sep);
>                 line++;
>         }
>
> On Windows that's appending "\r\n" to the end of every line of script
> output. But I think Windows automatically converts "\n" to "\r\n" for
> text streams (such as stdout), so perhaps it's becoming "\r\r\n".

That was the problem, and I fixed it by removing the WIN32 conditional
and just using "\n" everywhere.

David Fifield

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