platform/upstream/efl.git
11 years agoevas - and apply proxy_src_clip only when object and proxy is same in render_mapped
ChunEon Park [Mon, 29 Apr 2013 12:05:33 +0000 (21:05 +0900)]
evas - and apply proxy_src_clip only when object and proxy is same in render_mapped

11 years agoAdd start of drm engine
Chris Michael [Mon, 29 Apr 2013 12:02:00 +0000 (13:02 +0100)]
Add start of drm engine

NB: Nothing to see here, just basic evas_engine setup.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoAdd missing braces
Chris Michael [Mon, 29 Apr 2013 11:58:02 +0000 (12:58 +0100)]
Add missing braces

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoFix engine_info_get function call being in the wrong place.
Chris Michael [Mon, 29 Apr 2013 11:56:08 +0000 (12:56 +0100)]
Fix engine_info_get function call being in the wrong place.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoAdd logging macros to drm engine.
Chris Michael [Mon, 29 Apr 2013 11:40:24 +0000 (12:40 +0100)]
Add logging macros to drm engine.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoAdd start of Drm Engine header.
Chris Michael [Mon, 29 Apr 2013 11:37:58 +0000 (12:37 +0100)]
Add start of Drm Engine header.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoAdd drm engine build infrastructure.
Chris Michael [Mon, 29 Apr 2013 11:35:08 +0000 (12:35 +0100)]
Add drm engine build infrastructure.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoAdd "--enable-drm" for drm engine.
Chris Michael [Mon, 29 Apr 2013 11:34:39 +0000 (12:34 +0100)]
Add "--enable-drm" for drm engine.
Add evas engine check for drm

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoAdd EVAS_CHECK_ENGINE_DEP_DRM to m4
Chris Michael [Mon, 29 Apr 2013 10:54:26 +0000 (11:54 +0100)]
Add EVAS_CHECK_ENGINE_DEP_DRM to m4

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoWhen we free the swap buffer, do Not munmap the buffer data.
Chris Michael [Mon, 29 Apr 2013 10:30:44 +0000 (11:30 +0100)]
When we free the swap buffer, do Not munmap the buffer data.

Buffer data (in this instance) is from the swapper (ie: the creation
of the swapper allocates & mmaps the data). As such, we should not
unmap it when we free buffer(s), just when we free the swapper.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoUse the width & height from the buffer (actual image size) when
Chris Michael [Mon, 29 Apr 2013 10:29:33 +0000 (11:29 +0100)]
Use the width & height from the buffer (actual image size) when
calling evas cache image data.
When we push an updated region, be sure to account for the rectangle
position during the conversion function.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoBe sure to set the Render_Engine's info to the new engine info.
Chris Michael [Mon, 29 Apr 2013 10:29:12 +0000 (11:29 +0100)]
Be sure to set the Render_Engine's info to the new engine info.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoevas - fix doc
ChunEon Park [Mon, 29 Apr 2013 07:20:39 +0000 (16:20 +0900)]
evas - fix doc

11 years agoevas - add evas_object_image_source_clip_set()/get()
ChunEon Park [Mon, 29 Apr 2013 06:30:37 +0000 (15:30 +0900)]
evas - add evas_object_image_source_clip_set()/get()

11 years agoejde/edje_entry.c : commit_cancel should be EINA_FALSE when there is one or more...
WooHyun Jung [Mon, 29 Apr 2013 03:46:22 +0000 (12:46 +0900)]
ejde/edje_entry.c : commit_cancel should be EINA_FALSE when there is one or more characters in preediting string.

11 years agoEvas: rename Evas_Common.h to Evas_Common_Header.h
Daniel Zaoui [Sun, 28 Apr 2013 11:22:02 +0000 (14:22 +0300)]
Evas: rename Evas_Common.h to Evas_Common_Header.h

This change is due to EFL compilation failing in Windows. This file was
conflicting with evas_common.h. Case sensitive issue...

