client-side-windows merged

[Alexander Larsson]

The client-side-windows branch has now been merged into master.

The current state of this code is pretty good. There are two known
regressions on X11 (see http://live.gnome.org/GTK%2B/ClientSideWindows),
but both are corner cases. The quartz backend is supposed to be at least
as good as the old one. The win32 backend is in pretty bad shape, but
after some hacking today it at least builds and mostly works.

I will continue working on this code, fixing regressions and getting the
win32 backend up to speed. However, at this point we need broader
testing, and doing that work in master will help us get that.


_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

[Matthias Clasen-2]

On Wed, Jul 1, 2009 at 10:34 AM, Alexander Larsson<alexl@...> wrote:
> The client-side-windows branch has now been merged into master.

Thanks Alex. This should be exciting. I'll be looking at doing a
development release including this towards the end of the week, for
increased exposure...
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

[Alexander Larsson]

On Wed, 2009-07-01 at 17:06 -0400, Matthias Clasen wrote:
> On Wed, Jul 1, 2009 at 10:34 AM, Alexander Larsson<alexl@...> wrote:
> > The client-side-windows branch has now been merged into master.
>
> Thanks Alex. This should be exciting. I'll be looking at doing a
> development release including this towards the end of the week, for
> increased exposure...

May we live in interesting times...

_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

[Matthias Clasen-2]

On Thu, Jul 2, 2009 at 4:30 AM, Alexander Larsson<alexl@...> wrote:

> On Wed, 2009-07-01 at 17:06 -0400, Matthias Clasen wrote:
>> On Wed, Jul 1, 2009 at 10:34 AM, Alexander Larsson<alexl@...> wrote:
>> > The client-side-windows branch has now been merged into master.
>>
>> Thanks Alex. This should be exciting. I'll be looking at doing a
>> development release including this towards the end of the week, for
>> increased exposure...
>
> May we live in interesting times...
>

As I've just briefly discussed with alex on irc, we need some way to
tag bugs that are likely caused by the client-side-window merge. Here
is the proposal: please put 'csw' in the status whiteboard for likely
csw issues.

Thanks, Matthias
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

[Hans Breuer]

At 01.07.2009 16:34, Alexander Larsson wrote:
> The client-side-windows branch has now been merged into master.
>
> The current state of this code is pretty good. There are two known
> regressions on X11 (see http://live.gnome.org/GTK%2B/ClientSideWindows),
> but both are corner cases. The quartz backend is supposed to be at least
> as good as the old one.

> The win32 backend is in pretty bad shape, but
> after some hacking today it at least builds and mostly works.
>
Mostly works seems to be a bit exaggerated ;) I was able to fix the
positioning of menus, but to make them work again the pointer grab code
needs to be rewritten (I did not define TODO_CSW for building).

See testgtk::menus for a test case.

> I will continue working on this code, fixing regressions and getting the
> win32 backend up to speed.
Hopefully the above is first on your list, because it is the main issue
currently prohibiting real use.

> However, at this point we need broader
> testing, and doing that work in master will help us get that.
>
testgtk::panes does not change the cursor on mouse over the grip.

testgtk::shapes draw a gray backgound now, where it was previously transparent.

But these are to only additional regression I found with testgtk.

I'll continue to check things out with other test applications.

Regards,
        Hans

-------- Hans "at" Breuer "dot" Org -----------
Tell me what you need, and I'll tell you how to
get along without it.                -- Dilbert

_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

[Alexander Larsson]

----- Original Message -----
From: "Hans Breuer" <hans@...>
To: "Alexander Larsson" <alexl@...>
Cc: gtk-devel-list@...
Sent: Saturday, July 4, 2009 1:12:13 PM GMT +01:00 Amsterdam / Berlin / Bern / Rome / Stockholm / Vienna
Subject: Re: client-side-windows merged

At 01.07.2009 16:34, Alexander Larsson wrote:
> The client-side-windows branch has now been merged into master.
>
> The current state of this code is pretty good. There are two known
> regressions on X11 (see http://live.gnome.org/GTK%2B/ClientSideWindows),
> but both are corner cases. The quartz backend is supposed to be at least
> as good as the old one.

