[Octave-bug-tracker] [bug #48017] --no-window-system command line option does not disable graphics

[Rik]

Follow-up Comment #2, bug #48017 (project octave):

Our startup mechanism is so convoluted now that I'm not always sure what is
going on.

I see in src/main.cc the following:


int
main (int argc, char **argv)
{
  int retval = 0;

  bool start_gui = true;
  bool gui_libs = true;

  std::string octave_bindir = get_octave_bindir ();
  std::string octave_archlibdir = get_octave_archlibdir ();
  std::string octave_cli
    = octave_bindir + dir_sep_char + "octave-cli-" OCTAVE_VERSION;
  std::string octave_gui = octave_archlibdir + dir_sep_char + "octave-gui";

#if defined (HAVE_OCTAVE_QT_GUI)
  // The Octave version number is already embedded in the
  // octave_archlibdir directory name so we don't need to append it to
  // the octave-gui filename.

  std::string file = octave_gui;
#else
  std::string file = octave_cli;
#endif

  // Declaring new_argv static avoids leak warnings when using GCC's
  // --address-sanitizer option.
  static char **new_argv = new char * [argc + 1];

  int k = 1;

  bool warn_display = true;

  for (int i = 1; i < argc; i++)
    {
      if (! strcmp (argv[i], "--no-gui-libs"))
        {
          // Run the version of Octave that is not linked with any GUI
          // libraries.  It may not be possible to do plotting or any
          // ui* calls, but it will be a little faster to start and
          // require less memory.  Don't pass the --no-gui-libs option
          // on as that option is not recognized by Octave.

          start_gui = false;
          gui_libs = false;
          file = octave_cli;
        }
      else if (! strcmp (argv[i], "--no-gui"))
        {
          // If we see this option, then we can just exec octave; we
          // don't have to create a child process and wait for it to
          // exit.  But do exec "octave-gui", not "octave-cli", because
          // even if the --no-gui option is given, we may be asked to do
          // some plotting or ui* calls.

          start_gui = false;
          new_argv[k++] = argv[i];
        }
      else if (! strcmp (argv[i], "--silent") || ! strcmp (argv[i],
"--quiet")
               || ! strcmp (argv[i], "-q"))
        {
          warn_display = false;
          new_argv[k++] = argv[i];
        }
      else
        new_argv[k++] = argv[i];
    }

  new_argv[k] = 0;

  if (gui_libs || start_gui)
    {
      int dpy_avail;

      const char *display_check_err_msg = display_available (&dpy_avail);

      if (! dpy_avail)
        {
          start_gui = false;
          gui_libs = false;

          file = octave_cli;

          if (warn_display)
            {
              if (! display_check_err_msg)
                display_check_err_msg = "graphical display unavailable";

              std::cerr << "octave: " << display_check_err_msg << std::endl;
              std::cerr << "octave: disabling GUI features" << std::endl;
            }
        }
    }


For starters, I think if a user explicitly sets the --no-window-system option
then they don't need to be warned about the lack of a graphical display.  This
suggests that the following code needs to be extended to include
"--no-window-system" and "-W".  


    else if (! strcmp (argv[i], "--silent") || ! strcmp (argv[i], "--quiet")
               || ! strcmp (argv[i], "-q"))
        {
          warn_display = false;
          new_argv[k++] = argv[i];
        }


Second, 


      const char *display_check_err_msg = display_available (&dpy_avail);


Is this call made before, or after, the call to no_window_system()?  If it is
made before then it will screw things up.


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?48017>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/