Emmanuele Bassi [Wed, 19 Oct 2011 14:23:55 +0000 (15:23 +0100)]
Avoid the shadowing of 'clone'
The clone() call can creep into the build environment, so it's a good
thing to avoid using 'clone' as a variable name.
Neil Roberts [Wed, 19 Oct 2011 12:42:18 +0000 (13:42 +0100)]
tests: Use the portable cogl wrappers for GLSL builtins
Instead of directly using the GLSL names for the builtins in the
shaders for test-shader and test-pick, this makes it use the Cogl
wrapper names instead. That way it will be portable to GLES2 as well.
Reviewed-by: Emmanuele Bassi <ebassi@linux.intel.com>
Emmanuele Bassi [Mon, 17 Oct 2011 10:48:47 +0000 (11:48 +0100)]
docs: Improve documentation of LayoutManager
Especially with regards to layout properties and their usage inside a
ClutterScript UI definition.
Emmanuele Bassi [Mon, 17 Oct 2011 09:25:20 +0000 (10:25 +0100)]
docs: Clean up the 'running Clutter apps' chapter
Emmanuele Bassi [Mon, 17 Oct 2011 09:25:07 +0000 (10:25 +0100)]
Remove mentions of clutter_redraw()
Emmanuele Bassi [Mon, 17 Oct 2011 09:24:25 +0000 (10:24 +0100)]
debug: Clean up debugging notes
Emmanuele Bassi [Sat, 15 Oct 2011 17:35:45 +0000 (18:35 +0100)]
interactive: Use the newly added CLUTTER_ALIGN_BOTH value
We center a lot of stuff using ClutterAlignConstraints.
Emmanuele Bassi [Sat, 15 Oct 2011 17:33:49 +0000 (18:33 +0100)]
align-constraint: Add CLUTTER_ALIGN_BOTH
Just like ClutterBindConstraint has two shorthand enumerations for binding
position and size and avoid using multiple instances, ClutterAlignConstraint
should have a way to align an actor with the same alignment factor on both
axis at the same time; this is especially useful for centering actors.
Fran Dieguez [Sat, 15 Oct 2011 13:24:41 +0000 (15:24 +0200)]
Updaged Galician translations
Emmanuele Bassi [Fri, 14 Oct 2011 10:34:38 +0000 (11:34 +0100)]
Tag all deprecated symbols using CLUTTER_DEPRECATED
This requires some minor surgery in the build to disable the deprecation
warnings in the deprecated classes.
Emmanuele Bassi [Fri, 14 Oct 2011 10:34:26 +0000 (11:34 +0100)]
Bump up the dependency on gtk-doc
Emmanuele Bassi [Wed, 12 Oct 2011 17:04:23 +0000 (18:04 +0100)]
build: Strip out G_DISABLE_DEPRECATED
We have compiler warnings, now.
Emmanuele Bassi [Thu, 13 Oct 2011 16:50:25 +0000 (17:50 +0100)]
Disable deprecation warnings for the test suite
We test deprecated functionality as well as current one, so we need to
enable all symbols and disable the deprecation warnings.
Emmanuele Bassi [Thu, 13 Oct 2011 10:38:27 +0000 (11:38 +0100)]
debug: Move paint-deform-tiles to CLUTTER_PAINT
The CLUTTER_DEBUG class of debugging flags is meant for debugging notes,
while the CLUTTER_PAINT debugging flags are for changing the output of
the paint cycle. Painting the DeformEffect tiles should go in the latter.
Emmanuele Bassi [Thu, 13 Oct 2011 09:42:30 +0000 (10:42 +0100)]
docs: Include test-threads.c in the API reference
We can use the test-threads.c interactive test as part of the documentation
on how to use threads with Clutter.
Emmanuele Bassi [Thu, 13 Oct 2011 09:38:36 +0000 (10:38 +0100)]
interactive/threads: Clean up
Add some comments explaining why we use idle handlers to update the UI,
and update to the newest API in GLib master.
Piotr Drąg [Wed, 12 Oct 2011 18:33:16 +0000 (20:33 +0200)]
Updated POTFILES.in
Emmanuele Bassi [Wed, 12 Oct 2011 15:17:39 +0000 (16:17 +0100)]
Avoid using deprecated API internally
Use the proper API to implement deprecated functions.