11 years agoeo_ptr_ind: pack ptr, active flag and generation all together
Jérémy Zurcher [Sat, 27 Apr 2013 22:43:23 +0000 (00:43 +0200)]
eo_ptr_ind: pack ptr, active flag and generation all together

use of an array of the below struct instead of 3 separate arrays
leads to better cache performance and smaller memory usage
typedef struct
{
  _Eo *ptr;
  unsigned int active     : 1;
  unsigned int generation : BITS_FOR_GENERATION_COUNTER;
} _Eo_Id_Entry;

11 years agoeo_ptr_ind: fix hdr->size in _eo_id_mem_alloc
Jérémy Zurcher [Sat, 27 Apr 2013 22:17:07 +0000 (00:17 +0200)]
eo_ptr_ind: fix hdr->size in _eo_id_mem_alloc

11 years agoecore_audio: Test read-while-paused code path as well
Daniel Willmann [Fri, 26 Apr 2013 17:48:42 +0000 (18:48 +0100)]
ecore_audio: Test read-while-paused code path as well

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Test VIO input/output of base in/out class
Daniel Willmann [Fri, 26 Apr 2013 17:48:00 +0000 (18:48 +0100)]
ecore_audio: Test VIO input/output of base in/out class

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Test the looped signal/code path as well
Daniel Willmann [Fri, 26 Apr 2013 17:46:55 +0000 (18:46 +0100)]
ecore_audio: Test the looped signal/code path as well

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Add vio_set method / write cb to output
Daniel Willmann [Fri, 26 Apr 2013 17:32:18 +0000 (18:32 +0100)]
ecore_audio: Add vio_set method / write cb to output

ecore_audio_obj_out now also supports VIO. Add attribute need_writer so
we can generalize the idler creation for subclasses at a later time.

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Remove duplicate variables/move to appropriate place
Daniel Willmann [Fri, 26 Apr 2013 17:24:15 +0000 (18:24 +0100)]
ecore_audio: Remove duplicate variables/move to appropriate place

* The idler for the write callback is now in the output struct because
  different outputs might need it.
* Remove paused attribute from output - we have it in Ecore_Audio_Object
  already

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Move sndfile VIO into a file to access from in- and output
Daniel Willmann [Fri, 26 Apr 2013 17:17:03 +0000 (18:17 +0100)]
ecore_audio: Move sndfile VIO into a file to access from in- and output

The VIO wrapper functions are needed from the sndfile inputs and outputs
so move them to a separate file and access from both.

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Zero copy in pulse output
Daniel Willmann [Thu, 25 Apr 2013 16:27:13 +0000 (17:27 +0100)]
ecore_audio: Zero copy in pulse output

Using pa_stream_write_begin we can just request a buffer to write into
This reduces the amount of calloc()/free() and memcpy calls.

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoevas - fixed logic fault in proxy clipping.
ChunEon Park [Fri, 26 Apr 2013 11:05:46 +0000 (20:05 +0900)]
evas - fixed logic fault in proxy clipping.

11 years agoevas - For compatibility, Make proxy's source clip to be optional. Im checking partal...
ChunEon Park [Fri, 26 Apr 2013 08:19:50 +0000 (17:19 +0900)]
evas - For compatibility, Make proxy's source clip to be optional. Im checking partal code for zmike.

11 years agoFix variable usage in framespace get. Thanks Seb ;)
Chris Michael [Fri, 26 Apr 2013 06:41:17 +0000 (07:41 +0100)]
Fix variable usage in framespace get. Thanks Seb ;)

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoRemove blank line
Chris Michael [Thu, 25 Apr 2013 13:14:52 +0000 (14:14 +0100)]
Remove blank line

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoReenable ability for async render of shm engine.
Chris Michael [Thu, 25 Apr 2013 13:14:20 +0000 (14:14 +0100)]
Reenable ability for async render of shm engine.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
11 years agoevas - fix the proxy object to not be clipped by source clipper
ChunEon Park [Fri, 26 Apr 2013 05:45:41 +0000 (14:45 +0900)]
evas - fix the proxy object to not be clipped by source clipper

11 years agoevas/wayland_shm: Fix resize to the left or top when rotated.
Rafael Antognolli [Wed, 24 Apr 2013 22:32:49 +0000 (19:32 -0300)]
evas/wayland_shm: Fix resize to the left or top when rotated.

11 years agoevas/wayland_egl: Fix resize to the left or top when rotated.
Rafael Antognolli [Wed, 24 Apr 2013 21:33:43 +0000 (18:33 -0300)]
evas/wayland_egl: Fix resize to the left or top when rotated.

