Re: custom treemodel?

["Jonathon Jongsma" <jonathon jongsma gmail com>]

On 5/17/06, Murray Cumming <murrayc murrayc com> wrote:
> It's all to do with the fact that you don't have much control over when
> the per-iter data is deallocated. You just get to say "everything I have
> allocated before now is now invalid, so I'll release it and it's your own
> dam fault if you try to use it.". You do that by increasing the value of
> stamp.
>
> So I guess I keep a GlueList so I can release it when I do that
> invalidation. And I guess I have a GlueItem for each iter, to be neat and
> generic.
>
> I don't think I actually invalidate the iters. And I don't think that
> ListStore or TreeStore ever do that either, so it's a bit theoretical.
> Maybe ListStore gets away with this (without memory leaks when iters are
> forgetten) by just using basic types instead of extra memory for each
> iter. Just guessing.

OK, I think I get it now, more or less.  I've just noticed, however
that when I run the example from gtkmm CVS, I get the following
warning:

(lt-example:29959): GLib-GObject-WARNING **: cannot add interface type
`GtkTreeModel' to type `gtkmm__CustomObject_16ExampleTreeModel', since
type `gtkmm__CustomObject_16ExampleTreeModel' already conforms to
interface

It seems to be caused by the following lines in the constructor:
 GType gtype = G_OBJECT_TYPE(gobj()); //The custom GType created in
the Object constructor, from the typeid.
 Gtk::TreeModel::add_interface( gtype );

I don't have a lot of experience with GObject stuff, so I have no idea
whether these should be necessary or not.  But it seems to run fine
and doesn't issue a warning when those lines are commented out.  Any
thoughts?

Jonner