Emmanuele Bassi [Wed, 12 Oct 2011 11:36:41 +0000 (12:36 +0100)]
texture: Disable deprecation warnings
Sadly, ClutterTexture is still using internally deprecated API, like
clutter_actor_get_shader().
Emmanuele Bassi [Wed, 12 Oct 2011 09:36:17 +0000 (10:36 +0100)]
Remove _clutter_shader_release_all()
This function is called when the backend is being disposed - as a way
of releasing all ClutterShader. This doesn't take into account three
things:
- ClutterShader is deprecated
- the Backend is *never* disposed
- once the process terminates, all its resources are automatically
released by the OS
So the _clutter_shader_release_all() function is a pointless exercise
in futility.
Emmanuele Bassi [Wed, 12 Oct 2011 08:57:53 +0000 (09:57 +0100)]
tests: Fix some compiler warnings
Emmanuele Bassi [Wed, 12 Oct 2011 08:57:33 +0000 (09:57 +0100)]
tests: Disable deprecation warnings
Emmanuele Bassi [Tue, 11 Oct 2011 23:20:43 +0000 (00:20 +0100)]
interactive/threads: Use G_PRIVATE_INIT
Though strictly not necessary, we should be using the proper init macro
for GPrivate.
Emmanuele Bassi [Tue, 11 Oct 2011 22:42:23 +0000 (23:42 +0100)]
Remove some compiler warnings
Some issues found using clang as the compiler.
Emmanuele Bassi [Tue, 11 Oct 2011 22:11:56 +0000 (23:11 +0100)]
Fix included header
Emmanuele Bassi [Tue, 11 Oct 2011 22:06:06 +0000 (23:06 +0100)]
build: Disable deprecation warnings with --enable-deprecated
Just like we turn everything on with --disable-deprecated, we have to
turn everything off with --enable-deprecated. This means disabling the
deprecation warnings from the compiler as well.
Emmanuele Bassi [Tue, 11 Oct 2011 22:03:09 +0000 (23:03 +0100)]
Add a Clutter-specific deprecation macro for symbols
Just like GLIB_DEPRECATED and GLIB_DEPRECATED_FOR, Clutter should have
its own wrappers for G_DEPRECATED and G_DEPRECATED_FOR, to allow opting
out of deprecation warnings.
Deprecation warnings are enabled by default, now, even when building
Clutter.
Emmanuele Bassi [Tue, 11 Oct 2011 22:02:17 +0000 (23:02 +0100)]
Fix duplicate typedef of ClutterAlpha
Caught on OS X but not on my Fedora machine. Weird.
Emmanuele Bassi [Tue, 11 Oct 2011 21:43:54 +0000 (22:43 +0100)]
Do not leak strings from GKeyFile
The string getters always copy, so we need to free the returned
values.
Emmanuele Bassi [Tue, 11 Oct 2011 17:25:14 +0000 (18:25 +0100)]
docs: Fixes to build the API reference
Emmanuele Bassi [Tue, 4 Oct 2011 13:21:32 +0000 (14:21 +0100)]
Move more typedefs into clutter-types.h
Emmanuele Bassi [Tue, 4 Oct 2011 12:28:04 +0000 (13:28 +0100)]
Move all enumerations to a separate file
This should allow sharing types, and we can avoid glib-mkenums thrawling
the whole repository for enumerations.
Emmanuele Bassi [Tue, 4 Oct 2011 11:18:26 +0000 (12:18 +0100)]
Move clutter-fixed.[ch] to the deprecated section
Emmanuele Bassi [Tue, 4 Oct 2011 10:58:44 +0000 (11:58 +0100)]
Move ClutterBehaviour to the deprecated section
Emmanuele Bassi [Tue, 4 Oct 2011 10:52:47 +0000 (11:52 +0100)]
Minor header surgery to ClutterBehaviour
This moves a couple of definitions to the common types header, and makes
sure that ClutterBehaviour subclasses include clutter-behaviour.h first,
so that their types can be fully expanded without necessarily have the
ClutterBehaviour header header included by their public headers. This is
the necessary prelude to have clutter-behaviour.[ch] moved to the
deprecated section.
Emmanuele Bassi [Tue, 4 Oct 2011 10:32:36 +0000 (11:32 +0100)]
Move ClutterBehaviour subclasses to the deprecated section
Moving ClutterBehaviour itself will require some minor header surgery,
so we'll do that next.