11 years agoEvas textblock: Fixed line size calculation when using multiple fonts.
Tom Hacohen [Thu, 25 Apr 2013 14:16:13 +0000 (15:16 +0100)]
Evas textblock: Fixed line size calculation when using multiple fonts.

Thanks to MinSu Seo for reporting.

11 years agolcov-check: Enable branch coverage on newer lcov versions
Daniel Willmann [Thu, 25 Apr 2013 09:49:36 +0000 (10:49 +0100)]
lcov-check: Enable branch coverage on newer lcov versions

In newer versions branch coverage is disabled by default. Re-enable it
explicitly.

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoedje entry: remove duplicated ecore_imf_context_reset in adjusting the selection
Jihoon Kim [Thu, 25 Apr 2013 00:45:15 +0000 (09:45 +0900)]
edje entry: remove duplicated ecore_imf_context_reset in adjusting the selection

11 years agoEvas: Split Evas headers
Daniel Zaoui [Wed, 24 Apr 2013 19:56:24 +0000 (22:56 +0300)]
Evas: Split Evas headers

Now, Evas.h includes three new files:
- Evas_Eo.h: Eo API functions (functions defines, enums, base id).
- Evas_Legacy.h: contains the API functions related to objects
- Evas_Common.h: common data (structs, enums...) +
functions not related to objects.

This phase is needed for the EFL 1.8 release to disable Eo APIs if we
consider it is not enough mature to be used by applications.

11 years agoeina/evas: Fix includes for examples
Daniel Willmann [Wed, 24 Apr 2013 23:08:55 +0000 (00:08 +0100)]
eina/evas: Fix includes for examples

Some examples are still missing includes

11 years agoeo: Fix incorrect pointer allocation
Daniel Willmann [Wed, 24 Apr 2013 22:27:48 +0000 (23:27 +0100)]
eo: Fix incorrect pointer allocation

We want to allocate sizeof(Eo_Dbg_Info) - not sizeof(Eo_Dbg_Info *)
bytes.

11 years agoecore_evas/wayland: Consider the frame/border for pointer position.
Rafael Antognolli [Wed, 24 Apr 2013 20:56:53 +0000 (17:56 -0300)]
ecore_evas/wayland: Consider the frame/border for pointer position.

When the window is rotated, the logical pointer position is calculated
based on the window size (width or height) minus the current position,
depending on the rotation used. For wayland, we must add the window
decorations to the ecore_evas size, when doing this calculation.

11 years agoecore_evas/wayland: Fix rotation when going back to 0 or 180 degrees.
Rafael Antognolli [Wed, 24 Apr 2013 20:14:15 +0000 (17:14 -0300)]
ecore_evas/wayland: Fix rotation when going back to 0 or 180 degrees.

11 years agoedje: Split Edje headers
Aharon Hillel [Thu, 11 Apr 2013 09:48:20 +0000 (12:48 +0300)]
edje: Split Edje headers

Now, Edje.h includes three new files:
- Edje_Eo.h: Eo API functions (functions defines, enums, base id).
- Edje_Legacy.h: contains the API functions related to objects
- Edje_Common.h: common data (structs, enums...) + functions not related to
objects.

This phase is needed for the EFL 1.8 release to disable Eo APIs if we
consider it is not enough mature to be used by applications.

11 years agoecore_evas/wayland: Add support for rotation on wayland_egl backend.
Rafael Antognolli [Wed, 24 Apr 2013 18:21:44 +0000 (15:21 -0300)]
ecore_evas/wayland: Add support for rotation on wayland_egl backend.

Also move the common code for rotation_set from the shm engine to
wayland_common.

11 years agofix examples makefiles to find Efl_Config.h
Carsten Haitzler (Rasterman) [Wed, 24 Apr 2013 17:08:18 +0000 (02:08 +0900)]
fix examples makefiles to find Efl_Config.h

11 years agoEo: fixed an issue with the test_value test.
Tom Hacohen [Wed, 24 Apr 2013 16:42:24 +0000 (17:42 +0100)]
Eo: fixed an issue with the test_value test.

11 years agoEo: Eo_Class_Id should be uintptr_t not size_t.
Tom Hacohen [Wed, 24 Apr 2013 15:55:11 +0000 (16:55 +0100)]
Eo: Eo_Class_Id should be uintptr_t not size_t.

