Re: fltk window not resizable

[Ben Abbott]

On Feb 20, 2011, at 4:49 AM, Kai Habel wrote:

> On 19.02.2011 19:08, Ben Abbott wrote:
> 
>> On Feb 19, 2011, at 12:26 PM, Kai Habel wrote:
>> 
>>> On 15.02.2011 17:33, bpabbott wrote:
>>> 
>>>> On Feb 15, 2011, at 11:30 AM, Benjamin Lindner<address@hidden>  wrote:
>>>> 
>>>>> Hello list,
>>>>> 
>>>>> Using a 3-4-0 snapshot on mingw platform, the fltk plotting window is
>>>>> not resizable. It's movable, and I can minimize&restore, but I cannot
>>>>> resize.
>>>>> Is this intentional?
>>>>> Is it the same on other platforms, or just on windows?
>>>>> 
>>>>> benjamin
>>>> 
>>>> I see this on MacOS as well.
>>>> 
>>>> Ben
>>> 
>>> After having a closer look, I noticed that we already have a call to 
>>> size_range in the plot_window constuctor. The following changeset call the 
>>> size_range method early. After applying it, I can resize the window (on 
>>> MinGW). Does this work for both of you as well?
>>> 
>>> Kai
>> 
>> I now have a place to grab and *attempt* to resize the window. 
>> Unfortunately, it does not change size.
>> 
>> Even more bizarrre is that if I do ...
>> 
>>      print -dpdfwrite "-S1800,1000" test.pdf
>> 
>> ... I'm able to resize the window from its default size up to 1800x1000 
>> pixels.
>> 
>> If I then ....
>> 
>>      print -dpdfwrite "-S18,10" test.pdf
>> 
>> ... I'm able resize the window from (18 to 1800) x (10 to 1000) pixels in 
>> size.
>> 
>> I'm using FLTK-1.1.10.
>> 
>> Ben
>> 
> 
> Strange,
> 
> out of curiosity what does happen on macos if we call size_range twice? I can 
> still resize on linux and mingw with the attached changeset.
> 
> Kai

That fixed half the problem. Now I can decrease the window size, but not 
increase it.

The printing trick above fixe the problem because the window size changed. 
Thus, I can now get full functionality by ...

        close all
        graphics_toolkit fltk
        plot (1:10)
        sz = get (0, "screensize")
        p = get (gcf, "position")
        set (gcf, "position", sz)
        set (gcf, "position", p)

I tried searching for similar reports by MacOS users of FLTK, but didn't find 
anything. Looks like (1) its just me, or (2) Octave is doing something unusual. 
I have a second Mac at work. I'll do some experiments on that this week and see 
I can duplicate the problem.

Ben