Jean Guyomarc'h [Sat, 28 May 2016 11:50:13 +0000 (13:50 +0200)]
eina: make eina backtrace portable
Backtrace relied on non-portable code.
Platform-specific code is now enclosed
within macros defined at configuration-time.
Jean Guyomarc'h [Sat, 28 May 2016 11:35:54 +0000 (13:35 +0200)]
autotools: check for sched_getcpu()
sched_getcpu() is glibc-only, so not portable. Mac OS X
does not have it, and has no obvious replacement function.
This commit will allow future code to test for the existence
of this function, to provide fallbacks instead of making
compilation fail.
Jean Guyomarc'h [Sat, 28 May 2016 11:07:22 +0000 (13:07 +0200)]
autotools: improve libunwind detection
Libuwind may not be shipped with a pkg-config file.
It can be distributed on the system, but the autotools
would fail to detect it because it relied only on pkg-config.
We now first check with pkg-config, and then try to compile and
link a program using libuwind to see if it is supported anyway.
This is a first step towards a working eina_log_backtrace on
Mac OS X.
Jean Guyomarc'h [Sat, 28 May 2016 11:07:22 +0000 (13:07 +0200)]
autotools: improve libunwind detection
Libuwind may not be shipped with a pkg-config file.
It can be distributed on the system, but the autotools
would fail to detect it because it relied only on pkg-config.
We now first check with pkg-config, and then try to compile and
link a program using libuwind to see if it is supported anyway.
This is a first step towards a working eina_log_backtrace on
Mac OS X.
Hermet Park [Sat, 28 May 2016 10:08:03 +0000 (19:08 +0900)]
eio: fix build break.
there were missing eo headers.
fix and make it work again.
Jean Guyomarc'h [Sat, 28 May 2016 08:52:09 +0000 (10:52 +0200)]
autotools: fix configuration of Ecore_Cocoa
On Mac OS X, we are using OBJC, not GNU-OBJC.
This test seems unnecessary as well, as a longuage tests
and a linking test are performed later.
Fixes T3710.
Stephen Houston [Sat, 28 May 2016 02:38:06 +0000 (21:38 -0500)]
Elm Image: The introduction of elm_image_scale_type broke elm. Make TYPE_FILL the default.
The default scale type should not be none as this breaks elm icon and elm image as it doesn't follow the default view of elm image before elm image scale was added. The default behavior is fill so this commit makes fill the default.
Dave Andreoli [Fri, 27 May 2016 19:57:04 +0000 (21:57 +0200)]
Elm Win legacy: Fix wrong function names
This fixes errors from 2 commits:
ec464939d9b8e4daabb55fab07e369a7e00cc941
589eae9a8bff9aa160df61656a7faab8bd2b9169
Ji-Youn Park: SPANK!
Chris Michael [Fri, 27 May 2016 15:58:51 +0000 (11:58 -0400)]
Merge branch 'devs/devilhorns/ecore_drm2'
This new Ecore_Drm2 library is going to replace the existing Ecore_Drm.
This will refactor a lot of the code, bring improvements over the existing API,
and provide additional support for missing features such as
atomic modeset, nuclear pageflip, and hardware planes support.
@feature
Chris Michael [Fri, 27 May 2016 13:55:11 +0000 (09:55 -0400)]
ecore-drm2: Add API functions to support cacheing of context & keymap
@feature
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Thu, 26 May 2016 14:46:40 +0000 (10:46 -0400)]
ecore-evas-drm: Fix issue of mouse pointer not centering on startup
This patch fixes an issue where starting Enlightenment would not
center the mouse pointer. Basically the issue is that we cannot warp
the mouse pointer until After the ee->prop.window has been set else
E will not process the mouse_move event
@fix
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Thu, 26 May 2016 14:43:35 +0000 (10:43 -0400)]
ecore-drm2: Add API function to calibrate input devices
This adds a new API function that Ecore_Evas can call to calibrate
input devices given size of the output
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Mike Blumenkrantz [Tue, 24 May 2016 20:18:46 +0000 (16:18 -0400)]
ecore-drm2: update to latest elput
Mike Blumenkrantz [Mon, 23 May 2016 16:53:49 +0000 (12:53 -0400)]
ecore-drm2: perform logind connection after finding device path in ecore_drm2_device_find
Chris Michael [Tue, 24 May 2016 13:32:21 +0000 (09:32 -0400)]
ecore-drm2: Add API function to set left-handed mouse mode
This commit adds an API function used to set a mouse to be left-handed
@feature
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Derek Foreman [Fri, 20 May 2016 16:56:38 +0000 (11:56 -0500)]
evas-drm: Fix occasoinaly freezes
We need to continue to tick even if we can't find a buffer - otherwise
we'll stop updating forever.
We also need to reset buffer busy bits to 0 if we run out of buffers or
we'll never try to assign one. (The page flip handler would eventually
clear the busy bit for the current buffer, but a few lines ago we just
set that to NULL so it will never happen...)
Derek Foreman [Fri, 20 May 2016 16:46:46 +0000 (11:46 -0500)]
evas-drm: Don't re-use the current buffer immediately
If we don't block the current buffer from being assigned to the new
current buffer we'll end up rendering into the same buffer over and over
and never actually double buffer.
Chris Michael [Sat, 14 May 2016 08:13:38 +0000 (04:13 -0400)]
ecore-drm2: Check return of elput_manager_open
We should check to make sure that opening the device has succeeded
before trying to continue
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:45:38 +0000 (09:45 -0400)]
ecore-drm2: Initialize elput input when device gets opened
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:43:53 +0000 (09:43 -0400)]
ecore-drm2: Fix potential invalid access to output current_mode
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:42:42 +0000 (09:42 -0400)]
ecore-drm2: Add API function to get screen size range
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:41:21 +0000 (09:41 -0400)]
ecore-drm2: Add API function to return available crtcs
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:39:46 +0000 (09:39 -0400)]
ecore-drm2: Add API function to get possible crtc of a given output
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:37:50 +0000 (09:37 -0400)]
ecore-drm2: Add API to return current fb used on a given output
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:36:34 +0000 (09:36 -0400)]
ecore-drm2: Add API to return current resolution of a given output
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:34:32 +0000 (09:34 -0400)]
ecore-drm2: Add API function to return connector type of an output
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:33:15 +0000 (09:33 -0400)]
ecore-drm2: Add API function to get cloned & connected properties of an output
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:31:38 +0000 (09:31 -0400)]
ecore-drm2: Add API functions to get name & model of an output
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:29:48 +0000 (09:29 -0400)]
ecore-drm2: Add API function to set the mode of an output
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:28:20 +0000 (09:28 -0400)]
ecore-drm2: Add API function to return information about a given output mode
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:26:52 +0000 (09:26 -0400)]
ecore-drm2: Add API function to return list of output modes for a given output
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:25:36 +0000 (09:25 -0400)]
ecore-drm2: Add API function to return physical size of a given output
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:24:12 +0000 (09:24 -0400)]
ecore-drm2: Add API functions to get/set if an output is enabled
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:21:17 +0000 (09:21 -0400)]
ecore-drm2: Add API functions to get/set if an output is primary
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 18 May 2016 18:33:37 +0000 (14:33 -0400)]
elementary: Port elementary to use Ecore_Drm2
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Tue, 3 May 2016 16:03:00 +0000 (12:03 -0400)]
ecore-evas: Port ecore_evas drm engine to use Ecore_Drm2 library
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Tue, 3 May 2016 16:02:20 +0000 (12:02 -0400)]
evas-engines: Port evas drm engines to use Ecore_Drm2 library
This patch ports the evas drm and gl_drm engines to use the new
Ecore_Drm2 library
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 11 May 2016 13:07:48 +0000 (09:07 -0400)]
ecore-drm2: Add API functions needed to port Ecore_Evas drm
This patch adds 2 new API functions which are required by Ecore_Evas
in order for it to function with drm. These API functions allow for
restricting pointer movement, and for setting the window id which will
be used when sending input events
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Mon, 9 May 2016 16:47:02 +0000 (12:47 -0400)]
ecore-drm2: Add drm2 event for session activation
This patch adds a new Ecore_Drm2_Event_Activate that can be raised to
inform Enlightenment that a session has been activated or suspended so
that rendering can be paused/resumed
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Wed, 4 May 2016 15:03:15 +0000 (11:03 -0400)]
ecore-drm2: Add API functions to get and set pointer position
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Mon, 2 May 2016 15:45:39 +0000 (11:45 -0400)]
ecore-drm2: Add API functions to work with framebuffer objects
This patch adds support for creating, deleting, and manipulating
framebuffer objects via exposed API.
