platform/upstream/efl.git
9 years agoeolian-cxx: Get rid of class_prefix and related code.
Savio Sena [Thu, 25 Sep 2014 19:09:40 +0000 (16:09 -0300)]
eolian-cxx: Get rid of class_prefix and related code.

This patch fixes the use of eolian_function_full_c_name_get API also.

9 years agoEo: Change eo_add/del/unref behaviour.
Tom Hacohen [Thu, 25 Sep 2014 14:51:17 +0000 (15:51 +0100)]
Eo: Change eo_add/del/unref behaviour.

Before this change eo_add() used to create an object with 1 ref, and if
the object had a parent, a second ref.
Now, eo_add() always returns an object with 1 ref, and eo_add_ref()
    preserves the old behaviour (for bindings).

eo_unref now un-parents if refcount is 0, and eo_del() is an alias for
eo_unref (will change to be a way to ensure an object is dead and goes
        to zombie-land even if still refed).

9 years agoecore-drm: Don't crash if we get no devpath from Eeze
Chris Michael [Thu, 25 Sep 2014 16:04:23 +0000 (12:04 -0400)]
ecore-drm: Don't crash if we get no devpath from Eeze

Summary: During a device remove event, we should check that Eeze has
returned us a devpath to remove else we segfault when trying to strcmp
the device path later during actual remove.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Add some more debug information if a device fails to open
Chris Michael [Thu, 25 Sep 2014 15:46:46 +0000 (11:46 -0400)]
ecore-drm: Add some more debug information if a device fails to open

Summary: This is mainly for debugging, but lets print out a bit more
information if a device fails to open.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix adding an input device when we don't get the devpath
Chris Michael [Thu, 25 Sep 2014 15:36:24 +0000 (11:36 -0400)]
ecore-drm: Fix adding an input device when we don't get the devpath
from Eeze.

Summary: If we fail to get a devpath from Eeze for an input device,
then we need to error out gracefully and not add a failed input device.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoelua: update eolian bindings and generator
Daniel Kolesa [Thu, 25 Sep 2014 14:53:42 +0000 (15:53 +0100)]
elua: update eolian bindings and generator

9 years agoeolian: remove the need to pass prefix to eolian_function_full_c_name_get
Daniel Kolesa [Thu, 25 Sep 2014 14:37:18 +0000 (15:37 +0100)]
eolian: remove the need to pass prefix to eolian_function_full_c_name_get

9 years agoeolian: properly replace commas in event_c_name_get and cleanup generator
Daniel Kolesa [Thu, 25 Sep 2014 14:09:36 +0000 (15:09 +0100)]
eolian: properly replace commas in event_c_name_get and cleanup generator

9 years agoeolian: new API: eolian_event_c_name_get
Daniel Kolesa [Thu, 25 Sep 2014 13:58:53 +0000 (14:58 +0100)]
eolian: new API: eolian_event_c_name_get

9 years agoefreet: TODO++
Sebastian Dransfeld [Thu, 25 Sep 2014 11:54:13 +0000 (13:54 +0200)]
efreet: TODO++

9 years agoEo: Get rid of eo_add_custom.
Tom Hacohen [Thu, 25 Sep 2014 09:24:27 +0000 (10:24 +0100)]
Eo: Get rid of eo_add_custom.

We decided to deprecate this one in favour of setting thing in the
construction between the constructor and the finalizer.

9 years agoefreet: Fix leak
Sebastian Dransfeld [Thu, 25 Sep 2014 07:22:22 +0000 (09:22 +0200)]
efreet: Fix leak

When doing efreet_desktop_command_append_multiple we add all files from
command, whilst still looping command->files. So the command was created
with all files for all files.

Set l to NULL, since we parse all elements in command->files in
efreet_desktop_command_append_multiple to break the loop.

9 years agoefreet: remove debug fprintf
Sebastian Dransfeld [Thu, 25 Sep 2014 07:22:07 +0000 (09:22 +0200)]
efreet: remove debug fprintf

9 years agoefreet: Fix leak on error
Sebastian Dransfeld [Thu, 25 Sep 2014 07:01:12 +0000 (09:01 +0200)]
efreet: Fix leak on error

If we encountered an error, special hashes in internal desktop struct
wasn't free'd.