11 years agoEo: Added test to the special eina value type.
Tom Hacohen [Wed, 24 Apr 2013 15:45:34 +0000 (16:45 +0100)]
Eo: Added test to the special eina value type.

11 years agoand emove efl eo api macro.
Carsten Haitzler (Rasterman) [Wed, 24 Apr 2013 15:39:57 +0000 (00:39 +0900)]
and emove efl eo api macro.

11 years agoadd --with-api=XXX (both|legacy|eo) to define which api set to enable
Carsten Haitzler (Rasterman) [Wed, 24 Apr 2013 15:31:31 +0000 (00:31 +0900)]
add --with-api=XXX (both|legacy|eo) to define which api set to enable
by default (default == both).

11 years agoeeze/sensor/tizen: Fix typos to let it build for tizen
Stefan Schmidt [Wed, 24 Apr 2013 14:59:16 +0000 (15:59 +0100)]
eeze/sensor/tizen: Fix typos to let it build for tizen

Finally have a way to compile the merged efl tree against the latest tizen
things. Only three typos after all this changes it quite good imho. Zero
would have been better though. :)

11 years agoadd a global Efl_Config.h for everyone.
Carsten Haitzler (Rasterman) [Wed, 24 Apr 2013 14:43:33 +0000 (23:43 +0900)]
add a global Efl_Config.h for everyone.

* ned to replicate changes in other .pc.in files
* need to replicate changes in other E*.h installed header files

11 years agoeo: Lets fix make distcheck by including eo_ptr_indirection header
Daniel Willmann [Wed, 24 Apr 2013 12:59:53 +0000 (13:59 +0100)]
eo: Lets fix make distcheck by including eo_ptr_indirection header

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agorejig 32bit eoid allocation to 5.5.12.
Carsten Haitzler (Rasterman) [Wed, 24 Apr 2013 07:20:18 +0000 (16:20 +0900)]
rejig 32bit eoid allocation to 5.5.12.

11 years agoecore/evas wayland_shm: Add support for window rotation.
Rafael Antognolli [Fri, 1 Mar 2013 17:19:10 +0000 (14:19 -0300)]
ecore/evas wayland_shm: Add support for window rotation.

11 years agoevas/wayland: Remove framespace clipper.
Rafael Antognolli [Mon, 22 Apr 2013 19:29:50 +0000 (16:29 -0300)]
evas/wayland: Remove framespace clipper.

This clipper caused several bugs already, and there are some bugs still
not fixed. Let's remove it and try to fix any remaining with some other
kind of solution that does not depend on adding or clipping objects
during the evas render phase, which causes unexpected behavior.

11 years agoecore_evas/wayland: Change the default frame smart object.
Rafael Antognolli [Mon, 22 Apr 2013 19:44:43 +0000 (16:44 -0300)]
ecore_evas/wayland: Change the default frame smart object.

Add 4 rectangles to be used as border of the window, instead of a single
rectangle under the framespace. This allows to move the frame object to
the top layer, instead of staying on the lowest layer. With the frame
over the other objects, there's no need of framespace clipper object,
which causes several bugs.

11 years agoecore/wayland: Add support for engine animated cursor.
Rafael Antognolli [Mon, 22 Apr 2013 16:35:54 +0000 (13:35 -0300)]
ecore/wayland: Add support for engine animated cursor.

11 years agoexamples/ecore_audio: Require input and output file as argument
Daniel Willmann [Tue, 23 Apr 2013 16:11:48 +0000 (17:11 +0100)]
examples/ecore_audio: Require input and output file as argument

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Add more return values for methods, catch up with API
Daniel Willmann [Tue, 23 Apr 2013 16:10:16 +0000 (17:10 +0100)]
ecore_audio: Add more return values for methods, catch up with API

input_attach, input_detach, format_set, and source_set now return an
Eina_Bool

Tests, examples and edje_multisense adapted

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoRename edbus->eldbus
Lucas De Marchi [Tue, 23 Apr 2013 14:40:05 +0000 (11:40 -0300)]
Rename edbus->eldbus

git grep -l edbus2 | while read f; do sed -i 's/edbus2/eldbus/g' "$f"; done
find . -name '*edbus2*' -exec rename edbus2 eldbus {} \;

git grep -l "EDBUS" | while read f; do sed -i 's/EDBUS/ELDBUS/g' "$f"; done
git grep -l "EDBus" | while read f; do sed -i 's/EDBus/Eldbus/g' "$f"; done
git grep -l "edbus (v2)" | while read f; do sed -i 's/edbus (v2)/eldbus/g' "$f"; done
git grep -l "Edbus" | while read f; do sed -i 's/Edbus/Eldbus/g' "$f"; done
git grep -l "edbus" | while read f; do sed -i 's/edbus/eldbus/g' "$f"; done

find . -name '*edbus*' -exec rename edbus eldbus {} \;
find . -name '*EDBus*' -exec rename EDBus Eldbus {} \;

