On 02.05.2012, at 14:27, Edzard Egberts wrote: > Matthias Melcher schrieb: >> >> On 02.05.2012, at 12:03, Edzard Egberts wrote: >> >>> Edzard Egberts schrieb: >>>> Must be something more basic. >>> >>> I just found out - it was a problem of x-/y- coordinates. Need to >>> tell fl_read_image() the position of the window, otherwise it is >>> out of drawing area. >> >> Ah yes. The default coordinates are a magic value that tells the WM >> to position the window when it is mapped. Until then, the window is >> at some outlandish position. > > Oh no, it doesn't depend on coordinates, it depends on screen! I just > figured out, that on my left screen I get the whole window, but when > moving the test window to the right screen, the offscreen drawing starts > to get cut and when totally on right screen (my main screen) it vanishs. > Offscreen drawing was first step to save a jpeg screenshot and now I can > see, how it changes. It's the same, when using Fl_Window(x, y, w, h). > > Is it a known problem, that offscreen drawing doesn't work for both > screens of dual screen systems? I'm using Fedora 14, Kernel 2.6.35.14, > GNOME 2.32.0
No. I hear this for the first time. I wonder if this is related to the FLTK screen_xywh() function calls, or if this is some feature of the underlying OS. If you have a few minutes, could you try to set a breakpoint in the screen_xywh functions and the retrieve the snapshop again? If it does hit a breakpoint, we can then check who calls it and if that somehow relates to clipping. _______________________________________________ fltk mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk