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Mon, 2 May 2016 15:26:44 +0000 (11:26 -0400)]
ecore-drm2: Add API functions to work with outputs
This patch adds support for creating, destroying, and otherwise
working with outputs via exposed API functions.
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Mon, 2 May 2016 14:40:47 +0000 (10:40 -0400)]
ecore-drm2: Add API functions to work with a drm device
This commit adds new API functions to find, open, and close a drm
device, along with API functions to retrieve clock_id and cursor size.
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Mon, 2 May 2016 14:21:00 +0000 (10:21 -0400)]
ecore-drm2: Add start of Ecore_Drm2 library
This new library is going to replace the existing Ecore_Drm. This will
refactor a lot of the code, bring improvements over the existing API,
and provide additional support for missing features.
@feature
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Carsten Haitzler (Rasterman) [Fri, 27 May 2016 14:55:24 +0000 (23:55 +0900)]
efl - ecore c+ ecore-imf - fix odd case input when faking a real event
so ibus module for ecore-imf likes to send an x event back to
ourselves faking a key... this works unless we are looking at
modifiers that make make for odd combos with keysyms turned into
keycodes. so actualyl use the correct original keycode plus state
unless we are having to explicitly send a keysym from ibus core.
this fixes T3703
@fix
Chris Michael [Fri, 27 May 2016 13:17:46 +0000 (09:17 -0400)]
elput: Add API function to support caching of existing keymap
@feature
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Fri, 27 May 2016 13:12:53 +0000 (09:12 -0400)]
elput: Add API function to support caching of an xkb_context
@feature
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Ji-Youn Park [Fri, 27 May 2016 08:13:08 +0000 (16:43 +0830)]
elm_win: clean elm apis related with rotation.
elm_win only support two kind of apis.
1. elm_win_screen_rotation_get
this api is used to detect screen rotation before app doing something before rotation event.
for example, app can query rotation before deciding there layout.
2. elm_win_available_rotation_set/get
like video app, can set only landscape mode if they need.
Ji-Youn Park [Fri, 27 May 2016 05:55:50 +0000 (14:25 +0830)]
elm_win: remove elm_win_wm_manual_rotation_done feature from eo.
this kind of manual rotation_done api don't need to public api now.
under the window deal with rotation job automatically.
Jean-Philippe Andre [Wed, 27 Apr 2016 06:21:49 +0000 (15:21 +0900)]
elm_win: Code simplification
Simplify ELM_WIN_DATA_GET_OR_RETURN
Yeshwanth Reddivari [Fri, 27 May 2016 04:41:12 +0000 (10:11 +0530)]
Separtor: use orientation APIs
Summary: Use orientation APIs instead of horizontal get/set apis.
Test Plan: elementary_test -to separator
Reviewers: jpeg, cedric, Hermet, raster, singh.amitesh
Reviewed By: singh.amitesh
Differential Revision: https://phab.enlightenment.org/D3984
Felipe Magno de Almeida [Thu, 26 May 2016 19:47:20 +0000 (16:47 -0300)]
eolian-cxx: Added tests for wrapper conversion in C++ API
Felipe Magno de Almeida [Thu, 26 May 2016 19:47:04 +0000 (16:47 -0300)]
eolian-cxx: Fixed lookup of class types
Felipe Magno de Almeida [Thu, 26 May 2016 19:46:40 +0000 (16:46 -0300)]
eolian-cxx: Removed useless commented code
Felipe Magno de Almeida [Thu, 26 May 2016 19:46:23 +0000 (16:46 -0300)]
eina-cxx: Added crange_array template typedef
Felipe Magno de Almeida [Tue, 24 May 2016 20:36:26 +0000 (17:36 -0300)]
eina: Add benchmarks for eina promises
Chris Michael [Thu, 26 May 2016 16:36:51 +0000 (12:36 -0400)]
elput: Add APIs to support keyboard key remapping
This patch adds 2 new API functions which can enable keyboard key
remapping, and set which keys are to be remapped.