11 years agoeoid -> use mmap for allocating id tables. makes it a separate memory
Carsten Haitzler (Rasterman) [Tue, 23 Apr 2013 14:38:44 +0000 (23:38 +0900)]
eoid -> use mmap for allocating id tables. makes it a separate memory
regions with page separations... this allows us more direct control
over access and visibility.

11 years agoeoid -> rjid 64bit bit allocations for.. not so much bloat.
Carsten Haitzler (Rasterman) [Tue, 23 Apr 2013 09:48:35 +0000 (18:48 +0900)]
eoid -> rjid 64bit bit allocations for.. not so much bloat.

11 years agoecore: Split Ecore headers
Aharon Hillel [Wed, 3 Apr 2013 14:46:59 +0000 (17:46 +0300)]
ecore: Split Ecore headers

Now, Ecore.h includes three new files:
- Ecore_Eo.h: Eo API functions (functions defines, enums, base id).
- Ecore_Legacy.h: contains the API functions related to objects
- Ecore_Common.h: common data (structs, enums...) + functions not
related to objects.

This phase is needed for the EFL 1.8 release to disable Eo APIs if we
consider it is not enough mature to be used by applications.

11 years agoefl: add m4 macro to handle a way to enable/disable Eo API and EFL legacy API.
Cedric Bail [Tue, 23 Apr 2013 10:53:16 +0000 (19:53 +0900)]
efl: add m4 macro to handle a way to enable/disable Eo API and EFL legacy API.

11 years agogitignore: don't monitor generated file.
Cedric Bail [Tue, 23 Apr 2013 11:14:10 +0000 (20:14 +0900)]
gitignore: don't monitor generated file.

11 years agofix eeze scanner header install
Mike Blumenkrantz [Tue, 23 Apr 2013 09:21:43 +0000 (10:21 +0100)]
fix eeze scanner header install

11 years agoevas: take stride into account with swapbuf.
Cedric Bail [Tue, 23 Apr 2013 07:48:27 +0000 (16:48 +0900)]
evas: take stride into account with swapbuf.

11 years agoEo: tests to check eo ptrs indirection feature.
Daniel Zaoui [Thu, 18 Apr 2013 10:55:55 +0000 (13:55 +0300)]
Eo: tests to check eo ptrs indirection feature.

The test doesn't fail if the feature is disabled.

11 years agoEo: pointers indirection mechanism for objects and classes
Daniel Zaoui [Thu, 18 Apr 2013 11:19:02 +0000 (14:19 +0300)]
Eo: pointers indirection mechanism for objects and classes

Summary: This feature replaces Eo pointers with ids to prevent bad usage
or reuse of these pointers. It doesn't change API.
The mechanism uses tables storing the real pointers to the objects.
See the src/lib/eo/eo_ptr_indirection.c file for more details on the
mechanism.

11 years agoRevert "ummm dont set max based on max hint... this got added at some point"
Carsten Haitzler (Rasterman) [Tue, 23 Apr 2013 06:44:16 +0000 (15:44 +0900)]
Revert "ummm dont set max based on max hint... this got added at some point"

This reverts commit 116013320a941357a925441928ea7e10ed6c0ebd.

11 years agoEo: add benchmarks to measure eo_add.
Daniel Zaoui [Thu, 18 Apr 2013 10:55:11 +0000 (13:55 +0300)]
Eo: add benchmarks to measure eo_add.

The first one checks simple objects creation.
The second one creates objects, removes half of them and creates the same number.

