When the window is rotated, the logical pointer position is calculated
based on the window size (width or height) minus the current position,
depending on the rotation used. For wayland, we must add the window
decorations to the ecore_evas size, when doing this calculation.
Now, Edje.h includes three new files:
- Edje_Eo.h: Eo API functions (functions defines, enums, base id).
- Edje_Legacy.h: contains the API functions related to objects
- Edje_Common.h: common data (structs, enums...) + functions not related to
objects.
This phase is needed for the EFL 1.8 release to disable Eo APIs if we
consider it is not enough mature to be used by applications.
Finally have a way to compile the merged efl tree against the latest tizen
things. Only three typos after all this changes it quite good imho. Zero
would have been better though. :)
This clipper caused several bugs already, and there are some bugs still
not fixed. Let's remove it and try to fix any remaining with some other
kind of solution that does not depend on adding or clipping objects
during the evas render phase, which causes unexpected behavior.
Add 4 rectangles to be used as border of the window, instead of a single
rectangle under the framespace. This allows to move the frame object to
the top layer, instead of staying on the lowest layer. With the frame
over the other objects, there's no need of framespace clipper object,
which causes several bugs.
input_attach, input_detach, format_set, and source_set now return an
Eina_Bool
Tests, examples and edje_multisense adapted
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
Now, Ecore.h includes three new files:
- Ecore_Eo.h: Eo API functions (functions defines, enums, base id).
- Ecore_Legacy.h: contains the API functions related to objects
- Ecore_Common.h: common data (structs, enums...) + functions not
related to objects.
This phase is needed for the EFL 1.8 release to disable Eo APIs if we
consider it is not enough mature to be used by applications.
Summary: This feature replaces Eo pointers with ids to prevent bad usage
or reuse of these pointers. It doesn't change API.
The mechanism uses tables storing the real pointers to the objects.
See the src/lib/eo/eo_ptr_indirection.c file for more details on the
mechanism.
This follows the simplified handling of sensor object we are already
doing for the sync paths. Its a bit more complicated here as we need
to pass all data through the module specific async handling. But the
result should be the same.
Now instead of having ifdefs in the .c files we just don't compile the
ones we don't need. Much cleaner.
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
We rely on the app to provide a sensible object pointer and we now longer
need to have a copy of the object around to operate on it.
Simplifies code, maintenance and reduces mem copies. Win-Win :)
The frame should not stay in the same layer as the other objects. If the
application wants to allow it to be on top of the content, then it must
provide a custom frame object and set it to that layer.
NOTE: Should we make Elementary's be on a lower layer too?
These objects should be clipped only during rendering, since keeping
them clipped after that allows for unexpected behavior on the
application side. For instance, an application could check if objects
have clippers before doing something to them, assuming that some objects
should have no clipper, but under wayland, after the first render
iteration, there will be no objects without a clipper.
This commit fixes this behavior by unclipping objects that had no
clipper prior to the render iteration.
Additionally, it fixes a bug where a maximized/fullscreen window could
have not all of its content rendered immediately. This was occuring
because some objects could be clipped to the framespace clipper, but
considered invisible in the beginning of the render phase, where they
are evaluated. They were considered invisible because the framespace
clipper object was not resized at that phase yet, and thus these objects
were being clipped out from the viewport.
In case of short reads just make sure that playback is started
immediately. No need to drain just yet as that just causes issues with
delayed removals of streams.
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
A protected read function must now be implemented by the child class to
perform the actual reading.
Signals on playback loop and end are sent.
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
(udev registered it as fake)... and in the process i found that we
probably double free as sensor is calloced in the module and AGAIN in
eeze sensor core... oh and symbols probably might lak from modules..
so static them up yo.
Because of the way eo is dispatching method calls of objects the usual
error log you get if you mix up objects or try to call non-existent
methods is:
ERR<12404>:eo lib/eo/eo.c:362 _eo_dov_internal() Can't find func for op
0x24 (ecore_audio_obj_in:ECORE_AUDIO_OBJ_IN_SUB_ID_SPEED_GET) for class
'ecore_audio_obj_out_pulse'. Aborting.
