Re: Missing drag'n'drop methods in GtkTreeView

[Torsten Schoenfeld <kaffeetisch web de>]

On Wed, 13 Aug 2003 12:44:00 -0400 (EDT), muppet wrote:

> > [1] I'm not sure if I got the stack thingies (EXTEND, PUSHs) right.
> >     Please check it.
>
> you did it right, but there's an easier way:
>
> void gtk_tree_widget_to_tree_coords (GtkTreeView * tree_view, gint wx,
> gint wy, OUTLIST gint tx, OUTLIST ty)
>
> does what your hand-written XS does with a lot less typing.  this is
> one of the nifty features new to xsubpp in 5.8.0...

Is it safe to assume that everyone has 5.8.0?

> > [2] I'm not sure if newSVGtkTreePath or _copy or _own should be used
> >     since I simply don't know the difference.
>
> _copy should be used when returning an object that you are not allowed
> to modify.  _own should be used when returning an object that you are
> supposed to free.  if you use _copy where you should use _own, you get
> a leak.  if you use_own where you should use nothing, then you destroy
> something that's not yours.
>
> i haven't looked yet into which one is correct where.

I see. I couldn't find any documentation or usage exmaples on
gtk_tree_view_get_drag_dest_row. So I've no idea what would be The Right
Thing here.

> > I failed implementing gtk_tree_view_get_search_equal_func due to a
> > "missing" entry in the typemap (GtkTreeViewSearchEqualFunc). I can't
> > tell whether it's /really/ missing or I simply did something wrong.
>
> [...]
>
> that sucks.  this is why the callback stuff is a bit of grey magic. 
> (not black, because it is documented, but not white because there are
> inevitably sticky issues like this.)

I didn't find documentation on returning function pointers and the only
other occurence of this thingy I found[1] is
 
  gtk_color_selection_set_change_palette_hook

which, coincidentally, is still unimplemented, too. :) All that to say
that I'm still at a loss at how to implement this.

Bye,
-Torsten

[1] grep -r '^[# ]*Gtk[a-zA-Z]*Func' gtk2-perl-xs/*