Re: GTK+ 2.21.6

[Emmanuele Bassi <ebassi gmail com>]

On Wed, 2010-08-18 at 16:15 -0400, Paul Davis wrote:
> On Wed, Aug 18, 2010 at 2:01 PM, Jose Aliste <jose aliste gmail com> wrote:
> 
> > After looking at the git log, you interpret correctly, but  I fail to
> > see why do you perceive this as being bad. So, some GDK API is marked
> > as deprecated, it seems fair to me that Gtk+ maintainers ensure the
> > deprecated API is not used in gtk+ itself anymore. It is also a good
> > test that you are not deprecating something that can't be deprecated
> > (i.e. there is no equivalent functionality in cairo).
> 
> i think you missed my point. this is not about the deprecations. its
> about the fact that 2.22 will contain a complete reimplemenation of
> the GDK draw routines. this is happening in a branch this is
> supposedly "stable". i don't understand how this can reconciled. no
> API is changed in order for this to happen, but the implementations
> are all new.

by that line of reasoning, we would not be allowed to change
implementations to fix bugs *ever*, and we'd be forced to create a new
major release every time a refactoring of the code was performed.

in any case, this is not what's happening in the 2.x line...

>  existing programs will compile and run, but if linked
> against 2.22 they will be testing an implementation that was (i
> thought) aimed at the 3.0 release line.

... because the gdk_draw_* *implementations* have not been touched.

GTK+ has stopped using those calls internally and deprecated the API so
that application developers will be able to port their applications to
Cairo for GTK+ 3.0; this is merely finishing the long road of moving to
Cairo that started with GTK+ 2.8.

ciao,
 Emmanuele.

-- 
W: http://www.emmanuelebassi.name
B: http://blogs.gnome.org/ebassi