9 years agoefreet: Implement fields added by desktop spec 1.1
Sebastian Dransfeld [Thu, 25 Sep 2014 06:47:18 +0000 (08:47 +0200)]
efreet: Implement fields added by desktop spec 1.1

9 years agoefreet: Accept both only_show_in and not_show_in
Sebastian Dransfeld [Wed, 24 Sep 2014 20:15:17 +0000 (22:15 +0200)]
efreet: Accept both only_show_in and not_show_in

In one file it is allowed with both OnlyShowIn and NotShowIn, and it is
the user who has to ensure that these lists don' clash.

@fix

9 years agoevas: supplement missing fields in doc.
ChunEon Park [Thu, 25 Sep 2014 02:48:08 +0000 (11:48 +0900)]
evas: supplement missing fields in doc.

9 years agoefreet: add check for cache version
Sebastian Dransfeld [Wed, 24 Sep 2014 19:08:37 +0000 (21:08 +0200)]
efreet: add check for cache version

If version differs between new and old cache, signal change

9 years agopkg-config: Make Eina++ require and use Eo.
Savio Sena [Wed, 24 Sep 2014 16:02:29 +0000 (13:02 -0300)]
pkg-config: Make Eina++ require and use Eo.

Currently bindings/eina-cxx/{eina_clone_allocators.hh, eina_list.hh,
eina_array.hh} include Eo.h, so it's actually already a dependency. This
patch just reflects it in eina-cxx.pc.

9 years agoecore-drm: Add an extra blank line just to separate code
Chris Michael [Wed, 24 Sep 2014 13:59:23 +0000 (09:59 -0400)]
ecore-drm: Add an extra blank line just to separate code

Summary: This commit has no functional changes, just
cosmetic/formatting to make the code easier to read.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Added private functions to get brightness levels in backlight
vivek [Wed, 24 Sep 2014 13:58:57 +0000 (09:58 -0400)]
ecore-drm: Added private functions to get brightness levels in backlight

Summary:
 Added private internal functions to get various brightness levels
   1) Added _ecore_drm_actual_brightness_get to get Actual brightness
   2) Added _ecore_drm_max_brightness_get to get Max brightness
   3) Added _ecore_drm_brightness_get to get Brightness
 These functions will eventually be exposed with an API for ecore-drm
to get/set the backlight level for an output.

@feature

Signed-off-by: vivek <vivek.ellur@samsung.com>
Reviewers: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1482

9 years agoecore-drm: Remove need to use an extra variable for storing device.
Chris Michael [Wed, 24 Sep 2014 13:06:47 +0000 (09:06 -0400)]
ecore-drm: Remove need to use an extra variable for storing device.

Summary: This commit just modifies the internal code to remove the
need for an extra variable. There really is no need to store the
device twice. We already have it in the list of returned devices.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoelua: class unregister + keep track of Eo_Class
Daniel Kolesa [Wed, 24 Sep 2014 12:49:39 +0000 (13:49 +0100)]
elua: class unregister + keep track of Eo_Class

9 years agoEdje cmake: Add a way to define edje compilation target.
Tom Hacohen [Wed, 24 Sep 2014 12:47:06 +0000 (13:47 +0100)]
Edje cmake: Add a way to define edje compilation target.

9 years agoelua: remove the eo_methods stuff, improve class registration
Daniel Kolesa [Wed, 24 Sep 2014 12:37:26 +0000 (13:37 +0100)]
elua: remove the eo_methods stuff, improve class registration

9 years agoelua: Eo metatype with method table redirection
Daniel Kolesa [Wed, 24 Sep 2014 10:51:11 +0000 (11:51 +0100)]
elua: Eo metatype with method table redirection

9 years agoecore_x_vsync - fix tiny leaklet of drmversion
Carsten Haitzler (Rasterman) [Wed, 24 Sep 2014 10:33:10 +0000 (19:33 +0900)]
ecore_x_vsync - fix tiny leaklet of drmversion

@fix

9 years agoelua: ad the rest of Eo APIs to ffi cdef
Daniel Kolesa [Wed, 24 Sep 2014 10:09:46 +0000 (11:09 +0100)]
elua: ad the rest of Eo APIs to ffi cdef

9 years agoevas_events: fixed keyup with modifier
Andrii Kroitor [Wed, 24 Sep 2014 09:42:26 +0000 (11:42 +0200)]
evas_events: fixed keyup with modifier

Summary:
removed check that disallows exact modifiers set.
I.e. if we need Ctrl+s without any other modifiers:

mask = evas_key_modifier_mask_get(obj, "Control");
evas_object_key_grab(obj, "o", mask, ~mask, false);

This works with keydown, but was broken with keyup.

