platform/upstream/efl.git
6 years agoexamples: calendar_cxx: comment out unused variables
Stefan Schmidt [Mon, 15 Jan 2018 09:40:39 +0000 (10:40 +0100)]
examples: calendar_cxx: comment out unused variables

I normally would simply remove them, but they belong to some more
commented out code below. We should either remove both or get this
fixed.

6 years agoexamples: cxx: avoid tick in warning message
Stefan Schmidt [Mon, 15 Jan 2018 09:38:36 +0000 (10:38 +0100)]
examples: cxx: avoid tick in warning message

The compiler is not to happy about having this tick in the warning
message. Switch to the more formal can not and be done with it.

menu_cxx_example_01.cc:3:26: warning: missing terminating ' character

6 years agoecore_file_monitor: replace EINA_LIST_FOREACH to EINA_LIST_FOREACH_SAFE
WooHyun Jung [Mon, 15 Jan 2018 04:54:01 +0000 (13:54 +0900)]
ecore_file_monitor: replace EINA_LIST_FOREACH to EINA_LIST_FOREACH_SAFE

If ecore_file_monitor_del is called inside the file monitor callback function,
eina_list found from monitor_hash would be freed. (You can check this inside
eina_hash_list_remove.)
Then, EINA_LIST_FOREACH makes one more for loop with invalid eina_list pointer.

EINA_LIST_FOREACH_SAFE can prevent from this problem.

6 years agoEcore_Win32: move monitor window initialization to earlier in init func
Vincent Torri [Sun, 14 Jan 2018 13:33:10 +0000 (22:33 +0900)]
Ecore_Win32: move monitor window initialization to earlier in init func

Summary: This monitor window is just used to receive events when mutiple
monitors are available. it should not be managed by the ecore loop
(creation and destruction events), so initting it earlier means
ecore_win32 attaches less memory/overhead to it as it's just being
used for notifications for devices.

Test Plan: DrMemory to check used memory

Reviewers: cedric

Subscribers: jpeg

Differential Revision: https://phab.enlightenment.org/D5736

6 years agoPyolian: add tests for recently added funcs
Dave Andreoli [Sun, 14 Jan 2018 11:05:33 +0000 (12:05 +0100)]
Pyolian: add tests for recently added funcs

6 years agoCorrect module name for the interface context
Andy Williams [Sun, 14 Jan 2018 00:21:35 +0000 (00:21 +0000)]
Correct module name for the interface context

6 years agodocs: Show when classes really are top level EFL namespace
Andy Williams [Sun, 14 Jan 2018 00:05:59 +0000 (00:05 +0000)]
docs: Show when classes really are top level EFL namespace

6 years agoecore_audio wasapi - fix ecore win32 handler deletion and destruction
Carsten Haitzler (Rasterman) [Sat, 13 Jan 2018 05:44:07 +0000 (14:44 +0900)]
ecore_audio wasapi - fix ecore win32 handler deletion and destruction

vtorri pointed this out in T6607. this should fix this.

fixed T6607

@fix

6 years agoEvil: remove useless dirent code
Vincent Torri [Fri, 12 Jan 2018 21:33:47 +0000 (13:33 -0800)]
Evil: remove useless dirent code

Test Plan: compilation of the EFL

Reviewers: cedric

Subscribers: jpeg

Differential Revision: https://phab.enlightenment.org/D5735

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
6 years agoelm_tooltip: immediately show tooltip if callback is set for obj under pointer
Mike Blumenkrantz [Thu, 11 Jan 2018 21:28:16 +0000 (16:28 -0500)]
elm_tooltip: immediately show tooltip if callback is set for obj under pointer

@fix

6 years agoelm_tooltip: force immediate hint recalc during tooltip creation if 0x0 size
Mike Blumenkrantz [Thu, 11 Jan 2018 21:27:22 +0000 (16:27 -0500)]
elm_tooltip: force immediate hint recalc during tooltip creation if 0x0 size

try to avoid initial 0x0 sized tooltips

@fix

6 years agoeina: document quadtree
Bryce Harrington [Fri, 12 Jan 2018 19:24:59 +0000 (11:24 -0800)]
eina: document quadtree

Reviewers: cedric, ajwillia.ms

Subscribers: segfaultxavi, jpeg

Differential Revision: https://phab.enlightenment.org/D5522

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
6 years agoefl: remove inclusion of dirent.h where it is not used
Vincent Torri [Fri, 12 Jan 2018 19:06:08 +0000 (11:06 -0800)]
efl: remove inclusion of dirent.h where it is not used

Test Plan: compilation

Reviewers: cedric

Subscribers: jpeg

Differential Revision: https://phab.enlightenment.org/D5733

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
6 years agoefl: make efl build again
Daniel Kolesa [Fri, 12 Jan 2018 18:28:35 +0000 (19:28 +0100)]
efl: make efl build again

After a function pointer validation branch got enabled, it turned
out that people have been writing obviously incorrect eo files
all along.

