To: Francois.Pottier at inria.fr
Cc: lablgtk at kaba.or.jp
Subject: Re: Misc questions
In-Reply-To: <20030922100602.B31077 at pauillac.inria.fr>
References: <20030922085521.A30725 at pauillac.inria.fr>
	<20030922165104L.garrigue@kurims.kyoto-u.ac.jp>
	<20030922100602.B31077@pauillac.inria.fr>
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-Id: <20030922171934X.garrigue at kurims.kyoto-u.ac.jp>
Date: Mon, 22 Sep 2003 17:19:34 +0900
From: Jacques Garrigue <garrigue at kurims.kyoto-u.ac.jp>
Lines: 41

From: Francois Pottier <francois.pottier at inria.fr>
> On Mon, Sep 22, 2003 at 04:51:04PM +0900, Jacques Garrigue wrote:
> > 
> > OK. Gobject.obj is contravariant. Since for variants [`treemodel] is a
> > subtype of [`treemodel|`liststore], then a (liststore obj) is a
> > subtype of a (treemodel obj).
> 
> Hmm, I do see that
> 
>   [`treemodel|`liststore] obj   <:   [`treemodel] obj
> 
> is valid. But I still don't see why
> 
>   [`treemodel] obj   <:   tree_model obj
> 
> should be valid -- which was my original question :)

I see: there was a bug, since tree_model is an abbreviation of
[`treemodel] obj, and (tree_model obj) is just meaningless.
So you will have to fix it by replacing the definition of
RowReference.create by this one:
  external create :
    [> `treemodel ] Gtk.obj -> Gtk.tree_path -> Gtk.row_reference
    = "ml_gtk_tree_row_reference_new"

When you see a bug, call it a bug :)

(I suggested a coercion because I believed list_store#model would
return a [`treemodel|`liststore] obj. But this is not the case
anyway.)

> > You have to use ocamlbrowser to see all the contents directly.
> 
> Aah! I see. So browsing the .mli files by hand doesn't tell you all
> of the truth. That is subtle.

This is only for the Gtk* modules, and they have no .mli to start with.
But considering all the inheritance going around, this is useful with
classes too...

Jacques