Carsten Haitzler (Rasterman) [Sun, 8 Sep 2013 01:03:35 +0000 (10:03 +0900)]
eina - remove deadlock from share-common + remove extra lock/unlock
Tom Hacohen [Fri, 6 Sep 2013 16:19:57 +0000 (17:19 +0100)]
Evas object clouseau: Added object name to the property list.
Tom Hacohen [Fri, 6 Sep 2013 16:19:07 +0000 (17:19 +0100)]
Evas object clouseau: cleaned up the code a bit.
Cedric Bail [Fri, 6 Sep 2013 07:47:36 +0000 (16:47 +0900)]
edje: add border linked to Image sets.
It is quite common that in an image sets each image has different border size.
This patch permit to define the border value on a per image basis in the set.
Michael Bouchaud [Thu, 5 Sep 2013 21:32:23 +0000 (23:32 +0200)]
edje: fix recursive scale_set on edje_object
Carsten Haitzler (Rasterman) [Thu, 5 Sep 2013 11:30:51 +0000 (20:30 +0900)]
evas - fix render op reset when drawing map objs :)
Cedric Bail [Thu, 5 Sep 2013 10:30:38 +0000 (19:30 +0900)]
edje: forgotten patch to fix ABI.
Cedric Bail [Thu, 5 Sep 2013 10:22:51 +0000 (19:22 +0900)]
edje: fix function to use Eo API.
I am really unhappy to have to do this patch myself. This commit is in since
May and I have continuously request to be fixed. It has taken way to long. It's
done now. Next time I see such a patch in EFL, I will revert it right away.
Cedric Bail [Thu, 5 Sep 2013 08:15:58 +0000 (17:15 +0900)]
evas: add evas_object_smart_iterator_new().
This is going to break Eo ABI. Rebuild all software using Eo API on top
of EFL (Elementary for example).
Carsten Haitzler (Rasterman) [Thu, 5 Sep 2013 03:46:06 +0000 (12:46 +0900)]
eina - stringshare - strlen already would deal with optimization cases
Chris Michael [Wed, 4 Sep 2013 13:36:29 +0000 (14:36 +0100)]
Merge branch 'framespace'
Chris Michael [Wed, 4 Sep 2013 13:25:50 +0000 (14:25 +0100)]
Create any wayland canvases as Alpha by default now.
Remove calls to ecore_wl_window_update_size inside the alpha &
transparent_do functions (this call not needed here anymore).
NB: This is needed due to recent frame "fixes". Basically, even tho we
set the opaque region correctly on a surface now, if the canvas itself
is not alpha, then we will end up with a black "spacer" around the
window.
NNB: Tested Both engines with terminology (both translucent and not) and also the window state test in elementary.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 4 Sep 2013 13:24:04 +0000 (14:24 +0100)]
Set surface opaque region to NULL if window is transparent/alpha.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Michaël Bouchaud (yoz) [Wed, 4 Sep 2013 10:06:03 +0000 (12:06 +0200)]
edje: be recursive on edje_object_scale_set
Carsten Haitzler (Rasterman) [Wed, 4 Sep 2013 02:40:07 +0000 (11:40 +0900)]
ecore-input - compose seq handler now uses compact str - saves about 100k
Carsten Haitzler (Rasterman) [Mon, 2 Sep 2013 12:07:37 +0000 (21:07 +0900)]
evas - in dump also ensure an idle flush logic and proxy surface dump happens
Rafael Antognolli [Sat, 31 Aug 2013 17:23:44 +0000 (14:23 -0300)]
evas/gl: Do not reassign the surface during make current.
There was already a surface created by _evgl_tls_resource_create(). If
we assign a new one here, the wrong one will be destroyed at
_evgl_tls_resource_destroy(), and later the GL window will be destroyed
before the surface, causing invalid access errors.
This fixes https://phab.enlightenment.org/T326
Tom Hacohen [Tue, 3 Sep 2013 10:49:02 +0000 (11:49 +0100)]
Evas textblock: fixed *_markup_get issue with escaped chars.