Emmanuele Bassi [Tue, 4 Oct 2011 10:16:38 +0000 (11:16 +0100)]
docs: Move deprecated classes out of the way
The deprecated sections should be much more prominently separated from
the current API; we can use a new part inside the main reference index
for this.
Emmanuele Bassi [Tue, 4 Oct 2011 10:12:55 +0000 (11:12 +0100)]
Move ClutterShader to the deprecated section
Emmanuele Bassi [Tue, 4 Oct 2011 10:07:03 +0000 (11:07 +0100)]
Move ClutterScore to the deprecated section
Emmanuele Bassi [Tue, 4 Oct 2011 09:57:28 +0000 (10:57 +0100)]
Move timeout pool to the deprecated section
Emmanuele Bassi [Tue, 4 Oct 2011 09:51:09 +0000 (10:51 +0100)]
build: Move deprecated code into its own prefix
The code that has been deprecated should live into its own directory,
both in the repository and when installed. This should make it clear
which functionality is actually maintained and which is not.
We start with an oldie: the frame source API.
Emmanuele Bassi [Tue, 11 Oct 2011 16:51:14 +0000 (17:51 +0100)]
interactive/threads: Fix up after deprecations
The GThread API has undergone a massive restructuring, and the fallout
is still being processed.
Emmanuele Bassi [Tue, 11 Oct 2011 16:08:20 +0000 (17:08 +0100)]
docs: Document the settings.ini file
Describe the location, groups, and keys of the settings.ini file used by
Clutter.
Emmanuele Bassi [Tue, 11 Oct 2011 15:05:24 +0000 (16:05 +0100)]
Show lib and sysconfdir in the configure summary
Emmanuele Bassi [Tue, 11 Oct 2011 14:28:47 +0000 (15:28 +0100)]
Add more sections to the Clutter configuration file
On top of the existing "Settings" group in the settings.ini file we
should have two more groups:
Environment - contains all the configuration possible through
environment variables
Debug - contains all the possible debug variables
Emmanuele Bassi [Tue, 11 Oct 2011 13:42:31 +0000 (14:42 +0100)]
Add a configuration file for ClutterSettings
ClutterSettings should be able to load its initial state by using
configuration files in SYSCONFDIR and XDG_CONFIG_HOME. This allows
Clutter to have a system (and user) configuration on platforms that
do not have XSETTINGS bridges.
Emmanuele Bassi [Mon, 3 Oct 2011 13:12:48 +0000 (14:12 +0100)]
Deprecate clutter_[sg]et_font_flags()
We already have two mechanisms for controlling the font rendering
quality on a per-application basis:
• ClutterSettings properties
• clutter_backend_set_font_options()
The font flags were always a stop-gap solution, and one that tried to
simplify a fairly complex issue beyond the point of actually being
useful.
https://bugzilla.gnome.org/show_bug.cgi?id=660786
Emmanuele Bassi [Mon, 10 Oct 2011 16:00:28 +0000 (17:00 +0100)]
Deprecate clutter_clear_glyph_cache()
This is a wrapper function for API that has long since been public.
Emmanuele Bassi [Mon, 10 Oct 2011 15:57:18 +0000 (16:57 +0100)]
Deprecate clutter_get_input_device_for_id()
We have the ClutterDeviceManager methods for that.
Emmanuele Bassi [Mon, 10 Oct 2011 15:45:20 +0000 (16:45 +0100)]
Clean up grab implementation
The grab API is a relic of Clutter 0.6, and hasn't been through proper
vetting in a *long* time — mostly due to the fact that we don't really
like grabs, and point to the ::captured-event as a way to implement
"soft grabs" in toolkits and applications.
The implementation of full and device grabs uses weak references on
actors instead of using the ::destroy signal, which is meant exactly for
the case of releasing pointers to actors when they are disposed.
The API naming scheme is also fairly broken, especially for
device-related grabs.
Finally, keyboard device grabs are just not implemented.
We can, in one go, clean up this mess and deprecate a bunch of badly
named API by introducing generic device grab/ungrab methods on
ClutterInputDevice, and re-implement the current API on top of them.
Emmanuele Bassi [Mon, 10 Oct 2011 14:48:43 +0000 (15:48 +0100)]
Deprecate clutter_redraw()
It's just a badly named proxy to clutter_stage_ensure_redraw().