@feature
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Felipe Magno de Almeida [Thu, 26 May 2016 16:34:45 +0000 (13:34 -0300)]
ecore: eldbus: Fix prototype for promise then_cb new prototype
Added Eina_Promise parameter and removed indirection for Eina_Error.
Felipe Magno de Almeida [Thu, 26 May 2016 16:34:04 +0000 (13:34 -0300)]
eina: Make cancelled promises use EINA_ERROR_PROMISE_CANCELLED error code
Felipe Magno de Almeida [Thu, 26 May 2016 16:05:32 +0000 (13:05 -0300)]
eina: Use pointer directly instead of by unnecessary indirection
Felipe Magno de Almeida [Thu, 26 May 2016 16:04:20 +0000 (13:04 -0300)]
eina: Fix crashing when accessing past-the-end of the malloc'ed promise
The promise race composition was calculating its size wrongly, making
the promise's size smaller than it should, causing heap corruption.
Felipe Magno de Almeida [Thu, 26 May 2016 16:03:05 +0000 (13:03 -0300)]
eina: Fix then callbacks prototypes
Eina_Error is not passed by pointer anymore, which could cause invalid
pointer access in promise compositions (all and race).
Also added Eina_Promise* to prototypes.
Felipe Magno de Almeida [Thu, 26 May 2016 16:01:36 +0000 (13:01 -0300)]
eina: Add missing buffer_get function in Eina_Promise
Function buffer_get was available only in Eina_Promise_Owner, despite
being defined in both classes.
Felipe Magno de Almeida [Thu, 26 May 2016 15:59:50 +0000 (12:59 -0300)]
eina: Fix new prototype for then cb in test
Added Eina_Promise* parameter to function prototypes
Chris Michael [Thu, 26 May 2016 14:45:47 +0000 (10:45 -0400)]
elput: Add API function to calibrate input devices
This adds an API function that can be called to calibrate input
devices based on a given output size
@feature
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Thu, 26 May 2016 15:37:18 +0000 (11:37 -0400)]
elput: Remove badly named API
This patch removes an API function that basically "tried to do too
much". This was supposed to be used for centering mouse pointer and
calibrating an input device based on output size. Remove this
silliness. It would be better to have one API for calibrating, and a
separate API for pointer centering. These new APIs will come shortly
in pending patches.
@fix
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Chris Michael [Thu, 26 May 2016 15:35:42 +0000 (11:35 -0400)]
elput: Fix issue where double/triple click was not working
As we use the mouse threshold for determining double/triple clicking,
it would probably be a good idea if that threshold had a value
@fix
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Daniel Kolesa [Thu, 26 May 2016 13:36:41 +0000 (14:36 +0100)]
eolian: add core infra to support builtin strings and void pointers
Daniel Kolesa [Thu, 26 May 2016 13:17:07 +0000 (14:17 +0100)]
eolian: put ownable checks into its own func
Chris Michael [Thu, 26 May 2016 12:42:38 +0000 (08:42 -0400)]
elput: Minor formatting fixes
NB: No functional changes, just cleanup of formatting
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Jean-Philippe Andre [Thu, 26 May 2016 11:53:53 +0000 (20:53 +0900)]
Evas: Fix compilation warning
Missing initializers. Damn C and {0}.
I could also have used GCC's {} form.
Jean-Philippe Andre [Thu, 26 May 2016 11:08:04 +0000 (20:08 +0900)]
Evas GL: Fix compilation warning, use func pointer
GLES 3 support should be a runtime-only requirement, so
no direct call to a GLES3 function should appear in evas.
Use our dlsym() pointer instead.
Jean-Philippe Andre [Thu, 26 May 2016 05:06:52 +0000 (14:06 +0900)]
Elm Layout: Make box & table part proxies temporary
efl_part returns short lived objects, let's make them
die after each function call.
Jean-Philippe Andre [Thu, 26 May 2016 02:54:30 +0000 (11:54 +0900)]
Elm.Layout: Fix BOX and TABLE usage with efl_part
Thanks Dave for the report. This was work in progress...
