Re: ABI and API for g_object_ref_sink() (Re: GTK_FLOATI
- From: "ANDREW PAPROCKI, BLOOMBERG/ 731 LEXIN" <apaprocki bloomberg net>
- To: GTK-DEVEL-LIST GNOME ORG
- Cc:
- Subject: Re: ABI and API for g_object_ref_sink() (Re: GTK_FLOATI
- Date: Thu, 15 Dec 2005 19:36:01 -0500
>1. a flags field
>2. the floating reference (stored in the flags field, incidentally)
>3. the destroy signal
>4. a user-data property (this one is far surpassed by GObject's
>arbitrary data keys, but is there to ease porting from 1.x)
>5. the compat layer for GtkArgs (which were replaced by GObject
>properties, but that's not so important and will probably be removed
>entirely in 3.x)
Is it worth pushing the floating flag to GObject, pushing the destroy signal
into GObject, and eliminating the user-data and GtkArgs code all in 3.x?
If the 32-bit flags field can not move into GObject, GtkObject would simply
become a holder for this extra 32-bits and would not need any API other than to
manipulate the bits.
I agree that destroy would be useful in GObject as well, and given the API/ABI
problems of moving the floating flag into 2.x, should the above be considered as
a viable option if it was to wait entirely for 3.x?
Andrew Paprocki
Bloomberg LP
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]