Markup_get was misbehaving and returning wrong results with some escaped
chars. markup_to_utf8 was working correctly. Merged the code together
and now both are consistent and correct.
Thanks to WooHyun for reporting.
Ulisses Furquim [Tue, 3 Sep 2013 04:48:08 +0000 (01:48 -0300)]
evas/render: improve evas_render_sync()
Tracking only the async rendering canvases and just waiting for the last one
to finish rendering. This should be enough to sync all canvases since the
render thread orderly executes the commands.
Cedric Bail [Tue, 3 Sep 2013 03:58:51 +0000 (12:58 +0900)]
eina: forgotten lock, thanks coverity.
Mariusz Grzegorczyk [Tue, 3 Sep 2013 03:44:27 +0000 (12:44 +0900)]
ecore: fix segfault while adding new local or global data to thread.
Cedric Bail [Tue, 3 Sep 2013 03:39:49 +0000 (12:39 +0900)]
Revert "evas/render - this is insane to me since nowhere it release the render_pre_done FALSE pairly."
This reverts commit
fd7d1f9f98fae4700d8196c3a3f4a9f9f8264d66.
This commit is the one breaking Evas_Map use in shelf. Let's investigate why it was disable
in the first place.
Cedric Bail [Tue, 3 Sep 2013 02:08:12 +0000 (11:08 +0900)]
evas: simplify logic and API signature, thanks k-s.
Ulisses Furquim [Mon, 2 Sep 2013 23:03:23 +0000 (20:03 -0300)]
evas/render: sync all canvases in _render_dump()
Ulisses Furquim [Mon, 2 Sep 2013 23:01:37 +0000 (20:01 -0300)]
evas/async_render: remove block/unblock of render thread.
This reverts commit
42a46214c4f9b35c0e1f5a84c56ea76ba2235eae.
Ulisses Furquim [Mon, 2 Sep 2013 22:38:52 +0000 (19:38 -0300)]
evas/render: introduce evas_render_sync().
evas_render_sync() will loop through all canvases and wait for their
rendering to finish. Since this function will execute from the main
thread that will sync all of them.
José Roberto de Souza [Mon, 2 Sep 2013 20:07:16 +0000 (17:07 -0300)]
eldbus: Check if interface of signal match with signal handler
Thanks fgau for pointing this issue.
José Roberto de Souza [Fri, 30 Aug 2013 13:52:04 +0000 (10:52 -0300)]
eldbus: Add timeout message error as a macro
José Roberto de Souza [Fri, 30 Aug 2013 13:23:28 +0000 (10:23 -0300)]
eldbus: Add eldbus_object_manager_interfaces_added/removed()
José Roberto de Souza [Fri, 30 Aug 2013 13:15:35 +0000 (10:15 -0300)]
eldbus: Make properties_monitor() return a boolean indicating if props are already cached
José Roberto de Souza [Fri, 30 Aug 2013 13:15:10 +0000 (10:15 -0300)]
eldbus: Add a new proxy event to notify when properties are cached
Cedric Bail [Mon, 2 Sep 2013 11:28:23 +0000 (20:28 +0900)]
evas: focus_set can fail... internally handle that case properly now.
Jean-Philippe Andre [Tue, 27 Aug 2013 03:43:22 +0000 (12:43 +0900)]
evas: Fix sigsegv after font rehint
If a text object is rehinted after it is deleted,
baaad things happen. Mark Evas_Object_Text as dead by
setting font to NULL.
Note the crash happened with cserve2 and maybe not in normal
cases.
Cedric Bail [Mon, 2 Sep 2013 03:32:54 +0000 (12:32 +0900)]
evas: handle error case by properly deallocating memory.
Cedric Bail [Mon, 2 Sep 2013 03:26:40 +0000 (12:26 +0900)]
eina: don't try close dummy handle.
Shinwoo Kim [Mon, 2 Sep 2013 02:51:08 +0000 (11:51 +0900)]
[ecore_input_evas] check the state of last mouse event more properly.