Jean-Philippe Andre [Tue, 24 May 2016 11:40:18 +0000 (20:40 +0900)]
Efl: Remove part name in Efl.Container (use Efl.Part)
This touches Edje and also Elementary where part names are used.
This commit blew up in size since now all content part APIs
(get, set, unset) require to use Efl.Part instead.
This is a big refactoring commit, but no logic should
have been changed. Fingers crossed.
Jean-Philippe Andre [Thu, 26 May 2016 09:30:35 +0000 (18:30 +0900)]
Evas GL common: Simplify GLPIPES... trying to prevent a crash
An unknown crash happened in glDrawArrays() where color_ptr is
a non-reachable pointer that looks like a valid malloc() zone.
Unreproducible issue. I can only hope this patch doesn't actually
break anything.
This cleans up rectangle and line in particular, that were not
using common code to use GLPIPES. I honestly couldn't find a
problem though.
Jiwon Kim [Thu, 26 May 2016 08:19:34 +0000 (17:19 +0900)]
Elm entry: fix returning wrong focus region
Summary:
In on_focus_region function,
entry returns relative x,y position from edje object.
It should be calculated from elm object's position.
@fix
Test Plan:
1. $elementary_test "Entry on Page Scroll"
2. click under button
3. click upper text in entry
4. page should not be scrolled
Reviewers: tasn, cedric, woohyun, Hermet, herdsman, raster
Subscribers: Blackmole, id213sin, cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D3923
Ji-Youn Park [Thu, 26 May 2016 08:18:02 +0000 (16:48 +0830)]
elm_win: remove elm_win_conformant_set/get from eo
conformant set api give information to enlighten ment, this window
need to know the change of conformant size.
but in the future, window have to section background and forground.
if user set layout as content of background part, the layout size is
not changed by clipborad or keybord.
if user set layout as content of forground part of window, the layout size
is changed by clipboard or keyboard state.
and the window notify to the enlightenment it has interest soft input's state
automatically, without users api call
Ji-Youn Park [Thu, 26 May 2016 08:09:32 +0000 (16:39 +0830)]
elm_image: fix build break
Jaeun Choi [Thu, 12 May 2016 04:25:13 +0000 (13:25 +0900)]
Elm_image: implement 'scale_type' and 'scalable' property related functions
This patch implements the functions related to new eo properties
added in
a5a44b0104c7302b9e3ca2925d28b3212206c714
Ji-Youn Park [Thu, 26 May 2016 07:02:58 +0000 (15:32 +0830)]
elm_win: remove elm_win_keyboard_win_set/get from eo.
this api is not used by wayland side.
and this kind of API can be implemented name set or class set api.
Carsten Haitzler (Rasterman) [Thu, 26 May 2016 04:03:01 +0000 (13:03 +0900)]
eina test suite - ooops reenable tests
accidentally snukc in commented disables in test suite as i was
hunting why eina test suite crashes while dealing with feedback on my
compare fixes etc.
Carsten Haitzler (Rasterman) [Thu, 26 May 2016 03:49:58 +0000 (12:49 +0900)]
efl - new cmp "fix" code - remove one duplicated cmp
Vincent Torri [Thu, 26 May 2016 03:27:31 +0000 (12:27 +0900)]
Fix addition of Evil library in .pc files
Summary: lib/evil/libevil.la was added to .pc files, which broke the use of the EFL pc files
Test Plan: compilation
Reviewers: cedric, jpeg, zmike, raster
Reviewed By: raster
Differential Revision: https://phab.enlightenment.org/D3983
Carsten Haitzler (Rasterman) [Thu, 26 May 2016 03:17:07 +0000 (12:17 +0900)]
efl - fix many bounds over/underflow where we use int for ptr cmp
this addresses more things brought up in comments in
fixes T3638 commentd
@fix
Lauro Moura [Tue, 12 Apr 2016 19:00:23 +0000 (16:00 -0300)]
eio: Adds Eo-based Eio API
The legacy Eio_File factory functions are replaced by an Eo object
called Eo_Job that return promises wrapping the async file operations.
With this commit, the legacy Eio callbacks are replaced by the following
Eo/Promises counterparts :
* Done_Cb -> Promise then success callback
* Error_Cb -> Promise then error callback
* Main_Cb -> Promise progress callback
* Filter_Cb -> Job object event (more below)
Events are used to deliver and get the filter data. To differentiate
between the named and direct versions, they come in "filter,direct" and
"filter,name" versions.