Reviewers: cedric, seoz, Hermet, raster

Subscribers: cedric, reutskiy.v.v

Differential Revision: https://phab.enlightenment.org/D1474

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
Signed-off-by: Carsten Haitzler (Rasterman) <raster@rasterman.com>
9 years agoelua: add some more eo APIs to ffi cdef
Daniel Kolesa [Wed, 24 Sep 2014 09:39:47 +0000 (10:39 +0100)]
elua: add some more eo APIs to ffi cdef

9 years agoeina_value_util_type_offset: abs(unsigned int) is nonsense
Jérémy Zurcher [Wed, 24 Sep 2014 09:15:58 +0000 (11:15 +0200)]
eina_value_util_type_offset: abs(unsigned int) is nonsense

9 years agoeo: use the "EXTRA_DIST" macro to install .py files
Cedric BAIL [Wed, 24 Sep 2014 08:26:35 +0000 (10:26 +0200)]
eo: use the "EXTRA_DIST" macro to install .py files

Summary:
The SCRIPTS macro will add the executable flag (+x) to the Python scripts ; but as
these do not have a shebang (#!/bin/python) and we are executing them explicitly with
"python ...", it is useless. Plus, some build environments will issue a warning about
this.

Reviewers: tasn, JackDanielZ

Subscribers: mbachmann

Projects: #efl

Differential Revision: https://phab.enlightenment.org/D1469

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoautotools: modified autogen.sh to be able to build from any directory.
Srivardhan Hebbar [Wed, 24 Sep 2014 08:20:07 +0000 (10:20 +0200)]
autotools: modified autogen.sh to be able to build from any directory.

Summary:
Now we cannot have a designated build directory for building EFL. We have to build from the same source folder only. So have modified autogen.sh script
so that we can build from any directory. This is similar to the one which I had submitted for Enlightenment and Elementary.

Signed-off-by: Srivardhan Hebbar <sri.hebbar@samsung.com>
Reviewers: devilhorns, cedric, raster

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1480

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoefl build - fix parallel builds with new branch merge from dh
Carsten Haitzler (Rasterman) [Wed, 24 Sep 2014 00:06:42 +0000 (09:06 +0900)]
efl build - fix parallel builds with new branch merge from dh

fix tha broken stuff like missing dependencies and links for drm
engine module that cause parallel build breaks!

9 years agoeeze: Fix missing break in switch(s)
Chris Michael [Tue, 23 Sep 2014 23:18:58 +0000 (19:18 -0400)]
eeze: Fix missing break in switch(s)

Summary: Fixes Coverity CID1240226, CID1240225

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years ago+eina_strdup(), eina_streq()
Mike Blumenkrantz [Tue, 23 Sep 2014 20:15:34 +0000 (16:15 -0400)]
+eina_strdup(), eina_streq()

there are macros/inlines for these in most efl projects I've seen, may as well have them standardized here

@feature

9 years agomove eina_value_util_time_string_new() to not inline
Mike Blumenkrantz [Tue, 23 Sep 2014 19:51:51 +0000 (15:51 -0400)]
move eina_value_util_time_string_new() to not inline

uses strptime which will cause warnings for people who don't do includes correctly

9 years agofix config.h inclusion across the tree
Mike Blumenkrantz [Tue, 23 Sep 2014 19:48:16 +0000 (15:48 -0400)]
fix config.h inclusion across the tree

9 years agoadd eina_value_util.h to Eina.h
Mike Blumenkrantz [Tue, 23 Sep 2014 19:19:21 +0000 (15:19 -0400)]
add eina_value_util.h to Eina.h

9 years agoMerge branch 'devs/devilhorns/ecore_drm_eeze'
Chris Michael [Tue, 23 Sep 2014 19:50:44 +0000 (15:50 -0400)]
Merge branch 'devs/devilhorns/ecore_drm_eeze'

ecore-drm: Port existing ecore-drm to use Eeze

ecore-drm now uses Eeze library for udev device functions. This
removes the existing dependency on udev and makes use of our own
library (Eeze) to find, list, watch, and otherwise deal with udev
devices. This is the last of the ecore-drm code that needs porting

NB: Ecore-Drm Now uses Eeze for udev and Eldbus for dbus functionality.

