Re: [PATCH] Flush console input queue differently in nmap_tty.cc

[kx <kxmail () gmail com>]

Will FlushConsoleInputBuffer work?

http://windowssdk.msdn.microsoft.com/en-us/library/ms683147.aspx

Sorry I don't have time to try it myself.

-kx

On 8/22/06, Kris Katterjohn <kjak () ispwest com> wrote:
> This uses tcflush() in tty_getchar() under the !WIN32 section and does
> the "flush loop" inside tty_getchar() in the WIN32 section. It's
> definitely better for the UNIX part, and is probably better for WIN32
> because only _getch() is looped instead of the whole tty_getchar().
>
> I tried to find something like tcflush() for Windows, but I didn't see
> anything. Maybe kx knows of something.
>
> It's a diff against 4.20ALPHA4.
>
> Thanks,
> Kris Katterjohn
>
>
> --- x/nmap_tty.cc       2006-03-08 15:45:01.000000000 -0600
> +++ y/nmap_tty.cc       2006-08-22 18:26:07.000000000 -0500
> @@ -126,7 +126,21 @@ extern NmapOps o;
>
>  // Microsoft's runtime makes this fairly simple. :)
>  void tty_init() { return; }
> -static int tty_getchar() { return _kbhit() ? _getch() : -1; }
> +
> +static int tty_getchar()
> +{
> +       if (_kbhit()) {
> +               int c = _getch();
> +
> +               /* flush input queue */
> +               while (_getch() >= 0);
> +
> +               return c;
> +       }
> +
> +       return -1;
> +}
> +
>  static void tty_done() { return; }
>
>  #else
> @@ -167,6 +181,10 @@ static int tty_getchar()
>  #endif
>                c = 0;
>                 numChars = read(tty_fd, &c, 1);
> +
> +               /* flush input queue */
> +               tcflush(tty_fd, TCIFLUSH);
> +
>                if (numChars > 0) return c;
>        }
>
> @@ -232,9 +250,6 @@ bool keyWasPressed()
>     return false;
>
>   if ((c = tty_getchar()) >= 0) {
> -    // Eat any extra keys (so they can't queue up and print forever)
> -    while (tty_getchar() >= 0);
> -
>     // printf("You pressed key '%c'!\n", c);
>     if (c == 'v') {
>        o.verbose++;
>
>
>
> _______________________________________________
> Sent through the nmap-dev mailing list
> http://cgi.insecure.org/mailman/listinfo/nmap-dev
> Archived at http://SecLists.Org

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