Of course the problem is not really in lib/eo/eo.c, but in the function
calling eo_do()
Now the macros pass source file and line number on to the _internal
functions so we can log where the error originally happened:
ERR<1938>:eo lib/eo/eo.c:362 _eo_dov_internal() in
lib/ecore_audio/ecore_audio_obj_out_pulse.c:119: Can't find func for op
0x24 (ecore_audio_obj_in:ECORE_AUDIO_OBJ_IN_SUB_ID_SPEED_GET) for class
'ecore_audio_obj_out_pulse'. Aborting.
This makes debugging with eo a lot easier.
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
Eina value type does not support pointer so I used int as of now.
This needs to be converted into hexa later.
By using Clouseau, one can see this Clipper information.
- re-enabling split BiDi cursor tests, disabled in 0d68ffbe;
ligatures tests are still disabled;
- change "fail_if" to "ck_assert_int_*",
because it prints error message with values;
- fixing usage of embedding LTR/RTL codes in tests;
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
Switch from absolut microseconds since epoch to a monotonic clock with
realtive values. Switch from unsigned long long to double.
This aligns it with how we present time in efl. ecore_time_get is used
when possible. For the tizen modules we convert the the epoch timestamp
we get into a double. This is still a working monotone clock source.
As this will be released the first time with 1.8 we don't have any API
break here.
Together with eeze_sensor_module_register/unregister these are the functions
that are irrelevant for developer when they just want to use the eeze_sensor
API.
Still all three need to be marked with EAPI as they are used in the backend
modules and due to our visibility hidden usage the module loading would fail
if we remove EAPI from them.
Moving them into the private header should give an indication that these are
not meant for pure API users.
These functions let you pass an array of callbacks instead of just one.
It's more memory efficient to use this if you just add a bulk of events
on the same object.
This commits breaks ABI, and breaks API of the EO_EV_CALLBACK_ADD/DEL
signals (the event info passed).
Eeze sensor and Tizen types can't be in sync as the order have been changed
several times in Tizen already. We need to keep the order stable for ABI
compliance though after 1.8.
While we did the translation in one direction we did not in the other. Fix
this bug that shows now.
We use the typedef everywhere else. This makes sure we also use it
in the struct to avoid any confusion. The API is new for upcoming
1.8 release so no problem in changing it now.
It's implemented exactly the same way as on the X11 backend. It does
handle multiple devices, though may need some adjustments when
multi-touch is added, since there's no support for multi-touch on
EFL-wayland yet.
Some of tests fail on e5-jenkins-slave-x86_64-1.
Tests seem right and pass locally,
but not on the server.
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
If logical cursor is between LTR/RTL text two cursors will be shown.
Upper cursor is shown for the text of the same direction as
paragraph, lower cursor - for opposite.
NOT DONE YET
Signed-off-by: Tom 'TAsn' Hacohen <tom@stosb.com>
The wl_display_sync() request doesn't really wait for all the requests
to be processed, but instead sends a request for the "done" event.
Wayland relies on the fact that the requests are processed in order, so
when the "done" event is received, it means that all the other requests
requested prior to the respective "sync" have been processed already.
This commit makes the ecore_wl_sync() call actually wait for its "done"
event (thus blocking the ecore mainloop).
%f causes 0.0000 to always be printed... changed to %d
to get the desired/correct output result.
Signed-off-by: U. Artie Eoff <ullysses.a.eoff@intel.com>
If query at x coord, which points to rigth half of LTR char,
next position will be returned. The same for left half of RTL char.
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
We did use this 'begin' boolean to make eina_hash always allocate only
once per item it push in the hash. This boolean was alone at the end of
a structure. It would have costed us 4bytes on 32bits system and 8bytes
on 64bits. Removing it make elemines consume 100KB less on 32bits system.
We may have a speed impact on hash insertion here, but I don't think we
do use eina_hash_add and friends in any hot path, at the moment. If that
was the case there would be some way to mitigate this, just not worth it
at the moment.
Width and height should be swapped when the screen is rotated by 90 or
270 degrees.