Emmanuele Bassi [Mon, 10 Oct 2011 14:42:36 +0000 (15:42 +0100)]
Cosmetic fixes
Algimantas Margevičius [Sat, 8 Oct 2011 21:57:38 +0000 (00:57 +0300)]
Updated Lithuanian translation
Emmanuele Bassi [Fri, 7 Oct 2011 15:56:21 +0000 (16:56 +0100)]
animation: Use the timeline's new-frame signal
Instead of using the ClutterAlpha's notify::alpha. This makes the code
less dependent on GObject's property notification.
Emmanuele Bassi [Fri, 7 Oct 2011 14:53:50 +0000 (15:53 +0100)]
Deprecate clutter_threads_init()
GLib deprecated g_thread_init(), and threading support is initialized
by GObject, so Clutter already runs with threading support enabled. We
can drop the clutter_threads_init() call requirement, and initialize the
Big Clutter Lock™ on clutter_init(). This reduces the things that have
to be done when dealing with threads with Clutter, and the things that
can possibly go wrong.
Emmanuele Bassi [Fri, 7 Oct 2011 14:08:27 +0000 (15:08 +0100)]
Use a static mutex for the default Clutter lock
The Big Clutter Lock™ can now be a static GMutex, since GLib supports
them. We can also drop a bunch of checks given the recent changes in
GLib threading API.
Emmanuele Bassi [Fri, 7 Oct 2011 14:07:26 +0000 (15:07 +0100)]
Drop g_thread_init()
It's not necessary any more, and it has been deprecated.
Emmanuele Bassi [Fri, 7 Oct 2011 14:06:37 +0000 (15:06 +0100)]
texture: Drop G_MUTEX_INIT
The static initializer for GMutex has been removed from GLib.
The g_thread_supported() call can also be removed: threading is always
enabled in GLib ≥ 2.31.
Marc-André Lureau [Wed, 5 Oct 2011 20:03:25 +0000 (22:03 +0200)]
Mark clutter_state_set_key() source_state_name as nullable
https://bugzilla.gnome.org/show_bug.cgi?id=661022
Zan Dobersek [Fri, 7 Oct 2011 11:05:39 +0000 (12:05 +0100)]
Add clutter_unicode_to_keyval function
This commit introduces a unicode-to-keyval conversion function that
performs identical action as the gdk version of that function. Also
added is the necessary table holding all the conversion values.
https://bugzilla.gnome.org/show_bug.cgi?id=661015
Sunjin Yang [Fri, 7 Oct 2011 06:21:34 +0000 (15:21 +0900)]
box-layout: Plug memory leaks
https://bugzilla.gnome.org/show_bug.cgi?id=660985
Daniel Mustieles [Fri, 7 Oct 2011 08:13:08 +0000 (10:13 +0200)]
Updated Spanish translation
Emmanuele Bassi [Tue, 4 Oct 2011 09:32:27 +0000 (10:32 +0100)]
Deprecate the GParamSpec for CoglFixed
The fixed-point API is not used anywhere, and it's pretty much useless
to have a fixed-point type for properties.
Håkon Løvdal [Mon, 3 Oct 2011 14:56:26 +0000 (15:56 +0100)]
spelling fix: timeour -> timeout
Alexandre Franke [Mon, 3 Oct 2011 13:57:21 +0000 (15:57 +0200)]
Update French translation
Kristjan SCHMIDT [Sat, 1 Oct 2011 15:24:51 +0000 (17:24 +0200)]
Updated Esperanto translation
Adel Gadllah [Sat, 1 Oct 2011 10:36:36 +0000 (12:36 +0200)]
clutter-actor-meta: Fix _clutter_meta_group_has_metas_no_internal
This function is backwards i.e it is supposed to check for metas that
are *not* internal, hence the name.
https://bugzilla.gnome.org/show_bug.cgi?id=660623
Emmanuele Bassi [Fri, 30 Sep 2011 16:02:57 +0000 (17:02 +0100)]
paint-volume: Don't try to complete a completed volume
If we do project() → get_bounding_box(), we'll try to complete the
volume twice, which whacks out all the lazily computed vertices.
