Re: figure positioning problem

[Ben Abbott]

On Nov 19, 2011, at 11:18 AM, John W. Eaton wrote:

> diff --git a/src/graphics.cc b/src/graphics.cc
> --- a/src/graphics.cc
> +++ b/src/graphics.cc
> @@ -3237,9 +3237,9 @@
>     {
>       Matrix old_bb, new_bb;
> 
> -      old_bb = get_boundingbox ();
> +      old_bb = get_boundingbox (true);
>       position.set (v, true, do_notify_toolkit);
> -      new_bb = get_boundingbox ();
> +      new_bb = get_boundingbox (true);
> 
>       if (old_bb != new_bb)
>         {

With this change the problem I mentioned earlier with printing from a script is 
greatly improved. The only remaining problem is that when I run the script 
below ...

        clear all
        v = version;
        dpi = round (get (0, "screenpixelsperinch"));
        paperposition = get (0, "defaultfigurepaperposition");
        position = get (0, "defaultfigureposition");
        position(3:4) = paperposition(3:4) * dpi;
        for backend = {"gnuplot", "fltk"}
          close all
          graphics_toolkit (backend{1});
          figure (1, "position", position)
          plot (rand (2))
          set (gca, "activepositionproperty", "position")
          axis ([1, 2, 0, 1])
          print (sprintf ("-r%d", dpi), sprintf ("Octave-%s-%s.png", v, 
backend{1}))
          drawnow
        endfor

... the height of the FLTK result is off by the height of the status bar, but 
when I then create a second image from the command line ...

        print -r100 test_r100.png

... everything is ok (the FLTK result matched gnuplot).

Ben