> The win32 backend is in pretty bad shape, but
> after some hacking today it at least builds and mostly works.
>
Mostly works seems to be a bit exaggerated ;) I was able to fix the
positioning of menus, but to make them work again the pointer grab code
needs to be rewritten (I did not define TODO_CSW for building).

TODO_CSW is just meant for greping, not for defining.
I'll try to get time in the week for looking at the win32 stuff, although
I'm currently fixing some X11 bugs that have been reported.

> But these are to only additional regression I found with testgtk.

Thats unexpectedly good.
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

[Frederic Peters-5]

Alexander Larsson wrote:

> The client-side-windows branch has now been merged into master.

The http://git.gnome.org/cgit/gtk+/commit/?id=0b586a5a change to
gdkdrawable.h:

@@ -100,6 +100,7 @@ struct _GdkDrawableClass
   void (*draw_drawable)  (GdkDrawable  *drawable,
                          GdkGC        *gc,
                          GdkDrawable  *src,
+                         GdkDrawable  *original_src,
                          gint          xsrc,
                          gint          ysrc,
                          gint          xdest,

breaks a part of the API/ABI that is (at least) used in PyGTK:

gdk.c:7243: warning: passing argument 4 of ‘((struct GdkDrawableClass
  *)g_type_check_class_cast((struct GTypeClass *)klass,
  gdk_drawable_get_type()))->draw_drawable’ makes pointer from integer
  without a cast
gdk.c:7243: error: too few arguments to function ‘((struct
  GdkDrawableClass *)g_type_check_class_cast((struct GTypeClass *)klass,
  gdk_drawable_get_type()))->draw_drawable’


How should this matter be resolved ?



        Frederic
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

[Brian J. Tarricone]

On 07/12/2009 01:02 PM, Frederic Peters wrote:

> Alexander Larsson wrote:
>
>> The client-side-windows branch has now been merged into master.
>
> The http://git.gnome.org/cgit/gtk+/commit/?id=0b586a5a change to
> gdkdrawable.h:
>
> @@ -100,6 +100,7 @@ struct _GdkDrawableClass
>     void (*draw_drawable)  (GdkDrawable  *drawable,
>                            GdkGC        *gc,
>                            GdkDrawable  *src,
> +                         GdkDrawable  *original_src,
>                            gint          xsrc,
>                            gint          ysrc,
>                            gint          xdest,
>
> breaks a part of the API/ABI that is (at least) used in PyGTK:
>
> gdk.c:7243: warning: passing argument 4 of ‘((struct GdkDrawableClass
>    *)g_type_check_class_cast((struct GTypeClass *)klass,
>    gdk_drawable_get_type()))->draw_drawable’ makes pointer from integer
>    without a cast
> gdk.c:7243: error: too few arguments to function ‘((struct
>    GdkDrawableClass *)g_type_check_class_cast((struct GTypeClass *)klass,
>    gdk_drawable_get_type()))->draw_drawable’
>
>
> How should this matter be resolved ?

For starters, why isn't pygtk just using gdk_drawable_draw_drawable()
instead of dereferencing the class pointer?

Regardless, if you check out current master, it appears that change
that's causing your problem isn't there anymore (instead there's a
draw_drawable_with_src function pointer that's been added to
GdkDrawableClass).  See
http://git.gnome.org/cgit/gtk+/log/gdk/gdkdrawable.h.  This was fixed 11
days ago, it seems.

        -brian
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

[Matthias Clasen-2]

On Sun, Jul 12, 2009 at 7:50 PM, Brian J. Tarricone<bjt23@...> wrote:

> For starters, why isn't pygtk just using gdk_drawable_draw_drawable()
> instead of dereferencing the class pointer?
>
> Regardless, if you check out current master, it appears that change that's
> causing your problem isn't there anymore (instead there's a
> draw_drawable_with_src function pointer that's been added to
> GdkDrawableClass).  See
> http://git.gnome.org/cgit/gtk+/log/gdk/gdkdrawable.h.  This was fixed 11
> days ago, it seems.
>

Yes, the api breakage was discovered soon after the merge and is fixed in 2.17.4
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

[Frederic Peters-5]

Matthias Clasen wrote:

> Yes, the api breakage was discovered soon after the merge and is
> fixed in 2.17.4

Perfect; I didn't notice I was still tracking the csw branch; sorry
for the noise.


        Frederic
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list