Monitors were wrapped inside a new class Eo_Sentry.
The user creates a sentry object and adds monitoring targets to it,
listening to events on it.
The sentry event info is composed of two strings. The source string
is the path being monitored, i.e. the one passed to eio_sentry_add, and
the trigger string is the path that actually triggered the event, e.g.
a new file created in a monitored directory.
Felipe Magno de Almeida [Wed, 25 May 2016 21:14:29 +0000 (18:14 -0300)]
eina: Add promise parameter to then calllbacks
Added promise parameter to then callbacks so callbacks can steal
ownership of the value from the promise.
Lauro Moura [Mon, 25 Apr 2016 21:02:05 +0000 (18:02 -0300)]
eio: Call correct function to cleanup Eio_File.
The Eio functions operating on Eina_Files were just freeing
the Eio_File pointer on completion instead of calling eio_file_free
to unregister the thread.
Carsten Haitzler (Rasterman) [Thu, 26 May 2016 00:28:23 +0000 (09:28 +0900)]
eina hash - fix stringshare key comparison function
this fixes T3638
@fix
a note... thanks so much to aerodynamik for spotting this. i'm rather
surprised coverity didn't spot this... unless someone said to "shut up
coverity you're wrong" and they should not have.
i also might have expected compilers to spot this too... and add a
warning.
anyway ... this was a seriously subtle bug that could have caused all
kinds of havoc in efl. keys that are different may be compared to be
the same. it could get ordering wrong and sorting thus maybe insert
keys that cannot be found anymore and oh so much more besides.
Chris Michael [Wed, 25 May 2016 19:03:15 +0000 (15:03 -0400)]
elput: Fix formatting
NB: No functional changes, just making this readable
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Mike Blumenkrantz [Wed, 25 May 2016 16:54:25 +0000 (12:54 -0400)]
elput: use physical name of seat when finding existing seats
"seat0" is used as the default internal name, and this is a physical name.
using the logical name causes mismatches and allows devices to function
in unintended ways
Mike Blumenkrantz [Wed, 25 May 2016 16:49:44 +0000 (12:49 -0400)]
elput: further simplify device internals
'window' and pointer max w/h attributes are set for a manager object,
so let the attributes stay there. this also resolves issues where devices
created at a later point would fail to be initialized with these attributes
Mike Blumenkrantz [Tue, 24 May 2016 20:14:41 +0000 (16:14 -0400)]
elput: change some functions and internal flags to enable async input setup
this sets attributes on the Elput_Manager struct so that devices created at
a later point can then have relevant attributes applied to them
Mike Blumenkrantz [Tue, 24 May 2016 20:11:20 +0000 (16:11 -0400)]
elput: define and implement an async device opening interface for libinput
this adds an overly-complex method of removing blocking dbus calls from libinput's
synchronous device initialization architecture. libinput was clearly never meant
to be used in this way, but we're doing it anyway because we're efl.
#SamsungFeatures
Mike Blumenkrantz [Tue, 24 May 2016 20:18:26 +0000 (16:18 -0400)]
elput: remove 'sync' param from device management
unnecessarily complicates code
Carsten Haitzler (Rasterman) [Wed, 25 May 2016 08:34:24 +0000 (17:34 +0900)]
Revert "eo - now ref and unref objects on each eo call to keep things safe"
This reverts commit
4044fe650469d6b5910b62692236487630d8db9a.
Amitesh Singh [Wed, 25 May 2016 07:10:10 +0000 (12:40 +0530)]
slider: no need to destroy & create popups again
already theme_apply() instantiates popups
Amitesh Singh [Wed, 25 May 2016 06:15:10 +0000 (11:45 +0530)]
slider: theme - fix warnings
Fixes warnings related to "disabled" & "disabled_visible" states
not present in respective parts.
@fix
Jean-Philippe Andre [Wed, 25 May 2016 05:48:35 +0000 (14:48 +0900)]
fileselector button: Move EO APIs to legacy
Jean-Philippe Andre [Wed, 25 May 2016 02:56:59 +0000 (11:56 +0900)]
fileselector entry: Move EO API to legacy