On 08/27/2013 07:50 PM, Marc-André Lureau wrote: > Hi > > ----- Mensaje original ----- >> Similarly to spice_display_mouse_ungrab(), this is intedned to be used >> from outside code to force the display to grab the mouse (and >> keyboard). > > That sounds reasonable, although we probably want the function to return whether it succeeded. > I wasn't going to plug this in since try_mouse_ungrab is void and I'd have to either duplicate the code or re-make the function. No problem though if you want that to return the result. > What should happen if you try to grab with mouse in client mode? fail silentely? warn? > I was thinking that silent fail is OK since the caller can check the mode etc. to know whether this makes sense at all. However if try_mouse_ungrab is redone (or there is other function which returns non-void) than this can be changed to one of the return codes and left up to the caller whether it wants to ignore it or not. I forgot to mention why I sent this patch. We wanted to control this behavior in virt-manager, but it looks like we'll copy the behavior from virt-viewer and won't grab the mouse when not needed. I'll see whether we'll need this at all and will send a v2 in case nobody beats me to it and we need it at last. [...] >> @@ -83,7 +83,8 @@ static gboolean gtk_widget_get_realized(GtkWidget *widget) >> * focused if the properties #SpiceDisplay:grab-keyboard and >> * #SpiceDisplay:grab-mouse are #TRUE respectively. It can be >> * ungrabbed with spice_display_mouse_ungrab(), and by setting a key >> - * combination with spice_display_set_grab_keys(). >> + * combination with spice_display_set_grab_keys(). The grab can be >> + * also forcefully initiated using spice_display_mouse_ungrab(). >> * And yes, this should've been s/grab/ungrab as Christophe pointed out. Thanks, Martin _______________________________________________ Spice-devel mailing list Spice-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/spice-devel