Additionally, add a command to the ecore_evas_window_sizes_example which
returns the screen geometry.
Address sanitizer found this. Not really a serious error as text[i] will
be 0 in that case (I believe) and the loop is aborted in any case.
Still, better safe than sorry.
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
Since the objects are moved by the framespace offset, it must be
considered when populating map points. This is done when the map is
applied to an object (the map points are updated with the framespace
offset of the canvas that is parent of that object.
Additionally, a flag is set on the map struct to indicate that it had
its points updated already to avoid re-adding the offset.
Applications using these functions should not know of any offset. This
patch makes the canvas pointer position to be returned exactly the same
as on X11 backends.
This check makes no sense, since objects can be on outside of the
screen, with negative position, but they still need to be adjusted by
the framespace offset.
Framespace offset adjustment should be applied to every object not
marked with "is_frame".
Additionally, it should be applied only once. Since it is already being
applied on the *_feed_mouse_* functions, there's no need to apply it
again on the _evas_event_source_mouse_* functions, which are called by
the former ones.
Also add the missing adjustment to the feed_mouse_move one.
- mark all children of a given smart object as "is_frame" if the smart
object is also marked as a frame;
- when moving a smart object, use the originally requested move
coordinate to calculate the offset that the children should be moved
too;
- _smart_move_children_relative will fetch the child position with
geometry_get(), this way getting the corrected object position, before
adding the offset.
The real canvas size contains enough space for rendering the frame, but
ecore_evas_geometry_get() should be consistent accross all the engines
and report only the client area.
Implement ecore_audio_input_speed_set() in playback example
Fix volume limit off-by-one check while I'm at it
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
1. Added evas_obj_image_preload_begin/cancel APIs.
2. Removed evas_obj_image_preload. This accepts 'cancel' as a parameter and it's so confusing to developers.
3. No ChangeLog/NEWS for this change because Eo APIs were not released yet.
4. Discussed with Raster.
5. It's encouraged to use elm_image however. elm_image has elm_image_preload_disabled_set() API.
When user pressed home and ctrl+home, caret is normaly moved to first letter.
But, it is not common behavior for alt+home.
For example, gedit does not have any behavior and urlbar of firefox open
home page when alt+home is pressed.
No runtime changes. This makes it possible to set mouse_events on a
group, and it will use this value as the default for mouse_events on
every part of that group. It defaults to 1 if not set, to keep things
working as they do now.
That is, instead of setting mouse_events: 0; on every single part you
don't want to receive events, it's now possible to set it for the group
and only use mouse_events: 1; explicitly on those that should handle
events.
On some use cases, it is necessary to create a new wl_surface without
associating it to a wl_shell_surface. This new API makes it possible.
Signed-off-by: Eduardo Lima (Etrunko) <eduardo.lima@intel.com>
Apparently obj->layer and obj->layer->evas can sometimes be NULL. It is
checked in other objects, for example, image object. Add the checks here
to "fix" a crash reported by Christopher Michael.
Maybe we should switch to use -Wno-unused-parameter in our default cflags?
I mean unused parameters are really common and I do not see their harm.
Using -Wno-unused-parameter would let us remove all the cluttered EINA_UNUSED
and __UNUSED__ and also be safe when a paramter gets is use later as the
compiler could optimize it out when tagged with unused. Still we would avoid
the warnings to find the important ones.
The main eeze sensor log domain was EAPI as it was used in the modules as well.
(Gives some nice runtime error when loading the shared object). Switch to different
domains for the module and make the main domain private.
Spotted by zmike.
Jenkins as well as klocwork complained here as data is already used for
the parameter of the function and we shadow it here locally. Does not look
like it would really harm us here but better avoid it.
The way ecore_event_add() is used in the sensor modules takes care of
that already (not providing a free cb).
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
The event callback will take care of freeing the object.
This issue was found thanks to clangs -fsanitize=address option - coming
to a CI job near you.
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
This memset is not necessary as pack_it_copy can only be accessed when
the part type is a BOX or a TABLE and thus pack_it will be defined. Sadly
GCC 4.7 is more stupid than GCC 4.6 and think that it is an unitialized data
resulting in a massive number of useless warning that could hide real warning.