11 years agofix recursive proxy rendering to just show a black rectangle
Mike Blumenkrantz [Mon, 22 Apr 2013 15:41:24 +0000 (16:41 +0100)]
fix recursive proxy rendering to just show a black rectangle

11 years agoeeze/sensor: Simplify object handling in async paths.
Stefan Schmidt [Fri, 19 Apr 2013 08:55:02 +0000 (09:55 +0100)]
eeze/sensor: Simplify object handling in async paths.

This follows the simplified handling of sensor object we are already
doing for the sync paths. Its a bit more complicated here as we need
to pass all data through the module specific async handling. But the
result should be the same.

11 years agoummm dont set max based on max hint... this got added at some point
Carsten Haitzler (Rasterman) [Mon, 22 Apr 2013 15:05:59 +0000 (00:05 +0900)]
ummm dont set max based on max hint... this got added at some point
but it actually breaks older code that used these hints to store
sizing info but didnt expect it to be enforced.

11 years agofix 64bit ecore-x selection overrun bug.
Carsten Haitzler (Rasterman) [Sat, 20 Apr 2013 01:31:01 +0000 (10:31 +0900)]
fix 64bit ecore-x selection overrun bug.

11 years agonote in comments som get visual info from fbconfig stuff to mach
Carsten Haitzler (Rasterman) [Sat, 20 Apr 2013 00:13:38 +0000 (09:13 +0900)]
note in comments som get visual info from fbconfig stuff to mach
depths. doesn't help though.

11 years agoEfl: Remove libeo.so.1.7.99-gdb.py that got added in the last commit.
Tom Hacohen [Mon, 22 Apr 2013 14:42:15 +0000 (15:42 +0100)]
Efl: Remove libeo.so.1.7.99-gdb.py that got added in the last commit.

SPANK SPANK SPANK Woohyun.

11 years agoedje/edje_entry : When selected word exists, selection is cleared and cursor remains...
WooHyun Jung [Mon, 22 Apr 2013 14:34:25 +0000 (23:34 +0900)]
edje/edje_entry : When selected word exists, selection is cleared and cursor remains on the current entry by additional key input.

11 years agoevas/evas_gl - cast arguments properly according to the spec of the gles/gl APIs.
ChunEon Park [Sun, 21 Apr 2013 13:31:58 +0000 (22:31 +0900)]
evas/evas_gl - cast arguments properly according to the spec of the gles/gl APIs.

11 years agoeina_unicode.H: be nice. add more documentation for deprecated api.
Daniel Juyung Seo [Sat, 20 Apr 2013 10:01:02 +0000 (19:01 +0900)]
eina_unicode.H: be nice. add more documentation for deprecated api.

11 years agoecore_audio: Guard against attaching an input twice
Daniel Willmann [Fri, 19 Apr 2013 15:41:17 +0000 (16:41 +0100)]
ecore_audio: Guard against attaching an input twice

Add check in input_attach and change test case accordingly

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore/ews: Make this engine be aware of the wayland frame object.
Rafael Antognolli [Fri, 19 Apr 2013 15:53:45 +0000 (12:53 -0300)]
ecore/ews: Make this engine be aware of the wayland frame object.

Its background object must be resized to the framespace, and left at
least one layer available for the frame object to be under it.

11 years agoremoved useless mkdir in bould.
Tom Hacohen [Fri, 19 Apr 2013 15:57:04 +0000 (16:57 +0100)]
removed useless mkdir in bould.

11 years agoEo libeo-gdb.py: Fixed up autofoo to not create it as root.
Tom Hacohen [Fri, 19 Apr 2013 15:50:15 +0000 (16:50 +0100)]
Eo libeo-gdb.py: Fixed up autofoo to not create it as root.

11 years agoecore_audio: Don't fail configure with --disable-audio
Daniel Willmann [Fri, 19 Apr 2013 15:16:34 +0000 (16:16 +0100)]
ecore_audio: Don't fail configure with --disable-audio

Who knew I should test all these different possibilities...

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio test: Better test coverage for inputs
Daniel Willmann [Fri, 19 Apr 2013 14:30:26 +0000 (15:30 +0100)]
ecore_audio test: Better test coverage for inputs

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Remove unnecessary methods in tone input
Daniel Willmann [Fri, 19 Apr 2013 14:29:18 +0000 (15:29 +0100)]
ecore_audio: Remove unnecessary methods in tone input

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Fix seeking in tone input
Daniel Willmann [Fri, 19 Apr 2013 14:28:00 +0000 (15:28 +0100)]
ecore_audio: Fix seeking in tone input

Also make seekable be a property of the input as this doesn't make sense
for the output.