So while I have no idea if this is logically fully correct, at
least EFL builds again now...

cc @thiepha

6 years agoeolian: correctly validate function pointers
Daniel Kolesa [Fri, 12 Jan 2018 18:13:55 +0000 (19:13 +0100)]
eolian: correctly validate function pointers

6 years agoeo: fix tests build.
Cedric BAIL [Fri, 12 Jan 2018 17:56:34 +0000 (09:56 -0800)]
eo: fix tests build.

6 years agoRevert "promise: Add even simpler helper for main loop promise creation"
Cedric BAIL [Thu, 11 Jan 2018 19:21:55 +0000 (11:21 -0800)]
Revert "promise: Add even simpler helper for main loop promise creation"

This reverts commit e931fd698d26b8bec0e34239d2f79c059b339a51.

6 years agoRevert "efl-loop: Don't use 'main' as a variable name"
Cedric BAIL [Thu, 11 Jan 2018 19:21:00 +0000 (11:21 -0800)]
Revert "efl-loop: Don't use 'main' as a variable name"

This reverts commit 214dbdbd590a8f3fd9a0f3d5ef975d6ff9ce0e43.

6 years agoeolian: aliased_base_get funcs don't need unit
Daniel Kolesa [Fri, 12 Jan 2018 17:02:32 +0000 (18:02 +0100)]
eolian: aliased_base_get funcs don't need unit

6 years agoeolian: type_typedecl_get doesn't need a unit
Daniel Kolesa [Fri, 12 Jan 2018 16:52:44 +0000 (17:52 +0100)]
eolian: type_typedecl_get doesn't need a unit

6 years agoeolian: do not require unit for type_class_get
Daniel Kolesa [Fri, 12 Jan 2018 16:33:28 +0000 (17:33 +0100)]
eolian: do not require unit for type_class_get

6 years agoeolian: store typedecl/class in type
Daniel Kolesa [Fri, 12 Jan 2018 16:25:23 +0000 (17:25 +0100)]
eolian: store typedecl/class in type

6 years agoelementary: Fix constant qualifer warning message
Chris Michael [Fri, 12 Jan 2018 15:50:26 +0000 (10:50 -0500)]
elementary: Fix constant qualifer warning message

Gcc complains here due to _wl_default_seat_id_get not accepting a
'const' Evas_Object, so to avoid the warning just case it to a normal
Evas_Object when passing in.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
6 years agopotfiles.in - remove src files no longer in tree
Carsten Haitzler (Rasterman) [Fri, 12 Jan 2018 12:51:09 +0000 (21:51 +0900)]
potfiles.in - remove src files no longer in tree

6 years agoedje - multisense - only complain once about being inable to play audio
Carsten Haitzler (Rasterman) [Fri, 12 Jan 2018 09:47:21 +0000 (18:47 +0900)]
edje - multisense - only complain once about being inable to play audio

no p[oint creating streams of noise, so only once.

6 years agoecore_audio - only complain once about not loading pulse/sndfile
Carsten Haitzler (Rasterman) [Fri, 12 Jan 2018 09:46:04 +0000 (18:46 +0900)]
ecore_audio - only complain once about not loading pulse/sndfile

otherwise we get a complaint for everty time some audio needs/wants to
play and that's just noisy and ugly, so only do it once - the first
time sndfile/pulse are being loaded and it fails.

6 years agoecore_audio - don't about destruction because of failed sndfile load
Carsten Haitzler (Rasterman) [Fri, 12 Jan 2018 09:45:17 +0000 (18:45 +0900)]
ecore_audio - don't about destruction because of failed sndfile load

this means objetcs are only partly destroyed and that is wrong.

@fix

6 years agoeolian_cxx: support multiple function pointers
Thiep Ha [Fri, 12 Jan 2018 09:45:13 +0000 (18:45 +0900)]
eolian_cxx: support multiple function pointers

6 years agoeio: remove unecessary include.
Cedric Bail [Fri, 12 Jan 2018 05:30:16 +0000 (21:30 -0800)]
eio: remove unecessary include.

6 years agoecore: remove use of efl_main_loop_get from efl_net_socket_ssl_server_example.
Cedric BAIL [Thu, 11 Jan 2018 19:12:13 +0000 (11:12 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_socket_ssl_server_example.

6 years agoecore: remove use of efl_main_loop_get from efl_net_socket_ssl_dialer_example.
Cedric BAIL [Thu, 11 Jan 2018 19:05:30 +0000 (11:05 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_socket_ssl_dialer_example.

6 years agoecore: remove use of efl_main_loop_get from efl_net_session_example.
Cedric BAIL [Thu, 11 Jan 2018 18:19:24 +0000 (10:19 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_session_example.

6 years agommap memory allocation - do not used when under valgrind
Carsten Haitzler (Rasterman) [Thu, 11 Jan 2018 17:59:34 +0000 (02:59 +0900)]
mmap memory allocation - do not used when under valgrind

we can't sensibly use things like massif to track memory if we bypass
itr with mmaping -1 fd anonymous memory... so if built with valgrind
support and running under valgrind, use malloc/calloc and free so
these tools actually do something useful for these bits of memory.