Reviewed-by: Robert Bragg <robert@linux.intel.com>
Neil Roberts [Thu, 29 Sep 2011 19:14:26 +0000 (20:14 +0100)]
Add a conformance test for ClutterShaderEffect
This adds a simple conformance test which sets up a few shader effects
using both the old style with clutter_shader_effect_set_source and the
new style by overriding get_static_shader_source. The effects are then
verified to confirm that they drew the right pixel colour.
https://bugzilla.gnome.org/show_bug.cgi?id=660512
Reviewed-by: Emmanuele Bassi <ebassi@linux.intel.com>
Neil Roberts [Thu, 29 Sep 2011 19:05:39 +0000 (20:05 +0100)]
clutter-shader-effect: Add a get_static_shader_source virtual
This is used as an alternative to calling
clutter_shader_effect_set_shader_source. A ClutterShaderEffect
subclass is now expected to implement this method to return the source
for the effect that will be used for all instances of this
subclass. It is only called once regardless of the number of instances
created. That way Clutter can avoid recompiling the shader source for
every new instance of the effect.
https://bugzilla.gnome.org/show_bug.cgi?id=660512
Reviewed-by: Emmanuele Bassi <ebassi@linux.intel.com>
Adel Gadllah [Thu, 29 Sep 2011 16:34:12 +0000 (18:34 +0200)]
ClutterActor: Add clutter_actor_has_actions
Adds an efficent way to query whether an actor has any
applied actions.
https://bugzilla.gnome.org/show_bug.cgi?id=660471
Adel Gadllah [Thu, 29 Sep 2011 16:29:40 +0000 (18:29 +0200)]
ClutterActor: Add clutter_actor_has_contraints
Adds an efficent way to query whether an actor has any
applied constraints.
https://bugzilla.gnome.org/show_bug.cgi?id=660471
Adel Gadllah [Thu, 29 Sep 2011 13:31:30 +0000 (15:31 +0200)]
ClutterActor: Add clutter_actor_has_effects
Adds an efficent way to query whether an actor has any
applied effects.
https://bugzilla.gnome.org/show_bug.cgi?id=660471
Emmanuele Bassi [Thu, 29 Sep 2011 13:01:03 +0000 (14:01 +0100)]
Pass the remaining args to the executed interactive unit
We only support running one interactive test at a time; everything after
the unit name is to be considered an argument to the actual unit.
krishnababu k [Thu, 29 Sep 2011 06:39:11 +0000 (12:09 +0530)]
Updated Telugu Translations
Emmanuele Bassi [Wed, 28 Sep 2011 14:54:32 +0000 (15:54 +0100)]
Revert "Add lightness, brightness, and contrast effects"
Discussed a bit with Neil, and I might have jumped the gun a little.
We might want a different API for 1.10. See comment 12 on:
https://bugzilla.gnome.org/show_bug.cgi?id=656156
This reverts commit
4829ae1c79f1ae60d562621b81a7be800e0697bb.
Joseph Scheuhammer [Tue, 27 Sep 2011 17:01:36 +0000 (13:01 -0400)]
Add lightness, brightness, and contrast effects
Added three classes for new clutter effects: inverse lightness,
modify brightness, and modify contrast.
https://bugzilla.gnome.org/show_bug.cgi?id=656156
Emmanuele Bassi [Wed, 28 Sep 2011 12:32:57 +0000 (13:32 +0100)]
tests/interactive: Add some descriptions
Emmanuele Bassi [Tue, 27 Sep 2011 16:50:51 +0000 (17:50 +0100)]
test-textures -> test-texture-slicing
The unit is supposed to be testing the slicing support in CoglTexture.
Emmanuele Bassi [Tue, 27 Sep 2011 16:40:51 +0000 (17:40 +0100)]
test-interactive: Allow querying the interactive test for a description
It would be nice if the interactive tests had a way to be queried for a
description, instead of "Just Knowing" what they are meant to be doing.
Emmanuele Bassi [Tue, 27 Sep 2011 15:40:34 +0000 (16:40 +0100)]
test-texture-async: Clean up the test code
Force threading on, and stop using the default stage and behaviours:
let's try to use modern API.
Emmanuele Bassi [Tue, 27 Sep 2011 15:06:43 +0000 (16:06 +0100)]
texture: Simplify asynchronous loading code
The asynchronous loading code could do with some modernization.
First of all, we should drop the internal GMutex held when manipulating
the boolean flags: it's far too expensive for its role, and modern GLib
provides us with bitlocks that are quite a lot faster.
Then we should consolidate most of the implementation into something
smaller and more manageable.
Emmanuele Bassi [Tue, 27 Sep 2011 13:22:22 +0000 (14:22 +0100)]
texture: Remove GStaticMutex
It's now possible to statically initialize a GMutex.