@feature

9 years agoecore-drm: Port ecore_drm_output backlight init code to use Eeze
Chris Michael [Tue, 23 Sep 2014 19:45:53 +0000 (15:45 -0400)]
ecore-drm: Port ecore_drm_output backlight init code to use Eeze

Summary: This ports the current backlight_init code to use Eeze
library instead of udev functions directly.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Port ecore_drm_inputs code to use Eeze instead of udev
Chris Michael [Tue, 23 Sep 2014 19:11:43 +0000 (15:11 -0400)]
ecore-drm: Port ecore_drm_inputs code to use Eeze instead of udev

Summary: This changes all of our internal ecore-drm input code to use
Eeze library instead of udev directly.

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Remove udev from private header and add an Eeze_Udev_Watch
Chris Michael [Tue, 23 Sep 2014 17:28:06 +0000 (13:28 -0400)]
ecore-drm: Remove udev from private header and add an Eeze_Udev_Watch
for Ecore_Drm_Input

Summary: This commit removes all references to 'udev' in the private
header, and replaces udev_monitor (in Ecore_Drm_Input structure) with
an Eeze_Udev_Watch.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix ecore_drm_device_find function to use Eeze
Chris Michael [Tue, 23 Sep 2014 17:13:25 +0000 (13:13 -0400)]
ecore-drm: Fix ecore_drm_device_find function to use Eeze

Summary: This converts the internal code of ecore_drm_device_find
function to use Eeze function calls instead of udev in order to locate
the drm card.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Remove references to udev functions and init Eeze during
Chris Michael [Tue, 23 Sep 2014 14:34:00 +0000 (10:34 -0400)]
ecore-drm: Remove references to udev functions and init Eeze during
startup

Summary: This removes the variable & function calls to udev and adds
code to initialize/shutdown Eeze.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix comment to not mention udev
Chris Michael [Tue, 23 Sep 2014 13:06:17 +0000 (09:06 -0400)]
ecore-drm: Fix comment to not mention udev

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Add Eeze as an internal dependency for ecore-drm
Chris Michael [Tue, 23 Sep 2014 13:04:56 +0000 (09:04 -0400)]
ecore-drm: Add Eeze as an internal dependency for ecore-drm

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Remove internal dependency on libudev
Chris Michael [Tue, 23 Sep 2014 13:04:37 +0000 (09:04 -0400)]
ecore-drm: Remove internal dependency on libudev

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Remove dependency on libudev for ecore-drm
Chris Michael [Tue, 23 Sep 2014 13:03:46 +0000 (09:03 -0400)]
ecore-drm: Remove dependency on libudev for ecore-drm

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Add Eeze as a dependency for Ecore_Drm
Chris Michael [Tue, 23 Sep 2014 13:03:17 +0000 (09:03 -0400)]
ecore-drm: Add Eeze as a dependency for Ecore_Drm

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Move Eeze above Ecore_Drm
Chris Michael [Tue, 23 Sep 2014 13:02:28 +0000 (09:02 -0400)]
ecore-drm: Move Eeze above Ecore_Drm

Summary: Ecore_Drm is going to be using Eeze for udev functionality,
so we need to check for Eeze deps before Ecore_Drm

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Add cases for backlight and leds in get_syspath_from_watch.
Chris Michael [Tue, 23 Sep 2014 19:26:30 +0000 (15:26 -0400)]
eeze: Add cases for backlight and leds in get_syspath_from_watch.

Summary: This adds support for backlight and leds devices for use with
an Eeze_Udev_Watch

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Add cases for backlight and leds in eeze_udev_find_by_type
Chris Michael [Tue, 23 Sep 2014 19:25:29 +0000 (15:25 -0400)]
eeze: Add cases for backlight and leds in eeze_udev_find_by_type

Summary: Add code to find backlight and led devices inside
eeze_udev_find_by_type function. This will be used in Ecore-Drm to
find the backlight of an output.