6 years agoevas sw generic - don't dlopen osmesa until first needed.
Carsten Haitzler (Rasterman) [Thu, 11 Jan 2018 17:41:50 +0000 (02:41 +0900)]
evas sw generic - don't dlopen osmesa until first needed.

osmesa needs llvm. llvm apparently just by dlopening or linking to the
lib (libLLVM...) gets you 3.5mb of dirty pages just in this lib. that's
a whole lib entirely dirty pages. odd and horrible. in fact once i
stoppd dlopening OSMesa all the time on engine init (and only when gl
is needed)... the amount of dirty pages went from 17208 to 8860.
that's a whopping drop of 8mb! 8mb saved! in fact just dlopening
osmesa and doing the other gl init stuff led to more anonymuse
mappings with dirty pages. 2 of them (2072k and 2076k) which baffled
me as that didn't seem like heap or efl's own data. these disappeared
along with libLLVM-5.0.so (3520k + 60k dirty pages). we stopped
linking/loading libedit (12k dirty), libglapi (20k dirty),
libLLVM-5.0 (3580k dirty), libncursesw (72k dirty),
libOSMesa.so (260k dirty), libtinfo (20k dirty). ... or at least
stopped until absolutely needed. total 17208k of dirty pages went down
to 8860.

my test case was just launching terminology (and doing nothing with it).

@fix memory bloating

6 years agoelementary: Remove use of implicit declaration
Chris Michael [Thu, 11 Jan 2018 17:37:43 +0000 (12:37 -0500)]
elementary: Remove use of implicit declaration

evas_canvas_default_device_get used here leads to an 'implicit
declaration of function warning'. Use evas_default_device_get instead
to remove warning.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
6 years agoecore signal - since we arent going to use threads due to osx, remove
Carsten Haitzler (Rasterman) [Thu, 11 Jan 2018 14:13:13 +0000 (23:13 +0900)]
ecore signal - since we arent going to use threads due to osx, remove

remove thread code since osx is not happy with threads trapping
signals (or at least a thread setting up the handler and trapping
there with signal blocks...). this should now work universally.

6 years agocnp_dnd: make legacy APIs use new interfaces
Thiep Ha [Tue, 9 Jan 2018 09:50:23 +0000 (18:50 +0900)]
cnp_dnd: make legacy APIs use new interfaces

Legacy APIs can uses efl_selection, efl_ui_dnd, efl_ui_dnd_container interfaces
with helper functions.

6 years agodnd: add efl_ui_dnd_container
Thiep Ha [Tue, 9 Jan 2018 07:37:27 +0000 (16:37 +0900)]
dnd: add efl_ui_dnd_container

efl_ui_dnd_container is the interface of dnd for container objects.

6 years agodnd: add efl_ui_dnd
Thiep Ha [Tue, 9 Jan 2018 07:22:49 +0000 (16:22 +0900)]
dnd: add efl_ui_dnd

Efl_Ui_Dnd is the interface for drag and drop of elm_cnp.

6 years agoselection: add efl_selection interface
Thiep Ha [Tue, 9 Jan 2018 06:34:12 +0000 (15:34 +0900)]
selection: add efl_selection interface

Efl_Selection is the object interface for selection api of elm_cnp.
It allows get, set, clear, check selection.

6 years agocnp_dnd: add selection manager
Thiep Ha [Mon, 8 Jan 2018 10:13:10 +0000 (19:13 +0900)]
cnp_dnd: add selection manager

Selection manager implements functions for selection (cnp)
and drag and drop.
All objects can uses it APIs. It also supports multi-seat.

6 years agoefl_ui_focus_manager_calc: store node data before the node is freed
YeongJong Lee [Thu, 11 Jan 2018 08:23:29 +0000 (09:23 +0100)]
efl_ui_focus_manager_calc: store node data before the node is freed

Summary:
we can consider that the node is freed during focus_manager routine.
for example, efl_ui_focus_manager_redirect_set call edje event callbacks,
and a application can delete a object in the edje callback. if the object is
the focusable object of a node, focus_manager make the node freed.
the focus_manager is able to use freed node. (a good example is test_popup.c)

this prevent reusing freed pointers.

Test Plan:
1. elementary_test -to popup
2. popup-center-text + 1 button
3. Click the Close button
4. check that there is no erroe message

Reviewers: bu5hm4n

Reviewed By: bu5hm4n

Subscribers: cedric, woohyun, jpeg, Jaehyun_Cho

Differential Revision: https://phab.enlightenment.org/D5729

6 years agoecore signal write - save and restore errno since may be on mainloop
Carsten Haitzler (Rasterman) [Thu, 11 Jan 2018 08:25:24 +0000 (17:25 +0900)]
ecore signal write - save and restore errno since may be on mainloop

now i disabled the signal thread watcher ... save+restore errno.

