<html> <head> <base href="https://bugzilla.gnome.org/" /> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - wayland: connection flooded when dragging" href="https://bugzilla.gnome.org/show_bug.cgi?id=763298#c8">Comment # 8</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - wayland: connection flooded when dragging" href="https://bugzilla.gnome.org/show_bug.cgi?id=763298">bug 763298</a> from <a href="page.cgi?id=describeuser.html&login=carlosg%40gnome.org" title="Carlos Garnacho <carlosg@gnome.org>"> Carlos Garnacho</a> <pre>(In reply to Matthias Clasen from <a href="show_bug.cgi?id=763298#c7">comment #7</a>) > Here is another concern I came upwith this morning: > > The gdk dnd api lets you pretty freely call gdk_drag_status from whereever. Hmmm, wherever in response to DnD events, or crazier things like timeouts, etc? I thought the docs seemed clear enough (even if outdated/non-generic): "This function is called by the drag destination in response to gdk_drag_motion() called by the drag source." It's right I'm missing GDK_DRAG_ENTER in the switch though. > With this change, I'm afraid that this might not work anymore for Wayland, > since gdk_drag_status will not have any effect unless it is in response to a > drag motion event. > > Now, I don't know if such 'freestanding gdk_drag_status calls ever worked > with Wayland in the first place... According to the wayland DnD protocol, the wl_data_offer.set_actions request (which gdk_drag_status/commit_drag_status boil down to) are allowed "during dnd" (typically in response to wl_data_device.enter/motion) and once after wl_data_device.drop, only if an "ask" action was agreed previously. I guess we should also enforce the latter more pervasively.</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are on the CC list for the bug.</li> </ul> </body> </html>