This is particularly useful when using table and replicating the
same group all over the place. At least for many games I have in mind
this will save a lot of lines !
When creating windows directly with Ecore_Evas, instead of using
Elementary, a default smart object is provided for the frame (client
side window decorations).
Now this smart object inherits from the clipped smart object, which
means that all its member objects will be clipped to its main clipper,
being skipped when the framespace clipper checks for objects without a
clipper.
Fix a ticket, which I don't remember the number because trac doesn't
work.
fail_if() from libcheck always writes to a file created with tmpfile(),
so it is not a good idea to call it millions of times. This wastes
memory and time, let's use both for something better. :-)
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
At least for wayland, eglGetProcAddress seems to fail to get the
eglQueryString symbol. Not sure if it's the right fix, but brings
wayland_egl back to working again.
Fixes#2301.
As cedric said a couple years ago in commit 17c820ae:
"This initialisation is not needed, but will reduce llvm warning noise."
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
* Use an Eina_Hash for the garbage collector list.
* Turn off garbage collection on object that are unlikely to match.
This patch make 1.8 as fast as 1.7 again.
I don't really like this patch. I think it would be nicer to have mmap
been correctly detected when Evil or Exotic is there, but at this point
I don't feel at ease with configure.ac.
In the error case we 'goto' the error path directly without passing
through the declaration and initialization of the variable.
This doesn't work so move the declaration/initialization to the start.
See this example (compile with -Wall for the warning):
-----
#include <stdio.h>
int main(void)
{
goto bar;
int i = 15;
bar:
printf("Foo: %i\n", i);
return 0;
}
----
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
* Use Eina_Hash instead of Eina_List to remember what memory did change.
* Turn off Eina_Cow use of Eina_Magic when doing a release as it is only
used internaly and we should not make any mistake there.
We now need to pass the current class to eo_do_super. This is faster and
more memory efficient and generally lets us do things better.
Using the eo_benchmarks we get ~20% speed-up.
This reverts commit 8ecd30d578.
First of all, if it's not tested it shouldn't be committed. Seriously,
test your stuff before committing, particularly when you *add* an API.
If it slips in and we forget it there, once it's released we can no
longer remove it.
Second, EDBus is a complete wrapper over libdbus and is in our intention
to remove libdbus when the EDBus implements the wire protocol itself.
Therefore there's no point in adding a public function to create an
EDBus_Connection from a DBusConnection, even if it's passed as a void
pointer.
NOTE: if you start your process with Systemd it will automatically use
Journald API. You will need to overide the default logging function to
change that behavior.
This reverts commit c002d113f1.
This commit reliably breaks builds with clang. Please test with
export CC=clang before you commit again.
export CC=clang
./autogen.sh --enable-multisense && make -j 10
The error Jenkins gets (I confirmed locally):
/bin/bash: line 1: 13549 Segmentation fault (core dumped) EFL_RUN_IN_TREE=1 ./bin/edje/edje_cc -id . -fd . -id ./tests/emotion/data tests/emotion/data/theme.edc tests/emotion/data/theme.edj
Daniel Willmann
Rafael Antognolli
Aharon Hillel
Carsten Haitzler
Tom Hacohen
Stefan Schmidt
Daniel Willmann
Rafael Antognolli
Lucas De Marchi
Mike Blumenkrantz
Cedric Bail
Daniel Zaoui
WooHyun Jung
ChunEon Park
Daniel Juyung Seo
Valerii Kanunik
Jihoon Kim
Shinwoo Kim
Jiyoun Park
Deon 'PrinceAMD' Thomas
Yakov Goldberg
Igor Murzov
Sung W. Park
Dmytro Kaidalov
Vitaliy Venzak
Chris Michael
Jérémy Zurcher
Cedric Bail
U. Artie Eoff
Raphael Kubo da Costa
Vladislav Brovko
Ryuan Choi
Iván Briano
Eduardo Lima (Etrunko)
Jerome Pinot
Jonas Gastal
Boris 'billiob' Faure