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Add Eeze udev types for backlight and leds
Chris Michael [Tue, 23 Sep 2014 19:20:14 +0000 (15:20 -0400)]
eeze: Add Eeze udev types for backlight and leds

Summary: Add additional types to Eeze udev device types for backlight
and leds. Ecore-Drm will use these types to locate the backlight in
order to support setting/getting the backlight level on an output.

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-evas-drm: Fix calling ecore_drm_device_find twice in some cases
Chris Michael [Tue, 23 Sep 2014 19:09:47 +0000 (15:09 -0400)]
ecore-evas-drm: Fix calling ecore_drm_device_find twice in some cases

Summary: If no device name was passed in to ecore_drm_device_find
function, then if we fail to find the default drm device on the first
call, there is No point in calling the same function again to get the
same result.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Fix issue with eeze_udev_find_by_type not finding devices when
Chris Michael [Tue, 23 Sep 2014 18:58:37 +0000 (14:58 -0400)]
eeze: Fix issue with eeze_udev_find_by_type not finding devices when
name is NULL.

Summary: This was totally wrong !! eeze_udev_find_by_type can be
passed EEZE_UDEV_TYPE_NONE to list all devices and it can also be
passed name == NULL to find All devices listed of a certain type. To
check for !etype && !name is just plain Wrong !! We have checks below
for a valid name (and then do a comparison).

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agorename eina_value_util_timestamp_new -> eina_value_util_time_string_new
Mike Blumenkrantz [Tue, 23 Sep 2014 18:57:10 +0000 (14:57 -0400)]
rename eina_value_util_timestamp_new -> eina_value_util_time_string_new

9 years ago+eina_value_util api
Mike Blumenkrantz [Tue, 23 Sep 2014 18:37:48 +0000 (14:37 -0400)]
+eina_value_util api

merged from maelstrom/azy. commonly used functionality for eina_value with _util_ namespace to make them easier to find in the value header nightmare

@feature

9 years agoeeze: Fix return value for eeze_udev_syspath_get_sysnum
Chris Michael [Tue, 23 Sep 2014 17:04:57 +0000 (13:04 -0400)]
eeze: Fix return value for eeze_udev_syspath_get_sysnum

Summary: oopsie !! Big fat copy/paste error :( This fixes the return
value for the eeze_udev_syspath_get_sysnum function

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Add code for eeze_udev_syspath_get_sysnum function
Chris Michael [Tue, 23 Sep 2014 16:59:54 +0000 (12:59 -0400)]
eeze: Add code for eeze_udev_syspath_get_sysnum function

Summary: This adds an API function to return the sysnum of a device
based on syspath. This is needed in ecore-drm to return the id
(sysnum) of a drm device.

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Add eeze_udev_syspath_get_synum function
Chris Michael [Tue, 23 Sep 2014 16:51:23 +0000 (12:51 -0400)]
eeze: Add eeze_udev_syspath_get_synum function

Summary: This adds an API declaration for getting the sysnum of a
device from the syspath. This is needed in the ecore-drm code to fetch
the id of a device.

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoedje: fix borked edje_external.
Cedric BAIL [Tue, 23 Sep 2014 16:03:02 +0000 (18:03 +0200)]
edje: fix borked edje_external.

Someone fixed a CID without looking of the full use path a data, resulting
in a value being used after free and stored in edje file. Leading to a garbage
string with no meaning for those poor externals.

9 years agoeina: do no use umask on Windows in eina_file_mkstemp()
Vincent Torri [Tue, 23 Sep 2014 13:44:05 +0000 (15:44 +0200)]
eina: do no use umask on Windows in eina_file_mkstemp()

umask() sets the permissions of the file to read-only on Windows
(see umask documentation on MSDN).
This breaks the creation of .edj file (epp needs to modify the
created file).
Anyway, on Windows, permissions should be given to anybody.

@fix

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoMerge branch 'devs/devilhorns/eeze'
Chris Michael [Tue, 23 Sep 2014 14:04:53 +0000 (10:04 -0400)]
Merge branch 'devs/devilhorns/eeze'

eeze: Add ability for Eeze to find DRM cards

Ecore-Drm is going to use Eeze (instead of udev directly) to locate
drm cards and other inputs. As such, I needed the ability (inside
Eeze) to locate devices based on the drm subsystem. This merge adds
code to handle that.

