Re: #50213 - No accessor for GObject::property_specs
- From: Owen Taylor <otaylor redhat com>
- To: Tim Janik <timj gtk org>
- Cc: Gtk+ Developers <gtk-devel-list gnome org>
- Subject: Re: #50213 - No accessor for GObject::property_specs
- Date: 31 Mar 2001 22:34:07 -0500
Tim Janik <timj gtk org> writes:
> On 28 Mar 2001, Owen Taylor wrote:
>
> >
> > Tim Janik <timj gtk org> writes:
> >
> > > On 26 Mar 2001, Owen Taylor wrote:
> > >
> > > > Can I commit the following?
> > >
> > > no, this function needs to do quite a bit more work, mask out
> > > duplicate param specs in the ancestry
> >
> > How so? This function is meant for things like documentation generators
> > and GUI builders that probably want to split things up by the
> >
> > class where the property is defined, so it shouldn't include
> > parent param specs
>
> it has to, in order for parent class param specs to be overridable.
> you're right though in that the fucntion needs to honour a user argument
> of the topmost base class to walk to.
I think the possibility of overriding is something that the person
doing the querying has to be aware of, and we can't magically hide
it behind the scenes.
> > >(that allowes for overriding of default values and other property
> > >spec parts then)
>
> > > and some CONSTRUCT cleanups.
> >
> > I'm not sure what you are referring to here.
>
> G_PARAM_CONSTRUCT* flags.
What about G_PARAM_CONSTRUCT flags?
Owen
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]