6 years agoecore signal - for now disable the sig watcher thread and do on mainloop
Carsten Haitzler (Rasterman) [Thu, 11 Jan 2018 07:17:18 +0000 (16:17 +0900)]
ecore signal - for now disable the sig watcher thread and do on mainloop

we used to do signals on main loop. keep doing. the pipes should work
in cleanly serializing the signals irrespective of when/where they are
caught (because we do into kernel and back out again). hoping this
makes osx work again. can't test as i have no osx box or vm. works on
linux and freebsd though.

6 years agoElm_test panel: use eina instead of dirent to list files
Vincent Torri [Thu, 11 Jan 2018 06:47:19 +0000 (15:47 +0900)]
Elm_test panel: use eina instead of dirent to list files

Summary:
Short/middle term: use UTF-16 on Windows.

So I plan to remove most of external API (like dirent in Evil) and use only EFL to have less work later

Test Plan: compile and run elm_test

Reviewers: jpeg

Subscribers: raster, cedric

Differential Revision: https://phab.enlightenment.org/D5731

6 years agoecore: remove use of efl_main_loop_get from ecore_promise2_example.
Cedric Bail [Thu, 11 Jan 2018 02:15:47 +0000 (18:15 -0800)]
ecore: remove use of efl_main_loop_get from ecore_promise2_example.

6 years agoeina: add eina_promise_continue_new to create a promise using an existing dead future.
Cedric Bail [Thu, 11 Jan 2018 02:14:57 +0000 (18:14 -0800)]
eina: add eina_promise_continue_new to create a promise using an existing dead future.

6 years agoecore: no write after del for efl_loop_timeout.
Cedric Bail [Thu, 11 Jan 2018 02:14:25 +0000 (18:14 -0800)]
ecore: no write after del for efl_loop_timeout.

6 years agoecore: remove use of efl_main_loop_get from efl_net_server_simple_example.
Cedric Bail [Wed, 10 Jan 2018 19:23:52 +0000 (11:23 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_server_simple_example.

6 years agoecore: remove use of efl_main_loop_get from efl_net_server_example.
Cedric Bail [Wed, 10 Jan 2018 18:53:19 +0000 (10:53 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_server_example.

6 years agoecore: remove use of efl_main_loop_get from efl_net_dialer_windows_example.
Cedric Bail [Wed, 10 Jan 2018 18:46:35 +0000 (10:46 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_dialer_windows_example.

6 years agoecore_drm2: simplify device selection code
Derek Foreman [Wed, 10 Jan 2018 18:54:42 +0000 (12:54 -0600)]
ecore_drm2: simplify device selection code

The platform check was added for systems (like ARM) that don't generally
have PCI graphics devices.  However, now we pick a fallback device that
doesn't have a PCI constraint, so the platform check should no longer be
necessary.

6 years agoecore_drm2: Allow non-pci DRM devices to be selected
Derek Foreman [Wed, 10 Jan 2018 18:23:34 +0000 (12:23 -0600)]
ecore_drm2: Allow non-pci DRM devices to be selected

Summary: This is a tweak to c264ef264f0936632a1603fef45c59a50369631d for D5712 . chosen_dev in the loop was only being set for DRM devices attached to PCI devices. While this is useful for determining if the device is the preferred boot_vga device, There is no apparent requirement (comparing to Weston) for all DRM devices to be attached to a PCI device. (This is considering USB DisplayLink devices. I am not sure how the parent device tree is with these...)

Reviewers: devilhorns, ManMower

Reviewed By: devilhorns, ManMower

Subscribers: cedric, jpeg, #efl

Tags: #efl

Differential Revision: https://phab.enlightenment.org/D5727

6 years agoci: keep dockerfiles in sync with repo used for building on docker hub
Stefan Schmidt [Tue, 9 Jan 2018 21:14:50 +0000 (22:14 +0100)]
ci: keep dockerfiles in sync with repo used for building on docker hub

We have an extra repo for these files on our github project to allow
docker hub building these images automatically for us to be ready when
CI kicks in. We still want to have them here as well to aid developer
using them for testing.

6 years agotravis: add misc and wayland build options to build matrix
Stefan Schmidt [Fri, 5 Jan 2018 15:24:43 +0000 (16:24 +0100)]
travis: add misc and wayland build options to build matrix

6 years agotravis: drop older osx build, latest should be enough
Stefan Schmidt [Fri, 5 Jan 2018 15:27:11 +0000 (16:27 +0100)]
travis: drop older osx build, latest should be enough

6 years agotravis: test docker build with prebuild image for speedup
Stefan Schmidt [Tue, 2 Jan 2018 15:47:12 +0000 (16:47 +0100)]
travis: test docker build with prebuild image for speedup

6 years agoci: install wayland and other build options dependencies in Fedora image
Stefan Schmidt [Fri, 5 Jan 2018 14:52:38 +0000 (15:52 +0100)]
ci: install wayland and other build options dependencies in Fedora image

We will use this as base image for different build option testing.

