Re: units updater for figures is broken when using fltk

[Ben Abbott]

On Feb 7, 2012, at 10:40 AM, Michael Goffioul wrote:

> On Sun, Feb 5, 2012 at 5:00 PM, Ben Abbott <address@hidden> wrote:
>>> I think the problem is not in get_boundingbox, but in the incorrect
>>> interpretation of the position property in the FLTK code. Moreover,
>>> the code is probably not robust against feedback loops when changing
>>> properties, like:
>>> 1) the user change figure position
>>> 2) the backend is notified and move/resize the figure window
>>> 3) as in any toolkit, moving/resizing the window triggers corresponding 
>>> events
>>> 4) the backend receives events and propagates the changes back to the
>>> figure position (this is needed as the move/resize could be the result
>>> of a user action with the mouse)
>>> 5) if the newly computed position does not match the previous one, goto 1)
>>> 
>>> Step 4) above should not happen when the position change is coming
>>> from octave, but I suspect it does, which can explain the infinite
>>> loop you're seeing.
>>> 
>>> Michael.
>> 
>> The loop may have been due to my not understanding what I was doing. I used 
>> ...
>> 
>>        Matrix pos = fp.get_boundingbox (true);
>> 
>> But I don't understand what the boolean arg is intended for. In graphics.cc 
>> (figure::properties::get_boundingbox) the boolean is named "internal" and is 
>> references in the snippet below.
>> 
>>        Matrix pos = (internal ?
>>                      get_position ().matrix_value () :
>>                      get_outerposition ().matrix_value ());
>> 
>> I choose get_boundingbox (true) because I don't know the purpose of the 
>> outerposition property for figures (there's none in Matlab). I notice you 
>> added it  below.
>> 
>>        
>> http://hg.savannah.gnu.org/hgweb/octave/annotate/d99aa455296e/src/graphics.cc
>> 
>> What is the figure's outerposition property for ?
> 
> http://www.mathworks.co.uk/help/techdoc/ref/figure_props.html#OuterPosition
> 
>> Back to the need to obtain the figure position property in pixels, I 
>> attempted the change below, but ran into scoping problems.
>> 
>>  void draw (void)
>>  {
>>    Matrix screen_size = screen_size_pixels ();
>>    Matrix pos = fp.get_position ().matrix_value ();
>>    pos = convert_position (pos, get_units (), "pixels", screen_size);
>>    Fl_Window::resize (pos(0), pos(1), pos(2), pos(3) + status_h + menu_h);
>> 
>>    return Fl_Window::draw ();
>>  }
> 
> The code above does not flip the Y axis. If you also flip the Y-axis,
> it will be equivalent to get_boundingbox(true) and you'll probably run
> into the same endless loop.

I've been experimented with a few approaches, but always encounter the loop 
problem. 

Even the example below starts looping if the windows is moved with the mouse.

diff --git a/src/DLD-FUNCTIONS/__init_fltk__.cc 
b/src/DLD-FUNCTIONS/__init_fltk__.cc
--- a/src/DLD-FUNCTIONS/__init_fltk__.cc
+++ b/src/DLD-FUNCTIONS/__init_fltk__.cc
@@ -1173,9 +1173,14 @@
 
   void draw (void)
   {
+    // Get the figure position in pixels
+    octave_value funits = fp.get_units ();
+    fp.set_units (octave_value ("pixels"));
     Matrix pos = fp.get_position ().matrix_value ();
+    fp.set_units (funits);
+    // Adjust for the menu and status bars
     Fl_Window::resize (pos(0), pos(1), pos(2), pos(3) + status_h + menu_h);
-
+    // Draw the figure
     return Fl_Window::draw ();
   }

My impression is another place where a similar change is needed ... either that 
or I'm just missing something ? 

Ben