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Make efl compile with PA disabled again
Daniel Willmann [Fri, 19 Apr 2013 12:39:02 +0000 (13:39 +0100)]
ecore_audio: Make efl compile with PA disabled again

Now instead of having ifdefs in the .c files we just don't compile the
ones we don't need. Much cleaner.

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Fix possible memory leak in pulse main loop integration
Daniel Willmann [Fri, 19 Apr 2013 12:10:49 +0000 (13:10 +0100)]
ecore_audio: Fix possible memory leak in pulse main loop integration

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoeeze/sensor/tizen: Fix start and stop from sensors during init and shutdown
Stefan Schmidt [Fri, 19 Apr 2013 09:00:11 +0000 (10:00 +0100)]
eeze/sensor/tizen: Fix start and stop from sensors during init and shutdown

11 years agoeeze/sensor: Simplify sensor object handling.
Stefan Schmidt [Thu, 18 Apr 2013 16:35:37 +0000 (17:35 +0100)]
eeze/sensor: Simplify sensor object handling.

We rely on the app to provide a sensible object pointer and we now longer
need to have a copy of the object around to operate on it.

Simplifies code, maintenance and reduces mem copies. Win-Win :)

11 years agoecore: update ecore_test_ecore_main_loop_event test for more coverage code.
Valerii Kanunik [Fri, 19 Apr 2013 06:42:41 +0000 (15:42 +0900)]
ecore: update ecore_test_ecore_main_loop_event test for more coverage code.

Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
11 years agoefreet: shutdown in the right order.
Cedric Bail [Fri, 19 Apr 2013 06:41:17 +0000 (15:41 +0900)]
efreet: shutdown in the right order.

11 years agoedje: remove useless ecore_imf_context_reset call in case of pressing ESC and Return key
Jihoon Kim [Fri, 19 Apr 2013 05:46:42 +0000 (14:46 +0900)]
edje: remove useless ecore_imf_context_reset call in case of pressing ESC and Return key

11 years agoecore_evas/wayland: Actually, the frame object should be on LAYER_MIN.
Rafael Antognolli [Thu, 18 Apr 2013 22:39:46 +0000 (19:39 -0300)]
ecore_evas/wayland: Actually, the frame object should be on LAYER_MIN.

Also set the layer on the wayland_egl backend.

11 years agoecore_evas/wayland: Put the default frame object in a lower layer.
Rafael Antognolli [Thu, 18 Apr 2013 22:04:07 +0000 (19:04 -0300)]
ecore_evas/wayland: Put the default frame object in a lower layer.

The frame should not stay in the same layer as the other objects. If the
application wants to allow it to be on top of the content, then it must
provide a custom frame object and set it to that layer.

NOTE: Should we make Elementary's be on a lower layer too?

11 years agoevas/wayland: Unclip objects from the framespace after rendering.
Rafael Antognolli [Thu, 18 Apr 2013 19:10:53 +0000 (16:10 -0300)]
evas/wayland: Unclip objects from the framespace after rendering.

These objects should be clipped only during rendering, since keeping
them clipped after that allows for unexpected behavior on the
application side. For instance, an application could check if objects
have clippers before doing something to them, assuming that some objects
should have no clipper, but under wayland, after the first render
iteration, there will be no objects without a clipper.

This commit fixes this behavior by unclipping objects that had no
clipper prior to the render iteration.

Additionally, it fixes a bug where a maximized/fullscreen window could
have not all of its content rendered immediately. This was occuring
because some objects could be clipped to the framespace clipper, but
considered invisible in the beginning of the render phase, where they
are evaluated. They were considered invisible because the framespace
clipper object was not resized at that phase yet, and thus these objects
were being clipped out from the viewport.

11 years agotests/ecore_audio: Create temporary test files in build dir
Daniel Willmann [Thu, 18 Apr 2013 18:47:35 +0000 (19:47 +0100)]
tests/ecore_audio: Create temporary test files in build dir

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
11 years agoecore_audio: Fix warnings
Daniel Willmann [Thu, 18 Apr 2013 18:34:33 +0000 (19:34 +0100)]
ecore_audio: Fix warnings

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>