6 years agoci: enable examples build in for OSX target
Stefan Schmidt [Fri, 5 Jan 2018 14:52:12 +0000 (15:52 +0100)]
ci: enable examples build in for OSX target

6 years agoeo: Add "destruct" event
Jean-Philippe Andre [Wed, 10 Jan 2018 06:23:29 +0000 (15:23 +0900)]
eo: Add "destruct" event

Triggered after (almost) complete destruction of the object.

Not called "deleted" because the other event is already "del".
I don't like "destruct" much but this follows the terminology of
"constructor" / "destructor".

@feature

6 years agoelm_test: Fix invalid define for widget style api
Jean-Philippe Andre [Wed, 10 Jan 2018 05:15:58 +0000 (14:15 +0900)]
elm_test: Fix invalid define for widget style api

6 years agoecore: remove use of efl_main_loop_get from efl_net_dialer_websocket_example.
Cedric BAIL [Wed, 10 Jan 2018 00:28:59 +0000 (16:28 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_dialer_websocket_example.

6 years agoecore: remove use of efl_main_loop_get from efl_net_dialer_websocket_autobahntestee.
Cedric BAIL [Wed, 10 Jan 2018 00:20:44 +0000 (16:20 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_dialer_websocket_autobahntestee.

6 years agoecore: remove use of efl_main_loop_get from efl_net_dialer_unix_example.
Cedric BAIL [Wed, 10 Jan 2018 00:07:41 +0000 (16:07 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_dialer_unix_example.

6 years agoecore: remove use of efl_main_loop_get from efl_net_dialer_udp_example.
Cedric BAIL [Tue, 9 Jan 2018 23:51:44 +0000 (15:51 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_dialer_udp_example.

6 years agoecore: remove use of efl_main_loop_get from efl_net_dialer_simple_example.
Cedric BAIL [Tue, 9 Jan 2018 23:43:09 +0000 (15:43 -0800)]
ecore: remove use of efl_main_loop_get from efl_net_dialer_simple_example.

6 years agoecore: forgot to switch to use Efl_Net.h for efl_net_dialer_http_example.
Cedric BAIL [Tue, 9 Jan 2018 23:36:00 +0000 (15:36 -0800)]
ecore: forgot to switch to use Efl_Net.h for efl_net_dialer_http_example.

6 years agoecore-drm2: return supported rotations if not using hardware
Chris Michael [Tue, 9 Jan 2018 17:06:53 +0000 (12:06 -0500)]
ecore-drm2: return supported rotations if not using hardware

If we are Not using Atomic/Hardware support for output rotations, we
should return all available rotations as these will still work in
software mode.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
6 years agoecore-drm2: Fix enabling outputs
Chris Michael [Tue, 9 Jan 2018 17:43:15 +0000 (12:43 -0500)]
ecore-drm2: Fix enabling outputs

Commit 9d583b3fdbed686642dbc0d96d1e417485ca0077 broke
ecore_drm2_output_enabled_set function due changing order of setting
output->enabled value. This patch fixes both issues by checking the
'enabled' variable being passed in.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
6 years agoci: actually install systemd in the Ubuntu 17.10 Dockerfile
Stefan Schmidt [Fri, 5 Jan 2018 13:10:46 +0000 (14:10 +0100)]
ci: actually install systemd in the Ubuntu 17.10 Dockerfile

This allows us to build efl with systemd support on it and enables the
newer Ubunutu version by default.

6 years agoelm_access: mapi have to be set NULL after free(m->api) 38/167238/3
JunsuChoi [Tue, 16 Jan 2018 09:47:45 +0000 (18:47 +0900)]
elm_access: mapi have to be set NULL after free(m->api)

Summary: mapi have to be set NULL after free(m->api)

Reviewers: kimcinoo

Reviewed By: kimcinoo

Differential Revision: https://phab.enlightenment.org/D3756

Change-Id: I114fbe9e84aea06bbb739f2a3949fb2068805db4

6 years agoTIZEN_PROFILE_WEARABLE macro is not used 15/167215/4
JunsuChoi [Tue, 16 Jan 2018 07:36:42 +0000 (16:36 +0900)]
TIZEN_PROFILE_WEARABLE macro is not used

Change-Id: I48581bf4764839f82a307331a5365b684dde0aa3

6 years agoatspi: suppress runtime error, build warning 97/167197/5
JunsuChoi [Tue, 16 Jan 2018 06:42:54 +0000 (15:42 +0900)]
atspi: suppress runtime error, build warning

   [runtime error]
   (1) elm_atspi_bridge.c:2787 _bridge_path_from_object() safety check failed: eo == NULL
   (2) elm_atspi_bridge.c:6184 _bridge_object_register() Unable to register class w/o Elm_Interface_Atspi_Accessible!
   (3) lib/eo/eo.c:697 _eo_call_resolve() in elm_widget.eo.c:169: func 'elm_obj_widget_access_highlight_in_theme_get' (288) could not be resolved for class 'Edje_Object'

   - Note: The access_highlight_in_theme (3) should handle non Elm_LAYOUT_CLASS.
     Or the access_highlight_in_theme should be removed. If the screen-reader
     shows highlight, then the access_highlight_in_theme doesn't make sense.

