From taube at uiuc.edu  Thu Dec 15 02:12:16 2005
From: taube at uiuc.edu (Rick Taube)
Date: Wed, 14 Dec 2005 20:12:16 -0600
Subject: [lambda-gtk-devel] Re: lambda-gtk wrong return types in the
	generated ffi
In-Reply-To: <21f2dc700512141341g768f731fs4682ed2b1b527a06@mail.gmail.com>
References: <21f2dc700512141341g768f731fs4682ed2b1b527a06@mail.gmail.com>
Message-ID: <0c45016f13b45bd4c531a81050010aea@uiuc.edu>

thank you for reporting this. ill try to fix it over the break. ive 
been remiss updating some things, primarily because im thinking that i 
shuould now be generating cffi bindings but havent found the time to 
work on this.
--rick

On Dec 14, 2005, at 3:41 PM, Peter Gijsels wrote:

> Hi,
>
> I've been playing around with lambda-gtk, and I like it a lot.
> Sometimes it's a bit low-level (e.g. valist: yuk), but in general it's
> quite usable.
>
> However, I encountered a bug: the return type of some of the generated
> functions are wrong. For gtk_tree_path_to_string the generated return
> type is gchar while it should be c-string:
> hunk ./gtkffi-cmusbcl.lisp 13134
> - (defalien "gtk_tree_path_to_string" gchar (a (* t)))
> + (defalien "gtk_tree_path_to_string" c-string (a (* t)))
>
> It seems that there are other functions that are suffering from the
> same problem: e.g. gtk_tree_path_get_indices.
>
> I didn't immediately find how to fix this in the generating code.
>
> Regards,
> Peter