@feature

9 years agoeeze: Add ability to get syspath from a watch for DRM
Chris Michael [Tue, 23 Sep 2014 13:57:20 +0000 (09:57 -0400)]
eeze: Add ability to get syspath from a watch for DRM

Summary: This adds the EEZE_UDEV_TYPE_DRM to the switch for
_get_syspath_from_watch.

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Add case for EEZE_UDEV_TYPE_DRM in eeze_udev_find_by_type function
Chris Michael [Tue, 23 Sep 2014 13:56:12 +0000 (09:56 -0400)]
eeze: Add case for EEZE_UDEV_TYPE_DRM in eeze_udev_find_by_type function

Summary: This adds the ability to find drm cards using
eeze_udev_find_by_type function.

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Add missing drm subsystem & device type and add a new
Chris Michael [Tue, 23 Sep 2014 13:27:53 +0000 (09:27 -0400)]
eeze: Add missing drm subsystem & device type and add a new
Eeze_Udev_Type for drm

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoEo do: simplify eo_do macro.
Tom Hacohen [Tue, 23 Sep 2014 13:36:18 +0000 (14:36 +0100)]
Eo do: simplify eo_do macro.

This moves the mainloop check inside the function. There was never need
for it to be in client code (i.e a header/macro).
This is better suited inside eo_do_start because this is a macro some
bindings have to re-implement, and we definitely don't want it to be any
more complicated than it has to be.

This breaks ABI and makes elm 1.12 depend on efl 1.11. This is not an issue
as because of eolian and interfaces it's already the case.

9 years agoecore-drm: Fix minor formatting issue
Chris Michael [Tue, 23 Sep 2014 12:29:45 +0000 (08:29 -0400)]
ecore-drm: Fix minor formatting issue

NB: No functional changes

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoEcore_drm: Added initial support for backlight in ecore_drm module
vivek [Tue, 23 Sep 2014 12:28:08 +0000 (08:28 -0400)]
Ecore_drm: Added initial support for backlight in ecore_drm module

Summary:
Added initial support for backlight in drm by adding various structures
and provided api for initializing the backlight struct.

Signed-off-by: vivek <vivek.ellur@samsung.com>
Reviewers: devilhorns, stefan_schmidt

Subscribers: stefan_schmidt, cedric

Differential Revision: https://phab.enlightenment.org/D1448

9 years agoefl - vsync - deal with both broken and non-broken libdrm
Carsten Haitzler (Rasterman) [Tue, 23 Sep 2014 12:31:54 +0000 (21:31 +0900)]
efl - vsync - deal with both broken and non-broken libdrm

@fix

9 years agoevas/engines: avoid redefinition of typedef Outbuf
Jérémy Zurcher [Tue, 23 Sep 2014 09:15:11 +0000 (11:15 +0200)]
evas/engines: avoid redefinition of typedef Outbuf

'typedef struct _Outbuf Outbuf' is in software_generic/Evas_Engine_Software_Generic.h
that is always included

9 years agoRevert "edje: prevent resource leak."
Amitesh Singh [Tue, 23 Sep 2014 09:06:16 +0000 (11:06 +0200)]
Revert "edje: prevent resource leak."

Summary:
This reverts commit f6eb1f17a1b34b2f2fb49feb09bb03a8263be56f.

This will result into failure of edc compilation with recursive use of include.

Reviewers: cedric, raster, seoz

Reviewed By: seoz

Subscribers: chinmaya061, cedric, seoz

Differential Revision: https://phab.enlightenment.org/D1475

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoeolian: silence uninitialized var in goto end branch
Jérémy Zurcher [Tue, 23 Sep 2014 08:56:57 +0000 (10:56 +0200)]
eolian: silence uninitialized var in goto end branch

9 years agoeo: call stack depth is 1024
Jérémy Zurcher [Tue, 23 Sep 2014 08:51:05 +0000 (10:51 +0200)]
eo: call stack depth is 1024