   [build warning]
   (1) warning: passing argument 2 of '_object_get_bus_name_and_path' discards 'const' qualifier from pointer target type
   (2) warning: passing argument 3 of '_object_get_bus_name_and_path' from incompatible pointer type

Change-Id: I5dbeabf1224dbfdd17fd2d1eb9d72bfff120ccab

6 years ago[elementary] Fix warnings. 02/167302/3
Myoungwoon Roy, Kim [Wed, 17 Jan 2018 02:20:22 +0000 (11:20 +0900)]
[elementary] Fix warnings.

Original commit: 5bc8bced58157f29589d55af35ab1d96663bc3b2

Change-Id: I976d2405215e3c8d854d1beb574780a426581bd4

6 years agoFix wrong doxygen expressions 28/167228/2
Myoungwoon Roy, Kim [Tue, 16 Jan 2018 08:25:04 +0000 (17:25 +0900)]
Fix wrong doxygen expressions

This patch fixes typos and wrong doxygen expressions in elm_box API reference doxygen.
@fix @doc

Original commit: e4d9e041b3121be0ca19436cda0b842acff952fb

Change-Id: I12385763b02caf5fcf7b0aa476dc129c18b1393b

6 years agoatspi: Expose API 60/167260/2
Shinwoo Kim [Tue, 16 Jan 2018 11:12:16 +0000 (20:12 +0900)]
atspi: Expose API

┌───────────────────────────────────────────────────────────────────────┐
│ commit 4c5a44eb8849ebe7a4f2a11fd53f93abc56b486c                       │
│ Author: Shinwoo Kim <cinoo.kim@samsung.com>                           │
│ Date:   Mon Apr 17 11:35:13 2017 +0900                                │
│                                                                       │
│    atspi: Expose API for both Mobile and Wearable                     │
│    The exposed API could be used both profile Mobile and Wearable.    │
└───────────────────────────────────────────────────────────────────────┘

┌───────────────────────────────────────────────────────────────────────┐
│ commit 7ea67a6c7c6dc68c78684d784d0d7a65cc138801                       │
│ Author: Shinwoo Kim <cinoo.kim@samsung.com>                           │
│ Date:   Wed Jul 19 16:38:42 2017 +0900                                │
│                                                                       │
│    atspi: Expose API                                                  │
│    The following API is exposed on the Mobile profile.                │
│        - elm_object_part_access_register                              │
│        - elm_object_part_access_object_get                            │
│        - elm_access_object_register                                   │
│        - elm_access_object_unregister                                 │
│        - elm_access_object_get                                        │
│        - elm_atspi_bridge_utils_say                                   │
│        - elm_atspi_accessible_name_cb_set                             │
│        - elm_atspi_accessible_description_cb_set                      │
│        - elm_atspi_accessible_attribute_append                        │
│        - elm_atspi_accessible_reading_info_type_set                   │
│        - elm_atspi_accessible_reading_info_type_get                   │
│        - elm_atspi_accessible_gesture_cb_set                          │
│        - elm_atspi_accessible_can_highlight_set                       │
│        - elm_atspi_accessible_can_highlight_get                       │
│        - elm_atspi_component_highlight_grab                           │
│        - elm_atspi_component_highlight_clear                          │
└───────────────────────────────────────────────────────────────────────┘

Change-Id: I06238a020d9224fdd355e4274088f81f394dfd06

6 years agoatspi: enhance documentation 56/167256/2
Shinwoo Kim [Tue, 16 Jan 2018 10:52:30 +0000 (19:52 +0900)]
atspi: enhance documentation

 ┌────────────────────────────────────────────────┐
 │commit ebc57884d8f6fe908824b5cf350b86be56587678 │
 │Author: Shinwoo Kim <cinoo.kim@samsung.com>     │
 │Date:   Thu Apr 20 10:00:44 2017 +0900          │
 │                                                │
 │ atspi: enhance documentation                   │
 └────────────────────────────────────────────────┘

Change-Id: I02b651427088abbf6f650d7d70cecefebfebd040

6 years agowin/conform: Improve a behavior of indicator and conformant
Wonki Kim [Tue, 16 Jan 2018 13:04:47 +0000 (22:04 +0900)]
win/conform: Improve a behavior of indicator and conformant

this patch improves a behavior of indicator and conformant to meet tizen ux guidelines

@tizen_feature

Change-Id: I65b242d470f3e789edd1c3f73bfe0cecc1371d57
Signed-off-by: Wonki Kim <wonki_.kim@samsung.com>
6 years agoentry: Apply Tizen Copy & Paste feature with CBHM
Youngbok Shin [Tue, 16 Jan 2018 12:54:28 +0000 (21:54 +0900)]
entry: Apply Tizen Copy & Paste feature with CBHM

@tizen_feature

6 years agoelm_cnp: fix the wrong selection object 42/167242/1
Taehyub Kim [Tue, 16 Jan 2018 10:02:42 +0000 (19:02 +0900)]
elm_cnp: fix the wrong selection object

Change-Id: I2ac1ed31f4a5258a7c3d724c7cf8f1607f1ef86f

6 years agoconfig: reinitialize fontconfig for checking newly installed font
Youngbok Shin [Tue, 16 Jan 2018 08:19:49 +0000 (17:19 +0900)]
config: reinitialize fontconfig for checking newly installed font

@tizen_fix

6 years agoelementary config: support lazy mount of user data area
Youngbok Shin [Tue, 16 Jan 2018 08:13:40 +0000 (17:13 +0900)]
elementary config: support lazy mount of user data area

It includes the following patches about supporting lazy mounting.

