My suggestion would bet to add a column to the store containing an opaque, unique ID for each row (not unlike a database). To me, this seems conceptually simpler then trying to refresh TreeIters.
2011/9/12 Neil Muller <drnlmuller+...@gmail.com>: > I'm pretty sure that moving the row current_track points to will > invalidate the gtk.TreeIter you have, as the row gets deleted and > re-created elsewhere, which would be why you're seeing None for the > path. You should be able to check if this is the case by using > ListStore.iter_is_valid(). I was wondering the same thing, since TreeIter lifespans can be less than useful in some models. But the docs (and the flags) indicate that TreeIters survive changes. Of course, if you're right about it deleting and inserting (which you probably are), it doesn't matter. > I would add some logic to the drag_data_received signal to deal with this > case. Pardon my n00bishness, but isn't tree-row-drag code part of GTK and not the application? _______________________________________________ pygtk mailing list pygtk@daa.com.au http://www.daa.com.au/mailman/listinfo/pygtk Read the PyGTK FAQ: http://faq.pygtk.org/
