On Thu, May 12, 2011 at 07:59:28PM +0100, Ian MacArthur wrote: > > On 12 May 2011, at 19:29, Kurt Van Dijck wrote: > > >> @Kurt: Are there specific failures you are observing related to not > >> handling this X11 event? > > > > I'm incorporating an embedded X server, with very dedicated window > > manager. My first thing was to call XDestroyWindow, and then the FLTK > > appliction does not even know that something happened. > > Hm, OK. > What usually happens if a window is forcibly closed by the Xserver though? > > I don't think it (a regular Xserver) can be sending DestroyNotify, it must be > doing something else, or someone would have noticed this before, surely? > > > > Maybe raising FL_HIDE is as good too. I never thought on that. > > Now I suggested it, I'm not so sure... maybe that would work. I'm not sure > that making an FL_CLOSE event would be right either though. It would make > fltk try and close the window, perhaps, which might be bad if the window is > already gone... Though I imagine the intent is to fix it so that fltk's > internal state shows the window as destroyed, so... > > Did the patch actually work in your use-case?
That's the thing: it did work in that my application (only 1 window) would come out of it's Fl::run() and exit, without Segmentation faults. > > > > > _______________________________________________ > fltk-dev mailing list > [email protected] > http://lists.easysw.com/mailman/listinfo/fltk-dev _______________________________________________ fltk-dev mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk-dev