1.
    elm_config: Add fallback path to support lazy mount

    elm_config would access to user home directory to save their
    config information. However, elm_config couldn't accesss to
    the path because user data area could not be mounted
    (Lazy mount of user data area[below /opt/usr/] is applied)

    elm_config monitors some file which is created after mounting
    then if it doesn't exist, elm_config will access to the other path

2.
    config: update config data for processes which are launched before lazy mounting

@tizen_feature

6 years agoctxpopup: merge eo & header files for different profiles 22/167222/2
JinYong Park [Tue, 16 Jan 2018 08:06:15 +0000 (17:06 +0900)]
ctxpopup: merge eo & header files for different profiles

@tizen_fix

original patch : https://review.tizen.org/gerrit/#/c/110411/

Change-Id: Id73796e4eaf1e54c3a9fc0e6cad64a712c5fd324
Signed-off-by: JinYong Park <j4939.park@samsung.com>
6 years agoFix typo and some wrong expressions 08/167208/2
Myoungwoon Roy, Kim [Tue, 16 Jan 2018 07:17:58 +0000 (16:17 +0900)]
Fix typo and some wrong expressions

This patch fixes typo and some wrong expressions in doc, box, gesture, index, map, and toolbar API reference doxygen.
@fix @doc

Original commit: 1999872b1b69f85c68836622095b98206e8ea125

Change-Id: I2780915e9b4081b6e58ad7b07854bcd766b82857

6 years agoFix a wrong group info. 99/167199/3
Myoungwoon Roy, Kim [Tue, 16 Jan 2018 06:49:38 +0000 (15:49 +0900)]
Fix a wrong group info.

This patch fixes a wong group info and some wrong expressions in elm_win API reference doxygen.
@fix @doc

Original commit: 0d18cf765f4c7a50f813bfa6449fec27dee7108fe8e1bb8e2201584b4a38c953b97a3dff3f40f684

Change-Id: Ie7cb88fb14ec2dd6bf4608a4be0b6ac4332b7df7

6 years agoFix typos and some wrong expressions in Ecore API reference doxygen. 91/167191/2
Myoungwoon Roy, Kim [Tue, 21 Feb 2017 09:35:00 +0000 (18:35 +0900)]
Fix typos and some wrong expressions in Ecore API reference doxygen.

This patch fixes typos, doxygen warnings and some wrong expressions in Ecore and Edje API reference doxygen.
@fix @doc

Original commit: 4172e39135d5fa8dd46cfd67bff3706445a2f46d

Change-Id: I56a13a062fa662f956b4d69563f3e38fd50811e9

6 years agoelm_cnp: skip other process's send/receive event 93/166693/2
Taehyub Kim [Thu, 11 Jan 2018 11:41:58 +0000 (20:41 +0900)]
elm_cnp: skip other process's send/receive event

Change-Id: Ic97eef4058333dd57af7dd270f34f6179fbe8ead

6 years agoelm_cnp: support html type for copy&paste 09/166609/2
Taehyub Kim [Thu, 11 Jan 2018 06:18:54 +0000 (15:18 +0900)]
elm_cnp: support html type for copy&paste

Change-Id: Ia132b95016e9021734355883ae8e6c59f3a44665

6 years agoelm_ctxpopup: add EINA_DEPREACATED for deprecated elm_ctxpopup_direction_available_get 92/166492/2
Taehyub Kim [Wed, 10 Jan 2018 10:36:39 +0000 (19:36 +0900)]
elm_ctxpopup: add EINA_DEPREACATED for deprecated elm_ctxpopup_direction_available_get

Change-Id: I09bb70d80e841b33e3163a4f61741b305caf54a8

6 years agoevas: fix cursor_line_coord_set fail when y is 0.
Youngbok Shin [Tue, 16 Jan 2018 07:30:34 +0000 (16:30 +0900)]
evas: fix cursor_line_coord_set fail when y is 0.

Signed-off-by: Vladyslav Shevchenko <v.shevchenko@samsung.com>
origin: spin (tizen_2.4 branch)

6 years agoevas textblock: Fix RTL + LTR word wrap issue caused by wrong line start pos
Youngbok Shin [Tue, 16 Jan 2018 07:20:18 +0000 (16:20 +0900)]
evas textblock: Fix RTL + LTR word wrap issue caused by wrong line start pos

@tizen_fix

6 years agoedje: Add 4000x4000 size limit with reduced loop count restriction
Youngbok Shin [Tue, 16 Jan 2018 07:11:37 +0000 (16:11 +0900)]
edje: Add 4000x4000 size limit with reduced loop count restriction

The calc loop restriction [255] can cause error from
evas_object_smart_need_recalculate_set() function.
It also has [255] restriction for recalculate cycle.
So, we need to reduce calc loop restriction for
reducing *_need_recalculate_set() function call.