Cedric Bail [Mon, 2 Sep 2013 02:16:59 +0000 (11:16 +0900)]
eina: fix typo and use the right refcount.
davemds [Sun, 1 Sep 2013 16:43:03 +0000 (18:43 +0200)]
emotion_generic: tabs--
Carsten Haitzler (Rasterman) [Fri, 30 Aug 2013 00:09:08 +0000 (09:09 +0900)]
move render wait before the block to avoid deadlock.
Boris Faure [Thu, 29 Aug 2013 13:28:32 +0000 (15:28 +0200)]
export EFL_VERSION_MAJOR/MINOR
Boris Faure [Thu, 29 Aug 2013 13:11:12 +0000 (15:11 +0200)]
edje: export EINA_VERSION_MAJOR/MINOR when compiling edc files
Carsten Haitzler (Rasterman) [Thu, 29 Aug 2013 12:18:04 +0000 (21:18 +0900)]
other async render issue - sync ALL rendering canvases, not just one
Chris Michael [Thu, 29 Aug 2013 08:25:58 +0000 (09:25 +0100)]
Added functions for setting input and opaque regions on a window.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 08:25:34 +0000 (09:25 +0100)]
Add API functions for setting window input & opaque regions.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:54:37 +0000 (08:54 +0100)]
Remove 2 unused integer members from window structure.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:57:53 +0000 (08:57 +0100)]
Update ecore_evas wayland_egl for removal of server_allocation member.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:56:57 +0000 (08:56 +0100)]
Update ecore_evas wayland_shm routines for removal of
server_allocation member.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:56:04 +0000 (08:56 +0100)]
Update common routines for removal of server_allocation field.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:45:12 +0000 (08:45 +0100)]
Don't use Eina_Rectangles for server and saved allocation, can just
use ints for width & height.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:44:02 +0000 (08:44 +0100)]
Reduce size of Ecore_Wl_Window structure.
Removed 2 Eina_Rectangles from window structure as we only use width &
height anyway, so we can store those as ints.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:36:21 +0000 (08:36 +0100)]
Remove pending_allocation member of window structure (not used
anywhere).
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:46:23 +0000 (08:46 +0100)]
Remove calls to saved input & opaque region of window.
We don't record these regions anymore. They were removed from the
window structure.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:27:44 +0000 (08:27 +0100)]
Remove region members from Ecore_Wl_Window structure. No need to keep
these around.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 08:01:48 +0000 (09:01 +0100)]
Fix copy/paste issue.
Use proper width & height for the new tilebuffer.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:24:10 +0000 (08:24 +0100)]
Avoid potential memleak on rotation of tilebuf.
Port of
70f443d8a8e1dcf9e9f to wayland egl
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:21:36 +0000 (08:21 +0100)]
Avoid potential memleak on rotation of tilebuf.
Port of
70f443d8a8e1d to wayland_shm
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:11:16 +0000 (08:11 +0100)]
Port
1f82fbe14d2a to wayland shm engine
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 29 Aug 2013 07:05:47 +0000 (08:05 +0100)]
Fix formatting
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Carsten Haitzler (Rasterman) [Thu, 29 Aug 2013 05:58:09 +0000 (14:58 +0900)]
evas - avoid memleak on rotation of tilebuf in engine for gl and sw possibly.
Cedric Bail [Thu, 29 Aug 2013 05:25:23 +0000 (14:25 +0900)]
evas: let's allow intercept of other type of operation on the same object during an intercept.
Cedric Bail [Thu, 29 Aug 2013 02:35:35 +0000 (11:35 +0900)]
evas: simplify code for setting intercept callback.
Cedric Bail [Thu, 29 Aug 2013 02:21:20 +0000 (11:21 +0900)]
evas: add interceptor for focus_set.
Cedric Bail [Thu, 29 Aug 2013 01:53:48 +0000 (10:53 +0900)]
evas: cleanup intercept code and improve performance there a little.
Cedric Bail [Thu, 29 Aug 2013 00:54:17 +0000 (09:54 +0900)]
evas: define MAP_HUGETLB when it is not defined.
