On 26.03.2012 17:15, MacArthur, Ian (SELEX GALILEO, UK) wrote: > Here's a thing... > > Playing about with the fltk-1 compat layer in fltk3, I notice that when > running the fullscreen test app (at least on Win32) then the GL subwindow > that shows the shape is not drawn. > > However, if I remove the window.end(); call at line 196 of fluscreen.cxx > (r9302) then all is well (so long as the test is only run in single-window > mode of course...) > > So, it would appear that creating the parent window, then "end"ing it, > creating a GL window, then adding it as a subwindow of the parent window, > does not work. > > Creating the GL subwindow while the parent window is still "begin"'d however > does work. > > Note that the "same" (i.e. equivalent) test works fine in "pure" fltk3 and in > fltk-1.3, so this is some sort of interaction at the compatibility layer, > AFAICT. > > FWIW, I have a slightly modified "working" version of the fltk-1.3 test code > that works fine in either fltk-1.3 or fltk3-compat builds, but that was done > by avoiding "end"ing the parent window before creating the GL window, IFF the > GL window is to be nested as a subwindow... > > So, it *can* work, but that's not quite the same thing as proving that the > compat-layer is working! > > I have not tried on a non-WinXX system yet, so I do not know how generic this > problem is; it may be peculiar to the fltk3 port on Win32, or it may be more > general.
Thanks for doing all these tests. I hope that I will have some more time to look deeper into FLTK 3 in the near future... > I wonder how Matthias is these days; hope he's well! I think he's well. At least there are some news in German newspapers about his activities. He finished the first apartments in the old Dujardin building in January, and he opened a new restaurant there as well in March. So I believe that he still has much to do these days... Albrecht _______________________________________________ fltk-dev mailing list fltk-dev@easysw.com http://lists.easysw.com/mailman/listinfo/fltk-dev