@tizen_fix

6 years agoedje entry: set imf context input hint with MULTILINE if current entry is multiline
Youngbok Shin [Tue, 16 Jan 2018 06:45:59 +0000 (15:45 +0900)]
edje entry: set imf context input hint with MULTILINE if current entry is multiline

@tizen_fix

6 years agoedje entry: set imf context input hint with MULTILINE if current entry is multiline
Youngbok Shin [Tue, 16 Jan 2018 06:30:50 +0000 (15:30 +0900)]
edje entry: set imf context input hint with MULTILINE if current entry is multiline

@tizen_fix

6 years agoedje: force update anchors when a API is called to get list of anchors
Youngbok Shin [Tue, 16 Jan 2018 06:16:28 +0000 (15:16 +0900)]
edje: force update anchors when a API is called to get list of anchors

@tizen_fix

6 years agoatspi: "GetNeighbor" is using deputy object 88/167188/4
JunsuChoi [Tue, 16 Jan 2018 05:50:52 +0000 (14:50 +0900)]
atspi: "GetNeighbor" is using deputy object

   A01 object is deputy object, and exists in embedding process.
   A02, A03 object exist in embedding process.
   B01, B02 objects exist in embedded process.

   The “GetNeighbor” method should return A02 object as “next” object of B02,
   and return B02 object as “prev” object of A02.
   The “next” object of A01 is B01.
   The “prev” object of B01 is A01.
   This is default implementation.
   The following is default order: A01 - B01 - B02 - A02 - A03

   But more important thing is that the embedding process could use
   ELM_ATSPI_RELATION_FLOWS_TO/FROM for each objects A01, A02, and A03.
   So there are some cases should be considered.

   (1) If A01 is ELM_ATSPI_RELATION_FLOWS_FROM object of A03, then the
      “GetNeighbor” should find B02 first when it navigates backward from A03.

   (2) If A03 is ELM_ATSPI_RELATION_FLOWS_TO object of A01, then the
      “GetNeighbor” should find A03 first when it navigates backward from B02.

   (3) Even though the A01 has ELM_ATSPI_RELATION_FLOWS_TO relation information,
      the “GetNeighbor” should find B01 first. Because the
      ELM_ATSPI_RELATION_FLOWS_TO object is used when “GetNeighbor” finds
      next object of B02.

Change-Id: I36dde39af3e99efcd7168c146b065fc6dfa5d104

6 years agoatspi: "GetNavigableAtPoint" sends deputy object 67/167167/3
JunsuChoi [Tue, 16 Jan 2018 04:07:58 +0000 (13:07 +0900)]
atspi: "GetNavigableAtPoint" sends deputy object

   The deputy means embedding side object works for embedded object.
   If there is no object at point on embedded side, then the deputy object should
   have highlight.

   So far the deputy object is elm_access, and the elm_access takes highlight.
   So there is no chance to have highlight on embedded side.

Change-Id: I0e48ad65273ffc0cb39601bd6a83de3469873beb

6 years agoatspi: enhance finding the first,last obj 65/167165/4
JunsuChoi [Tue, 16 Jan 2018 03:31:55 +0000 (12:31 +0900)]
atspi: enhance finding the first,last obj

   There are two kind of first(or last) object.
   (1) The first object in viewport of its scrollable parent.
   (2) The first object in accessibilty tree regardless of scrollable parent.

   Looping navigation should use the first object of type (2)
   Other cases should use the first object of type (1)

   If view is changed, then there is no highlight. In this case the first object
   of type (1) should be used. For example, if view is changed to 1 depth from
   2 depth, then the highlight should use the first object of type (1).
   If the following is 1 depth view after 2 depth view disapears, then there is
   not currently highlighted object. In this case the item 4 should grab highlight
   for next object, and the item 6 should grab highlight for the previous object.

   +--------+
   | item 4 |
   |--------|
   | item 5 |
   |--------|
   | item 6 |
   +--------+

Change-Id: Ie4b4e8599232ef614d9dbe5921220af3578f54c1

6 years agoenhance build message(error:red, warning:ylw) 90/167090/3
JunsuChoi [Mon, 15 Jan 2018 09:32:19 +0000 (18:32 +0900)]
enhance build message(error:red, warning:ylw)

Change-Id: I27d34def125eea0dc4985e987cf93eb30780ffd3

6 years agoedje: Add edje_object_part_text_freeze, thaw APIs for freezing cursor movements
Youngbok Shin [Tue, 16 Jan 2018 05:24:25 +0000 (14:24 +0900)]
edje: Add edje_object_part_text_freeze, thaw APIs for freezing cursor movements

It also apply the internal functions to elm_entry.

@tizen_feature