Rafael Antognolli [Thu, 29 Aug 2013 20:25:40 +0000 (17:25 -0300)]
evas/gl_x11: eng_window_use will not always force makecurrent.
If the egl_surface is different from the current one, it may be that the
it has been destroyed already. Removing the below check (and just
checking for different contexts) will avoid calling makecurrent when
destroying a window. That was always failing anyway.
Should fix https://phab.enlightenment.org/T311 for gl_x11 too.
Cedric Bail [Wed, 28 Aug 2013 08:16:48 +0000 (17:16 +0900)]
evas: use mmap/munmap on system that have it for image data.
Carsten Haitzler (Rasterman) [Wed, 28 Aug 2013 07:43:58 +0000 (16:43 +0900)]
evas - restore ability for evas_render_dump() to dump out referenced data
WooHyun Jung [Wed, 28 Aug 2013 04:40:27 +0000 (13:40 +0900)]
evas_event: Nomally, the point's state is maintained POINT_MOVE before it is released.
Jaehyun Cho [Tue, 27 Aug 2013 22:40:12 +0000 (07:40 +0900)]
[edje_embryo] Fix play_sample's parameter count check
play_sample function requires 2 parameters however it checks parameter count as 1.
Carsten Haitzler (Rasterman) [Tue, 27 Aug 2013 14:42:24 +0000 (23:42 +0900)]
ecore-audio - expose ready and fail events so multisense can avoid leaks
well well well. i WASN'T crazy. there WAS a leak. it was
ecore-audio+multisense. if a pulse audio output doesn't connect... NO
ONE KNOWS. all audio streams keep being appended to it forever just
consuming more memory indefinitely. there is no way to handle it.
expose events so it CAN be handled.
Tom Hacohen [Tue, 27 Aug 2013 13:11:16 +0000 (14:11 +0100)]
Evas bidi: Fixed a bug causing BiDi not to work in some cases.
This issue cause non-letter RTL characters not to be detected as RTL
which in turn turned off bidi for those strings.
Example broken text: <RLM><LRO>[PAS<RLO>[--DES<PDF><PDF>.
Thanks to Yakov Goldberg for reporting this issue.
Jean-Philippe Andre [Thu, 22 Aug 2013 12:13:38 +0000 (21:13 +0900)]
eina: Fix invalid check on Eina_File::global_map
global_map is set to MAP_FAILED in case of error after mmap.
So, it is initialized to MAP_FAILED and considered valid
otherwise.
So, we don't want to set the map to NULL or even check again NULL.
Carsten Haitzler (Rasterman) [Tue, 27 Aug 2013 08:21:10 +0000 (17:21 +0900)]
evas - unify partial update in sw engine swapper so it is like gl
Carsten Haitzler (Rasterman) [Tue, 27 Aug 2013 08:20:56 +0000 (17:20 +0900)]
fix apparent tb item leak (definitely lost)
Carsten Haitzler (Rasterman) [Tue, 27 Aug 2013 08:20:25 +0000 (17:20 +0900)]
fix ethumb client segv's i've been hunting...
Cedric Bail [Tue, 27 Aug 2013 07:52:24 +0000 (16:52 +0900)]
evas: try to also cancel multi button to.
Cedric Bail [Tue, 27 Aug 2013 07:52:00 +0000 (16:52 +0900)]
ecore: add infrastructure to get info from malloc_info to.
Cedric Bail [Tue, 27 Aug 2013 07:05:15 +0000 (16:05 +0900)]
ecore: fix build without mallinfo.
Cedric Bail [Tue, 27 Aug 2013 06:57:05 +0000 (15:57 +0900)]
ecore: fix function call order to actually make mallinfo call work.
José Roberto de Souza [Mon, 26 Aug 2013 19:11:31 +0000 (16:11 -0300)]
eldbus: Handle correclty the last unref of a object path or conn in service callback
This allow user remove the last reference of service object path or last
last reference of connection be removed inside of a method callback.
