It is not a regression or anything, but is there any chance someone could check this?: http://hub.qgis.org/issues/3330 It is a one line fix for a long-standing bug. Thanks, Alister