Emmanuele Bassi [Tue, 27 Sep 2011 13:21:41 +0000 (14:21 +0100)]
Depend on GLib 2.31
We need to switch away from GStaticMutex and use GMutex instead, as the
former has been deprecated.
Tomeu Vizoso [Wed, 28 Sep 2011 12:24:57 +0000 (14:24 +0200)]
Add paint-deform-tiles option to CLUTTER_DEBUG
To make easier to visualize the triangles in a ClutterDeformEffect
https://bugzilla.gnome.org/show_bug.cgi?id=660354
Мирослав Николић [Mon, 26 Sep 2011 23:55:26 +0000 (01:55 +0200)]
Added Serbian translation
Carles Ferrando [Mon, 26 Sep 2011 12:41:42 +0000 (14:41 +0200)]
[l10n]Updated Catalan (Valencian) translation
Emmanuele Bassi [Tue, 16 Aug 2011 15:01:22 +0000 (16:01 +0100)]
Remove internal usage of ClutterGeometry in StageWindow
The ClutterGeometry type is a poor substitute of cairo_rectangle_int_t,
with unsigned integers for width and height to complicate matters.
Let's remove the internal usage of ClutterGeometry and switch to the
rectangle type from Cairo.
https://bugzilla.gnome.org/show_bug.cgi?id=656663
Chao-Hsiung Liao [Sun, 25 Sep 2011 08:16:10 +0000 (16:16 +0800)]
Updated Traditional Chinese translation(Hong Kong and Taiwan)
Ask H. Larsen [Sat, 24 Sep 2011 15:28:11 +0000 (17:28 +0200)]
Updated Danish translation
Mario Blättermann [Sat, 24 Sep 2011 08:44:08 +0000 (10:44 +0200)]
[l10n] Updated German translation
Colin Walters [Fri, 23 Sep 2011 01:22:53 +0000 (21:22 -0400)]
clutter-main: Reword translator comment for default:LTR
This makes it much clearer, and fixes some typos in the original.
https://bugzilla.gnome.org/show_bug.cgi?id=659893
Alexandre Franke [Fri, 23 Sep 2011 09:23:26 +0000 (11:23 +0200)]
Mark forgotten string as translatable
https://bugzilla.gnome.org/show_bug.cgi?id=659928
Bruno Brouard [Fri, 23 Sep 2011 16:26:19 +0000 (18:26 +0200)]
Updated French translation
Neil Roberts [Fri, 23 Sep 2011 14:12:44 +0000 (15:12 +0100)]
Add a CLUTTER_PAINT debug option to enable continous redrawing
When testing the performance of an application, it's often useful to
force it to continuously redraw instead of going idle to help measure
the frame rate. This just adds a CLUTTER_PAINT=continuous-redraw which
causes the master clock to queue a redraw on all of the stages
just before it prepares its source.
Reviewed-by: Robert Bragg <robert@linux.intel.com>
Reviewed-by: Emmanuele Bassi <ebassi@linux.intel.com>
krishnababu k [Fri, 23 Sep 2011 12:31:37 +0000 (18:01 +0530)]
Updated Telugu Translations
Manoj Kumar Giri [Fri, 23 Sep 2011 07:16:22 +0000 (12:46 +0530)]
Updated Oriya Translation
I Felix [Fri, 23 Sep 2011 07:12:15 +0000 (12:42 +0530)]
Updated Tamil Translations
A S Alam [Fri, 23 Sep 2011 01:54:56 +0000 (07:24 +0530)]
update Punjabi Translation
Neil Roberts [Wed, 21 Sep 2011 15:36:30 +0000 (16:36 +0100)]
clutter-offscreen-effect: Apply matrix expansion to proj not mv
When the viewport gets expanded because the actor extends off the edge
of the screen, instead of applying the transformation to the root of
the modelview transformation it is now applied to the end of the
projection transformation. This should end up with the same
transformation. This fixes a problem when the offscreen effects are
nested and the inner effect would try to pick up the current modelview
transformation to rescale it to fit the new viewport size. In this
case the modelview would have already been scaled for the size of the
outer viewport so it would end up wrong.
https://bugzilla.gnome.org/show_bug.cgi?id=659601
Reviewed-by: Emmanuele Bassi <ebassi@linux.intel.com>
Reviewed-by: Robert Bragg <robert@linux.intel.com>