9 years agoeo: call stack can grow/shrink when not using mmap
Jérémy Zurcher [Tue, 23 Sep 2014 08:47:24 +0000 (10:47 +0200)]
eo: call stack can grow/shrink when not using mmap

- if HAVE_MMAP call stack do not shrink and abort() when should grow
- otherwise it's a growing/shrinking stack using realloc

9 years agoeo: remove stack->max_size
Jérémy Zurcher [Tue, 23 Sep 2014 08:43:36 +0000 (10:43 +0200)]
eo: remove stack->max_size

- define EO_CALL_STACK_SIZE instead of stack->max_size
- we are talking about size here not maxsize

9 years agoeo: remove XXX, it's ok now ...
Jérémy Zurcher [Tue, 23 Sep 2014 08:41:13 +0000 (10:41 +0200)]
eo: remove XXX, it's ok now ...

9 years agoeo: unify error msgs
Jérémy Zurcher [Tue, 23 Sep 2014 08:29:44 +0000 (10:29 +0200)]
eo: unify error msgs

9 years agoeo: call stack remove stack->dropcount
Jérémy Zurcher [Tue, 23 Sep 2014 08:26:33 +0000 (10:26 +0200)]
eo: call stack remove stack->dropcount

stack->shrink_frame does the same but more efficiently
change the value of EO_CALL_STACK_SHRINK_DROP if needed

9 years agoeo: fix call stack shrink_frame
Jérémy Zurcher [Tue, 23 Sep 2014 08:24:06 +0000 (10:24 +0200)]
eo: fix call stack shrink_frame

- do never shrink under EO_CALL_STACK_DEPTH_MIN size
- set shrink_frame at (current_size/2) - EO_CALL_STACK_SHRINK_OFFSET

9 years agoeo: remove FIXME: Thread Local Storage
Jérémy Zurcher [Tue, 23 Sep 2014 08:21:26 +0000 (10:21 +0200)]
eo: remove FIXME: Thread Local Storage

done in d39d7050

9 years agoAdd example code of ecore_imf_context_bidi_direction_set in ecore_imf_example
Jihoon Kim [Tue, 23 Sep 2014 01:20:49 +0000 (10:20 +0900)]
Add example code of ecore_imf_context_bidi_direction_set in ecore_imf_example

9 years agoecore_imf: Add ecore_imf_context_bidi_direction_set/get API
Jihoon Kim [Tue, 23 Sep 2014 01:18:35 +0000 (10:18 +0900)]
ecore_imf: Add ecore_imf_context_bidi_direction_set/get API

Some Input Methods want to know the bidi direction (LTR/RTL) at the current cursor position.

9 years agoMerge branch 'devs/devilhorns/ecore_drm'
Chris Michael [Mon, 22 Sep 2014 20:04:20 +0000 (16:04 -0400)]
Merge branch 'devs/devilhorns/ecore_drm'

ecore-drm: Merge rewrite of ecore-drm input/dbus handling to use
Eldbus library.

Previous ecore-drm code relied on dbus itself. This changes the code
(internally) to rely on Eldbus library and thus makes ecore-drm input
handling asynchronous.

@fix

9 years agoecore-drm: Rework internal dbus code to use Eldbus
Chris Michael [Mon, 22 Sep 2014 19:57:12 +0000 (15:57 -0400)]
ecore-drm: Rework internal dbus code to use Eldbus

Summary: This changes all of our internal dbus code for ecore-drm to
work with async Eldbus calls. This adds a potential delay to startup
in that input devices may not be immediately available for use, but is
unavoidable due to Eldbus not having a dbus function call to block for
a reply.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Rework input code to function with Eldbus
Chris Michael [Mon, 22 Sep 2014 19:54:37 +0000 (15:54 -0400)]
ecore-drm: Rework input code to function with Eldbus

Summary: Eldbus runs async so we needed to rework the internal input
code to function with it. This commit changes the iternal code to
issue a device open, and then wait for Eldbus to return us a callback
from the proxy that the device open is finished. This Does add a delay
during startup (which means devices will not always be immediately
ready for use) but thus is the nature of Eldbus.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix some error/warn message formatting
Chris Michael [Mon, 22 Sep 2014 19:44:50 +0000 (15:44 -0400)]
ecore-drm: Fix some error/warn message formatting