José Roberto de Souza [Mon, 26 Aug 2013 17:57:27 +0000 (14:57 -0300)]
eldbus: Send all object manager signals of all childrens before delete path or detach object manager
José Roberto de Souza [Fri, 23 Aug 2013 23:44:43 +0000 (20:44 -0300)]
eldbus: ecore_idler_del() return is the data
José Roberto de Souza [Fri, 23 Aug 2013 23:06:56 +0000 (20:06 -0300)]
eldbus: Only a explicit call should unregister a object path
José Roberto de Souza [Fri, 23 Aug 2013 22:47:40 +0000 (19:47 -0300)]
eldbus: Empty list of interfaces add/del if there is no parent object manager
José Roberto de Souza [Fri, 23 Aug 2013 22:45:12 +0000 (19:45 -0300)]
eldbus: Use EINA_LIST_FREE instead of EINA_LIST_FOREACH to send interfaces add
José Roberto de Souza [Fri, 23 Aug 2013 19:54:12 +0000 (16:54 -0300)]
eldbus: Send property changed signals before close connection
Carsten Haitzler (Rasterman) [Mon, 26 Aug 2013 03:08:15 +0000 (12:08 +0900)]
try and fix up valgrind+segv issues with ethumb and freed data access
discomfitor [Sun, 25 Aug 2013 16:15:19 +0000 (17:15 +0100)]
ecore-x xlib errors now only print when sync is enabled
I give up. people continually complain about these and submit them as crashes, so now they're only visible if you set ECORE_X_SYNC. the next step is an #ifdef ISCOMFITOR.
José Roberto de Souza [Fri, 23 Aug 2013 19:00:44 +0000 (16:00 -0300)]
eldbus: Correctly append name of interface removed
José Roberto de Souza [Mon, 19 Aug 2013 18:26:01 +0000 (15:26 -0300)]
ecore system: If by some reason name of locale changed dont have a '=' skip it
José Roberto de Souza [Mon, 19 Aug 2013 18:24:17 +0000 (15:24 -0300)]
eldbus: Mark introspection of parent object as dirty when adding a child object
Chris Michael [Fri, 23 Aug 2013 07:34:22 +0000 (08:34 +0100)]
Fix redraw rectangle when framespace changes to not be Outside viewport.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Rafael Antognolli [Fri, 23 Aug 2013 19:19:42 +0000 (16:19 -0300)]
evas/wayland_egl: eng_window_use will not always force makecurrent.
If the egl_surface is different from the current one, it may be that the it has
been destroyed already. Removing the below check (and just checking for
different contexts) will avoid calling makecurrent when destroying a window.
That was always failing anyway.
Should fix https://phab.enlightenment.org/T311 for wayland_egl.
Mike Blumenkrantz [Thu, 22 Aug 2013 12:59:05 +0000 (13:59 +0100)]
better fix for gettext macro versioning which doesn't break legacy support...
Dan Mashal [Thu, 22 Aug 2013 09:58:20 +0000 (02:58 -0700)]
Update for automake 1.13+
Chris Michael [Wed, 21 Aug 2013 07:31:29 +0000 (08:31 +0100)]
Reduce duplicated code for setting resize edges.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 21 Aug 2013 07:31:11 +0000 (08:31 +0100)]
Reduce duplicated code for setting engine resize edge
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 21 Aug 2013 07:30:52 +0000 (08:30 +0100)]
Add function prototypes for resize_edge_set.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 21 Aug 2013 07:28:48 +0000 (08:28 +0100)]
Add function for setting/resetting window resize edges to remove
duplicated code.
Resize the frame object before we update the window saved size.
Remove (again) call to _ecore_evas_wayland_resize and set the resize
edge of the window.
NB: The call to _ecore_evas_wayland_resize ends up sending duplicate
configure events here, hence whey it is removed.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 21 Aug 2013 07:27:45 +0000 (08:27 +0100)]
When updating window size, if the stored allocation size is the same
as being set, then get out.
This reduces unnecessary calls to resetting the input & opaque regions
if nothing has changed in terms of size.
Signed-off-by: Chris Michael <cp.michael@samsung.com>