NB: Nothing functional here, just alignment of error messages in output.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix private function prototypes for new dbus functions
Chris Michael [Mon, 22 Sep 2014 19:43:37 +0000 (15:43 -0400)]
ecore-drm: Fix private function prototypes for new dbus functions

Summary: In porting Ecore_Drm to use Eldbus, I had to change some
internal functions to work with the async nature of Eldbus. As a
result, I had to modify some private local functions.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Include config.h in private header
Chris Michael [Thu, 18 Sep 2014 15:39:29 +0000 (11:39 -0400)]
ecore-drm: Include config.h in private header

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Add Eldbus header to private header file
Chris Michael [Thu, 18 Sep 2014 15:35:20 +0000 (11:35 -0400)]
ecore-drm: Add Eldbus header to private header file

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Add Eldbus as a dependency for ecore-drm
Chris Michael [Thu, 18 Sep 2014 15:32:37 +0000 (11:32 -0400)]
ecore-drm: Add Eldbus as a dependency for ecore-drm

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Remove direct dbus dependency from Ecore_Drm
Chris Michael [Thu, 18 Sep 2014 15:31:37 +0000 (11:31 -0400)]
ecore-drm: Remove direct dbus dependency from Ecore_Drm

Sumary: Ecore_Drm will be using Eldbus library now, so remove dbus-1
dependency

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoefl: Move Eldbus above Ecore_Drm
Chris Michael [Thu, 18 Sep 2014 15:30:56 +0000 (11:30 -0400)]
efl: Move Eldbus above Ecore_Drm

Sumary: Ecore_Drm will use Eldbus library for dbus functionality, so
we need to check for Eldbus first before Ecore_Drm

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore_x: No need to free NULL value.
Amitesh Singh [Mon, 22 Sep 2014 17:42:36 +0000 (13:42 -0400)]
ecore_x: No need to free NULL value.

NB: This reintroduces the Coverity Resouce Leak detected with CID1039276

Summary:
When ecore_x_window_prop_property_get returns 0, data is already NULL hence
no need to free NULL value. This fault was introduced in 45a2296218b0f297
while fixing coverity defects.

Reviewers: seoz, cedric, raster, devilhorns

Subscribers: cedric, seoz

Differential Revision: https://phab.enlightenment.org/D1473

9 years agoAdd missing cmake config for efl libs.
Tom Hacohen [Mon, 22 Sep 2014 15:08:21 +0000 (16:08 +0100)]
Add missing cmake config for efl libs.

9 years agoepp: fix memory leak.
Amitesh Singh [Mon, 22 Sep 2014 12:52:10 +0000 (14:52 +0200)]
epp: fix memory leak.

Reviewers: cedric, seoz

Subscribers: cedric, seoz

Differential Revision: https://phab.enlightenment.org/D1470

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoedje: Edje_Edit - get the buffer with sound's data
Igor Gala [Mon, 22 Sep 2014 12:24:32 +0000 (14:24 +0200)]
edje: Edje_Edit - get the buffer with sound's data

Summary:
add opportunity get the buffer with sound data for certain sound
from the given edje object. It is needed to play some sound with logic API.
@feature

Reviewers: Hermet, raster, seoz, reutskiy.v.v, NikaWhite, cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1422

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoevas: Gl_Cocoa - fix early segfault caused by unloaded symbols
Jean Guyomarc'h [Mon, 22 Sep 2014 12:20:31 +0000 (14:20 +0200)]
evas: Gl_Cocoa - fix early segfault caused by unloaded symbols

Summary:
evas_gl_symbols() (introduced in commit 9a9d78d) was not dlsym()ed
which led to unloaded symbols, so NULL function pointers which were happily
dereferenced, leading to an early segfault, and therefore to a broken engine.

@fix

Reviewers: cedric, raoulh

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1440

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoeina: Remove TODO, add code to delete the modules array
kabeer khan [Mon, 22 Sep 2014 12:17:26 +0000 (14:17 +0200)]
eina: Remove TODO, add code to delete the modules array

Summary:
As mentioned in TODO in eina_bench.c I added the code to loop through _modules array and delete each item.:

Signed-off-by: kabeer khan <kabeer.khan@samsung.com>
Reviewers: cedric, raster

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1446

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>